http://www.lumagen.com/testindex.php?module=news

The Lumagen RadianceXG™ Processor Provides High-Performance Video Switching, Processing and Calibration for the Home Theater and Commercial Markets.

The RadianceXG provides a versatile backbone for high-definition (HD) theater systems. With 12 video, and 17 audio input sources, it offer extensive switching capabilities, including PiP/POP support. Two HDMI outputs, and two COAX audio outputs, are provided. Together these enable the RadianceXG to act as the central switching hub for large theater systems, at home or in a corporate environment.

Feature list:
Inputs (12 video, 17 audio including audio on the five HDMI inputs):
- 5 HDMI
- 2 Component/SCART
- 1 VGA (accepts RGBHV or Component)
- 2 SVideo
- 2 Composite
- 4 Stereo analog
- 6 COAX audio
- 2 Optical audio
- 1 IR command (TTL-level)

Outputs (2 video, 4 audio including audio on HDMI outputs):
- 2 HDMI
- 2 SPDIF

Functionality:
- Picture-in-Picture (PiP) and Picture-Outside-Picture (PoP)
- Two HDMI outputs allow two digital HDCP displays to have independent setups
- Second HDMI output can be copy of the first, or it can be audio only at a lower rate for receivers not capable of accepting 1080p inputs
- 10-bit, or higher, processing input to output
- Modular design. Video processing can be upgraded to 16-bit when 16-bit depth video processing chip becomes available, by replacing small daughter-card.
- Per-pixel detail enhancement
- MPEG artifact reduction, including mosquito and block artifacts
- Temporal noise reduction for SD and HD sources
- Per-pixel noise adaptive, directional interpolation video deinterlacing
- Film inverse-telecine with cadence processing including bad-edit detection
- Film genlock for correct 24, 48, 60, 72-Hertz output (24fps), and 50, 75 Hertz (25fps)
- Inherently low-latency video and film deinterlacing (1.25 frame if genlocked)
- Game mode for lowest latency (approximately 0.25 fames if genlocked)
- MPEG mosquito and block artifact noise reduction
- Lip-sync matching for audio and video
- HDMI and analog support up to 1080p input
- Adaptive diagonal filter for best quality video source processing
- Full-resolution Chroma processing (4:4:4)
- Direct selection of 4:3, 4:3 NLS, 16:9, 1.85 and 2.35 source aspect ratio
- User definable Preciseview™ Non-Linear-Stretch (NLS) of 4:3 source on 16:9 screen
- Software and video processing algorithms are user upgradeable using free updates
- 11-point parametric gamma grayscale calibration
- Color and Hue offsets for Red and Green push correction
- Primary color correction
- Extensive test patterns for display/projector setup
- Extensive input and output sizing, color-space, etc, controls provide an optimal setup for all combinations of input sources and displays.
- Rack mountable case (17 by 3.5 by 10.5)
- Optional external HDMI to analog converter
- Optional external SDI to HDMI converter
- Optional rack mount ears

Description:
The RadianceXG is perfectly suited for refining and enhancing the images presented by front projectors and top-level rear-screen displays. It combines the award winning Lumagen switching, calibration, and proprietary No-ring™ scaling, with the exceptional deinterlacing and noise reduction of the Gennum VXP™ video processing technology. Together the Lumagen and Gennum technologies provide 10-bit processing, per-pixel SD and HD video deinterlacing with enhanced adaptive diagonal filtering, edge enhancing scaling technology that does not introduce ringing – unlike other competing processors, plus MPEG mosquito and block artifact reduction. Finally, the most important image quality feature is temporal noise reduction for both SD and HD sources. Simply stated, the RadianceXG delivers extensive switching, advanced calibration features, and higher performance than other video processors.

The RadianceXG video inputs include 5 HDMI, with HDCP, 2 SD/HD Component, 1 VGA, 2 S-Video, and 2 composite. The HDMI, VGA, and component, inputs accept input resolutions of up to 1080p at 60 Hertz. For audio, in addition to audio using the 5 HDMI inputs, there are four analog stereo pairs, 6 coax and 2 optical audio inputs.

The RadianceXG’s two HDMI outputs support up to 1920 by 1080, 60-Hertz, progressive resolution. The second HDMI output can be a copy of the first output, or it can be used as an audio only output to send audio to receivers that do not support a full 1080p for their HDMI inputs. Additionally, for systems with two displays, an independent output configuration can be used for each.

The RadianceXG supports PiP, and POP, for any two inputs. Other key features include user-adjustable image enhancement, four configuration memories for each input, a programmable output resolution from 480p to 1080p, plus 1080i; the ability to zoom in on any source by up to 33 percent; selectable aspect ratios of 4:3, 4:3 NLS, 16:9, 1.85, and 2.35. The 4:3 NLS uses Lumagen’s Preciseview™ user programmable non-linear-stretch mode to allow the user to program exactly how 4:3 sources fill a 16:9 display. For film sources 2:2, 3:2 and 3:3 pull-down detection is supported (along with other less common cadences) to allow for correct cadence output at 24, 48, 60 and 72 Hertz for 24 fps film sources, plus 50 and 75 Hertz for 25 fps film source.

The RadianceXG is a modular platform. It allows for upgraded capabilities without the need to buy an entirely new video processor. For example, the current VXP processor from Gennum utilizes 10-bit pixel-depth processing, but when a 16-bit pixel-depth video processor becomes available, it can be installed by simply replacing a small daughter card. And the RadianceXG is designed to be able to take maximum advantage of the 16-bit processor’s pixel depth. The HDMI revision 1.1 inputs and outputs can also be upgraded to new HDMI revision levels as they become available, while keeping the same video processor in place.

The RadianceXG automatically detects input format for SD composite, and SVideo, as well as HD component and HDMI signals. There are four configuration memories per input with unique set-up parameters for each memory; an extensive suite of test patterns; 11-point parametric grayscale and gamma calibration; Color and Hue offsets; Y/C delay; CUE/ICP filtering, input sizing, and programmable input and output color-space. An RS232 interface is provided for software upgrades and control functions. An external universal power supply enhances the processor’s reliability, placement options, and it allows for quiet – no fan - operation.

The RadianceXG is the first product in Lumagen’s Radiance™ family of video processors. Lumagen plans other models in the Radiance line that use the VXP™ technology, and other video processors, such as the Silicon Optix Realta™.

Lumagen also produces the Vision series of video processors. These award-wining products were first to provide affordable per-pixel HD deinterlacing for video sources, and inverse-telecine for 1080i film-based sources, allowing a true 1080p viewing experience. Members of the Vision family include VisionHDP, and the Vision HDQ.

The RadianceXG video processor will be available in 2007 Q1 from authorized Lumagen dealers at a suggested price of $3499. Rack-ears for mounting the RadianceXG in a 19” rack are available at a suggested price of $50. Pricing for the external HDMI to analog converter and external SDI to HDMI converter is TBD.

What I like:

- Video processing CPU can be upgraded
- HDMI 1.1 can be upgraded to HDMI 1.3
- very fair price
- no fan

What I don't like:

./.

What I'd like to know:

- is MPEG/BAR noise reduction limited to SD?
- which algorithms are used for MPEG/BAR (Gennum, Algolith, something else)?
- how big is the RadianceXG case?
- will the Realta based Radiance be better or worse?
- why a Realta *and* a Gennum VPX based Radiance?

[- how big is the RadianceXG case? Rack mountable case (17 by 3.5 by 10.5)

Rack mountable case (17 by 3.5 by 10.5)
Ah, posted it myself, but missed this line - thanks... :)

Well, now I'm really confused. I was expecting the Radiance "JR" to come in at a much higher price, so I went ahead and ordered the VP50. Now I'm wondering whether to cancel and wait for the RadianceXG. Anyone else thinking the same thing?

Well, you'll have to wait at least 4 months for the RadianceXG. Even if Lumagen is able to deliver on the target release date (Q1 2007), it could be as much as 6 months. Are you willing to wait that long? Personally, I've ordered the VP50. Thanks to the great international upgrade program it was no difficult decision. If the Lumagen is as good as it sounds, I can still sell the VP50 next year and buy the Lumagen instead.

No SDI inputs -- neither SD-SDI or HD-SDI. :eek:

StooMonster

What a surprise! I was waiting for an announcement about their Realta-based Radiance. I didn't even know that they had a Gennum-based unit under development.

This sounds like a killer device. The calibration features, test patterns, world-class scaling, deinterlacing, and noise reduction and a healthy excess of inputs and outputs, all mated (I hope) to a user-friendly interface makes this thing sound irresistable.

I use SDI now so I would miss that. I don't think I would go for a $400 converter either. It should be no more than $200 ;)

- Rich

It was known for a long time that neither SDI in nor analog out would be available in the Radiance box. Lumagen plans to offer external converters for these purposes, which are said to be affordable - when bought together with the Radiance.

This looks like a winner. I was gonna order the HDp/HDQ ...now am in a quandry. This looks like exactly what I am looking for - 2 HDMI out - a big plus to drive both a plasma and PJ.

Actually thought the pricing was pretty good as well. Guess I will have to make my mind upon the PJ first.......too many decisions...but given the shipping time frame - apprently will have the time...if not the will power:)
Therese

wow, total surprise (I'm not kidding), this Gennum vs. HQV vs. DVDO seems a lot like the BD vs. HD DVD camps.

A switch like this from one manufacturer to another is definitely a surprise.

Feature list looks amazing!

A switch like this
But is it really a switch? I'm a bit confused about that. They're saying:

"Lumagen plans other models in the Radiance line that use the VXP™ technology, and other video processors, such as the Silicon Optix Realta™."

Modular upgrades = very impressive.

StooMonster

I really like the upgradeable chips ability!

What can this unit do that the Crystalio 2 can't do already?

Radiance XD (HQV chip)taken off the lumagen site, only shows Radiance XG

Bob I was wondering the same. If one was in the market for getting this unit would they be better off grabbing a CrystalioII now instead of having to wait? They seem very much alike. The Radience does have 5 HDMI inputs but that is not a great difference. I just wonder if Lumagen will implement a nicer user interface (like the CrystalioII) because I was not very impressed with the HDQs. Also entering many different codes does get to be cumbersome. If the radience had a better menu system then it would be worth looking at if not then I will be more than happy with the CrystalioII... Just a quick thought... :)

What can this unit do that the Crystalio 2 can't do already?Save you $1000.
Seriously, it does seem to offer color decoding adjustments in addition to full white balance control.

I am all for saving 1,000 but if this unit is seriously 6 months away is it worth it?

What I like:

- Video processing CPU can be upgraded
- HDMI 1.1 can be upgraded to HDMI 1.3
- very fair price
- no fan

What I don't like:

./.

What I'd like to know:

- is MPEG/BAR noise reduction limited to SD?
- which algorithms are used for MPEG/BAR (Gennum, Algolith, something else)?
- how big is the RadianceXG case?
- will the Realta based Radiance be better or worse?
- why a Realta *and* a Gennum VPX based Radiance?

Keep in mind that the cost of upgrading a DVD V50 to a V60 at some point may be less than replacing the main CPU on the Radiance! Being able to upgrade may not be the bargain it appears to be on first view.

Modular upgrades = very impressive.

StooMonster

My Onkyo TX-NR1000 made the same claim for a receiver. I fell
for all the propaganda about "future proofing" my home theater
as one would be able to get upgraded card modules and just
slide them in to replace existing one. Well after a couple of years
I finally got rid of the unit because Onkyo never released any
modules in the US-- just alot of promises as they focused
instead on releasing new receivers instead.

Hmmmmm..... Guess Lumagen felt they needed to use the Gennum to release something cheaper so they can compete with DVDO. It's nice but confusing, what happened to the ultimate Realta processor at 6K??!!

OK... what’s the trade in value for my old Lumegen HDP to get this puppy!


My Onkyo TX-NR1000 made the same claim for a receiver. I fell
for all the propaganda about "future proofing" my home theater
as one would be able to get upgraded card modules and just
slide them in to replace existing one. Well after a couple of years
I finally got rid of the unit because Onkyo never released any
modules in the US-- just alot of promises as they focused
instead on releasing new receivers instead.

Um, this is not Onkyo, it's Lumagen. They have kept their promises till this point, no matter how lofty they were.

I really like the upgradeable chips ability!

What can this unit do that the Crystalio 2 can't do already?
- it can be upgraded to the latest Gennum VXP chip
- it can be upgraded to HDMI 1.3
- it can do MPEG/BAR noise reduction
- it uses Lumagen scaling, which is said to be the best scaling algorithm on the market
- it is prepared for 16bit throughout the whole processing chain (HDMI 1.3 DeepColor)

I love the upgradeability!! Why? Because current Gennum VXP chips don't support Anime cadence detection yet. And because I believe Gennum is working on many new algorithms which may be worthwhile (e.g. reducing judder on 24fps camera pans). The RadianceXG will hopefully be able to make use of new features of newer Gennum chips, while the Crystalio II can only make use of new Gennum chips as long as the pin layout is compatible, which is unlikely to be the case with future Gennum chips, so I've been told. So PixelMagic will have to redesign their mainboard if they want to make use of newer Gennum chips.

Keep in mind that the cost of upgrading a DVD V50 to a V60 at some point may be less than replacing the main CPU on the Radiance! Being able to upgrade may not be the bargain it appears to be on first view.
Well, we don't know that yet, of course. I've no idea what upgrading is going to cost on the Lumagen. But my impression is that Gennum is very agressive in its chip development and I expect many great things to come out in their new chips sooner or later. I'm not sure whether DVDO/ABT can keep the pace. Gennum is much bigger in terms of engineering capacity, I believe. But we'll have to wait and see.

I had to read the opening paragraph twice to make sure it wasn't a joke!

Personally I think the Gennum chip has much better price/performance than the Realta, and I do love all the features and support Lumagen offer, but I would have liked to have seen the SRP in the range of $500-1000 less, especially considering the SDI inputs and DACs are optionally, and probably pricy extras. Once those have been factored in you're basically looking at or near the price of the Crystalio 2, something you don't need to wait until Q1 2007 for.

Does the Radiance do color gamut mapping? (You measure the primaries of the display and the Radiance makes sure you see correct NTSC, PAL or HD color if one wishes to and it's possible).

Could this be the chip Lumagen will be using?

http://www.avsforum.com/avs-vb/showthread.php?t=724013

I wonder if this is another indication of how difficult it is for VP designers to get comprehensive Realta solutions to market. This may be a stopgap solution if it is indeed taking longer than expected for Lumagen to get the Realta-based products out. But I suppose the Reon could have been used instead to get a lower-end product to market sooner. I must admit that this announcement is a bit surprising & confusing.

Could this be the chip Lumagen will be using?

http://www.avsforum.com/avs-vb/showthread.php?t=724013
I'd assume so. How it'll compare to the 9351 is still to be acertained.

Could this be the chip Lumagen will be using?

http://www.avsforum.com/avs-vb/showthread.php?t=724013

I saw this chip today at CEDIA. They had the chip on a board under plexiglass. It was a working model board with DVI in/out connectors. No HDMI on the board itself and I did not see what the connection on the display was, so it could have been HDMI on that end. They were doing a comparison between with and without the processing and the differences were very noticeable, of course.
Interesting I had just visited Lumagen yesterday and had asked about the Radiance and got no indication that this announcement was close to being made. The folks at the booth only said and announcement would be made sometime in the future, but they did not say when.

Thanks to Gordon (Elliots PC) for the first photos:

http://img.photobucket.com/albums/v221/x3els/DSC00112.jpg

Taken from this thread:

http://www.avforums.com/forums/showpost.php?p=3480106&postcount=31

The one with the LCD display is the top model, which will cost more. The one without the LCD display is the RadianceXG this thread is about.

Prettier than a plain black box, at least. What are they made out of? ABS plastic?

I must say, as nice as this unit sounds, since I do not want to use my scaler to switch audio - just video - I am loath to spend extra $$ for something I do not need. The Lumagen HDQ has proven perfect for me - will see, I suppose, whether the better video deinterlacing on the Radience vs HDQ (I imagine the film deinterlacing is identical) make the extra cost worthwhile.

Also, since I have no more room in my rack, appearance for me is irrelevent - my HDQ ended up 'hidden' above my drop ceiling.

Simon

fantastic design!!!

Wow, what a surprise! This doesn't bode well for SO. All the hype and then boom Gennum and DVDO come along and eat their lunch lol.

It'll be interesting to see how practical chip replacement really is. It sounds good in practice, but the reality can be something else. I assume the chip replacement will really be a board replacement and then the cost will likely be higher than the DVDO upgrade program...

I must say, as nice as this unit sounds, since I do not want to use my scaler to switch audio - just video - I am loath to spend extra $$ for something I do not need. The Lumagen HDQ has proven perfect for me - will see, I suppose, whether the better video deinterlacing on the Radience vs HDQ (I imagine the film deinterlacing is identical) make the extra cost worthwhile.

Also, since I have no more room in my rack, appearance for me is irrelevent - my HDQ ended up 'hidden' above my drop ceiling.

Simon

Hi Simon,

Unfortunately (or fortunately depending on how you look at it), audio switching is becoming a necessity in VP's because of the long audio delay the VP adds and also the fact that HDMI is a combo audio and video interface. So I look at it as a positive.

I must say, as nice as this unit sounds, since I do not want to use my scaler to switch audio - just video - I am loath to spend extra $$ for something I do not need. The Lumagen HDQ has proven perfect for me - will see, I suppose, whether the better video deinterlacing on the Radience vs HDQ (I imagine the film deinterlacing is identical) make the extra cost worthwhile.

Also, since I have no more room in my rack, appearance for me is irrelevent - my HDQ ended up 'hidden' above my drop ceiling.

Simon


Mark Petersen beat to the punch above regarding switching audio with VP's. All I'll add is if you have any desire to use one of the new generation of video processors you'll have no choice but to use the audio switching, that is unless you like your audio and video out of synch, way out of synch. Audio equipment doesn't have a large enough delay to accomodate the delay these processors introduce.

All I'll add is if you have any desire to use one of the new generation of video processors you'll have no choice but to use the audio switching, that is unless you like your audio and video out of synch, way out of synch. Audio equipment doesn't have a large enough delay to accomodate the delay these processors introduce.I have the Vantage HD Realta-based processor. I notice no problem with lip synch delay, except on occassion with HD DVD. Pausing the movie and then resuming resolves the issue. With regular DVD or cable, no problems at all.

I assume the chip replacement will really be a board replacement
A small daughter card needs to be replaced. Cost? No idea. But it's definitely not a full mainboard swap. Just a little daughter card.

I have the Vantage HD Realta-based processor. I notice no problem with lip synch delay, except on occassion with HD DVD. Pausing the movie and then resuming resolves the issue. With regular DVD or cable, no problems at all.

Are you running the audio through the Vantage? If so, you'll not notice much of a lip sync problem because the Realta dynamically varies the time delay based on the the video delay. From what I understand the Realta video delay can be as much as 4 fields but can be less than that.

"If so, you'll not notice much of a lip sync problem because the Realta dynamically varies the time delay based on the the video delay."

It would be nice if the video processors would keep the video delay constant. Basically just buffer the video if the current processing is quicker then the max. latency through the unit running different video processing. That would make it easier for those delaying the audio in other procesors to keep things always in sync since the audio delay would not need to be adjusted based on what the video processor is doing.

Shawn

It would be nice if the video processors would keep the video delay constant. Basically just buffer the video if the current processing is quicker then the max. latency through the unit running different video processing. That would make it easier for those delaying the audio in other procesors to keep things always in sync since the audio delay would not need to be adjusted based on what the video processor is doing.


That's true although for some applications such as games it's advantageous to minimize the latency altogether so some VP's like the new DVDO VP50 allow for various latencies.

Are you running the audio through the Vantage?No.

No.

Interesting. The delay in the Realta is about 100ms so it should be noticeable and it is with my Algolith Dragonfly. Perhaps you're compensating with delay in the audio prepro or your deinterlacing and NR settings are such that the delay isn't maxed out...

Mark Petersen beat to the punch above regarding switching audio with VP's. All I'll add is if you have any desire to use one of the new generation of video processors you'll have no choice but to use the audio switching, that is unless you like your audio and video out of synch, way out of synch. Audio equipment doesn't have a large enough delay to accomodate the delay these processors introduce.

Mark:

Yes, that is a good point. My Meridian gear can take care of the lip-synch adjustment, however, Meridian do not yet have an HDMI input on the 861 to accept the HDMI-transmitted audio. Since they are about the cleverest bunch out there, I suspect they will come up with something that allows me to keep audio and video separate - I hope!

Simon

this Gennum vs. HQV vs. DVDO seems a lot like the BD vs. HD DVD camps.

Nope :) HQV vs. ABT vs. Gennum is healthy and good for consumers. These can evolve. HD DVD vs. BD is mad and bad for consumers. They are fixed specs with split studio support.

ca1ore - From my reading those with Meridian, Lexicon and lots of other high end gear haven't been too successful with the delay issue. There is simply too much delay between audio and video after the Realta and Gennum based processors are done with the video. This is the very reason why we are seeing all of these vendors do audio switching. Another issue as I recall is having different audio inputs to your pre-amp and needing different delay settings. Different sources (cable, sat dvd, hd-dvd, etc.) introduce different delays. Some equipment allows for different delays per input and others don't.

The Lumagen outputs the audio via the HDMI or via optical or other connector. I believe the other competing processors do as well.

Gennum's VXP processing is surely moving forward at a more rapid pace than Realta's HQV processing so far.

The top end Marantz ($20,000 11S1), the top end projectiondesign ($28,000 Action! Model Three 1080), the top end Barco ($95,000 FLM-HD18)[not sure on this one though] are all bundled with Gennum's processing algorithms.

They have PixelMagic and now Lumagen has joined them as well. I think Gennum has gotten on their work more quietly with substance while HQV has been more tilted towards the hype so far.


On a personal note, I think that the Crystalio II is still the best looking VP out there. The Lumagen is a plain box and is not impressive looking. As far as what is in the box and how it performs, well time will tell. This Radiance series VP is already coming out a quarter late even if it is the Gennum version and we will see if the upgrades are offered briskly.

A Crystalio III in the future could rule all.

sspears,

Good point, I was referring to the emotions that seem to be involved...

Umberto Eco, the Italian semiologist, once famously compared Macs and PCs to the two main branches of the Christian faith: Catholics and Protestants.

The Mac is Catholic, he wrote in his back-page column of the Italian news weekly, Espresso, in September 1994. It is "cheerful, friendly, conciliatory, it tells the faithful how they must proceed step by step to reach -- if not the Kingdom of Heaven -- the moment in which their document is printed."

The Windows PC, on the other hand, is Protestant. It demands "difficult personal decisions, imposes a subtle hermeneutics upon the user, and takes for granted the idea that not all can reach salvation. To make the system work you need to interpret the program yourself: A long way from the baroque community of revelers, the user is closed within the loneliness of his own inner torment."

He was joking of course, but the point is that people do buy into technologies in an almost religious way -- and the trinity of VRS vs VXP vs HQV will keep us discussing the "number of angels on a pinhead" for a long time coming.

StooMonster

ca1ore - From my reading those with Meridian, Lexicon and lots of other high end gear haven't been too successful with the delay issue. There is simply too much delay between audio and video after the Realta and Gennum based processors are done with the video. This is the very reason why we are seeing all of these vendors do audio switching. Another issue as I recall is having different audio inputs to your pre-amp and needing different delay settings. Different sources (cable, sat dvd, hd-dvd, etc.) introduce different delays. Some equipment allows for different delays per input and others don't.

The Lumagen outputs the audio via the HDMI or via optical or other connector. I believe the other competing processors do as well.

My Meridian does a pretty good job of lip-synch delay, however, I do not have a Realta/Gannum based scaler. Time will tell on this I suppose.

Simon

I'm not interested in running audio through a VP, all I need is for the VP to send the audio delay out the RS232 port every times it changes so my Prepro can be adjusted accordingly. Can the new Lumagen do this?

Stoo,

coffee vs. tea perhaps?

So the Radience will be available in the first quarter of 2007? Does anyone have any guesses as to which month it will actually come out?

I think it will be in beta perhaps in December and it will stay in beta for several months. The pictures with LCD displays will be later models, not the XG. While Lumagen and Gennum won't say, the XG will clearly be using the coming new Gennum chip, not the present one. I really like the concept of daughter boarding the chip. This will make it possible to use later Gennum chips as well as chips by other manufacturers such as ABT and Silicon Optics. The Lumagen press release at Cedia specifically mentions Silicon Optics. Why did Lumagen switch? The Gennum chip is cheaper, has a much easier to use interface, and offers the big three in NR, Mosquito, Block, and 3D al for SD and HD.

I really like the concept of daughter boarding the chip. This will make it possible to use later Gennum chips as well as chips by other manufacturers such as ABT and Silicon Optics.
Oh, I thought the daughter board would only be capable of holding Gennum chips. I thought there would be other Radiance models for the Realta (if Lumagen will ever be using Realta after all). But what you're saying does make a lot of sense - and it makes the daughter board idea even *more* attractive than it was before!

The Gennum chip is cheaper, has a much easier to use interface, and offers the big three in NR, Mosquito, Block, and 3D al for SD and HD.
Not to forget: It's cooler and thus doesn't need a fan!!

One daughter board could be for the current pin config of the Gennums another could be for the Realta, and another for later Gennums. One could have as many different daughter boards as needed. Clearly,Lumagen is designing for the long term and the possibility of using different chips in the future without the customer needing to replace the main board.

Remember too the Gennum chip does not have the same ability to be upgraded as a gated chip. That is why the Radiance has so many other gated chips in it. Remember too that Lumagen has the tendency to use most of its own algs. Clearly Lumagen's scaling is the best re the competition. So I suspect Lumagen will only use the SD and HD deinterlacing algs in someone else's chip plus a few other features. It might use someone's elses NR but then again it could develop and use it's own. From the get go it probably will use whatever it doesn't have such as NR with the future perhaps bringing better with only a down load to those gated chips. If some future chip has better deinterlacing algs and/or better NR etc,, change out the daughter board to one holding the new chip and viola!

God. I sound like a Lumagen shill. Fear not, I am getting a VP50 and will get the VP60 when it comes out too. So much good stuff out there now and coming. I hope the DVDO folks come out with some NR that can be added for free or cheaply. One VP is just not enough. We live with multiple displays an each deserves a great VP. Bring em on.

The Gennum chip is cheaper, has a much easier to use interface, and offers the big three in NR, Mosquito, Block, and 3D al for SD and HD.

Just to be sure, u mean NR/Mosq/BR/etc for BOTH SD and HD material ?

Still the Radience XG sounds like one heck of a VP! ;)

The new Gennum chip offers Mosquito, Block, and 3D noise reduction for both SD and HD. This is from the Gennum literature handed out at the Cedia Expo.

Hi Mark,

Is that true that the present Gennum cannot be ugraded via firmware? (What does not being a gated chip actually mean? )

God bless..

Mark

The new Gennum chip offers Mosquito, Block, and 3D noise reduction for both SD and HD. This is from the Gennum literature handed out at the Cedia Expo.

thanks! now this makes it stand out since the realta based VPs so far have limited NR capabilities on HD.

Is that true that the present Gennum cannot be ugraded via firmware? (What does not being a gated chip actually mean? )
As far as I know some changes are possible via firmware upgrades. But you can't fully reprogram it to do something completely different. In contrast the Realta can be fully reprogrammed, I believe - which sounds great, but comes at a price: It gets much hotter when producing the same image quality (specialized chips are almost always more efficient than general processing chips) and the reprogramming (= development of new firmware) seems to be extremely complicated. Furthermore the Gennum chip can run all algorithms at the same time with full processing speed, while the Realta shares resources and each algorithm has less processing power available when a lot of algorithms are used at the same time.

Hi Mark,

Is that true that the present Gennum cannot be ugraded via firmware? (What does not being a gated chip actually mean? )

God bless..

Mark

Gated chip are chip with the actual logic define by a downloadable software. The FPGA being the more popular one. With such a chip you could develop specialized high speed logic that is totally reprogrammable. DVDO is using FPGA in their new VP50.

I don’t know what Gennum is using but some manufacturer are using ASIC what custom logic chip that are hardwired and therefore non reprogrammable.

Gated chip are chip with the actual logic define by a downloadable software. The FPGA being the more popular one. With such a chip you could develop specialized high speed logic that is totally reprogrammable. DVDO is using FPGA in their new VP50.

I don’t know what Gennum is using but some manufacturer are using ASIC what custom logic chip that are hardwired and therefore non reprogrammable.

So is lumagen the question is how many of them (3 i am guessing) and what do they do, im sure it does the scaling and calibration.

Daniel.

JP told me but I do not remember. I am sure there is enough large aray field programable gated chip capacity built in to do most anything. If Lumagen develops algs that do something better than the Gennum algs do, I am sure they will be put into the programable gated chips and the Gennum algs abandoned. Initially, I suspect the Gennum deinterlacing, noise reduction, and detail enhancement algs will be used. Most everything else will be Lumagen proprietary in those gated chips. Gennum's Viper Applcation Builder software will be used to interface and customize the Gennum algs.

JP told me but I do not remember. I am sure there is enough large aray field programable gated chip capacity built in to do most anything. If Lumagen develops algs that do something better than the Gennum algs do, I am sure they will be put into the programable gated chips and the Gennum algs abandoned. Initially, I suspect the Gennum deinterlacing, noise reduction, and detail enhancement algs will be used. Most everything else will be Lumagen proprietary in those gated chips. Gennum's Viper Applcation Builder software will be used to interface and customize the Gennum algs.

This would be a lot of flexibility but it add to the cost.

We need to open one. :D

JP told me but I do not remember. I am sure there is enough large aray field programable gated chip capacity built in to do most anything. If Lumagen develops algs that do something better than the Gennum algs do, I am sure they will be put into the programable gated chips and the Gennum algs abandoned. Initially, I suspect the Gennum deinterlacing, noise reduction, and detail enhancement algs will be used. Most everything else will be Lumagen proprietary in those gated chips. Gennum's Viper Applcation Builder software will be used to interface and customize the Gennum algs.
According to my information, Gennum's detail enhancement can only be used if Gennum's scaling is used, too, because Gennum's detail enhancement is done after scaling. So it's either both or none. I've been told that Lumagen hasn't fully decided yet which algs they'll be using where. I've suggested to make these things user adjustable. Let's just wait and see. I'm sure Lumagen will come up with a good solution for us.

Hey madshi, I agree that having the Gennum chip on a daughterboard is great as you can always upgrade to the latest and greatest Gennum chip. But taking this further, we get the following possibilities:
1) Your current daughterboard contains a Gennum chip. If Lumagen produce a hardware upgrade with boards containing either a Realta or Gennum chip, you can choose what you want.
2) If you don't feel like upgrading to the next Gennum, you can choose to skip an upgrade and go for the next one. The one downside I see with this is if you cannot skip upgrades, but let's see.

Cheers.

HI madshi,

I think that would be great if Lumagen allows us to choose to use their scaling or the scaling and detail enhancement on the Gennum.

God bless...

mark

Someone here skipping a chip upgrade? Like a drug addict with a little money skipping a hit?

Lumagen not using it's own scaling? Not a chance. Lumagen's press release says it will be using its own ringing free scaling. So I guess Lumagen will be using its own detail enhacement should that feature be included.

I agree with Mark's statement, Lumagen's scaling is second to none - it would be a real deal breaker not to use it for something like detail enhancement. It would be making a compromise in order to implement another compromise...

I very much doubt they would do that.

Maybe you will be able to choose what scaling to use, they seem big on providing a lot of options in their current firmware. Pretty much what htguy1 suggested, if you choose detail enhancement on, you get gennum scaling, otherwise you get Lumagen scaling.

HI

I can't think of a better way to prove just how great their scaling is than to go head to head on the same machine and let the customer person see it for themselves. I also am pretty certain that their scaling is unsurpassed, and this would be the proof. Plus people tend to love to have options to play with.

God bless...

Mark

I can't think of a better way to prove just how great their scaling is than to go head to head on the same machine and let the customer person see it for themselves. I also am pretty certain that their scaling is unsurpassed, and this would be the proof.
How EXACTLY are you judging scaling quality? What patterns or tests are used? Is there an objective measure of scaling quality or is completely subjective?

So what do you do, buy a CII now or a VP50 or wait for the XP in Jan/Feb to come out with the potential of having inter changeable chips and upgradeability???????????

How EXACTLY are you judging scaling quality? What patterns or tests are used? Is there an objective measure of scaling quality or is completely subjective?
So no one knows? Then how are statements made that scaler A is hands down better than scaler B? Or that the scaling in prduct XYZ is far superior than anything the competition has to offer?

How EXACTLY are you judging scaling quality? What patterns or tests are used? Is there an objective measure of scaling quality or is completely subjective?
Ok, let me try:

A good scaling algorithm should give you a picture which looks good. Doh!

No, seriously, the secret is just in comparing and checking which image looks better. Things to look for are:

- how sharp is the image?
- how much mosquito noise is visible?
- how much EE is visible?
- are any jaggies or other artifacts visible?
- overall subjective PQ impression

From what I've heard the Lumagen is supposed to offer a sharp picture without any additional EE (over what the source has). The Lumagen scaling is even supposed to reduce source EE and mosquito noise a tiny bit, while other scaling algorithms sometimes may increase EE and/or mosquito noise.

The differences may be small, though, so you may need a very big screen to see any noticable differences. Furthermore upscaling may behave differently compared to downscaling. One scaling algorithm may be better for upscaling while another one is better for downscaling.

Bob,

They must be using the Sencor scaleometer and getting perfect readings :).

Good question and one I've often wondered how quality of scaling is asessed. I can only guess that it is a matter of features and results that make everyone rate Lumagen so well. Better control?

I have seen devices that do poor scaling but I still don't know how one measures the quality of say Lumagen scaling against DVDO?

I would suggest waiting on the C2 since it is still going thru growing pains. A big one is the 1080p 48hz struggle. They are still working on a firmware fix for that. I am intrigued by the VP50 and plan to play with one of them. Then eventually I will get the Radience XG and compare. That sounds like the best way to go for now... :)

Thanks, madshi, for taking the time to explain things to me. The reason I ask is that I have been able to detect very little (if any) differences from one scaler to the next, including my HTPC, which theoretically has some of the best algorithms available, especially when ffdShow is used. To take your points one at a time:

- how sharp is the image? - At 120" screen size, I have found virtually NO difference unless the scaler is of particularly poor quality, like the one in the Optoma H-79. Every other scaler I have seen, including the ones in upscaling DVD players, is identical in this regard.
- how much mosquito noise is visible? I didn't know this was directly related to scaling and instead thought this was caused by MPEG encoding. I also thought it was directly controlled by noise reduction and was related to the amount of tradeoff one would be willing to accept - that is, the more noise reduction, the less mosquito noise but also the softer the image.
- how much EE is visible? Here is an area in which I have been seeking the Holy Grail, but so far I have not found it. In standard DVDs, I can't get rid of it no matter what I have tried, but I have admittedly not had the pleasure of seeing a Lumagen unit at work. If Lumagen has actually solved this problem and gotten rid of EE, then I will gladly purchase one of their products, but I have my doubts at this point, as NO SCALER I have ever seen, including the expensive Teranex VPs I've seen at shows, has ever accomplished this feat.
- are any jaggies or other artifacts visible? I thought those were deinterlacing artifacts, not scaling artifacts. The only such scaling artifact that I was aware of was "line twitter", something that again very seldom is a problem even with low cost solutions.
- overall subjective PQ impression - Well, yeah, I am always concerned about overall PQ, but I am trying to nail down what parts of the PQ are directly related to scaling and what parts are related to deinterlacing, optics quality (sharpness), and display technology (mosquito noise vs. low level dithering).

I'm not trying to give you a hard time , madshi, but rather to find out what specific attributes are directly related to scaling. My theory here is that a lot of people are mistaking some of the other problems and artifacts inherent in video processors and displays and simply blaming these problems on "scaling", when "scaling" as a single concept is actually very much the same from one unit to the other.

Now maybe the problem is that I just haven't seen the Lumagen scaling, and maybe they know something that the HTPC guys and Teranex don't know, so I will reserve final judgement until I get a chance to evaluate the Lumagen scaling for myself.

So one question still remains - Is there some specific test pattern or video clip that will clearly demonstrate the superiority of one scaler over the other? You know, some sort of grueling scaler torture test that one scaler will fail while another passes, much like the HQV disc does for other aspects of video reproduction? Right now it seems like people just look at image quality in general and then proclaim one scaler to be better than another, unless of course they are using the term "scaler" when they actually mean "video processor", as that term would incorporate many more parameters than simple scaling alone.

Scaling is not deinterlacing. It is not edge enhancement. It is line multiplication hopefully without ringing. AVIA PRO has several useful test paterns for measuring degree of ringing etc. I'll post the pattern numbers after I look them up again. Spears on Secrets makes reference to a test pattern he uses.

Scaling is not deinterlacing. It is not edge enhancement. It is line multiplication hopefully without ringing. AVIA PRO has several useful test paterns for measuring degree of ringing etc. I'll post the pattern numbers after I look them up again. Spears on Secrets makes reference to a test pattern he uses.
Thanks, Mark. Now that's what I'm talking about!

I thought a very simple test is the multiburst pattern ? They are supposed to be black/white/black/white strips. See on your projected (upscaled) screen whether it's still clearly black/white/black/white, and not white/darker white/lighter black/black/etc, and they are still one pixel width. I know the VP30 does not give me perfect strips, whereas the lumagen does.

Yes. That's a good test. Stacy, I believe, uses a multiburst pattern for this.

That's why I keep suggesting that DVDO needs to improve its already good scaling and hoping it does. But they have such a cool name for their present scaling and have sold it, on a ASIC chip to, for example, to D&M holdings for use in some of Denon's better DVD players, I think they may be happy to rest where they are as to scaling. That would be a pity.

You can yuk up a processor up all you want, but deinterlacing and scaling is the heart and everything else is fluf. ABT has the deinterlacing, so does Gennum and Silicon Opyics. Lumagen is the scaling champ, but it doesn't yet have high quality motion adaptive deinterlacing. It will soon.

It is all coming together for us and hopefully soon we will have video processors with ring free scaling, extremely good film and video deinterlacin, all sorts of NR that makes PQ overall better, and providing enough tweaks to keep an ISFer chained to a display under calibration.

I think I've finally found the 'uber' VP I'm willing to throw some hard earned dollars at with the RadianceXG. This is great news indeed.

In my experience modular designs haven't always worked out ideally and end up being more expensive in the long run. Unless you can guarantee the physical interface between the motherboard and daughterboard will never change for future (ie unknown) chips/specifications...let's hope that's the case here. It sounds like they've designed it for at least one future spec in mind though (16-bit processing) which is good.

Now what is the projected MSRP?

UA:....from last paragraph in first post...... $3499

Gordon

UA:....from last paragraph in first post...... $3499

Gordon
Thanks :o My brain automagically stops reading after I see a lot of marketing speak...third paragraph from the bottom....;)

You have to keep your brain in gear when reading marketing speak, often good laughs there. Now start reading some techno babble and one would think most here's manhood is challenged. :)

- how much mosquito noise is visible? I didn't know this was directly related to scaling and instead thought this was caused by MPEG encoding. I also thought it was directly controlled by noise reduction and was related to the amount of tradeoff one would be willing to accept - that is, the more noise reduction, the less mosquito noise but also the softer the image.
Yes, mosquito noise is caused by MPEG encoding. However, some people claim that Lumagen's scaling slightly reduces mosquito noise - without a real mosquito noise reduction algorithm running. I'm just repeating here what I heard elsewhere, though. So please don't take my word for it.

- how much EE is visible? Here is an area in which I have been seeking the Holy Grail, but so far I have not found it. In standard DVDs, I can't get rid of it no matter what I have tried, but I have admittedly not had the pleasure of seeing a Lumagen unit at work. If Lumagen has actually solved this problem and gotten rid of EE, then I will gladly purchase one of their products, but I have my doubts at this point, as NO SCALER I have ever seen, including the expensive Teranex VPs I've seen at shows, has ever accomplished this feat.
I didn't say that Lumagen's scaling totally eliminates EE/ringing. I just said that after scaling there may be differences in how much EE/ringing is noticable, depending on the scaling algorithm. Some scaling algorithms seem to increase or intensify EE. Lumagen's scaling is said to slightly reduce EE. Again: "Slightly reduce" is not "totally get rid of". Furthermore, just as above, I'm just repeating what I've been told in other thread in other forums. But Lumagen calls its own scaling "proprietary No-ring™ scaling". That should give us a hint which aspect of the scaling algorithm is most important to them.

- are any jaggies or other artifacts visible? I thought those were deinterlacing artifacts, not scaling artifacts. The only such scaling artifact that I was aware of was "line twitter", something that again very seldom is a problem even with low cost solutions.
If you use MS Paint to zoom into an image 800%, then you do see jaggies, don't you? When using a good scaling algorithm you usually don't. Well, probably no scaling algorithm used in today's devices is bad enough to produce visible jaggies, but theoretically it's possible.

I thought a very simple test is the multiburst pattern ? They are supposed to be black/white/black/white strips. See on your projected (upscaled) screen whether it's still clearly black/white/black/white, and not white/darker white/lighter black/black/etc, and they are still one pixel width. I know the VP30 does not give me perfect strips, whereas the lumagen does.
What was your source, what resolution was your source outputting, and over what type of connection?

Ok, I tried playing some multiburst patterns on 4 DVD player setups:

1. the Toshiba HD-A1 (upconverted, so the scaling takes place in the player) - PASS
2. the Oppo 970 HD (at 480i over HDMI - Crystalio 2 does the scaling) - FAIL
3. the Oppo 971h (upconverted, so the scaling takes place in the player) - PASS
4. the Panasonic RP-82 (at 480i over SDI - Crystalio 2 does the scaling) - PASS

Three of the four players came out clean as a whistle, while the fourth player (the 970HD) was the only player did that not display the higher frequency bursts (above 6.0 mhz) cleanly. Since the scaling was done in the C2 with both the RP-82 and the 970HD, and the RP-82 output was VERY clean, I would have to conclude that the cloudiness I observed from the 970HD was inherent in the player's HDMI output, not the C2, so that test does not seem to me to be a reliable indicator of scaling quality as it can fail the test despite the fact that the multibursts were superb with the RP-82/C2.
Some scaling algorithms seem to increase or intensify EE. Lumagen's scaling is said to slightly reduce EE.
Thanks, madshi. Maybe my next video processor will be a Lumagen!

The purpose of the multiburst pattern is to test whether the display is resolving detail at specified resolutions. The higher the MHz the system cleanly resolves the alternating lines the better able it is to resolve detail at various resolutions.

1080i: 37.1 MHz
720p: 18.5 MHz
480p: 13.5 MHz
480i: 6.75 MHz

So if a particular combination of source, input, and display does not cleanly resolve a multiburst pattern at a specified MHz, then this is an indication that the frequency response is clipped at that resolution and you are losing picture detail.

I only go over this information as a way of agreeing with Bob. How would you know if what seemed like weak scaling wasn't really just a frequency response problem?

A better approach would be to set the scaler to output the display's native rate, set the display to Native, and then look at 1-pixel horizontal and/or vertical lines and see how cleanly the system resolves these.

The maximum horizontal resolution of 720p of 1280 pixels across also needs 37 MHz circuit frequency response for a baseband analog video signal.

Video hints:
http://members.aol.com/ajaynejr/video.htm

What was your source, what resolution was your source outputting, and over what type of connection?

We had tried 2 src, I did HDMI from a pio59 480i, a friend did a SDI from Denon 2900.

Btw when u say "PASS", you really have identical intensity for each black/white strips ? and how thick is each strip (i.e. how many pixels ? I guess this depends on the resolution.) ? How is the transition from black to white ? (we are really just interested in the last band)

Btw when u say "PASS", you really have identical intensity for each black/white strips ?
Yup, on all the ones I labeled "PASS". The one labeled "FAIL" varied in intensity.
and how thick is each strip (i.e. how many pixels ? I guess this depends on the resolution.) ?
I have a 1080p display (Sony Ruby). I have no way of telling for sure, but on the highest frequency bursts the lines appear to be 1 pixel wide. I have also tried the internal 1 pixel wide strip pattern from the C2 and it displays perfectly, so I know that the projector is up to the task...:)
How is the transition from black to white ? (we are really just interested in the last band)
Very sharp and very crisp...like I said extremely clean. The only DVD player that was not clean was the 970HD above 6.0 mhz. To be perfectly honest with you, the only reason I tried the 2 upconverting DVD players was because I expected them to fail and wanted to see just how bad it would look, but to my surprise it was the 970HD that failed and the other 3 were extremely clean.

I am now convinced that multiburst patterns have too much interaction with other components in the chain (as Tom mentioned) and are not a good way to judge scaling quality as a single entity.

A few points:
1. Lumagen's algorithm definitely reduces mosquito noise. You don't really notice it until you switch to something else that lacks it. It also does not ring. I'm very sensitive to ringing and I have a lot of sharp content (subtitles, letterbox) that normally rings way too much.

2. I can definitely see scaling algorithm differences. Just put on a film mode source and scale it up to native res. The higher the ratio between source and target resolutions, the harder the scaling works and the problems become more pronounced. BTW, I find 768p 42" plasmas and 37" FHD LCDs running at NR to pronounce these issues more than 720p projectors at 100-120". Don't know if it's just me or what.

3. I wont get into the Lumagen vs. DVDO algorithm thing - but I think that Lumagen's algorithm, as implemented in HDQ and HDP Pro does a heck of a job. I've tested various scalers on a variety of displays (most recently the 1080P pioneer) and the difference can be quite astounding when you have a low res source.

http://www.lumagen.com/testindex.php?module=news

The RadianceXG video processor will be available in 2007 Q1 from authorized Lumagen dealers at a suggested price of $3499. Rack-ears for mounting the RadianceXG in a 19” rack are available at a suggested price of $50. Pricing for the external HDMI to analog converter and external SDI to HDMI converter is TBD.

Did the price just go up. I read the new at widescreenreview.com and the price was listed as $3999?

- Rich

Did the price just go up. I read the new at widescreenreview.com and the price was listed as $3999?
No idea. Lumagen's "news" section on their homepage still lists $3499 as of today.

No one told me it's going up.....

Gordon

No one told me it's going up.....

Gordon

Here is a The Lumagen RadianceXG™ Processor Provides High-Performance Video Switching, Processing And Calibration For The Home Theater And Commercial Markets. (http://www.widescreenreview.com/news_detail.php?id=12389)

The last paragraph states:

The RadianceXG video processor will be available in 2007 Q1 from authorized Lumagen dealers at a suggested price of $3999. Rack-ears for mounting the RadianceXG in a 19” rack are available at a suggested price of $50.

- Rich

A lot of people have been asking us, Lumagen, about an update on Radiance, so I thought I would post one on this thread before doing a press release this week. This info is hot off the presses, as we made the decision this past week.

Many people on this forum have followed the Radiance series since well before we even had a name for it. They will recall that we originally planned on the RadianceXD being the first family member, that it was to cost $5999, and that it would have six HDMI inputs and two HDMI outputs.

Then as the prices fell dramatically on 1080p projectors, we were trying to have 4 HDMI inputs at $4499, then at $3999 (this got into a Widescreen review of the Vision HDQ). We were working on further design refinements that we thought would allow a $3499 price with 5 HDMI inputs, 2 component, 2 SVideo, 1 VGA and 2 composite. This is what we talked about at CEDIA.

Three months have gone by and we have made some additional decisions on the Radiance series. We were not able to come up with enough cost savings to make the $3499 price with five HDMI inputs (or even just four) fly for us. We really wanted this to be the best video processor available, so, after a lot of thought, and even some work on a non-modular design, we have come to the decision to increase the input complement to six HDMI inputs plus add additional analog video inputs, and to list the product at $3999.

In addition, since this product has virtually all the features we originally proposed for the $5999 RadianceXD, we decided it should bear that moniker. We believe the second letter being related to the number of outputs (S=single, D=dual) will keep confusion down some as we add more products to the line.

I believe it’s okay for me to now confirm that we are using the new Gennum chip (as opposed to the last generation currently used by others) in the RadianceXD. Originally our NDA prevented us from publicly confirming this. We believe the silicon Gennum already has available for this new chip has all the functionality we need or will use. Since we – unlike pretty much everyone else - use the Gennum only for deinterlacing/inverse-telecine, edge enhancement and noise reduction, and these functions were demonstrated publicly at CEDIA, I hope we can put to rest some unfounded rumors about availability of the Radiance series Beta. If some issue was later uncovered, we – since our system is modular – could replace the Lumagen RadianceXD Beta unit’s Video-Processor-Daughter-Card containing the Gennum chip. However, we do not think this will be necessary based on the data we now have.

We are planning on the Beta shipments starting at the end of January with a few key distributors and dealers getting a look at the unit. We will follow with limited Beta through March. We are hopeful that first production will be in April. Of course we have a lot of ground to cover before then, and we might not make this date. However, as with our Vision series, if production release were to be delayed, we would continue the sale of Beta units. Beta units are again to be production hardware with pre-production software. I think our history shows we continue to actively update our software before and after production. So, for us, production release is a softer boundary than for other companies in this business.

RadianceXD Video Inputs (18):
6 HDMI
4 Component
4 SVideo
4 Composite

RadianceXD Audio Inputs (18):
6 HDMI (companion to video on same HDMI connection)
4 Stereo Analog
6 COAX
2 Optical

RadianceXD Outputs:
2 HDMI
2 COAX Audio

Other features are pretty much as described earlier. There will be a press release on our website Monday. We will be demonstrating the Alpha level RadianceXD at CES booth South 21666.

If anyone is interested in being a RadianceXD Beta test site, please email sales@lumagen.com and we will try to accommodate you. As with the Vision series, since Beta units have production hardware, and can be updated to full production with a software update, we do charge the same price for Beta as production. We do require Beta testers to sign a license agreement that these are not (yet) production units, that Beta results are confidential to Lumagen, and to actively provide feedback to Lumagen. We hope this is a win-win since it provides us with much enhanced testing, and provides Beta testers early access to new technology.

A lot of people have been asking us, Lumagen, about an update on Radiance, so I thought I would post one on this thread before doing a press release this week. This info is hot off the presses, as we made the decision this past week.

Many people on this forum have followed the Radiance series since well before we even had a name for it. They will recall that we originally planned on the RadianceXD being the first family member, that it was to cost $5999, and that it would have six HDMI inputs and two HDMI outputs.

Then as the prices fell dramatically on 1080p projectors, we were trying to have 4 HDMI inputs at $4499, then at $3999 (this got into a Widescreen review of the Vision HDQ). We were working on further design refinements that we thought would allow a $3499 price with 5 HDMI inputs, 2 component, 2 SVideo, 1 VGA and 2 composite. This is what we talked about at CEDIA.

Three months have gone by and we have made some additional decisions on the Radiance series. We were not able to come up with enough cost savings to make the $3499 price with five HDMI inputs (or even just four) fly for us. We really wanted this to be the best video processor available, so, after a lot of thought, and even some work on a non-modular design, we have come to the decision to increase the input complement to six HDMI inputs plus add additional analog video inputs, and to list the product at $3999.

In addition, since this product has virtually all the features we originally proposed for the $5999 RadianceXD, we decided it should bear that moniker. We believe the second letter being related to the number of outputs (S=single, D=dual) will keep confusion down some as we add more products to the line.

I believe it’s okay for me to now confirm that we are using the new Gennum chip (as opposed to the last generation currently used by others) in the RadianceXD. Originally our NDA prevented us from publicly confirming this. We believe the silicon Gennum already has available for this new chip has all the functionality we need or will use. Since we – unlike pretty much everyone else - use the Gennum only for deinterlacing/inverse-telecine, edge enhancement and noise reduction, and these functions were demonstrated publicly at CEDIA, I hope we can put to rest some unfounded rumors about availability of the Radiance series Beta. If some issue was later uncovered, we – since our system is modular – could replace the Lumagen RadianceXD Beta unit’s Video-Processor-Daughter-Card containing the Gennum chip. However, we do not think this will be necessary based on the data we now have.

We are planning on the Beta shipments starting at the end of January with a few key distributors and dealers getting a look at the unit. We will follow with limited Beta through March. We are hopeful that first production will be in April. Of course we have a lot of ground to cover before then, and we might not make this date. However, as with our Vision series, if production release were to be delayed, we would continue the sale of Beta units. Beta units are again to be production hardware with pre-production software. I think our history shows we continue to actively update our software before and after production. So, for us, production release is a softer boundary than for other companies in this business.

RadianceXD Video Inputs (18):
6 HDMI
4 Component
4 SVideo
4 Composite

RadianceXD Audio Inputs (18):
6 HDMI (companion to video on same HDMI connection)
4 Stereo Analog
6 COAX
2 Optical

RadianceXD Outputs:
2 HDMI
2 COAX Audio

Other features are pretty much as described earlier. There will be a press release on our website Monday. We will be demonstrating the Alpha level RadianceXD at CES booth South 21666.

If anyone is interested in being a RadianceXD Beta test site, please email sales@lumagen.com and we will try to accommodate you. As with the Vision series, since Beta units have production hardware, and can be updated to full production with a software update, we do charge the same price for Beta as production. We do require Beta testers to sign a license agreement that these are not (yet) production units, that Beta results are confidential to Lumagen, and to actively provide feedback to Lumagen. We hope this is a win-win since it provides us with much enhanced testing, and provides Beta testers early access to new technology.

Jim,

Thanks for the update before the press release. I look forward to the reviews of the Radiance Beta. I'm looking for this type unit for my G-90.

Tim

Jim,

Thanks for the update before the press release. I look forward to the reviews of the Radiance Beta. I'm looking for this type unit for my G-90.

Tim

Tim:
You should check out the G90 DVI input (with HDCP) module that I believe is legally available (from what I have heard). This would be an excellent interface to use with the Radiance or one of our Vision series.

Not sure who is sellin gthis though, so you would need to search some.

Jim, any plans to come out with a version supporting fewer I/O options, at a lower cost, as with the Vision series?

SDI support?

no analogue out?
gamma correction per primary?

Tim:
You should check out the G90 DVI input (with HDCP) module that I believe is legally available (from what I have heard). This would be an excellent interface to use with the Radiance or one of our Vision series.

Not sure who is sellin gthis though, so you would need to search some.
Jim,

I just missed a run of these; http://www.avsforum.com/avs-vb/showthread.php?t=703554&page=1&pp=30

A few of us are requesting another run.

T

I vote for HD-SDI input

Jim,

When the Radience was announced, an external DAC was promised for us CRT owners. Your advice to the G90 owner above has me concerned that you have shelved plans for this. Can you reassure me?

Martin

Jim, any plans to come out with a version supporting fewer I/O options, at a lower cost, as with the Vision series?

It is our intension to do a lower priced unit with fewer IO, but we are not ready to discuss any details.

SDI support?

We plan an SD-SDI external module.

We are not planning to do an HD-SDI box at this time. I believe there are one or two of these being produced already, and given that we believe the market for these is very small, we would suggest people who need HD-SDI look at one of these. Of course, if we see a lot of demand for this we could change our minds.

Jim,

When the Radience was announced, an external DAC was promised for us CRT owners. Your advice to the G90 owner above has me concerned that you have shelved plans for this. Can you reassure me?

Martin

We still plan to do an external DAC. It will probably not be available until we get to production status.

PLEASE NOTE: HDCP inputs will not be able to be output on analog from this external DAC. This is an HDCP legal issue, rather than a technical one. We are restricted from doing this by our license agreement for HDCP.

no analogue out?
gamma correction per primary?

Analog out is an external option (see my last post).

You can do gamma per primary already in the Vision series using the parametric grayscale/gamma menu. Of course this affects the grayscale, since the two are closely tied.

I don't personally think of this as gamma-per-primary. Instead I do an overall gamma, then calibrate grayscale at all the points and finally I iterate these calibrations if needed.

If you are asking if we have a single gamma control "knob" per color like a PC, we are considering this. This would allow a rough approximation of accurate grayscale and gamma before switching to our 11-point mode.

Hi Jim, will there be "Primary color correction", as advertised in the XG press release? I interpreted that to mean adjustment of a display's primaries in the CIE diagram.

Thanks.

Hi Jim, will there be "Primary color correction", as advertised in the XG press release? I interpreted that to mean adjustment of a display's primaries in the CIE diagram.

Thanks.

This is also a feature I am very interested in.

Mark

Jim,

First of all - congratulations!!!

This looks like an excellent product and I am confident that you will meet your usual high standards.

Few questions:
1. Will the analog inputs continue to be configurable BNC based as in previous models?
2. Will you continue to support RGBS through component + composite input?

Thanks!!!

2. Will you continue to support RGBS through component + composite input?I think that's why there are 4 composite inputs (to go with the 4 component inputs). Can't think of any other reason why anyone would want 4 composites! :D

JRP. Thanks for a very informative series of posts. I like the naming decision. A great way to get the moniker off your back. :)

Hi Jim, will there be "Primary color correction", as advertised in the XG press release? I interpreted that to mean adjustment of a display's primaries in the CIE diagram.

Yes, we plan on having primary correction. This may not be until after production release, but we will get it in as soon as we can.

Please note that primaries can only be corrected to the extent they are outside the primary color triangle. In other words if the display/projector has a color space that is larger than the standard, it can be corrected. If it is smaller than the standard, it can not be corrected.

Jim,

First of all - congratulations!!!

This looks like an excellent product and I am confident that you will meet your usual high standards.

Few questions:
1. Will the analog inputs continue to be configurable BNC based as in previous models?
2. Will you continue to support RGBS through component + composite input?

Thanks!!!

Thanks Ofer.

The analog inputs will be RCA and SVideo standard connections. This was a tough decision, but I think most people prefer the standard connections over BNC's at this point, so we decided to go with them. It also allowed us to have as many inputs as we do since they are smaller.

Inputs are going to continue to be configurable. We group YPbPr and composite again so it is a little cleaner to have RGBcvS inputs. All four of the component inputs will be capable of RGBcvS when combined with the respective composite input.

We will allow a single RGBHV input using one of the component and two of the composite inputs. That would leave the ability to, in addition, have two component and one RGBcvS (or one component and two RGBcvS).

While most people do not need four component, composite or SVideo inputs, we do get requests for that many, and so decided that to meet our "Best of Breed" product goal we thought it appropriate to go a little wild on the total number of inputs. We don't expect many people to use all 18 video inputs of course.

Jim,

Although I personally prefer coaxial digital audio connections I'm finding that many of today's HD set top boxes only offer Optical digital audio outputs. I have 3 boxes with only Optical and only 2 that offer coaxial. I would suggest you offer 4 and 4 instead of 6 and 2.
thanks
michael

Jim,

Although I personally prefer coaxial digital audio connections I'm finding that many of today's HD set top boxes only offer Optical digital audio outputs. I have 3 boxes with only Optical and only 2 that offer coaxial. I would suggest you offer 4 and 4 instead of 6 and 2.
thanks
michael
What he said. I prefer optical because it avoid ground loops. Furthermore many devices only have optical out and no coax out. There are only very few devices which only have coax out.

Congrats on the new product Jim!

For digital audio Optical verses Coaxial input count:

I understand what you are asking.

For a number of reasons we ended up with 2 optical and 6 COAX. We are pretty much locked in for the RadianceXD.

I will look at having more optical inputs in a future product. In the mean time you would have to use an external optical to COAX converter box.

One I found with a quick search is: CABLES TO GO 40019 Optical-coax Audio Converter on amazon.com for $17.45. Not sure how this one performs and if there are better options.

So will the HDMI in/out portion of the Radiance still be modular? Will it release with HDMI 1.1 and be upgradeable to HDMI 1.3? Or might it release with HDMI 1.2 or later? I am anxiously awaiting beta since all these newer devices seem to HATE DVI connections.

A question and suggestion:

Will there be any upgrade program for current owners of the vision products???

Also, It would be nice to have a 21 point Cal with the lower Stim. points having a 0.01 resolution instead of 0.1. If the 21 point is not do-able at least the increase to 0.01 for say 30 and below.

One more question, When do you expect to be able to ship if one wanted to beta test???
This one is for planning perposes.

Keep up the Great Work as a owner of almost 2 years I have to say You Guys are OUTSTANDING and would not hesitate buying again!

Thanks
Rew

So will the HDMI in/out portion of the Radiance still be modular? Will it release with HDMI 1.1 and be upgradeable to HDMI 1.3? Or might it release with HDMI 1.2 or later? I am anxiously awaiting beta since all these newer devices seem to HATE DVI connections.
From the press release:

The RadianceXD is a modular platform that allows for upgrading capabilities without the need to purchase an entirely new video processor. For example, the unit’s current video processor chip utilizes 10-bit pixel-depth processing, however when a 16-bit pixel-depth video processor becomes available from Lumagen, it can be installed by simply replacing a small daughter card and updating the unit’s software.I imagine the answer to your question is yes as there's not much point in being able to swap in a 16-bit pixel-depth processor if you can't output Deep Color...

Jim, I'm a bit worried after reading the very latest press release. There's no talk about detail enhancement, anymore. Furthermore earlier in this thread you didn't mention that you'd be using the Gennum scaling. This sound to me as if: (1) you've decided to not offer Gennum scaling as an option and (2) as a consequence there's no way for Radiance users to get access to the Gennum detail enhancement algorithms. Could you comment on that? I was really looking forward to Gennum's detail enhancement. So I'd love to have the option to choose between either using Lumagen's scaling and no detail enhancement or Gennum's scaling and Gennum's detail enhancement. Thanks!

madshi,

IMHO, detail enhancement usually tends to ruin a perfectly good picture. I can think of no better scaling algorithm than Lumagen's current scaling, so I would gladly give up DE in favor of better scaling (note: I have not compared Lumagen's scaling algorithm to Gennum's so this might be a moot point if Gennum reaches that level of PQ).

madshi,

IMHO, detail enhancement usually tends to ruin a perfectly good picture. I can think of no better scaling algorithm than Lumagen's current scaling, so I would gladly give up DE in favor of better scaling (note: I have not compared Lumagen's scaling algorithm to Gennum's so this might be a moot point if Gennum reaches that level of PQ).
Ofer, I've read that you do activate detail enhancement in your Mosquito (although only at very low levels). In a chat with a Gennum guy I asked whether detail enhancement should be done before or after scaling. He told me during/after scaling was noticably better. The Mosquito does it before scaling. Now if you like the Mosquito's detail enhancement which is done before scaling, how good can detail enhancement be if it's done in a good way during/after scaling? Btw, the Gennum guy claimed the detail enhancement in the latest chip was improved and might actually beat Algolith's algorithm. Well, we'll have to wait and see. But I'd really like to have access to these algorithms.

About the quality of the scaling: I've read from various sources that the Crystalio II scales better than the iScans. Maybe it's not as good as Lumagen's scaling, but I don't expect the difference to be extremely big. One of my sources told me he likes Lumagen's upscaling ever so slightly more than the CII's upscaling, but he also said he likes CII's downscaling noticably better than Lumagen's downscaling. Which is kinda interesting for me, as my display is 768p.

As I said, I'd prefer to have the option:

(1) Either Lumagen's scaling and no detail enhancement
(2) Or Gennum's scaling and Gennum's detail enhancement

I don't see how having this option can hurt anybody. After all the Gennum chip is there, why not offering the option to use it. Well, ok, if the current revision of the Gennum chip has problems with scaling and/or detail enhancement (which I don't know but which you could eventually read between the lines in one of Jim's statements) then this might be a reason to drop the Gennum scaling/DE option I'm asking for.

I wish lumagen well, but introducing another video processor for $4000 no matter how many inputs you have may be a recipe for a short lifetime as a company.

The $4000 price range for a video processor is going to be increasingly difficult to justify. Anthem is the first but likely NOT the last company with a AV pre/pro/reciever with quality video processing/deinterlacing builtin. The AV receiver is the logical place for video processing since the appropriate audio delay for the deinterlacing algorithm being used on the video content is probably already implemented in the audio portion of the receiver.

Most new 1080p TV's / Projectors will include some sort of deinterlacing/ scaling. The quality of course will vary greatly depending upon the chipset used. The appropriate audio delay for the deinterlacing (if any is in effect) is difficult if not impossible to automatically set up in the associated AV pre/pro/receiver.

At the low end Gefen with their 2 HDMI input scaler/deinterlacer at $450. Will bracket the video processor market.

Given the ongoing problems with HDMI any company selling a pre/pro/receiver with good customer service and timely firmware updates should do well. I do not see how a company that specializes in $4000 video processors can have anything but an infintesimally small market.

Does anyone else have thoughts on why the video processor companies are introducing products that cost nearly as much as the projectors they might be paired with? I don't get it.

dheiskel - Be careful what you wish for.
IMO we are very lucky to have companies who are prepared to take the high ground. As soon as you lower the price, you also pander to the lowest common denominators.
Keeping the price at the level it is now is obviously providing us with the features we desire. Lower price products featuring these chips always always compromise the feature-list.

I say to companies like Lumagen, Pixel Magic, Algolith, previously Immersive, Theatertek, etc, etc - keep doing what you're doing and I'll be glad to support you whenever funds allow. It's always an education being involved in discussions about your products - long may it continue.

I would add that only a few years ago (I know, I bought one), the only processor that did 1080P was the F* 5000 at $20K. With the Cryst* at $5K, I'm looking forward seeing what this puppy can do. Besides, I happen to know that if you know what you're doing and can install it yourself, it can be had for a substantial discount.

As to "hoping" it would be cheaper, the market will decide if it's successful. While hope is a nice word, it's no substitute for a sound marketing strategy.

I think Lumagen's RadianceXG is at a very attractive price point. Being perceived as expensive or not, personally I think all products coming from Lumagen are the result of a tremendous amount of work and dedication for which hourly development and research time is surely not account into reflecting the final price. This is not exactly a mass-produced unit. It is developed by a small crew dedicated to their enthusiasm for "superior video", knowing perfectly well they will never be exactely filling their wallets by making products like this that have an extremely limited market, but because they believe in trying to produce a superlative creation for the enthusiast. If these guys were doing this for money only, they would be doing it at all. My hat off to people like the Lumagen crew that passionately puts together quality devices like this, that none of the "big" names in the industry would even dare considering putting their time on at this point in this industry. In a world of standards and rebadges, its people like these German folks that push the envelope for the "better" products out there.

I wish them the best with this new unit (I do hope it has detail enhancement capabilities) and I will be among the first to get one as soon as its available.

The $4000 price range for a video processor is going to be increasingly difficult to justify. Anthem is the first but likely NOT the last company with a AV pre/pro/reciever with quality video processing/deinterlacing builtin. The AV receiver is the logical place for video processing since the appropriate audio delay for the deinterlacing algorithm being used on the video content is probably already implemented in the audio portion of the receiver.Denon is expected to announce and demonstrate HDMI 1.3 receivers with video processing from Silicon Optix at CES next month. It will be interesting to see whether this high-quality scaling and deinterlace is limited to the $2000+ 4x0x7 series, or whether it will also find its way into the sub-$1500 3807.

I knew when I made my post that many would be offended and take the observation the wrong way.

I don't want these video processor companies to fail! I want them to succeed but as DVDO and lumagen have transitioned to providing HDMI switching they have dramatically increased their prices.

Other than properly delaying the audio stream from an HDMI source, there are not many features that are being added to justify the price increase. In fact most video processor vendors will now be using third party chipsets (Genum etc.). These chipsets contain most of the intellectual property that these companies used to embed in their own custom silicon.

So why are all these vendors products increasing in price? Because they see the writing on the wall that much of their old market will very soon no longer exist. Manufacturers of TV's , projectors, and mid to high end AV Receivers/ pre/pro's will all be using the same intellectual property (if not the same chips) that lumagen et al will now be using to provide scaling/deinterlacing. Lumagen seems to have decided to create the ultimate processor for the ultimate price.

The point of my post is that none of the vendors are fully taking advantage of the new silicon (intellectual property) and designing lower priced products. Any of these companies could have chosen to do what Gefen is attempting to do which is to design the "good enough" scaler/deinterlacer. Frankly, I would rather see one of the video processor companies try this than Gefen. These companies probably all have design teams better capable of implementing a decent low cost video processor.

madshi,

good memory :)

Yes, I indeed do that, but I use it only to counter the softening effects of the MPEG artifact removal. It's on a very low setting (1-2 out of 10).

dheiskel,

If you look at this industry, it's pretty small to begin with. The only real way to make some real dough is to sell your company off or convert your IP into silicon and sell that off through a big chip company (SI, Genesis, etc.).

Scalers were never a major force in the marketplace. Any HT shop guy will immediately tell you what 1080P was, but will likely get stumped if you ask what technology it uses for deinterlacing, scaling, etc. Processors alter functions and will be needed a lot less if and when projectors standardize on 1080p and source material standardizes on 1080p too.

but then, it'll probably move on to do some other stuff. MPEG artifact removal is already a pretty important step. Calibration correction, keystone correction (which will become more relevant when HT projectors start having 2K and 4K resolutions that will allow a software solution to become relevant without ruining the source material too much).

This industry changes. You can't assume that a 500$ product will give you what others do for $2000 or $4000 until we actually see it with our own eyes. Such changes have been done in the past. DVDO did the first jump when they started selling doublers based on the 503/504 chipsets. Lumagen, Focus Enhancements, ABT and SO all contributed massive drops in costs in recent years. So, it's definitely possible that a "new" player like Hagai Geffen produces something like that. Still, it's important to separate hype from fact and see what you actually get for that kind of money. I can't see the feature set that the Radiance XG has (or the VP50, for that matter) being implemented for that kind of money.

In any case, that is a discussion that belongs elsewhere, not on this thread. Create a new thread with discussions about the future of scalers and their prices, that's where that kind of discussion belongs.

So will the HDMI in/out portion of the Radiance still be modular? Will it release with HDMI 1.1 and be upgradeable to HDMI 1.3? Or might it release with HDMI 1.2 or later?

Rodger:
Current plan is to intro with HDMI 1.1.

We have a 1.3 upgrade program on the drawing board but do not know when that will be done.

A question and suggestion:

Will there be any upgrade program for current owners of the vision products???

Also, It would be nice to have a 21 point Cal with the lower Stim. points having a 0.01 resolution instead of 0.1. If the 21 point is not do-able at least the increase to 0.01 for say 30 and below.

One more question, When do you expect to be able to ship if one wanted to beta test???
This one is for planning perposes.

Keep up the Great Work as a owner of almost 2 years I have to say You Guys are OUTSTANDING and would not hesitate buying again!

Thanks
Rew

Since we ended up at $3999, which was the price point we will be giving a 33% of retail trade-in credit for older Lumagen models. Of course you can still sell on the used market if you can get more than that value.

We are talking about having a 21 point grayscale for the Radiance line. We might go to a 0.04 (1 part in 256) to refine the current 0.1 IRE steps as well. We will update everyone as we get further along.

We plan for the very first Beta ship at the end of January. This is tight. We would then follow with the first larger batch of Beta at the end of February, but these are spoken for at this point. be able to open up Beta some at the end of March.

If you want to be on the Beta list please email sales@lumagen.com and we can add you.

Thanks for the kind words. It is very gratifying to have such great support from our customers.

Ofer, I've read that you do activate detail enhancement in your Mosquito (although only at very low levels). In a chat with a Gennum guy I asked whether detail enhancement should be done before or after scaling. He told me during/after scaling was noticably better.

It depends what you mean by 'detail enhancement'. Do you mean noise reduction? (like MPEG noise reduction, mosquito noise reduction, etc). In this case its highly recommended, in theory, to do it before scaling (even before deinterlacing) so that you can process the original field images. If you scale you will alter the size & shape of the MPEG blocks (for example) and make it harder to properly detect them.

Detail Enhancement (for example, an unsharp mask algorithm) would be more correctly done after scaling.

Put another way, too-aggressive noise reduction will tend to soften the picture. Too-aggressive detail enhancement will tend to introduce ringing (or other grain-type artifacts)

Andy K.

Jim, I'm a bit worried after reading the very latest press release. There's no talk about detail enhancement, anymore. Furthermore earlier in this thread you didn't mention that you'd be using the Gennum scaling. This sound to me as if: (1) you've decided to not offer Gennum scaling as an option and (2) as a consequence there's no way for Radiance users to get access to the Gennum detail enhancement algorithms. Could you comment on that? I was really looking forward to Gennum's detail enhancement. So I'd love to have the option to choose between either using Lumagen's scaling and no detail enhancement or Gennum's scaling and Gennum's detail enhancement. Thanks!

The Gennum detail enhancement is available in the various cases. Info from Gennum suggests that they designed it to work best at the output resolution. So, if we use Lumagen scaling for SD material, the amount of detail enhancement available might be reduced some, but it could still be turned on. For HD input, the Gennum detail enhancement should fully available.

We have not yet decided if we will support both our scaling and Gennum's scaling. It is possible but I make no commitment at this point.

I tend to agree with Ofer about detail enhancement having a negative overall image quality effect, but I have not had time to critically evaluate this with the Gennum, and will reserve judgement for a time.

I wish lumagen well, but introducing another video processor for $4000 no matter how many inputs you have may be a recipe for a short lifetime as a company. ...

I understand this is a sincere concern you have. I, of course, think about this type of issue all the time, especially as we develop new technologies and have to make the requisite price and performance trade offs.

I believe there will always be people who will pay a premium for the best. I believe that we can produce a better product than the mainstream companies. I especially believe that we can do a better job than those integrating a piece of silicon into an amp without much experience in video processing.

We are talking to several amplifier companies about integrating our technology into their audio amp/receivers, so some of these companies, at least, understand it is not as simple as throwing a chip into their boxes.

We also have extensive calibration and setup features that solve problems in large systems. Many of these problems occur only after a band new product is integrated into an existing system due to bugs in the new product, or bugs that the new product reveals in older products in that system. We support our customers and have often added a special feature or added support in our products to work around the bugs in the major brand products. Try to get that level of support from a larger company and you will be disappointed.

So, while I agree your concern is valid, I believe we can carve a small niche out for ourselves with people who really know about, and care about, video quality.

Thanks for the update Jim. I am really pleased with my HDQ and even more so by the support I have received. Not only did you folks work to get the HDQ to work with the HD-A1 but also went out and bought a MYHD card to help me resolve some issue with it.

It is hard to put a price on that kind of support. I cannot wait to see the RadianceXG.

-- Rich

P.S. The scaling is really fantastic. I appreciate the improvement every day :D

Jim - Could it or would it be possible to assign one of the digital audio input from Coax or Toslink to a particular input of the HDMI? I still have a OTA HD receiver that has a DVI output that I intend to plug into one of the HDMI port for Video and Coax/Toslink for audio thru the Radiance. I would very much like the Radiance to be the switching hub for both A/V, it would really ease the usage for wife and kids.

I'm very interested in this unit and hope to see/acquire it in April 07. Would like to use this for both Sony Pearl and 65" 1080P plasma in the same room. This is gonna be a long wait. Hoping to get a decent deal from AVS for this unit.

Thanks for the updates JRP.

You and the others at Lumagen are like a breath of fresh air in this industry.

You make every Lumagen owner somebody special and treat his/her needs accordingly.

It depends what you mean by 'detail enhancement'. Do you mean noise reduction? (like MPEG noise reduction, mosquito noise reduction, etc). In this case its highly recommended, in theory, to do it before scaling (even before deinterlacing) so that you can process the original field images. If you scale you will alter the size & shape of the MPEG blocks (for example) and make it harder to properly detect them.

Detail Enhancement (for example, an unsharp mask algorithm) would be more correctly done after scaling.
Thank you for your comment. I meant texture/edge enhancement. I'm aware that noise reduction should be done before scaling. Your comment confirms what the Gennum person told me, which is not surprising, as you both know very well what you're doing... :)

The somewhat surprising thing for me is that the Mosquito is doing detail enhancement before scaling and still several people like it. So I'm thinking if a similarly good algorithm would be used after scaling, it should be even better.

Put another way, too-aggressive noise reduction will tend to soften the picture. Too-aggressive detail enhancement will tend to introduce ringing (or other grain-type artifacts)
Yes, agreed. I'm planning to use all these algorithms with caution and at rather low settings.

The Gennum detail enhancement is available in the various cases. Info from Gennum suggests that they designed it to work best at the output resolution. So, if we use Lumagen scaling for SD material, the amount of detail enhancement available might be reduced some, but it could still be turned on. For HD input, the Gennum detail enhancement should fully available.
But since detail enhancement is more important for SD than for HD, I'm still hoping you'll consider offering Gennum scaling + detail enhancement as an option. I don't want to put any pressure on you, though, I'm just posting my opinion/wishes. I think my point is that I'd like to be able to test things out myself and decide for myself which algorithms I like best. I might decide (or not) that I can live with a bit of ringing if I get a noticably sharper picture in return.

We have not yet decided if we will support both our scaling and Gennum's scaling. It is possible but I make no commitment at this point.

I tend to agree with Ofer about detail enhancement having a negative overall image quality effect, but I have not had time to critically evaluate this with the Gennum, and will reserve judgement for a time.
Ok, thank you!

-----

One technical question, if you don't mind: Will the Radiance do FRC before or after scaling? I'm asking because I hope that in a year or two maybe Gennum might come out with a new chip which will do motion compensated FRC. Would the Radiance with its modular design be able to make use of this?

Thanks very much!!

Since we ended up at $3999, which was the price point we will be giving a 33% of retail trade-in credit for older Lumagen models. Of course you can still sell on the used market if you can get more than that value.

We are talking about having a 21 point grayscale for the Radiance line. We might go to a 0.04 (1 part in 256) to refine the current 0.1 IRE steps as well. We will update everyone as we get further along.

We plan for the very first Beta ship at the end of January. This is tight. We would then follow with the first larger batch of Beta at the end of February, but these are spoken for at this point. be able to open up Beta some at the end of March.

If you want to be on the Beta list please email sales@lumagen.com and we can add you.

Thanks for the kind words. It is very gratifying to have such great support from our customers.

Jim: Before I get my hopes up, please clarify what the 33% applies to i.e. is it 33% off the $3999 or 33% of the original MSRP for my HDP? My hopes obviously are for the former. Secondly, how long will the offer last? I don't think my very limited technical skills would make me a good beta tester candidate.

Finally, thanks for your and the other Lumagen folks hard work in developing these products and the great support that has followed. Based, in large part, upon my experience with Lumagen products I plan to continue using a VP with every large screen TV I own.

madshi,

I'm not at all sure that the Algolith detail enhancement is strictly an unsharp mask. It certainly does not behave like one.

What it does is basically adds a kind of "protection" mask around edges from being smeared accidentally by the MPEG artifact remover...

. I'm not really sure it's applicable to anything except Algolith's (and HQV's) noise removal.

Dsinger: I think it's a safe bet to say it'll be 33% of the retail price you paid for your previous Lumagen product.....

Gordon

I'll pay 33% extra to get one NOW :D

Mark

Mark: I think you may have just moved up my beta tester list....hahahaha

Gordon

Dsinger: I think it's a safe bet to say it'll be 33% of the retail price you paid for your previous Lumagen product.....

Gordon

You Brits are such pessimists!

Could it or would it be possible to assign one of the digital audio input from Coax or Toslink to a particular input of the HDMI?

Yes, we plan to have this feature for exactly the reason you state.

[QUOTE=madshi] Will the Radiance do FRC before or after scaling? I'm asking because I hope that in a year or two maybe Gennum might come out with a new chip which will do motion compensated FRC. Would the Radiance with its modular design be able to make use of this?QUOTE]

The simple answer is yes.

We have a Field Programmable Gate Array (FPGA) available to process the video before and after the video processor chip. In addition, we can have up to a 16-bit bus to and from the video processor. With the dynamic gates in the FPGA, we can completely change how we do things and what we support -- within the gate-count limit of the FPGA.

The Radiance FPGA has less work to do than the VisionHDQ FPGA and there are nearly twice as many gate available. So we do have significant room to work.

The real question comes down to one of whether it makes sense for us to make the change to how we do the software.

As an aside: Motion compensated frame interpolation is a very difficult, and in some cases an impossible, problem to do well. One example is that in the case of a reveal of an object in motion, the needed image information may not exist in any frame. So, there is no information to properly interpolate the non-exisistant frame one is trying to construct.

In conclusion, I would just say that we listen to our customers, and meet their requests as best we are able given our constrained resources and time.

Jim: Before I get my hopes up, please clarify what the 33% applies to i.e. is it 33% off the $3999 or 33% of the original MSRP for my HDP? My hopes obviously are for the former. Secondly, how long will the offer last? I don't think my very limited technical skills would make me a good beta tester candidate.

Finally, thanks for your and the other Lumagen folks hard work in developing these products and the great support that has followed. Based, in large part, upon my experience with Lumagen products I plan to continue using a VP with every large screen TV I own.

It is 33% of the product you are trading in. So for a VisionHDP which has a $1499 retail, it would be $495 credit toward a RadianceXD. I realize you might get more from selling on the used market.

Thanks for your support and kind words.

It is 33% of the product you are trading in. So for a VisionHDP which has a $1499 retail, it would be $495 credit toward a RadianceXD. I realize you might get more from selling on the used market.

Thanks for your support and kind words.

Jim: Thanks. I thought that would be the answer but had to try. How long will the trade in offer be open after the beta period ends? I have spent at least $15k improving my video enjoyment since July so a liberal time horizon would be nice. Thanks

The simple answer is yes.

We have a Field Programmable Gate Array (FPGA) available to process the video before and after the video processor chip. In addition, we can have up to a 16-bit bus to and from the video processor. With the dynamic gates in the FPGA, we can completely change how we do things and what we support -- within the gate-count limit of the FPGA.

The Radiance FPGA has less work to do than the VisionHDQ FPGA and there are nearly twice as many gate available. So we do have significant room to work.
Thanks for this information, much appreciated! It sounds like a very foresighted design. :)

I hope I'm not going on your nerves, but I'm very curious. So here's another question: Would you mind dropping a small hint on how the OSD of the Radiance will be like?

What would really be cool if the added blending functions to one of there products some day! With all this blendzilla stuff i see on here i'm getting an itch for it and I love Lumagen scalers so it seams like the next big upgrade for them !

How about it JP !

A quick question about the Radiance.

1) Will it output 1080p/24sf from 1080i/60 and 1080p/60 film sources?

2) Will it auto-detect film sources vs. video sources and automatically change the output to either 1080p/24sf or 1080i/60?

Thanks

I hope I'm not going on your nerves, but I'm very curious. So here's another question: Would you mind dropping a small hint on how the OSD of the Radiance will be like?

Beta builds will most likely be the same as the Vision series as we have started with that as a base.

We are still working on the look and feel of the new OSD, which should be out about the time we go to production.

It is a "hardware overlay" based design. The Vision series design had to BitBlt (i.e. draw) all characters onto the screen at 60 fps -- thus the limited number of characters. With the overlay design we can have as many characters as we want.

We will not be going for the most beautiful OSD, but rather a good looking, mostly textual, OSD that has a lot of functionality --- and longer names for functions.

The menu tree concept will be used as it is really a pretty efficient design.

How long will the trade in offer be open after the beta period ends?

We plan for the trade-in program to be on-going. However, we do reserve the right to change our mind in the furture. We really want to show our appreciation to our current customers, some of which prefer to trade in rather than sell their old units on the used market.

What would really be cool if the added blending functions to one of there products some day! With all this blendzilla stuff i see on here i'm getting an itch for it and I love Lumagen scalers so it seams like the next big upgrade for them !

How about it JP !

It is a possibility. Please email us at support @ lumagen.com with how you would want this to work and be set up.

No promises though.

A quick question about the Radiance.

1) Will it output 1080p/24sf from 1080i/60 and 1080p/60 film sources?

2) Will it auto-detect film sources vs. video sources and automatically change the output to either 1080p/24sf or 1080i/60?

Thanks

1) 1080i/60 will be able to be output at 24sF. No plans for 1080p60 to be output at 24sF at this point. Could be done, though.

BTW: This is effectively the same task as 1080i inverse telecine. So the technology is really old-hat. I am not sure it is worth doing though. HD DVD and Blue Ray will output at the native 24p or 1080i. HD satellite and cable transmit 1080i or 720p. So this feature is totally not needed for all the major sources, and I really think it is a non-feature. Sounds like some satellite providers in Europe get it wrong and output 576p60 from film but I think this is a temporary issue until HDMI is more prevalent and the box vendors get it right. ---- I'm sure to get comments on this of course. This is my opinion, others may have a different opinion. I'm not really interested in trying to convince others.

2) This is problematic given how bad many sources are. There is the case of film from a DVD or HD-DVD from one source disc and then video from another. This could be done as the change is infrequent. From a satellite/cable source, these get mixed often. In my opinion, the re-sync time of the HDMI interface and display would make this horribly irritating.

We prefer to allow the user to manual set this up. If we did this (auto switch), we would probably put a lot of time hysteresis in to not allow it to change out of video mode very often. Something to think about though.

What would be way cool is 1080p120 (or 1080p100 for 50Hertz countries). This is a multiple of both film and video and would give you the best of both worlds.

2) This is problematic given how bad many sources are. There is the case of film from a DVD or HD-DVD from one source disc and then video from another. This could be done as the change is infrequent. From a satellite/cable source, these get mixed often. In my opinion, the re-sync time of the HDMI interface and display would make this horribly irritating.

We prefer to allow the user to manual set this up. If we did this (auto switch), we would probably put a lot of time hysteresis in to not allow it to change out of video mode very often. Something to think about though.
I've invested some thoughts into this problem some days ago. Here's what I came up with. Don't know if it makes sense to you:

(1)
- there's pure film content -> should be shown in 24p
- there's pure video content -> should be shown in 60p
- there's mixed content (film with video overlays or similar) -> should be treated as video content -> should be shown in 60p

(2)
As long as video and film mode changes only every few minutes, there's no real problem. It may not be very pretty, but just changing output modes every few minutes would be acceptable, I think. The problem occurs only if video/film mode changes all the time.

(3)
Showing video content in 24p is BAD. Showing movie content in 60p is acceptable.

(4)
Because of (3) the VP should always start with 60p. Automatic switching to 24p output should only be done if film mode is detected for more than "x" seconds without interruption. Switching back to 60p should be done as soon as video mode is recognized again. Switching to 24p should only be done if the last switch to 24p happened more than "y" minutes ago.

(5)
It should be possible to turn the automatic switching on/off in the setup. Additionally there should be a temporary setting which can be activated through the remote control. E.g. if a video stream is played where the auto mode behaves unsatisfying, it should be possible to force either 60p or 24p output via remote control (I imagine having one key which can be pressed repeatedly to choose between the 3 choices "auto/60/24"). Changing the temporary setting should be valid only until either the VP is turned off or another input is chosen. At that point in time the setting of the setup menu should be activated again.

What would be way cool is 1080p120 (or 1080p100 for 50Hertz countries). This is a multiple of both film and video and would give you the best of both worlds.
Ben Waggoner (Microsoft) repeatedly mentioned he'd love to go 1080p120 sooner or later for HD-DVD output, because that way he could get rid of 3:2 judder while still optimally being able to fit interlaced PIP content in... :) Unfortunately HDMI 1.3 doesn't allow 1080p120 with more than 8bit RGB, I believe. So we'll have to choose between either 1080p120 with 8bit RGB or 1080p60 with more than 8bit RGB.

Beta builds will most likely be the same as the Vision series as we have started with that as a base.

We are still working on the look and feel of the new OSD, which should be out about the time we go to production.

It is a "hardware overlay" based design. The Vision series design had to BitBlt (i.e. draw) all characters onto the screen at 60 fps -- thus the limited number of characters. With the overlay design we can have as many characters as we want.

We will not be going for the most beautiful OSD, but rather a good looking, mostly textual, OSD that has a lot of functionality --- and longer names for functions.

The menu tree concept will be used as it is really a pretty efficient design.
Thanks for the information!

I hope you can/will invest some time into making the OSD a bit more intuitive and easier to understand for first time Lumagen users. When I tried setting up my HDQ I had more trouble understanding the whole OSD concept than I think is necessary. Here are some thoughts:

(1) It should be possible to give everything a name. Output profile "0" is not a good thing if you look into the setup menu only once in a while. You won't remember what output profile "0" was after a few days.

(2) It should be possible to change all settings without having to do special work to get access to the settings first. E.g. IIRC with the HDQ you can only edit the currently active output profile.

(3) For every item in the OSD there should be information available whether this item is globally valid or whether it's part of a profile or only valid for specific input formats or inputs or whatever.

(4) Maybe you can add a little "help" text for each OSD item, too?

(5) I remember to have seen some "shall I copy the changes to xyz?" questions. I was never sure what effect exactly I got when pressing "yes" or "no". I'm not sure, maybe it was me, but I didn't fully understand some of these things.

Madshi:

The issues that you refer to are what I call/consider "paradigm issues" meaning that it is a matter of the user acclimatizing to the operation/setup.

Take, as an example, Windows -- moving from DOS to Windows took, for most people, a fair amount of time to get comfortable with BUT, with effort and time, we ALL became comfortable with the interface REGARDLESS of its deficiencies.

Sure there are things about the Lumagen interface that can be improved -- just like there are with every other product -- BUT, that said, I would prefer more investment in video processing design that in interface design because the best interface is useless without the underlying performance.

Hope this helps.

Jim and/or Pat (from Lumagen):

A thought -- have you considered building an "overlay input/function" which I would describe as follows:

-- A composite or s-video input from the pre/pro which would recieve OSD information from the pre/pro

-- Pre/pro OSD information received by the foregoing input would then be overlayed onto the output regardless of the format (i.e. componet, HDMI, etc.) or resolution (i.e. 720p, 1080i, 1080p) etc.

The above would be a terrific and VERY USEFUL function, particularly when it comes to calibrating items such as LipSync.

Over to you...

The reason for 1080p/24sf support is for a device like the Qualia. It should be able to feed it 1080i/60, 1080p/24. 1080p/60 and have the scaler pull out the proper 1080p/24sf. I assume it is a simple task to do this for 1080p/24 and 1080p/60 into 1080p/24sf.

Of course if you are showing video, it would be nice to simply tell it, or have it detect it and have it output 1080i directly.

Jim and/or Pat (from Lumagen):

A thought -- have you considered building an "overlay input/function" which I would describe as follows:

-- A composite or s-video input from the pre/pro which would recieve OSD information from the pre/pro

-- Pre/pro OSD information received by the foregoing input would then be overlayed onto the output regardless of the format (i.e. componet, HDMI, etc.) or resolution (i.e. 720p, 1080i, 1080p) etc.

The above would be a terrific and VERY USEFUL function, particularly when it comes to calibrating items such as LipSync.

Over to you...

yes ...such an overlay would be a very useful feature

Madshi:

The issues that you refer to are what I call/consider "paradigm issues" meaning that it is a matter of the user acclimatizing to the operation/setup.

Take, as an example, Windows -- moving from DOS to Windows took, for most people, a fair amount of time to get comfortable with BUT, with effort and time, we ALL became comfortable with the interface REGARDLESS of its deficiencies.

Sure there are things about the Lumagen interface that can be improved -- just like there are with every other product -- BUT, that said, I would prefer more investment in video processing design that in interface design because the best interface is useless without the underlying performance.
That's quite true. But the best performance can leave a shallow impression or might not even be fully unlocked if the interface is difficult to use.

I've experience with the Lumagen HDQ and various iScan models and I have to say that I found it easier to setup the iScans. I think the Lumagen setup is probably targetted at professionals (e.g. ISF calibrators) and at people who are working with the OSD everyday. For those people the Lumagen OSD is probably great. But I think for a hobby home cinema freak like me who wants to setup everything himself and touches the VP only once in a while the Lumagen OSD is more difficult to use than that of the competition.

The only thing that would have me looking at the Crystallio is the User Interface!

I've experience with the Lumagen HDQ and various iScan models and I have to say that I found it easier to setup the iScans. I think the Lumagen setup is probably targetted at professionals (e.g. ISF calibrators) and at people who are working with the OSD everyday. For those people the Lumagen OSD is probably great. But I think for a hobby home cinema freak like me who wants to setup everything himself and touches the VP only once in a while the Lumagen OSD is more difficult to use than that of the competition.


Madshi, I have to disagree with you on this one as I, like you, am a hobby home theater fanatic (and not a professional calibrator, installer, etc.) and actually find the interface very comfortable and easy to use.

That said -- one thing did help me understand the eleganec of it...take a look at the last few pages of the manual which provide a map/"tree diagram" as this really helped me understand things.

There is one point of yours thatI agree with...there is a lack of diocumentation as to which variables are global, input specific or even input resolution specific (i.e. some variables for input 1A are different/independent for 480i, 720p, and 1080i).

Madshi, I have to disagree with you on this one as I, like you, am a hobby home theater fanatic (and not a professional calibrator, installer, etc.) and actually find the interface very comfortable and easy to use.
Ok, that's fair. But do you have experience with other VPs, too, to compare the Lumagen interface with? And does your setup have to handle both 50Hz and 60Hz sources?

Let me just list my first steps how I setup my VP50:

(1)
I go into the "Output Configuration".
(2)
Under "Format" I choose "1366x768".
(3)
I change "VSync Front" timing to get 1:1 pixel mapping with my plasma. Test patterns can be easily turned on/off and switched by the remote control without leaving the "Ouput Configuration" menu tree at all.
(4)
I tell the iScan "50Hz input -> 50Hz output (locked)" and "60Hz input -> 60Hz output (locked)".

If my stupid plasma would do 1:1 pixel mapping this way in both 50Hz and 60Hz, I would already be done without even having to use output profiles at all. Unfortunately my plasma needs different timings for 1:1 pixel mapping in 60Hz, so I need to dig a bit deeper:

(5)
I activate output profiles.
(6)
I save the current output settings as output profile 0.
(7)
I change the "VSync Front" timing again for 60Hz 1:1 pixel mapping and save the whole configuration as output profile 1. This all worked without ever having to leave the output menu tree.
(8)
Now for every input and every input format I can simply choose whether I want to use output profile 0 or 1, which the iScan automatically remembers.
Done. Very quick, straightforward and easy to understand. Now if I wanted to do this with the HDQ, I need to do this:

1)
I go into the output tree.
(2)
I change vres to 768p.
(3)
I change hres to 1366. The HDQ changes that to 1367. I enter 1366 again. The HDQ changes that to 1367.
(4)
I go into custom timings and manually enter 1366 there. This time it works.
(5)
I need to change some more custom timing settings to achieve 1:1 pixel mapping. Doing so is more time consuming compared to the iScan for me. I have to always enter full numbers. Can't just increase a timing parameter by one. Furthermore I always have to "execute" before I can see the result of my timing changes. Furthermore I don't know how to activate and switch test patterns while I'm in the custom timings edit screen. Is that possible at all? Don't know.

But finally, after some minutes, output profile 0 (50Hz output) done. Now I need to setup output profile 1:

(6)
I need to go into the input menu tree and change there to output profile 1. Of course first I must have understood that I need to do this because it's the only way to be able to edit output profile 1.
(7)
I go back into the output profile tree.
(8)
I change vres to 768p.
(9)
I change hres to 1366. The HDQ changes that to 1367. I enter 1366 again. The HDQ changes that to 1367.
(10)
I go into custom timings and manually enter 1366 there. This time it works.
(11)
I need to change some more custom timing settings to achieve 1:1 pixel mapping.
(12)
Now I need to lookup the manual and read the section about MEM A, MEM B, MEM C and MEM D three or four times, and try to understand how auto switching for NTSC/PAL sources works. Which memory bank is active when PAL input sources are active and when NTSC input sources are active? Furthermore I need to understand how to link the two output profiles I created above to the different memory banks. The remote only has buttons for MEM A and MEM B, but not for MEM C and MEM D. So how can I edit MEM C and MEM D? And what areas of the setup tree do the memory banks effect? The output profiles or the input settings or both? And what happens if my HTPC goes with 50p into the Lumagen? Is that the same as 50i for PAL/NTSC auto switching? What happens if I go in with 24p from a Sony Blu-Ray player? Which memory bank is used for that? Can I tell the Lumagen to automatically use output profile 2 for 24p output when I input 24p sources? Will banks MEM E and MEM F be used then or how is this supposed to work?

Finally, I must remember to save everything, or else things will be lost. This is unlike any other CE device I've ever used.
These are just some examples of how in my experience the iScan was much easier to setup compared to the HDQ. The iScan logic of how which profiles are used and activated is straightforward and easy to understand and is fit for 24p input etc. The Lumagen method with the different memory banks is utterly confusing to me and I still don't see how it can handle 24p inputs "properly"...

Madshi:

I am familiar with some, but by no means all, other video processors as a means for judging the user innterface. As it relates to your second point perhaps it is a case of the Lumagen offering greater flexibility and hence more items to change/tweak....

Just my $0.02

Madshi: In your example on setting output res of 1366 x 768 I am at a loss as to why you think the scaler creates a 1367 res automatically. The default is 1366 x 768..... As to how it can hande 24p inputs properly....that functionality isnt implimented yet, so we have to wait and see.

Both devices are equally unintuative for the layman to set up, you prefer one I have no preference although having done so many Lumagens I do find it easy to get around. Even those devices with wonderful slick on screen graphics are not intuative to set up. I've set up ALOT of video processors. I've sat and read through all their manuals. None are simple.

Once they are all set up and configured it is irrelevant for the most part. All that matters is the image quality and usebility after set up. A SETUP guide for end users with an explanation of how the menu systems work in the device is definately worthwhile, as are help menu's. I also think that removing some of the advanced features from the user menu's might be worthwhile. In the UK I''ve taken the step of making it a condition of being a Radiance dealer that they MUST instal the product and send me the config files so I can check it's done properly. This isn't because I'm assuming it's going to be really complicated to set up but because I want to ensure that anyone in UK who buys one gets the best performance for their investment. I've visited folk all over UK who have not just Lumagen products, DVDO, PMS, Vigatec, Faroudja, Key Digital, Extron, Focus Enhancements, TAW and a couple of others whose names escape me......more often than not the end user has not managed to work out how to set up their product optimally and they are not realising the full potential of the device. It is the nature of the beast.

I think that Jim and probably all the other manufacturers of these devices, realise that as they become more powerfull in what they can do and the set up options they can have, they become harder to make simple to set up. Anything that any of them can do to help is good and I would imagine that Lumagen realise this as much as anyone else.

User interface and useability interfacing is a whole other industry. I remember going to a course in 98 at CEDIA all about it. An eye opener.

Gordon

madshi,

The GUI is definitely nicer on the iScan units, no doubt about it.

On the other hand, the Lumagen provides more functionality. This is a tradeoff you can't really work around. If you want 100 different options, the GUI has to be more complex.

I personally prefer complexity for features. Stuff like the A-D memories, manual assignments of output profiles, how to customize aspect ratios, etc.

It definitely takes more time on the Lumagen processors (because you have to set more stuff up), but I can get it to 100% of what I want. With the iScan features, there are still some loose ends.

So, there are benefits to both approaches. I think Lumagen shouldn't necessarily make things "simpler" (not when it comes at the expense of less features). Alternatively, a simple vs. Advanced mode might just be what the doctor ordered.

Madshi: In your example on setting output res of 1366 x 768 I am at a loss as to why you think the scaler creates a 1367 res automatically. The default is 1366 x 768.....
It's just some days ago where this happened. I began with factory default. Then I changed OUT->RES->VRES to 768. I'm not sure what HRES was set to by the HDQ after that, I think it was 1280 or something like that. Anyway, I changed OUT->RES->HRES to 1366, but the HDQ kept changing HRES to 1367. I was able to fix it by going into OUT->RES->TIMING. It was really not a problem, but it got me by surprise.

Both devices are equally unintuative for the layman to set up, you prefer one I have no preference although having done so many Lumagens I do find it easy to get around. Even those devices with wonderful slick on screen graphics are not intuative to set up. I've set up ALOT of video processors. I've sat and read through all their manuals. None are simple.
Maybe, but I can clearly differ between the VPs by rating how much reading I have to do in the manual. I've had to read quite a lot in the HDQ manual. I only needed a quick glance at the iScan manual. And I'm sure I wouldn't ever need to look into the Crystalio II manual.

Once they are all set up and configured it is irrelevant for the most part. All that matters is the image quality and usebility after set up.
I have to agree with that.

A SETUP guide for end users with an explanation of how the menu systems work in the device is definately worthwhile, as are help menu's. I also think that removing some of the advanced features from the user menu's might be worthwhile.
Generally I'm fine with having many options. The thing I had the most trouble with was the NTSC/PAL memory bank concept used by the HDQ combined with the fact that specific settings can only be edited if you "make them active" first. And that I never knew exactly which settings were valid globally or per input or per refresh rate or per input format or whatever. Apart from these things I don't really dislike the Lumagen menu logic, but I think it could be made a bit more user friendly by making some small changes (e.g. see the suggestions I made earlier in this thread).

I've visited folk all over UK who have not just Lumagen products, DVDO, PMS, Vigatec, Faroudja, Key Digital, Extron, Focus Enhancements, TAW and a couple of others whose names escape me......more often than not the end user has not managed to work out how to set up their product optimally and they are not realising the full potential of the device. It is the nature of the beast.

I think that Jim and probably all the other manufacturers of these devices, realise that as they become more powerfull in what they can do and the set up options they can have, they become harder to make simple to set up. Anything that any of them can do to help is good and I would imagine that Lumagen realise this as much as anyone else.
I'm just thinking, maybe it would make sense to offer a PC software to configure the VP and then download the final configuration into the VP? With a PC software you could make things easier to understand, you could easily offer context sensitive help etc.

On the other hand, the Lumagen provides more functionality. This is a tradeoff you can't really work around. If you want 100 different options, the GUI has to be more complex.

I personally prefer complexity for features. Stuff like the A-D memories, manual assignments of output profiles, how to customize aspect ratios, etc.

It definitely takes more time on the Lumagen processors (because you have to set more stuff up), but I can get it to 100% of what I want. With the iScan features, there are still some loose ends.
I agree with you in that I don't mind added complexity if it buys me worthwhile features. But I think that more complexity doesn't necessarily have to mean that the whole setup concept must automatically be harder to understand.

What I'm looking/asking for is not reduced complexity, but a logical, intuitive and self explaining setup concept. And this is where IMHO the HDQ could use some improvement. I do notice, though, that I may be alone in my view here. So maybe we should move on to more pleasent topics.

Did I mention that I do love the modular and forward looking hardware design of the Radiance? :)

I assume the Radiance will be able to convert 1080P24 to 1080P24sf??

I like to think I am relatively clued up on video processors but I had to read the manual to set up the Crystalio2.....having done a few now I do seem to have a handle on it as they say. It's not intuative though.....at least not to me.

Gordon

I like to think I am relatively clued up on video processors but I had to read the manual to set up the Crystalio2.....having done a few now I do seem to have a handle on it as they say. It's not intuative though.....at least not to me.
Ok, I have to admit I haven't really setup a CII myself yet. So maybe I was a bit over enthusiastic about their OSD... :o

I like to think I am relatively clued up on video processors but I had to read the manual to set up the Crystalio2.....having done a few now I do seem to have a handle on it as they say. It's not intuative though.....at least not to me.
I'm a VP newbie. I bought the Vantage-HD and then the Crystalio 2 and I have yet to read either manual. Both are so incredibly intuitive that I could figure out everything simply through playing with the OSD.

What did you need the C2 manual for?

Merry Christmas: I needed to read it to find out why it was selecting the wrong video level on HDMI input and to work out how dynamic VXP worked....and in the another case why Dynamic VXP wasn't working as I assumed it would. I guess if I'd just mucked around with the menu systems I'd have got there eventually....but reading the manual seemed a more logical and faster approach.

The BIG clue that these things need manuals is that they are provided in the box.....no one wants to waste money producing unneccessary documentation.

Merry Christmas everyone!

Gordon

How about does it output 2 different resolutions to two different displays? How many HDMI out and can they be individually configured?

@jrp
One more question about the inputs and outputs:

You suggest to use external interfaces for SDI Input or Analog Output.

Those interfaces I know that make a SDI-HDMI-Conversion or a HDMI-YPbPR-Conversion are at least about 500 $ each. So that's a 1.000 € additional for the Radiance. Do You plan an own solution from Lumagen or do You prefer 3rdParty solutions ?

Im sorry if this has already been discussed but what will the tradein value be for those like me with the HDQ? also will there be an improvement in the image on my marantz 11-s1 which already uses the gennum VXP 9351?

Already asked and answered but here goes....trade-in will be 1/3rd of pricing....chip will be the newer gennum offering (i.e. better/newer than 9351)...

Dr. Kimble. From what I know, Lumagen will manufacture (or have made), its own SD-SDI to HDMI transcoder. The existing one on the market that I know of is by Algolith however that won't pass blacker than black.

Manufacturing a HDMI (HDCP protected) to analog output box would violate the terms of the HDMI license. The HDMI folks go after even the offshore boys who have built stand alone boxes. It seems to be OK for them to build projector specific HDMI input cards though. These are more or less available for the most popular CRT projectors. I doubt Lumagen would be manufacturing input cards for CRT projectors.

@Mark
Perhaps ther was an misunderstanding on question 2:

My intention is NOT to copy HDCP-protected digitals sources on an analog ouput !! I know and understand the problems and if my idea is to sopy a HDMI signal, I certainly wouldn't do it that way.

What I would like to do is to have the possibilty to record a non copy protected signal to an analog display or recorder.

The Crystalio II e.g. can manage this.

And a HDMI/DVI to analog converter (e.g from cypress) is about 500 $

Coming to my question one, ther is e.g. a SDI to DVi Converter from ramelectronics for about 1.000 $.

And if You need this features, it makes a difference to buy 2 external boxes for 1.500 $ or to have them integrated like in the CII.

So my question is still, if those or similar boxes will be offered by Lumagen an waht will be the estimate price.

I'm planning on purchasing the Panny 65" commercial plasma screen next year. After reading the Lumagen posts, it seems that the Radiance might be the best VP for this unit. My question relates to audio time lag due to visual processing in the VP. Currently, I pass the video and audio cable stream through my pre-pro Integra Research RDC 7.0 to my 50" Panny plasma (bought in 2001). I haven't had to fiddle with any time lag problems (other than in the very early HD days of ABC when there was a problem). If the TW/Comcast cable goes directly into the VP, do I then direct the audio stream into the RDC for adjustment (which will be annoying due to its OSD)? I've considered upgrading the pre/pro to the RDC 7.1, but hate the negative customer service of IR so much I'd rather not spend another dime there! However, the 7.1 will have an HDMI 1.3 card next year (allegedly - but one can't count on it). Can you advise how I can handle any possible timing problems?

I'm definitely interested on getting my hands on a Lumagen XD for beta testing.

Speaking of which, has anyone noticed that the Dec. 11 press release on Lumagen's website now is reverting to the XD nomenclature, and it makes no specific mention of either Gennum or HQV processing? At the bottom the legal disclaimers mention both, but other than that Lumagen doesn't specify which processor they'll be using - only that they'll be showing a Radience XD at CES.

Am I completely clueless or is there a reason for these recent changes?

I'm definitely interested on getting my hands on a Lumagen XD for beta testing.

Speaking of which, has anyone noticed that the Dec. 11 press release on Lumagen's website now is reverting to the XD nomenclature, and it makes no specific mention of either Gennum or HQV processing? At the bottom the legal disclaimers mention both, but other than that Lumagen doesn't specify which processor they'll be using - only that they'll be showing a Radience XD at CES.

Am I completely clueless or is there a reason for these recent changes?

HogPilot,

Check page 4 of this thread.

http://www.avsforum.com/avs-vb/showthread.php?p=9132581&&#post9132581

In short, Lumagen is creating a scaler that can have a modular deinterlacing system and you will later be able to upgrade or replace the deinterlacing daughterboard to one that might better suit your needs, tastes or purposes.

I haven't found any mention of it here or in any other thread: did Lumagen in fact demo the RadianceXD at CES as promised?

If so, and if anyone here saw it, surely there must be something worth reporting.

Apparently it was there, but demoed only to insiders.
Gordon mentioned taking photos, but nothings appeared yet.

The pics from CES were pretty crappy. I took a handfull of my unit on my Dining room table yesterday. Will upload them and link later today but my image host site is down just now.

Jim has alot of great ideas for this product. Exciting times ahead.

Gordon

Images here:
http://www.pbase.com/gmfism/image/73545679

http://www.pbase.com/gmfism/image/73545677

http://www.pbase.com/gmfism/image/73545678

Do not like the silver at all...hope that this pup comes in black...

Jim,

You have a huge following here. You have been very responsive in the past. An update of the Luninance XG would be greatly appreciated. Thanks.

Tim

I'm planning on purchasing the Panny 65" commercial plasma screen next year. After reading the Lumagen posts, it seems that the Radiance might be the best VP for this unit. My question relates to audio time lag due to visual processing in the VP. Currently, I pass the video and audio cable stream through my pre-pro Integra Research RDC 7.0 to my 50" Panny plasma (bought in 2001). I haven't had to fiddle with any time lag problems (other than in the very early HD days of ABC when there was a problem). If the TW/Comcast cable goes directly into the VP, do I then direct the audio stream into the RDC for adjustment (which will be annoying due to its OSD)? I've considered upgrading the pre/pro to the RDC 7.1, but hate the negative customer service of IR so much I'd rather not spend another dime there! However, the 7.1 will have an HDMI 1.3 card next year (allegedly - but one can't count on it). Can you advise how I can handle any possible timing problems?


Anyone care to answer this question? I am interested to the answer to this. You have people here discounting the benefits of a standalone processor in favor of an integrated solution like the Anthem D2. Is there a clean solution to the timing issue?

Gordon,

Thanks!!!

VERY VERY VERY NICE DESIGN, IMO!!!

I hope they have it both in black and in silver (I'll take the silver one, of course).

WOW!

No readout on the front?! That's a pain as my video processor (Lumagen ProHD) is in a separate room and I often need to switch it on first and confirm the current input (eg if I plan to use the Toshiba HD-XA1 and want the EDID handshake to work correctly)...

The overall design is very nice though...

Mark

What the heck, half of my equipment is black, half silver, so how about two tone? :p

Are there any lights at all on the unit to indicate power or IR receipt?

- Rich

Jim and/or Pat (from Lumagen):

A thought -- have you considered building an "overlay input/function" which I would describe as follows:

-- A composite or s-video input from the pre/pro which would recieve OSD information from the pre/pro

-- Pre/pro OSD information received by the foregoing input would then be overlayed onto the output regardless of the format (i.e. componet, HDMI, etc.) or resolution (i.e. 720p, 1080i, 1080p) etc.

The above would be a terrific and VERY USEFUL function, particularly when it comes to calibrating items such as LipSync.

Over to you...

The RadianceXD will have PiP. You could select the Pre/pro OSD overlay as the PiP input to display this OSD info.

I assume the Radiance will be able to convert 1080P24 to 1080P24sf??

Yes.

Also, with the latest release, the Vision series will do this.

@jrp
One more question about the inputs and outputs:

You suggest to use external interfaces for SDI Input or Analog Output.

Those interfaces I know that make a SDI-HDMI-Conversion or a HDMI-YPbPR-Conversion are at least about 500 $ each. So that's a 1.000 € additional for the Radiance. Do You plan an own solution from Lumagen or do You prefer 3rdParty solutions ?

We plan a SD SDII to HDMI box. At this time we are not planning a HD-SDI to HDMI box. We also plan and external HDMI to analog DAC box (but it won't strip the HDCP encryption).

Yes these do add additional cost, but we feel that is is best not to burden each box with this extra cost, when most people don't need these features.

How about does it output 2 different resolutions to two different displays? How many HDMI out and can they be individually configured?

The RadianceXD has two HDMI outputs.

They can be on at the same time with exactly the same video timing.

If you had two displays of different resolutions you would only have one output on at any given time.,

Im sorry if this has already been discussed but what will the tradein value be for those like me with the HDQ? also will there be an improvement in the image on my marantz 11-s1 which already uses the gennum VXP 9351?

Trade-in for older Lumagen products toward the RadianceXD is 33% of the retail of the Lumagen being traded-in as credit toward the RadianceXD.

----

The RadianceXD uses the GF9450, which in my opinion is much better than the GF9350. I choose not to use the GF9350, but Gennum addressed the issues I had with it in the GF9450.

Also, the RadianceXD will use the best of the Gennum features in addition to the best of the Lumagen features. For example, we will use the Lumagen no-ring scaling, which I believe is the best out there.

I'm planning on purchasing the Panny 65" commercial plasma screen next year. After reading the Lumagen posts, it seems that the Radiance might be the best VP for this unit. My question relates to audio time lag due to visual processing in the VP. Currently, I pass the video and audio cable stream through my pre-pro Integra Research RDC ...

The RadianceXD, using the Gennum, will have approximately 1.25 fields of delay for video and film, when in genlock mode. This is about 20 mS in the US and 25mS in PAL countries. By comparison, current products using the SiI504 and other current products, such as DVDO, have between 55 and 80 mS in US and 70 and 100 in PAL countries, of video delay. The Silicon Optix is about 150 mS.

Most digital displays add one or two more fields of delay.

Generally, from feedback from many people I have talked to, most people do not notice a lip-sync error of 50 mS but do notice the lip-sync at 75 mS.

So, even without running audio through the RaidanceXD, and using a display with two fields additional delay (the most I have encountered driving at the displays native resolution as we do), the RadianceXD will not have visible lip-sync error. For all other processors mentioned, lip-sync can be noticed unless the audio is delayed to match (which can be done by a good quality receiver or amp).

The RadianceXD can do the processing. but will not be HDMI 1.3, so if you want the HDMI 1.3 audio, you can run the HDMI to the receiver first, then to the RaidanceXD.

Hope this answers your question.

I'm definitely interested on getting my hands on a Lumagen XD for beta testing.

Speaking of which, has anyone noticed that the Dec. 11 press release on Lumagen's website now is reverting to the XD nomenclature, and it makes no specific mention of either Gennum or HQV processing? At the bottom the legal disclaimers mention both, but other than that Lumagen doesn't specify which processor they'll be using - only that they'll be showing a Radience XD at CES.

Am I completely clueless or is there a reason for these recent changes?

The RadianceXD will use the Gennum GF9450.

We went back to the RadianceXD (and RadianceXS) nomenclature because we felt it was appropriate to denote the number of outputs in the name rather than the processor used. The "D" is for dual output, and the "S" is for single output. Of course we will need to add to this as we intro more products in the line, but we think it's a good starting naming convention.

Do not like the silver at all...hope that this pup comes in black...

We hoped that having the front be both black and silver (really a light-gray) would satisfy both camps. Apparently at least one decenting vote.

Not impossible to strip the silver etch and redo in black, but no plans at this time. Would have to be an extra cost option as we are buying the front panels in the colors shown.

Always open to hearing how many votes there are for an all black front plate option.

Jim,

You have a huge following here. You have been very responsive in the past. An update of the Luninance XG would be greatly appreciated. Thanks.

Tim

We are hard at work on it ... not enough? ;)

We had a hit of five weeks due to a major error in the spec of one of the chips we are using in the design. We had to spin one of the boards. The company admitted their error, but it is still costing us a lot of time. This is holding up software development until we get the assembled boards back in about two weeks.

I am hoping to ship the very first Beta unit at the end of February, with more shipments at the end of March. Of course we have a lot of work of our own to do so I can't say for sure we will make this.

As with the Vision series Beta, we think the Beta units will already have great processing of video, but will lack a number of features, and there will be some user interface foibles.

We will be improving the user menus and interface. Not sure if this will be in early Beta units. There will also be context-sensitive help screens, but these won't appear for some time.

At this point we are still working to decide what user interface changes we actually make. The structure will remain similar, but we might change the way the input memories interact with the output configurations.

One change is that we plan on the four memories each having sub memories for NTSC (60 Hertz) and PAL (50 Hertz) inputs. So our users in Europe will have four memories, instead of having to pair memory A and C (or B and D) for NTSC/PAL.

What the heck, half of my equipment is black, half silver, so how about two tone? :p

Are there any lights at all on the unit to indicate power or IR receipt?

- Rich

The RadianceXD is really two-tone. Silver (again really light gray) in themiddle with black end-caps and case.

There will be a two-color LED. Red will indicate off. When popwer is on, we will have the default as a green "on" indicator (default on the Vision series is no-light except on command receipt).

The RadianceXD will have PiP. You could select the Pre/pro OSD overlay as the PiP input to display this OSD info.

Jim:

1. Specifically wrt the above quote how this work in that a) could the PiP input be setup to automatically display when it received information or would one have to manually engage the PiP feature meaning that it would operate as a "poor man's OSD" and b) would it not be the case that the PIP would function more as a "small window" or "separate window" than an overlay? Not that this is a bad thing but I just want to understand how it works.

2. Noting that I am the decenting vote on the color I should also add that I would pay a premium, albeit a small one, to get a unit inblack as everything else, other than mu STB, is black.

3. Dual outputs...I undersatnd that they can be implemented to drive two different devices BUT because they be used to drive the video disaply on one unit while passing the audio into the pre/pro...in other words, does the output HDMI signal still contain the audio information or has it been stripped out with only the "processed" video being output.

4. And finally, on behalf of all the other users here many thanks for taking the time to answer all our questions...as I am sure you can appreciate, we are anxious to receive our beta units.

As always, much thanks.

2. Noting that I am the decenting vote on the color I should also add that I would pay a premium, albeit a small one, to get a unit inblack as everything else, other than mu STB, is black.

I also prefer black and would have no issue buying a user-installable replacement faceplate (similar to DVDO offering an optional silver front face).

We will be improving the user menus and interface. Not sure if this will be in early Beta units. There will also be context-sensitive help screens, but these won't appear for some time.

At this point we are still working to decide what user interface changes we actually make. The structure will remain similar, but we might change the way the input memories interact with the output configurations.
Thank you, that sounds good.

One change is that we plan on the four memories each having sub memories for NTSC (60 Hertz) and PAL (50 Hertz) inputs. So our users in Europe will have four memories, instead of having to pair memory A and C (or B and D) for NTSC/PAL.
Hmmmm... Just some thoughts: My receiver does both 50Hz SD and 50Hz HD, depending on which channel I'm watching. I'm sure I'd want to use different noise reduction, detail enhancement etc settings depending on whether I'm watching a SD or HD channel. Have you taken that into consideration? Also what about 24Hz sources? Shouldn't they have their own sub memory, too?

Have you thought about splitting specific setting areas off into profiles? E.g. just as an example you could bundle all noise reduction, detail enhancement, sharpening etc settings into "image adjustment profiles". Personally, I'd then make one profile for SD broadcasts, one for HD broadcasts, one for SD DVDs and one for HD DVDs. Then there'd need to be a way to assign such profiles to specific circumstances, e.g. a specific input resolution and frame rate on a specific input port. Does that all make sense to you? Or would it differ too much from your planned setup structure?

We hoped that having the front be both black and silver (really a light-gray) would satisfy both camps. Apparently at least one decenting vote.

Not impossible to strip the silver etch and redo in black, but no plans at this time. Would have to be an extra cost option as we are buying the front panels in the colors shown.

Always open to hearing how many votes there are for an all black front plate option.

I'm another vote for an all black front plate... :)

Jim,

The Crystallio II's with their older Gennum chip continue to have MAJOR problems with the TIVO Series 3 HD recorder. The image drops frequently into banding and color distortion. After many months, PMS has still been unable to resolve the problem so far.

Will you have access to a TIVO S3 for your alpha testing of the new Gennum chip?

Images here:
http://www.pbase.com/gmfism/image/73545679

http://www.pbase.com/gmfism/image/73545677

http://www.pbase.com/gmfism/image/73545678

I just got the latest issue of The Perfect Vision that has a blub on the Radiance. It looks much darker gray their. Which one is more accurate?

- Rich

My images were taken in a bright conservatory with a bright white backgrond. In a darker area less well lit it does look more midtoned.

Gordon

Edit - please disregard. Sorry.

Jim,

Any plans for a PRO version with a front LCD panel?

And any plans for a LIGHT version with a price in the 1500-2500$ ?

I think that a version without analog output and with only 2 or 3 hdmi input will be sufficient for many of us...

Richard.

Jim has stated that it could be say 6 months or more for it to morph out of beta and geared up production and you're asking him for future plans for cheaper and more expensive versions?

Jim has stated that it could be say 6 months or more for it to morph out of beta and geared up production and your asking him for future plans for cheaper and more expensive versions?

Time is not important, if I remember, Radiance has been announced in 2005, no ? ;)
So a cheaper version can be announced now for a 2008 release date, a 2009 beta version and a 2010 production one. :D

Richard.

What is the cost diff bertween the XD & XG?

the bland, it is my understanding that the XD and XD are one in the same..that is, when Lumagen announced the XD (X=extreme, D=dual output) it was to be based on the Realta chip...when Lumagen decided to move form the Realat chip to the Gennum chip XD became XG (G=Gennum)..and then, after some consideration, it was decided to return to the original XD nimenclature...

as far as the reason for the change see jrp's quote above in which he states that Gennum addressed all of his concerns wrt the 9350 chip in the newer 9450 chip which Lumagen will be using...

Joel,
Good to know.

I think a dual output would come in handy for me. I only wish you could output 2 different resolutions from each HDMI output.


IT appears not possible per Jim.

Jim,

"If you had two displays of different resolutions you would only have one output on at any given time.,"

Does that mean your earlier plans for a lower resolution output for audio equipment that can't handle 1080p is no longer in the works?

Shawn

rlemesle,

What was the point of your post exactly?

rlemesle,

What was the point of your post exactly?

I just wanted to say that the sooner a cheaper version is announced, the sooner we'll see it. Even if Radiance is only beta "now", a cheaper (or more expensive) version can already be announced.

Concerning dates, perhaps french humour isn't understood here...

Richard.

Concerning dates, perhaps french humour isn't understood here...

Richard.

It is..Mark H. is just a grumpy, old man with an analog system... ;)

I am not sure, but I think The Perfect Vision blurb did not mention the option to upgrade to HDMI 1.3. Is that still a feature of the XG?

- Rich

Hmmmmmmmmmmmmmm 6 months :rolleyes: :rolleyes:

Is there currently a target release date? I was previously expecting March time-frame.

Thanks

Lore. I will soon have a digital input for my FP CRT. Plus I will shortly have a DLA-RS1 to play with. I won't keep it long.

Now JP has stated that his first batch of beta will be 5 machines. All to beta testers. Then later another batch of beta's, say 20. I think he stated that a morph from beta to final by perhaps Cedia. Obviously a lot depends on the beta testing. On of the nice things about Lumagen's is that the company continually strives to improve their product thru software downloads. New features continually appear without charge and discovered bugs are quickly remedied. You can pick up the phone and talk to the guy who owns the company and the guys who write the programming. No one gets blown off.

Post 213:
Jim Peterson writes.....
"I am hoping to ship the very first Beta unit at the end of February, with more shipments at the end of March. Of course we have a lot of work of our own to do so I can't say for sure we will make this.

As with the Vision series Beta, we think the Beta units will already have great processing of video, but will lack a number of features, and there will be some user interface foibles."

I believe that Marks numbers are no longer accurate.

Gordon

Jim:

1. Specifically wrt the above quote how this work in that a) could the PiP input be setup to automatically display when it received information or would one have to manually engage the PiP feature ...

2. Noting that I am the decenting vote on the color I should also add that I would pay a premium, albeit a small one, to get a unit inblack as everything else, other than mu STB, is black.

3. Dual outputs...I undersatnd that they can be implemented to drive two different devices BUT because they be used to drive the video disaply on one unit while passing the audio into the pre/pro...in other words, does the output HDMI signal still contain the audio information or has it been stripped out with only the "processed" video being output.

1. We are not planning an "automatic" detection for the case wherre a amplifier OSD is input to the RadianceXD. It would be manually enabled by the user using the PiP buttons as with any other PiP source.

2. I will look into having a all-black face plate option. Not too difficult, but it is one more thing to get done.

3. The main HDMI output would have both audio and video. The second one, (when told to drop the clock rate for a receiver that can't take, say 1080p) would output blank video at 720p plus audio.

1. We are not planning an "automatic" detection for the case wherre a amplifier OSD is input to the RaidnaceXD. It would be manually enabled by the user using the PiP buttons as with any other PiP source.

I don't know if this is really a bid deal. All of this can be accomplished with remote control macros.

1. My receiver does both 50Hz SD and 50Hz HD, depending on which channel I'm watching. I'm sure I'd want to use different noise reduction, detail enhancement etc settings depending on whether I'm watching a SD or HD channel. Have you taken that into consideration? Also what about 24Hz sources? Shouldn't they have their own sub memory, too?

2. Have you thought about splitting specific setting areas off into profiles? E.g. just as an example you could bundle all noise reduction, detail enhancement, sharpening etc settings into "image adjustment profiles". Personally, I'd then make one profile for SD broadcasts, one for HD broadcasts, one for SD DVDs and one for HD DVDs. Then there'd need to be a way to assign such profiles to specific circumstances, e.g. a specific input resolution and frame rate on a specific input port. Does that all make sense to you? Or would it differ too much from your planned setup structure?

1. We already have sub-memories for each input resolution in the Vision series. When we added 1080p24, it was assigned to the "other" sub-memory. So you can have indipendent setups for 480i/p (or 576i/p), 720p, 1080i, and 1080p24/30.

This will be continued in the RadianceXD. So you will be able to have noise reduction, etc. settings based on input resolution on a per-memory basis.

2. This is similar to how we work multiple output resolutions. There are eight output configurations (i.e. profiles). So I understand your idea. The downside is that this is our biggest support nightmare as it seems most people do not understand the concept of "indirection" (as used by programmers) in building a profile/configuration and then selecting it as needed. In addition, most users like to tweak each input and input resolution, so there is not really a common profile for the various inputs.

So I think the way we have it now will be the way we will go. That is, you setup one input and then, if desired, copy all of it's parameters to another input(s) and make tweaks as desired.

1. We are not planning an "automatic" detection for the case wherre a amplifier OSD is input to the RaidnaceXD. It would be manually enabled by the user using the PiP buttons as with any other PiP source.

2. I will look into having a all-black face plate option. Not too difficult, but it is one more thing to get done.

3. The main HDMI output would have both audio and video. The second one, (when told to drop the clock rate for a receiver that can't take, say 1080p) would output blank video at 720p plus audio.

jrp, many thanks for the feedback..much appreciated...one follow up, as I assume not many of us will be sending the output to our pre/pros before being passed along to our display devices I am guessing that most will use the first output to drive video and the second output to drive audio is this assumption the rationale why the second output will output "blank video at 620 plus audio"?


I don't know if this is really a bid deal. All of this can be accomplished with remote control macros.

Yes, but not nearly as elagantly...

Jim,

The Crystallio II's with their older Gennum chip continue to have MAJOR problems with the TIVO Series 3 HD recorder. The image drops frequently into banding and color distortion. After many months, PMS has still been unable to resolve the problem so far.

Will you have access to a TIVO S3 for your alpha testing of the new Gennum chip?

Is this the new HD 20 (might have the model number off a bit) HD TiVO recoder from Direct TV?

If so we should be able to get one for our testing. If not, please let me know the specifics of this HD TiVo as to make and model.

BTW: We have an FPGA that can process the video before and after it runs through the Gennum. This, I believe, is a huge advantage for us not shared by the competition. We develop our own algorithms on top of the video chip we use and so can fix problems, and add value in other ways. It remains to be seen what this problem is caused by, whether the GF9450 fixes the issue verses the GF9350, or if it is not a Gennum issue at all.

Is this the new HD 20 (might have the model number off a bit) HD TiVO recoder from Direct TV?

If so we should be able to get one for our testing. If not, please let me know the specifics of this HD TiVo as to make and model.


The TiVo series 3 is the new OTA/CableCard HD DVR. I have mine working fine with a Lumagen HDQ.

- Rich

Any plans for a PRO version with a front LCD panel?

We are looking at doing a Radiance product with both a front panel video display and with a LAN interface. Might even have a couple processing improvements as well. No schedule/price/feature-list at this point as we are concentrating on the RadianceXD.

We also are looking at a lower priced unit with fewer IO, and maybe not including audio (except HDMI would still pass audio of course). Since the video processing delay is much shorter than other video-processor units on the market, it is not really needed for the Raidnace. Also, most people using this level of processor have audio delay in their audio processor. No schedule/price/feature-list at this point as we are concentrating on the RadianceXD.

jrp, many thanks for the feedback..much appreciated...one follow up, as I assume not many of us will be sending the output to our pre/pros before being passed along to our display devices I am guessing that most will use the first output to drive video and the second output to drive audio is this assumption the rationale why the second output will output "blank video at 720p plus audio"?...

The second output can be a copy of the first (video and audio), or blank video at 720p with audio.

The first output could be sent through your Pre/Pro, but there are Pre/Pro's that will not accept 1080p, so we added the second output option for audio only to accomodate these.

The TiVo series 3 is the new OTA/CableCard HD DVR. I have mine working fine with a Lumagen HDQ.

- Rich

If it works with the HDQ, I think it will work well with the RadianceXD, but we would need to test it to be sure.

We'll try to test this unit with the RadianceXD as soon as we can.

The downside is that this is our biggest support nightmare as it seems most people do not understand the concept of "indirection" (as used by programmers) in building a profile/configuration and then selecting it as needed. Jim,

I love your products, but...I've been writing device drivers and embedded software for over 20 years and I must say I find the documentation to be inscrutable w.r.t. these settings. The addition to the documentation of some common examples with diagrams would be a big help: steps performed and their effects in firmware. Doesn't have to be anything fancy; boxes and arrows would probably do it. Also, as mentioned by another poster, documenting whether a setting has global or local scope would be a big help.

Just my $0.02.

2. This is similar to how we work multiple output resolutions. There are eight output configurations (i.e. profiles). So I understand your idea. The downside is that this is our biggest support nightmare as it seems most people do not understand the concept of "indirection" (as used by programmers) in building a profile/configuration and then selecting it as needed.
I understand your concern. But I think much depends on how intuitive things are to use. If things are hard to grasp without reading the manual, users will not understand the concept. But if you can manage to make things intuitive/self-explaining, I think users will understand such a concept with ease.

E.g. here's how I could imagine how you could setup the factory default:

Input ->
- ...
- Image Adjustment Profile ->
- SD broadcasting
- SD DVDs
- HD broadcasting
- HD DVDs / Blu-Rays
- ...
Output ->
- ...
Image Adjustment Profiles ->
- create new profile
- Profiles ->
- SD broadcasting ->
- SD DVDs ->
- analog noise reduction -> 2 (0..10)
- mosquito noise reduction -> 2 (0..10)
- block noise reduction -> 2 (0..10)
- texture enhancement -> 0 (0..10)
- edge enhancement -> 0 (0..10)
--------
- rename this profile
- delete this profile
- copy/duplicate this profile
- HD broadcasting ->
- HD DVDs / Blu-Rays ->

Additionally there should be a context help on "Input -> Image Adjustment Profile" which explains that the user can setup/add/rename/change/delete profiles in the menu root under "Image Adjustments Profiles". I think with such a menu structure most if not all people would understand the concept. Important factors for me are that profiles can be named and that all profiles can be seen/changed without having to activate them first.

I'd welcome if you could extend output profiles to work in a similar way. That is: Allow people to give them names. Don't use a fixed number of profiles, but let users add/delete/copy/rename output profiles. And make all existing profiles available for browsing and editing in the menu structure, without having to activate them first. No "hidden" profiles, anymore, please, which are visible only if certain circumstances are met first. IMHO these changes would help a lot to make the OSD more user friendly.

Just my thoughts, though.

P.S: If all profiles (even the ones which are not currently active) are listed in the menu structure, there should be some indication which profile is the currently active one. E.g. you could add "(active)" to the profile name. Or you could change the front style/color.

E.g. here's how I could imagine how you could setup the factory default:

As long as the source types are not hard-coded (ie. I might not be using DVD, Blu-Ray, or broadcast TV).