Just to be perfectly clear: While Lugamen doesn't allow the unit to be sold under MSRP other companies are able to bundle the unit with other items as a value add. Sencore also does not not have exclusive distribution rights. This has all changed. Here's the proof (http://www.curtpalme.com/CalMAN_RadianceXD_kit.shtm).

Kal

I noticed the Calman deal over at curtpalme.com and am going to update the FAQ appropriately. I also didn't know that it was Lumagen that had the MSRP policy (I thought this came from Sencore), so I'll update that part as well. Thanks kal...

No problems Dave. Let me know if you have any more questions.

The RadianceXD is one mean video processor... it needs the respect it deserves. ;)


Kal

Just to be perfectly clear: While Lugamen doesn't allow the unit to be sold under MSRP other companies are able to bundle the unit with other items as a value add. Sencore also does not not have exclusive distribution rights. This has all changed. Here's the proof (http://www.curtpalme.com/CalMAN_RadianceXD_kit.shtm).

Kal

I noticed the Calman deal over at curtpalme.com and am going to update the FAQ appropriately. I also didn't know that it was Lumagen that had the MSRP policy (I thought this came from Sencore), so I'll update that part as well. Thanks kal...

Are you certain. The S in MSRP is supposed to mean suggested ;)

As far as I know, you can still buy them from AVScience and their pricing has always been below MSRP. Hast that changed?

- Rich

I can't comment on AVScience pricing (past or current), but from what I understand their below MSRP pricing was only for a set of pre-release boxes that they had received. I understand that to now be over. My Lumagen dealers have indicated that under no circumstance is anyone allowed to sell under MSRP for final versions of the product. They will be flogged & dragged through the streets if they do so... ;)

If the CalMAN/SpectraCal guys were able to discount at all you can bet they would. Just look at their Hubble pricing (around $3000) vs others (Sencore OTC1000 at around $7000) as an example.

Kal

I can't comment on AVScience pricing (past or current), but from what I understand their below MSRP pricing was only for a set of pre-release boxes that they had received. I understand that to now be over. My Lumagen dealers have indicated that under no circumstance is anyone allowed to sell under MSRP for final versions of the product. They will be flogged & dragged through the streets if they do so... ;)

Kal

Apparently you can comment: p

I suggest that if it aint broke don't fix it. ;)

- Rich

I can't comment on AVScience pricing (past or current), but from what I understand their below MSRP pricing was only for a set of pre-release boxes that they had received.

Correct, the only thing I'll add is that those 'pre-release' boxes were the final hardware, exact same box you will get if you buy one today. Only the software was beta.

In the previous version of the FAQ that you can see above, I suggested people take advantage of the pre-release price; I had no idea how prescient I was, in a way...

In any case, I have updated the pricing question re: Lumagen/Sencore/MSRP pricing, and also mentioned the curtpalme deal. And what a frakking deal that is. I got my own probe, a Chroma 5, from curtpalme.com and I am pretty amazed (in a good way) by certain prices they have going there. (In all fairness I'm an ex-CRT guy so I'm glad Curt has found a way to expand his business.)

Finally; it would be nice if Jason or Alan could comment re:availability of the XD via AVS...

Are you certain. The S in MSRP is supposed to mean suggested ;)

They're very persuasive. :p

Sencore owns distribution rights to the Radiance. Its been renamed and labeled Sencore by Lumagen and the MSRP is now $4495. Sencore is expanding the dealer base established by Lumagen. It will supposedly come down hard on mail order sellers who use to sell the product at $300 to $500 over dealer cost. Obviosly they will tolerate minor discounting say in the range of 10 to 15%. The Calman pro software and eye one sell for about $1150 through Curt's site. Curte doesn't sell them or know anything about them. Its sort of like a paid ad and selling service through his site. He does get a commisiion on those sold through his site. I just purchased the Calman pro software. I purchased it directly from Spectracal but I mentioned I learned about it on Curt's site. They gave me the discount offered through Curt's site and said they would credit Curt with his commision. Now the Lumagen package on Curt's gives you the software and Eye One for free. This equates to a 20% discount on the Radiance assuming you would use the discount to buy the software and Eye One. A free product discount costs the seller a lot less than a pure discount of the same value and of course Curt's site could offer the Eye One and Calman pro for cost without flak from Sencore so in realty you are getting about say a 15% discount out of pocket. Sort of like a calibrator who sold you a Radiance for $4500 and gave you a free calibration.

I did some more digging and Mark is right: All units are distributed by Sencore to resellers who then sell them to end users. Sorry of the confusion.

Regardless, they are not allowing any discounting at all as mentioned earlier. Nada. Zip. Not even $1, and definitely not 10-15%. I'm told that anyone caught doing so will no longer be allowed to sell the Radiance. They are being extremely strict about it. Same way that Sony sells their products really.

Kal

Update Gennum code for 480i/576i/1080i to 24p/48p

We are getting close to a first release for interlaced in to 24p/24sF/48p out for film sources. It should be within about a week.

We discovered an unpleasant fact while doing this work. It turns out there is a "feature" in the Gennum that prevents it from being updated using the normal "4-wire" interface we use for commands, UNLESS a particular signal is pulled high at reset. Seems like a bad idea to me (we can do everything but update the code), but that's the way it works.

Well, randomly we got it right to start, made a design tweak and then, randomly, didn't. So, for some of you there will be a need to open up your unit and plug in a small square-pin header we will provide at no charge (or your dealer/distributor if you are outside the US and Canada).

The way the Gennum update works is, you download the code and our flash ROM is re-written. On next power-on the new software looks to see if the GF9450 code is current. If not, it trys to update it, and if it can't (do to the lack of this resistor) it puts up a pesky little message each time you power on. You can continue to use your system, but the interlaced to 24p/24sF/48p film inverse telecine will not work well. After you install the header, the next time you power on the GF9450 will get updated.

For Serial numbers with last four digits of:
<1000: You should be good fine for the update, unless your board was swapped for some reason.

>=1000 and <1570: You win. You need the header.

>=1570: We have added a resistor to pull up the needed signal. So, unless we missed a unit (which is certainly possible), you should be good to go.

Of course, if you have trouble with a serial number that should allow the update, or trouble after installing the header, email support at Lumagen and we will help you. Note the current software (090208) is not a good test of whether your system will update the GF9450 properly since systems with the issue already have the matching GF9450 code and so no attempt by our software is made to update the GF9450 code.

If you are in the USA or Canada and do not want to install the header on your own (even knowing it's really easy, and knowing that even if you installed it backwards no harm would come to your system), we will install it for you under standard warranty service. That is, you pay shipping an insurance to Lumagen, we pay return shipping and insurance. If you are outside the US, please contact you dealer (who you may need to show the post to), and they can contact their distributor.

If you have a winning number, want to install the header yourself, and are in the USA or Canada, please email support@lumagen.com with your serial number and current address. We will mail the needed header to you.

====

For those of you not using the 24p/24sF/48p inverse telecine, we would still like you to get the header if you have a winning number. This is so everyone has the same level of Gennum code and we don't need to go chase phantom problems caused by old Gennum code. The benefit for you is that there are some minor tweaks for normal output modes (50p, 60p) to improve the quality -- and you can avoid that pesky message we would otherwise put on the screen at power-on telling you to contact our support team.

Randy Freeman
support@lumagen.com

Thanks, Randy. I look forward eagerly to the release (and I have a non-winning serial number so it should be even easier in my case to install the update).

One thing I'm not clear on is how to configure the Radiance for this once it's available. Ideally I want to tell the Radiance to output 24p for film sources and 60p for video sources, but is it going to be possible to have this done automatically?

If manual 24p/60p switching is needed, how easy it is to set it up so that this can be achieved with (say) just a single button push? (I suspect it's easy but I haven't thought it through yet.)

There are two ways to find your Radiance serial number.

1. Look at the label on the bottom of your unit.

2. The serial number is on the first line of the info screen that you can display by pressing the "Ok" button on the remote.

One thing I'm not clear on is how to configure the Radiance for this once it's available. Ideally I want to tell the Radiance to output 24p for film sources and 60p for video sources, but is it going to be possible to have this done automatically?

This is easily done. Each physical input has a number of virtual inputs based on the incoming signal which can be configured for different outputs. All you need to do is make sure you have output configs setup for 24p and 60p, then link them to the appropriate input resolutions and you are good to go. When Radiance senses 60i coming in, it will send 60p out; when it senses 24p coming in, it will send 24p out.

This is easily done. Each physical input has a number of virtual inputs based on the incoming signal which can be configured for different outputs. All you need to do is make sure you have output configs setup for 24p and 60p, then link them to the appropriate input resolutions and you are good to go. When Radiance senses 60i coming in, it will send 60p out; when it senses 24p coming in, it will send 24p out.I already have it set up for different input resolutions in this way (50p out for 50Hz sources, 60p out for 60Hz sources, 24p out for 24Hz sources).

The problem is that in this case we want to choose between 24p and 60p output depending not on the input resolution (1080i, say) but on whether the program was derived from a film source, i.e. whether it has the characteristic 2:3 cadence. If I set up the Radiance to output 24p for all 1080i sources then the result will not likely be correct for a program that originates on 60Hz video (Leno, Letterman, Conan, live sports, news, etc.).

Ideally there will be a way to tell the Radiance to choose 24p or 60p based on whether it detects the 2:3 cadence in the source. If there isn't any way to do it automatically then we'll want to have an easy way to switch manually between 60p and 24p output.

Ron: There will be no ability built in to Radiance to do what you want automatically. The easy way to do it is just use one of the three other memories, A,B,C,D to create the same as you have but with different output res for your 1080i60/p24 source

Ron: There will be no ability built in to Radiance to do what you want automatically. The easy way to do it is just use one of the three other memories, A,B,C,D to create the same as you have but with different output res for your 1080i60/p24 sourceHmm. I'm already using two memories (A and B) to provide a one-button switch between multi-channel audio (viewing on the living room projector) and two-channel audio (viewing on a remote plasma display). It sounds like I'll have to use both of the remaining memories to get a one-button switch to 1080p24 output for both situations.

Is there no simple button sequence that will just toggle between 1080p/60 and 1080p/24 with no other changes, and without having to follow the steps on the menu display?

Is there no simple button sequence that will just toggle between 1080p/60 and 1080p/24 with no other changes, and without having to follow the steps on the menu display?

Yes there is.. and Gordon highlighted it... for you, Buttons C + D will simply toggle you between 1080P24 and 1080P60.

If you need more sophistication (or memory)... do what I do, get two Blu Ray players - one for 1080P60, the other for 1080P24. This way you are covered and you use a second physical input for each (so, potentially, 8 full memories between them).

Or (cheaper!) simply use two virtual HDMI inputs mapped to the same physical HDMI input. I think this can be done...

Yes there is.. and Gordon highlighted it... for you, Buttons C + D will simply toggle you between 1080P24 and 1080P60.Well, yes, but I was hoping for something that wouldn't completely use up all the available memories, in case I need them for something more substantial than just audio setting and 24p/60p. Like, say, different configurations for the different displays, or daytime/nighttime, etc. Those are more conventional/expected uses of the memories, I think.

If you need more sophistication (or memory)... do what I do, get two Blu Ray players - one for 1080P60, the other for 1080P24. This way you are covered and you use a second physical input for each (so, potentially, 8 full memories between them).I'm not sure I understand (apart from the fact that I have far too many HDMI sources already and I'm trying to cut down, not add more). The problem I'm trying to solve is mainly for TV sources being provided by one device (a TiVo). Some (perhaps most) will be film-sourced and can use 24p output, but many will originate as 60Hz video and I'll need 60p for that. Blu-Ray sources I would expect to do the right thing anyway, i.e. output 1080p24 if that's what's on the disc, or 1080p60 otherwise.

I understand what you're looking for, but auto-switching between 24p and 60p depending on if cadence is detected or not is not feasible. We are not dealing with a flag-reader here, but a complex de-interlacer. For many DVDs, you would have frequent scanrate switching, causing much more annoyance than occasional judder or frame drop/repeat.

It's doable, and I think one of the other vp's does it, but the problem becomes that every time the resolution switches, it forces a new HDCP handshake. This means you'd have handshaking going on during commercials (video) if you were watching a film source.

it is do-able but Jim has already pointed out that they wil not do it due to the limitiations pointed out above. You'd have the thing swapping resolutions on every advert break on tv.

You could put a request in to Lumagen to ask them to consider putting an IR code shortcut in to swap OUTCONFIG's ie "ALT" "1" to swap to output 1, "alt""2" to swap to output2....that may well be do-able and could go on list...but it's not for me to say. Ask Jim.

Gordon

Is the Sigma Designs GF9452 any different than the Gennum VXP in the RadianceXD?

Here is the overview page:

http://www.sigmadesigns.com/public/Products/GF9452/GF9452.html

The Crystalio II allows this behavior through Dynamic VP. As stated, this forces a new handshake during cadence changes. On a CRT PJ, there is also an addition delay when the PJ re-scans for the proper memory block.

Is the Sigma Designs GF9452 any different than the Gennum VXP in the RadianceXD?

Here is the overview page:

http://www.sigmadesigns.com/public/Products/GF9452/GF9452.html

Yes, 12 bit vs 10 bit precision and the 9452 also says "Flexible output architecture to support single, dual and twin channel outputs"

9452 also has "extended" and progressive cadence detection. 9450 doesn't.

So the question, then, is when will the Radiance upgrade?

So the question, then, is when will the Radiance upgrade?
Exactly. They should throw in the new GF9452 as well as support HDMI 1.3 for the XE.

Exactly. They should throw in the new GF9452 as well as support HDMI 1.3 for the XE.

Unless the 9452 is completely backward compatible hardware AND software wise, I doubt very much that's going to happen. Also this would probably significantly impact the price of the XE.

Don't hold your breath. As announced, the XE is an XD with hdmi 1.3. That's it. There's ALWAYS going to be some shinier toy on the horizon.

All I can hope is that others agree with me when I say I am not going to buy it if it's not the absolute latest and greatest. Especially for the ridiculous asking price.

Donoman. Just about everyone here already own a Radiance. Many of us could care less about HDMI 1.3 at this point.

BTW I doubt many of us could ever afford the latest and greatest in video processing. The Radiance does not represent that but it does represent what many of us feel is the best high end consumer VP.

Now the new Sigma VXP chip. The Radiance (REMEMBER) has a daughter board for VP chip upgrades. Will a VP daughter board upgrade ever really happen? The folks at Sencore at Cedia had absolutely no clue that the VP chip could ever be upgraded by a daughter board switch.

It would seem to me that there are many desirable features of the new chip.

i should know this but I don't remember, would 1.3 be required to employ the increased bit depth of the new chip? If so, then a package of the two would seem quite in order. Remember 1.3 alone does not give one that much. But the new chip would seem to offer a lot.


And the Sigma press release says with a simple FGPA the chip could output 4K (with 2k per stream). Maybe those crazy rumors that Lumagen is secretly working on a 4K processor for Sony to be used with the new coming some day Sony 4K consumer projector are true. I just heard this from a consumer with some pretty good connections with a certain non Lumagen non Sony source. I am pretty sure this rumor will be denied and I really have no clue as to whether it is true or not. But in theory, if it were true, could a single modified Radiance using the new chip put out 4K using the two HDMI outs. I think not and two Lumagenswould be required and therefore two processing chips. Not very elegant. Better a redesigned mother board using a one new chip daughter and sufficient output to handle at the same time two different 2K streams, one on each HDMI output.

Anyway most of us will probably never need 4K anyway.

The old Gennum chip is getting dated. It does somethings very well and other chips do other things better. This is a fact, not an opinion. It would seem that the new chip has some very desirable features that I and others would want on their Radiances with or without HDMI 1.3. The bit depth increase is a serious upgrade and I think all would concede that if your projector can use it, PQ would be noticeably improved. That's a large reason many of us own Radiances to begin with not to mention oall the other features. Now with the new CMS 8 axis feature, the 21 step gray scale possibility, yada yada, wow. But what a nightmare to perhaps implement (particularly the 8 axis CMS?). Anyway I want it and certainly would be willing to pay something like $2K for it and 1.3 if 1.3 is required to use the increased bit depth. Do you think Sencore would be able to sell the upgrade for say $2K or keeping in line with normal Sencore pricing, the upgrade might cost 4K? If that much, I'd pass.

Donoman. Just about everyone here already own a Radiance. Many of us could care less about HDMI 1.3 at this point.


Good for you guys, but it's not acceptable to some of us to pay 4+k, not have the latest, and then have to pay more to upgrade it to the latest technology if I want it. Just because you, or even a large group of you, do not care about a feature, it is NO reason to post saying not to bother.

Good for you guys, but it's not acceptable to some of us to pay 4+k, not have the latest, and then have to pay more to upgrade it to the latest technology if I want it. Just because you, or even a large group of you, do not care about a feature, it is NO reason to post saying not to bother.

Are you talking about hdmi 1.3? Why do you need it so bad?

Are you talking about hdmi 1.3? Why do you need it so bad?

Primarily to bitstream HD Audio.

Yes. I know it *should* be the same...

But from an HTPC even a BD player I am beta testing, the speaker settings and mixers get involved when converted to LPCM. I have also noticed cases when the Dolby True-HD normalization was picked up by my AVR in bitsream mode only and really helped the center channel.

I ran out of HDMI inputs on my Onkyo Preamp so I would like to use some of them on the Radiance.

Also, newer HDMI chips can mean better compatibility. Although, at this point the Onkyo is a pain and the Lumagen works well.

Lastly, I seem to resell these thing and upgrade periodically. I think the HDMI 1.3 version will have a better resale value ;)

- Rich

Are you talking about hdmi 1.3? Why do you need it so bad?

I don't "need" it right now. I do want the ability to bitstream audio should I want to. I do want the capability to use xvYCC color in the future should I get a source supporting it (most likely my computer will be the first). I am not going to even consider spending this kind of money without the latest and greatest, including HDMI 1.3, and including the GF9452.

domo. the industry has brain washed most consumers as to the need to have HDMI 1.3. Of course, no question, at present it offers practically nothing but its needed, I am a retailer, to sell much of anything because of the brain washing. I am sorry, but of the the purposes of this forum is to be speak the truth. that'swhy i posted what i did and that is I think what Dave G. is saying. If 1.3 is necessary for the increased bit depth of the now sigma new VXP, then i would agree its now needed but not without the increased bit depth of the new chip.

the new Anthem processor (which includes audio0 has the new chip and 1.3.Its a paltry $8k compared to the $4.5k of the Sencore by Lumagen.

if you are brain wahed, I am not up to an intervention. :)

lumagen has a nice upgrade path for present radiance owners who want (or mistakenly believe) they need 1.3. a $400 bluray player can supply the new audio format. 1.3 offers nothing for present real world video sources. Maybe something for gamers soon. I dunno nor do I really care. i play all the freeking games i need by posting here. :)

With the current HDMI in the XD you could send your HD audio sources direct to your amp and then take the output of the amp to the XD, using the virtual input mapping function to get custom settings for each source fed to amp, just as you would if they went direct to scaler.

Also...are we sure that 12 bit processing would bring any benefits to our 8 bit sources?

I can see that the newer de-interlacing solution may be of use along with the dual output channels.

As to cost. If you feel that the cost of the product is beyond what you expect to spend then don't buy one. If you want the best video performance you can get from a domestic video processor and probably signifcantly better than most commercial ones then you need an XD.

RICH: I can see that if you have used up all the HDMI ins on your amp then sending some through the scaler would be useful.

I do think it's too expensive, yes, but I do want one. It's a consideration (not right this second, admittedly) if it's the latest technology, but without 1.3 and the GF9452, it's not even a consideration. I play games and use my PC with my projector quite often, and those will benefit from the additional precision far sooner than consumer video formats will. Hell, even Blu-ray is sticking with the horrid idea of 4:2:0.

Maybe you guys are so rich that you upgrade your stuff every 6 months, but if I'm going to spend this kind of money on a product, frankly, I don't give a s*** if the features are useful now. I care if they will be useful in a few years. You all need to understand that most of the market buys stuff for the long run.

if you are brain wahed, I am not up to an intervention. :)

lumagen has a nice upgrade path for present radiance owners who want (or mistakenly believe) they need 1.3. a $400 bluray player can supply the new audio format. 1.3 offers nothing for present real world video sources. Maybe something for gamers soon. I dunno nor do I really care. i play all the freeking games i need by posting here. :)

Entertaining enough, but do you have an HTPC?
Do you understand the difficulty in getting LPCM and protected path?

At this point, Lumagen must support HDMI 1.3 so what's the point in belaboring it. I am willing to bet it wont be long before the 1.1 version is discontinued ;)

The real villian here is the miserable HDMI security and handshaking.

- Rich

I do think it's too expensive, yes, but I do want one. It's a consideration (not right this second, admittedly) if it's the latest technology, but without 1.3 and the GF9452, it's not even a consideration. I play games and use my PC with my projector quite often, and those will benefit from the additional precision far sooner than consumer video formats will. Hell, even Blu-ray is sticking with the horrid idea of 4:2:0.

Maybe you guys are so rich that you upgrade your stuff every 6 months, but if I'm going to spend this kind of money on a product, frankly, I don't give a s*** if the features are useful now. I care if they will be useful in a few years. You all need to understand that most of the market buys stuff for the long run.

I didn't pay the current price for the XD. In the first version of the FAQ, I tried to hammer the point that people wanting one should have bought it RIGHT THEN. In hindsight, it was an incredible deal. I wouldn't be able to afford it at today's price. Looks like you can't either (and funny thing is, you want extra features for less money!) - well, tough break. I'd like a new 5000 sq ft house with all the trimmings and high end finishes. Guess what?

On the other hand, regardless of price, basing your purchasing decision on features that might or might not get used in the future (and my money is on 'not' seeing how deep color is NOT in BD's specs and never will be - as for bitstreaming, the workaround is so easy it's barely worth mentioning)... well, that's just silly. I too try to future proof my purchases. But I use common sense as well. I'm enjoying my XD/RS1 combo today, and it's heavenly. What's in YOUR ht?

Dan: I think that I and Lumagen understand that folk buy things for the long run. That's why they support their products with continuous firmware upgrades and why the Radiance also has hardware upgradeability.

I am sure that if there is a quality benefit and if the cost to impliment is worthwhile then Lumagen are the most likely folk to make updates available.

I concur that PC's will benefit from higher bit depth processing as they are not 8bit...but then the XD is to have a game mode added to the firmware that will bypass the gennum and allow this to happen in the Lumagen FPGA anyway.

I'm enjoying my XD/RS1 combo today, and it's heavenly. What's in YOUR ht?

Well, it's in my profile. Not AS good, but VP50 and Marantz VP12S4 ain't half-bad, either.

Well, it's in my profile. Not AS good, but VP50 and Marantz VP12S4 ain't half-bad, either.

I'm sure it's great. I've seen Marantz projectors in action, they're top notch for sure. So what's missing in that chain that the XD could provide?

I thought that the original modular concept of the Radiance allowed for an upgrade of the chip. Is it not possible to do this in a similar way to 1.1XD can be upgraded to 1.3XE?

Martin

I'm sure it's great. I've seen Marantz projectors in action, they're top notch for sure. So what's missing in that chain that the XD could provide?

Good noise reduction, support that actually gives a damn about what their customers want, etc.

Honestly, even if I do get a Radiance, it's not #1 on my list. But I absolutely plan to get one sometime if they keep things current.

I thought that the original modular concept of the Radiance allowed for an upgrade of the chip. Is it not possible to do this in a similar way to 1.1XD can be upgraded to 1.3XE?

Martin

Should be but IMO we shouldn't have to pay extra if we haven't bought anything yet. They should just be available with the new parts for almost the same price. The original price included R&D cost for the unit, and an upgrade cost is the cost of having ADDITIONAL pieces of equipment installed when you already bought the older stuff. If I say I want to buy it WITH the new equipment, the price should only go up as much as the prices of the new parts. They should have done a complete switch to 1.3 already. It makes no sense to continue making them with 1.1.

Good noise reduction, support that actually gives a damn about what their customers want, etc.

Honestly, even if I do get a Radiance, it's not #1 on my list. But I absolutely plan to get one sometime if they keep things current.

Re: price - try to find a used one. This thing is built like a tank. Try to agree with the buyer on a 30-day try out, that would leave you plenty of time to see if something's amiss, and it should be easy if you buy from someone here on AVS. Hell, I would even call Lumagen or Sencore see if they have some open box/B stock deals. The absolute worst that can happen is that they will say 'No we don't'. Worth a try IMHO.

Re: features. Look, I get the hdmi 1.3/new chip thing. I really do. I myself am a bit worried about the used buyer mentality, or at least I would be if I planned to resell my XD in the next couple years. But I'm going to keep if for a really, really long time. I think as far as formats are concerned, we're stuck with BD and hdmi for a good long while, and 1.1 is a lot more than enough for the foreseeable future. The next big thing is probably a long, long while away, and who says hdmi 1.3 will be sufficient by then anyway? As for the new chip, I haven't seen it in action, so I can't really say. All I know is that I'm plenty satisfied with the XD's performance right now, and I can't imagine anything coming after it will be a big improvement; just an incremental step. Remember also that most features on the XD (CMS, scaling, switching) are NOT A function of the vxp.

Martin: The VXP chip is on a small daughterboard. The interface between that daughter board and the Lumagen main pcb is designed in such a way that they could swap it out for A.N.OTHER chip vendors product instead of or including VXP. That was the original intention of the hardware upgradeability.

Changing from 1.1 to 1.3 is not simply a chip replacement. It is quite involved. Lumagen is offering a fair retrofit price.

Re the deinterlacing in the old Gennum chip, does somethings well. Film is good. Video is a mixed bag. For a high video watcher like me who still has some 480i sports channels (TVG etc), I think the Silicon Optics chip would be a better deinterlacer. Who knows if the new VXP has been improved here too. My guess is that someday we will see a daughter board chip upgrade to the new chip. Another brand of chip/ I doubt it, probably too much work to integrate it.

I can't speak for HQV because I don't own anything with it, but my VP50 isn't all that great for cable, either.

Processing video vs film, the algs used involve trade offs. Its a little like a nude woman trying to hide her private parts with a small towel. You can cover the top and expose the err or cover the bottom and expose the mountain peaks. Or you might cover both a little but expose enough to keep top and bottom lovers happy. Nothing is perfect or can it be when deinterlacing video.

Re the deinterlacing in the old Gennum chip, does somethings well. Film is good. Video is a mixed bag. For a high video watcher like me who still has some 480i sports channels (TVG etc), I think the Silicon Optics chip would be a better deinterlacer. Who knows if the new VXP has been improved here too.Bear in mind that the firmware in the current Radiance VXP chip is updatable. I own one of the very first Radiances (May 2007) which, unsurprisingly, had older VXP firmware. The recent VXP update improved things considerably in "Auto" deinterlacing mode.

Is there a generally accepted range of settings that usually produces the best results for DVD playback?

I think (still experimenting) that I see too much loss of detail if I use Noise Reduction settings higher than about "2" and that images start looking over-processed with the 3 sharpness and 3 texture settings higher than about "4". With the ranges of some of these controls going up to 31, I wonder if settings as low as the ones I think I'm settling on are typical/common of what others have found to work best for their DVD playback.

Any other DVD settings that are useful for making DVD look better?

Is there a generally accepted range of settings that usually produces the best results for DVD playback?For DVD, all of my NR and Enhancement settings are set to zero. Sure, poorer quality DVD transfers could benefit with some tweaking, but the best transfers don't. With movies, I don't want to risk losing any detail so I'll take what the DVD offers. OTA broadcasts are a different kettle of fish and mosquito and block NR are important to me.

Processing video vs film, the algs used involve trade offs. Its a little like a nude woman trying to hide her private parts with a small towel. You can cover the top and expose the err or cover the bottom and expose the mountain peaks. Or you might cover both a little but expose enough to keep top and bottom lovers happy. Nothing is perfect or can it be when deinterlacing video.


So Mark,

where do you see problems in the video deinterlacing on VXP ?

Michael

Sticky thread"Unofficial Radiance XD FAQ" has been merged with this thread.

Thank you!

Sticky thread"Unofficial Radiance XD FAQ" has been merged with this thread.

Thank you!

What was the point of that? The FAQ post is now buried in the middle of the thread and impossible to find!

What was the point of that? The FAQ post is now buried in the middle of the thread and impossible to find!Indeed. I don't think merging the two threads was the right call, Alan.

Oh crap. Now only 3060+ posts to dig through for those starting with questions...

What was the impetus?

I've privated Alan and he's open to suggestions so we're trying to find a solution. No malice on his part of course, he was just doing some super-extra-early Spring cleaning :)

Unofficial RadianceXD Facts



Dave, time to change your signature. With the two threads merged,
your signature thread is a "bridge to nowhere"!!@@

I have had audio problems since the day my Radiance arrived. I am unable to stream both normal bitstream/dolby prologic and dolby digital/AC3 to my Lexicon mc-1 processor from multiple sources.

If I plug my sources audio output directly to the Lexicon I get perfect sound. If I send them (both toslink and coax digital) through the Radiance I get horrific digital noise only when playing a DD source. I can fix this by setting the lexicon to "AC3 only" mode, but then I cannot hear any non-DD sound (so normal Tivo listening or non AC3 movies I hear nothing).

I have contacted the manufacturer on several occasions only to be told it might be lexicon's fault (which I don't buy, they are top notch and it works perfectly without the radiance) and they give me lip service about looking into it, but never get back to me with any information.

I'm going to give it another 30 days and then ask for a refund as the product does not work as advertised. I'm sad about that as I'm very happy with the video processing, but one of the main reasons I bought this particular unit instead of the competition was to also be able to switch audio simultaneously.

I post this only because I personally spent quite a bit of time in this thread and others in this forum before I made my purchase (which was a LOT of money to me - I'm not wealthy) and it's one of the few things I've bought in recent years that has been a major disappointment.

If I get my issue resolved, I'll post back. And I'm aware that many users here are happy. My experience however has been completely frustrating and if I had to do it again I would spend my money elsewhere, buyer beware.

In case our Email got caught in your spam filter here is a response we sent to you. We haven't heard from you recently. You should try loading the latest Radiance software. Please contact us if you continue to have an issue with your Lexicon MC1.

Best regards,
Randy Freeman

*********************
I just spoke to you about your coax audio issue on the Radiance. So I've tried some combinations this morning:

1. DVD player HDMI audio -> Radiance COAX (spdif) audio -> MC12 and switched the DVD between Auto (which gave AC3) and PCM. This worked, does that work on your system using 100207? If you're not using 100207 which software are you using now an we'll set up a system with that version for testing.

2 DVD player COAX (spdif) audio -> Radiance COAX (spdif) audio -> MC12 and switched the DVD between Dolby and 2 ch pcm. This worked also.

I then switched to using a JVC D412 AV receiver with cases 1 & 2, again it worked switching between AC3 and PCM on the DVD player and was being decoded correctly by the JVC.

Looking on the net I see a number of threads related to the MC-1 not locking properly:
http://www.pixelmagicforum.com/forum/archive/index.php?t-1718.html
http://www.avsforum.com/avs-vb/showthread.php?t=796234
http://www.smr-archive.com/forum_3b/tree2.html

One user mentioned that upon upgrading from a DC1 to the MC1 he got the problem. Some users mentioned upon cycling power to the MC1 that it would then lock to the audio, might be interesting to try that on your system. Although the MC1 has some kind of an issue it obviously works with a number of sources too since you've tried it direct. Unless we can reproduce it here though, I doubt we'll be able to fix it on our end. So there are several options:
1) Send us your MC-1 so we can reproduce the problem and fix it (probably wouldn't take us more than a few days if we have a bug that we can directly work on).
2) Feed your coax audio to the MC-1 directly to avoid the locking issue.
3) Get a new audio receiver---we have a lot of Radiance users and we've got quite a variety of equipment including the MC-12 and it all works fine switching between AC3 and PCM using the coax output from the Radiance..

Any word on the "Pro" version or SDI support in Radiance?

The RadiancePro has been discussed but there is currently no schedule for the production of this unit. The preliminary spec, which is subject to change, includes a SDI input.

Some customers have used the Black Magic SDI to HDMI mini-converter in order to connect a SDI signal to their Radiance. They report that the Black Magic converter works well with SDI signals that have jitter within the SDI specification. Many DVD players with SDI mods have very high jitter on the SDI output. In at least one reported case, the Black Magic converter didn't work because the moded player added too much jitter to the SDI signal. This same Black Magic converter worked very well when connected to a good SDI video source.

I am confident there are no competitors for Black Magic on the forum, but if there are and this gives them any heartburn, let me know and I'll remove.

I set myself up as a BlackMagic dealer just for this issue and (for right now) I am willing to get the BlackMagic mini comverter box for any AVS member at my cost + $20 + shipping (and there isn't much margin on them).

BUT, the distributor's return policy is such that there is not enough time to allow you to test and return to me and me to them if it won't work. I don't want to stock them so, the caveat is, you get it at cost, but the risk is yours as to whether or not your player's SDI mod is jitter free enough so it will work. All I ask is if you do get one, you post here and let everyone know if it worked, or not, what mod you have (i.e. Lumagen, JB Digital, Pixelmagic, etc.) and who did the mod (you, them).

Note that this converter does accept SD-SDI, HD-SDI and even dual link HD-SDI, outputs to HDMI.

PM if I can help..

Does the BM converter maintain 480i on the HDMI output?

It is supposed to be a "simple" transport method converter. what goes in it what goes out, only from a coax, to a HDMI cable. No conversions, no clipping, no screwing with the input in any way. They have products that allow interception and adjustment, this isn't supposed to change it. That's wht they say.

The RadiancePro has been discussed but there is currently no schedule for the production of this unit. The preliminary spec, which is subject to change, includes a SDI input.

Some customers have used the Black Magic SDI to HDMI mini-converter in order to connect a SDI signal to their Radiance. They report that the Black Magic converter works well with SDI signals that have jitter within the SDI specification. Many DVD players with SDI mods have very high jitter on the SDI output. In at least one reported case, the Black Magic converter didn't work because the moded player added too much jitter to the SDI signal. This same Black Magic converter worked very well when connected to a good SDI video source.


Yea, I have had major SDI funk over the years.

Some years ago had original Lumagen VisionPro (analog) with SDI.
Got a Panasonic RP-91 SDI modded by JVB. Worked fine. But some months later started getting wavy purple lines.

First shipped Lumagen to them and they checked it and said it was fine. Got it back and same problem. They had me ship both Lumagen and DVD player.
They determined too much jitter. Jim Peterson of Lumagen then did Lumagen's SDI mod to the DVD player and all was fine - and didn't charge me one penny. He sure didn't have to do that!

Then replaced Panasonic RP-91 with a Theta Compli built with SDI on board from factory. Worked fine. Then replaced VisionPro with VisionPro HDP (digital) with SDI.

Theta Compli SDI worked absolutely fine the next two years, until - replaced VisionPro HDP with Lumagen Radiance about March this year, when I said bye bye to Dwin CRT and got Sim2 C3X 1080.

I got an AJA S15 SDI to HDMI converter. Problems right from the start, jittery picture, would work then wouldn't work, then stopped working. Got tired of SDI crap and got an Oppo DVD player which puts out pure component video 480i over HDMI with no conversion and fantastic picture. Sound pretty good, too, but not as good as Theta Compli.

Recently, after months of not using AJA, I contacted them and they replaced the unit. Still doesn't work. Course AJA may have its own problems, as I found out too late it converts 480i component to RGB, and even when it did work, its picture didn't compare to using the Oppo DVD player instead.

I'd like to try the Black Magic, but I am concerned that I'll still have this same problem.

I wonder why the SDI converters built into the Lumagens have worked fine with my Theta Compli when the AJA doesn't???

The RadiancePro has been discussed but there is currently no schedule for the production of this unit. The preliminary spec, which is subject to change, includes a SDI input.

Some customers have used the Black Magic SDI to HDMI mini-converter in order to connect a SDI signal to their Radiance. They report that the Black Magic converter works well with SDI signals that have jitter within the SDI specification. Many DVD players with SDI mods have very high jitter on the SDI output. In at least one reported case, the Black Magic converter didn't work because the moded player added too much jitter to the SDI signal. This same Black Magic converter worked very well when connected to a good SDI video source.

Thanks for the response. Both my SDI devices are factory fitted SDI so jitter should not be a problem. Does the Black Magic output 480I 4:2:2 over HDMI?

Yes it does, I have one and it's great

I cannot find the FAQ. Alan says he merged the threads, but even searching all my posts in this thread I can't find it. Anyone has a link to it?

Edit: nevermind, found it. The Advanced Thread Search does NOT work. What a waste of time. :mad:

Dave, start a separate post with your Lumagen FAQ and I will sticky it.
Send me a PM once you have made the post.
Thank you!

I will Alan, need to take a few screenshots, I will work at that this week-end.

(I will also post in the forum support subforum, because I wasn't kidding, the advanced thread search really is broken.)

Thanks for the response. Both my SDI devices are factory fitted SDI so jitter should not be a problem. Does the Black Magic output 480I 4:2:2 over HDMI?They say whatever comes in, goes out, no conversions, except for physical connection type.

To clear up some confusion on the GF9452, I thought I would chime in.

The GF9452 has a 12-bit interface. HOWEVER, the internal processing is still 10-bit. There is NO picture quality advantage to it's 12-bit interface. I believe they did the 12-bit interface only to better connect to HDMI 1.3 chips. Since we have an FPGA between the GF9450 (or GF9452 if we did make a change) and the HDMI chips, there would not be any advantage to the GF9452's 12-bit interface for us.

We use our scaling, since we believe our scaling is much better than the GF9450 (or GF9452). So, the max output resolution of the GF9452 is of no importance to us. We set it to output at the native resolution of the source.

There is only one advantage we can see for the GF9452 and I'm not even sure it's public info so I don't want to mention it here. Not sure if this feature would be worth the large effort and extra cost for the GF9452, but we are considering if we should make a change. If the Return-On-Investment plays out we may offer the GF9452 in the future. If we did, you could upgrade the RadianceXD, or the upcoming RadianceXE HDMI 1.3, to this since we have the GF9450 on a daughter card for such an occasion.

What we really want to see from Sigma (Formerly the Gennum VXP division) is a true 12-bit chip, but for now the GF9450 still provides industry leading performance.

Is the HDMI 1.3 upgrade now available for current Radiance XD owners for $1,000?

Once upgraded, how will the upgraded Radiance compare to the upcoming
RadianceXE HDMI 1.3? Will then be identical or what will be the differences?

Thanks.

If you do move to a true 12-bit chip, does the architecture of the Radiance XE allow for 12-bit processing all the way through, or would there be another hardware upgrade required?

jrp, your own scaler chip is only 10-bit internally, too, then?

Is anything/everything that comes across either Dish, Cable or Direct TV
"Video"? Would that include a movie on HBO for example.

I understand that DVD's, and BluRay movies are "film" unless they are from a TV show like PBS.

thanks

Is the HDMI 1.3 upgrade now available for current Radiance XD owners for $1,000?

Once upgraded, how will the upgraded Radiance compare to the upcoming
RadianceXE HDMI 1.3? Will then be identical or what will be the differences?

We have a few first production boards in house now and should be shipping these out in the next coule weeks. We have prioritized the RadianceXD 24p output mode and are fighting some issues with this which is delaying the XE. The XE software is at the same level as the XD at this point, but we need to add the new HDMI formats. We expect quantities to become available at the beginning of the year. Initially RadianceXE Beta is only being offered to current RadianceXD owners and the price is $1000 in the US and Canada direct from Lumagen. Gordon Fraser of Convergent-AV (+44 (1483) 571500) is handling upgrades in Europe. Please contact him for more info if you live in Europe.

When upgraded, the performance, features, and interface, of a RadianceXD upgraded to a RadianceXE will be identical to a new RadianceXE unit.

If you do move to a true 12-bit chip, does the architecture of the Radiance XE allow for 12-bit processing all the way through, or would there be another hardware upgrade required?

From the palette output to the Radiance output the pipeline is already 12+ bits. However, the rest of the calculations in our FPGA are 10-bit. When we added the 3D palette it took up a lot of room in the FPGA. We would need to do quite a bit of work before we knew the full answer to the question of a full 12-bit path, which would include a lot of optimization. Certainly we could support some modes at full 12-bits, but perhaps not all modes. Sorry I can't be more specific but the 3D palette took a lot of the headroom we designed into the Radiance and could limit what can be done with the current hardware.

Is anything/everything that comes across either Dish, Cable or Direct TV
"Video"? Would that include a movie on HBO for example.

I understand that DVD's, and BluRay movies are "film" unless they are from a TV show like PBS.

These can be either video or film. Sports and many shows are probably video. Movies should be film.

We have a few first production boards in house now and should be shipping these out in the next coule weeks. We have prioritized the RadianceXD 24p output mode and are fighting some issues with this which is delaying the XE. The XE software is at the same level as the XD at this point, but we need to add the new HDMI formats. We expect quantities to become available at the beginning of the year. Initially RadianceXE Beta is only being offered to current RadianceXD owners and the price is $1000 in the US and Canada direct from Lumagen. Gordon Fraser of Convergent-AV (+44 (1483) 571500) is handling upgrades in Europe. Please contact him for more info if you live in Europe.

When upgraded, the performance, features, and interface, of a RadianceXD upgraded to a RadianceXE will be identical to a new RadianceXE unit.

Aside from HDMI 1.3, how will the XE vary from the XD? Or is XE simply the new model number of the XD with HDMI 1.3? Will the XE support any additional features?

Aside from HDMI 1.3, how will the XE vary from the XD? Or is XE simply the new model number of the XD with HDMI 1.3? Will the XE support any additional features?

The only difference is that the RadianceXE will support the new video formats and uncompressed audio.

The only difference is that the RadianceXE will support the new video formats and uncompressed audio.

Do you mean the advanced bitstream codecs? The XD already supports PCM = uncompressed audio. :confused:

There is only one advantage we can see for the GF9452 and I'm not even sure it's public info so I don't want to mention it here. Not sure if this feature would be worth the large effort and extra cost for the GF9452, but we are considering if we should make a change. If the Return-On-Investment plays out we may offer the GF9452 in the future. If we did, you could upgrade the RadianceXD, or the upcoming RadianceXE HDMI 1.3, to this since we have the GF9450 on a daughter card for such an occasion.

What we really want to see from Sigma (Formerly the Gennum VXP division) is a true 12-bit chip, but for now the GF9450 still provides industry leading performance.

If you think about :

Extended film cadence detection
Progressive film cadence detection
Vertical keystone correction
Twin output channels

Yes those info are public (http://www.sigmadesigns.com/public/Products/vxp_selection_guide/vxp_selection_guide.html)

I find them worth enough to be used in the next Radiance generations.

If you think about :

Extended film cadence detection
Progressive film cadence detection


What does that do?

Is there a setting for Adjusting the audio delay, and can it be set at different amounts per input, or is it a global setting.

Thanks

Audio delay is an announced feature of the Radiance. The current plan is to store the value for audio delay per input and per input resolution. My guess is that we are probably at least a month away from implementing this feature.

Best regards,
Randy Freeman

Thats great news and looking forward to it!

I have not been keeping up with my Radiance XD, is it now possible to pass-thru unprocessed 1080P24. IE 1080P24 in 1080p24 out

It's been possible for a while, yes.

Dave:

True straight thru no processing at all, same as a switcher?

Correct, except you have to make sure that scale bias is set to off both horizontally and vertically in that case. I don't have the menu path off the top of my head, but it's no hard to find.

Correct, except you have to make sure that scale bias is set to off both horizontally and vertically in that case. I don't have the menu path off the top of my head, but it's no hard to find.Scale bias *should* be off by default for the same input and output resolutions. Definitely worth checking though.

Quote:
Originally Posted by jackox
If you think about :

Extended film cadence detection
Progressive film cadence detection


What does that do?


I wonder !

I wonder !
I have the feeling that the people who want these so bad don't know either. But hey, if it sounds good it's gotta be great.

The hot feature of the new VXP is indeed the two independant out channels that can fe fed simultaneously with different resolutions.

"progressive film cadence detection" reminds me of Prep, I don't know why ... but I would love to get tech information in addition to this marketing naming.

One of the most impressive features of the Radiance is the 11-point gamma/gray scale adjustment tool. Using this, you can ensure that the luma and the chromaticity of white is correct at 10-100 % stim.

However, the CMS offers only a single level of adjustment for each of the primary/secondary colors. The problem is that the color performance of displays can be non-linear in just the same way as is the white point. Using the CMS, you can get the color nearly perfect at whatever level of stimulus you use for test patterns (I use 75%). But that's no more effective for color than would be a single-point white balance adjustment be for the gray scale.

I measured RGBCYM at 25%, 50%, 75%, and 100% and here's what I got.

http://home.comcast.net/%7Etlhuffman/linear_chroma.jpg

Green and blue look reasonably good, though they could use improvement, but red is quite non-linear. Interestingly, the non-linear behavior of the primaries is in the saturation plane, whereas the secondaries show non-linear response in the hue plane.

It would be nice if the CMS offered adjustment at different levels of stimulus, just like the gamma/grayscale tool does.

Is this feasible?

Hi,

i'm not sure I understand what you describe :o.


Are you saying when you measured the 25 %, 50 % or 75 % on primaries and secondaries colors , you have a CIE as your picture, with any another modifications ?

You don't have something like this ?

http://img528.imageshack.us/img528/2619/clip5resizecm2.jpg

Différents points on the CIE are the modifications at differents saturation levels (25 %, 50 % and 75 %), 100 % don't move, but "my" setting don't come from a scaler as the radiance. It's not CMS correction, it's only saturation level.

In you case, I don't understand (maybe it's a misunderstand (?) why you don't have modification (or it don't appears) at différent level ?

Sorry, but I'm interested in this scaler and I try to know if it can correct all points (75 %, 50 % or 25 % for exemple) on each primaries and secondaries colors.

i'm not sure I understand what you describe :o.


Are you saying when you measured the 25 %, 50 % or 75 % on primaries and secondaries colors , you have a CIE as your picture, with any another modifications ?

Différents points on the CIE are the modifications at differents saturation levels (25 %, 50 % and 75 %), 100 % don't move, but "my" setting don't come from a scaler as the radiance. It's not CMS correction, it's only saturation level.

In you case, I don't understand (maybe it's a misunderstand (?) why you don't have modification (or it don't appears) at différent level ?

Sorry, but I'm interested in this scaler and I try to know if it can correct all points (75 %, 50 % or 25 % for exemple) on each primaries and secondaries colors.The Radiance CMS works by color mixing. You desaturate a primary color by adding the other primaries to it. But the results are non-linear. You can adjust with test patterns at any level of stimulus you prefer, but then the results will be good only for that level of stimulus. I use 75% patterns, so if I measure the gamut using 75% patterns, the colors look spot on. Were I to measure using 100% patterns or 50% patterns, it would look very different.

I made the analogy to gray scale tracking. The goal there is to get x0.3127, y0.329 at all levels of stimulus throughout the gray scale. The same should be true for color tracking. A SMPTE-C red should measure x0.630, y0.340 at 25%, 50%, 75%, and 100% stimulus.

I was asking if it would be possible to engineer the CMS to accomplish this. It would seem that you'd need 4 levels of adjustment for each color, instead of just one as they have it now (eleven points as it has with the gray scale tool would be overkill, I think).

Hi,

thank for the answer. It's more clear in my mind :o.

When you said "non-linear", the effect could be like this.

http://i278.photobucket.com/albums/kk102/shacard1/gamut_vs_sans_gamut-1.jpg

The correction come from an "old" lumagen and it was only on 100 %.

The picture at right is without correction and the picture at left is with correction at 100 %, but we can see there are "some" problems under :confused:.

I think you ask if it's possible to obtain a correction on 11 points, but if the problem are the same (non-linear effect), it is not possible to have a "minor" error between these 11 points, no ?

The hot feature of the new VXP is indeed the two independant out channels that can fe fed simultaneously with different resolutions.

"progressive film cadence detection" reminds me of Prep, I don't know why ... but I would love to get tech information in addition to this marketing naming.Doesn't it simply mean that it can detect the 2:3 cadence in film-sourced material which has been converted to 1080p60 or 720p60?

The current VXP in the Radiance can only do cadence detection for interlaced sources, which means that (for instance) 720p60 film-sourced TV material (like ABC and Fox HD broadcasts) can't be converted to judder-free 24p. I would certainly welcome this capability.

On the other hand, it doesn't even work for interlaced sources yet, in my experience. Whenever I try to use the genlock feature with a 1080i source it works for a while and then becomes noticeably jerky, so it's not really usable with TV sources at the moment.

Lumagen says that TV sources aren't "clean" enough. For me, the irony is that (eventual) 1080i60->1080p24 for TV (which is most of my viewing) was one of the main reasons I bought the Radiance a year and a half ago.

The current VXP in the Radiance can only do cadence detection for interlaced sources, which means that (for instance) 720p60 film-sourced TV material (like ABC and Fox HD broadcasts) can't be converted to judder-free 24p. I would certainly welcome this capability.



In a related note, I just noticed that ABC went 1080i which I think is great. Football on Fox is definitely is fuzzy HD. I hope Fox goes 1080i too.
If that happens, progressive cadence detection becomes pretty much useless.

- Rich

In a related note, I just noticed that ABC went 1080iI don't think that's true. A few affiliates may have switched (like WCVB in Boston--I see that you're in Massachusetts) but I can't find anything to suggest that it's a network-wide change.

I don't think that's true. A few affiliates may have switched (like WCVB in Boston--I see that you're in Massachusetts) but I can't find anything to suggest that it's a network-wide change.

Hmm, well that means the Radiance might have a chance.

- Rich

A full bandwidth over the air broadcast can be deinterlaced much easier than an overly compressed program from cable or satellite. The cable and satellite companies are much more concerned with squeezing in more channels than giving you a high quality image. All the compression artifacts they add to the image make it harder for every deinterlacer, and this applies to all brands of deinterlacer. If over the air HD broadcasts are available to you, you can put up an antenna to receive the over the air broadcasts. These broadcasts have much higher bandwidth and can be much better quality.

You should try turning genlock on and off to see which setting gives you better results.

Hi,

thank for the answer. It's more clear in my mind :o.

When you said "non-linear", the effect could be like this.


The correction come from an "old" lumagen and it was only on 100 %.

The picture at right is without correction and the picture at left is with correction at 100 %, but we can see there are "some" problems under :confused:.

I think you ask if it's possible to obtain a correction on 11 points, but if the problem are the same (non-linear effect), it is not possible to have a "minor" error between these 11 points, no ?

i have the radiance and use it with my pj as it was set up by jason at avs, but i could never do this stuff that you are showing. if the image ever needs adjusting how does one go about finding a radiancexd-ologist?

Hi,

thank for the answer. It's more clear in my mind :o.

When you said "non-linear", the effect could be like this.

The correction come from an "old" lumagen and it was only on 100 %.

The picture at right is without correction and the picture at left is with correction at 100 %, but we can see there are "some" problems under :confused:.

I think you ask if it's possible to obtain a correction on 11 points, but if the problem are the same (non-linear effect), it is not possible to have a "minor" error between these 11 points, no ?The images you show reveal very large errors. Ideally, all of the readings should fall on the same spot on the chart regardless of whether the level of stimulus is 25%, 50%, 75%, or 100%.

Here's what I got:

http://home.comcast.net/%7Etlhuffman/nonlinear.gif

This is actually a little better than I had originally thought. Assuming that the 75% stimulus readings are the reference and 1.5 is the maximum acceptable CIE94 color error, the primaries were fine, except red at 25%. It was the secondaries where problems began to arise, especially with yellow. I highlighted in red all the results that exceeded the maximum acceptable error.

We are considering making a change to the Radiance software that would give you the option of calibrating the primary and secondary colors at 75% or 100%. I added this to the Radiance list of requested features.

The idea of adding a 25% color calibration point to the RadianceXD is something we would need to look into further. It is an idea that is already under consideration for a Radiance Pro model.

Best regards,
Randy Freeman

We are considering making a change to the Radiance software that would give you the option of calibrating the primary and secondary colors at 75% or 100%. I added this to the Radiance list of requested features.

The idea of adding a 25% color calibration point to the RadianceXD is something we would need to look into further. It is an idea that is already under consideration for a Radiance Pro model.Randy, This is very interesting. Thanks for the update.

New Radiance software rev 110508 has been posted which implements many new RS232 commands. We also posted "Tech Tip 11 - RS232 Control For The Radiance".
http://www.lumagen.com/testindex.php?module=manuals

Let us know if there are any other RS232 commands or queries that would be useful.

Best regards,
Randy Freeman

We are considering making a change to the Radiance software that would give you the option of calibrating the primary and secondary colors at 75% or 100%. I added this to the Radiance list of requested features.
So at present what is the best level of stimulus to be calibrating colours?

So at present what is the best level of stimulus to be calibrating colours?Because of non-linear performance, best to avoid the extremes on either end. 75% stimulus is probably the best. It is just about exactly at the 50% output point of the display.

We are considering making a change to the Radiance software that would give you the option of calibrating the primary and secondary colors at 75% or 100%. I added this to the Radiance list of requested features.

The idea of adding a 25% color calibration point to the RadianceXD is something we would need to look into further. It is an idea that is already under consideration for a Radiance Pro model.

Best regards,
Randy Freeman


Randy

I'm not sure I follow the "75% or 100%". Does that mean at the moment there are additional issues using external 75% stim colour patterns as opposed to 100% stim ?

I can't speak for others but one of the primary reasons for my purchase of the Radiance XD was full CMS. If this is "flawed" in real world execution I would expect it to be fixed and not require the puchase a new "pro version" to deliver what was expected in the XD.

To avoid any confusion can we have statement from Lumagen re what should be expected of the Radiance XD CMS ? Jim commented on the Vision HDx range offering gamut "mitigation" but not full CMS does this now also extend to the Radiance in real world use ? :(

D

I'm not sure I follow the "75% or 100%". Does that mean at the moment there are additional issues using external 75% stim colour patterns as opposed to 100% stim ?

I can't speak for others but one of the primary reasons for my purchase of the Radiance XD was full CMS. If this is "flawed" in real world execution I would expect it to be fixed and not require the puchase a new "pro version" to deliver what was expected in the XD.

To avoid any confusion can we have statement from Lumagen re what should be expected of the Radiance XD CMS ? Jim commented on the Vision HDx range offering gamut "mitigation" but not full CMS does this now also extend to the Radiance in real world use ?Randy can speak for himself and Lumagen, but since I started this, I feel I should say something in Lumagen's defense. First, if you review my previous post (http://www.avsforum.com/avs-vb/showpost.php?p=15045760&postcount=3111) on this, the non-linear behavior I called attention to is a non-issue for the primaries. They are all within tolerances, except very dark reds. There is more of a problem with the secondaries, especially yellow for some reason.

Second, and more importantly, what the Radiance offers is the current SOTA for external color correction in a consumer device. There is nothing better, other than perhaps some tools intended for professional post production. Does that mean it is perfect and has no room for improvement? No.

Providing a multi-level CMS that made independent corrections at 25, 50, 75, and 100% would require significant additional work and I suspect perhaps some hardware changes. The improvement, though measurable, would be very small, relevant to only absolute purists.

I consider myself pretty anal about this stuff, and I am quite pleased with what the Radiance offers now. Although the added capability would certainly be nice, I doubt I'd be willing to pay for it. But, that's me

Randy can speak for himself and Lumagen, but since I started this, I feel I should say something in Lumagen's defense.

Tom

That's fine providing it's clear from the marketing material that it only works with displays that process colour in a certain fashion enabling an informed purchasing decision. I don't recall this being made clear from any of Lumagen's or Sencores marketing material.

This is from the mareting material on the Sencore site -

"Primary / Secondary Color Gamut Calibration (3D CMS Palette) – means: precise matching of SMPTE standard color space like 601, 709, xvYcc, etc."

"Exclusive 3D CMS Palette™ seven-point input/output color gamut calibration provides accurate gamut correction for all input signals and displays, with no control interaction"

Little point having "precise matching" if it's only accurate for the point that is measured. :confused:

Any idea what displays operate linear versus on non linear with regard to the Radiance CMS ?

D

That's fine providing it's clear from the marketing material that it only works with displays that process colour in a certain fashion enabling an informed purchasing decision. I don't recall this being made clear from any of Lumagen's or Sencores marketing material.

This is from the mareting material on the Sencore site -

"Primary / Secondary Color Gamut Calibration (3D CMS Palette) – means: precise matching of SMPTE standard color space like 601, 709, xvYcc, etc."

"Exclusive 3D CMS Palette™ seven-point input/output color gamut calibration provides accurate gamut correction for all input signals and displays, with no control interaction"

Little point having "precise matching" if it's only accurate for the point that is measured. :confused:

Any idea what displays operate linear versus on non linear with regard to the Radiance CMS?All of those marketing claims are true and not merely technically true but misleading the way many manufacturers advertise Deep Color and xvYCC color support are. The non-linear behavior I measured is small and is not something that is generally even measured.

All else being equal, a CMS built into the display will probably work better than an external device. But then you lose the flexibility of having a CMS you can hook up to any display.

The non-linear behavior I measured is small and is not something that is generally even measured.

All else being equal, a CMS built into the display will probably work better than an external device. But then you lose the flexibility of having a CMS you can hook up to any display.

Tom

Have you measured any display that had a linear response using the Radiance CMS ?

D

Have you measured any display that had a linear response using the Radiance CMS ?As an example of what I said below, no. This is not because I have measured several and none of them have. I just haven't taken the measurements. However, my guess is that this phenomenon is not display-dependent.

As an example of what I said below, no. This is not because I have measured several and none of them have. I just haven't taken the measurements. However, my guess is that this phenomenon is not display-dependent.

Tom

Thanks for clarifying.

D

Providing a multi-level CMS that made independent corrections at 25, 50, 75, and 100% would require significant additional work and I suspect perhaps some hardware changes. The improvement, though measurable, would be very small, relevant to only absolute purists.

Quick question. Do any non-Radiance implementations of CMS allow independent corrections at multiple stim levels or is this a general problem across all CMS systems??

Cheers,

Mark

Tom,

"Assuming that the 75% stimulus readings are the reference and 1.5 is the maximum acceptable CIE94 color error"

Does that 1.5 maximum acceptable CIE94 error take into account lumens? I'd assume perceptible color error varies quite a bit depending upon the overall brightness of an image. IOW one could likely see color problems at 1000 lumens easier then they could at 1 lumen for example.

If that doesn't take this into account then you do not accurately know what the color error is at 25 IRE and if it is outside of the perceptible range or not.

"a CMS built into the display will probably work better than an external device. "

Have you measured any built in CMS systems at 25,50,75 and 100 IRE to see what happens?

Shawn

Does that 1.5 maximum acceptable CIE94 error take into account lumens? I'd assume perceptible color error varies quite a bit depending upon the overall brightness of an image. IOW one could likely see color problems at 1000 lumens easier then they could at 1 lumen for example.dE only analyzes the brightness of a color relative to the brightness of white at the same level of stimulus. So, though 25% is a very dim color, it is represented as a percentage of 25% gray, so the value is the same as a 75% color. The eye's sensitivity to color at lower levels of stimulus is a different matter. It is most sensitive to green under these conditions.

Also, at lower levels of stimulus, there is less colorfulness to color. So in that sense the eye sees less color overall. This is partly because light is a component to colorfulness and partly because the human eye begins to lose color vision when light levels get low (at night the world appears more or less black-and-white). Nonetheless, saturation should measure the same at lower levels of stimulus. But your point is well-taken. Measurements of color are not typically taken at low levels of stimulus precisely because color perception at low light levels is less critical.

One might argue that, because of this, performance at 25% stimulus is not very important. However, at a minimum, a linear response at 75% and 100% stimulus would be desirable. My measurements indicate that the Radiance provides that for all colors except yellow and cyan.

Have you measured any built in CMS systems at 25,50,75 and 100 IRE to see what happens?As I indicated below, no I haven't taken this measurement before. But since this discussion seems to have stimulated interest in this, I will measure the Sharp 20K tonight and see what I get. I am a little concerned that a FP will have insufficient light output to measure a 25% signal, certainly with blue and perhaps with red and magenta as well.

Quick question. Do any non-Radiance implementations of CMS allow independent corrections at multiple stim levels or is this a general problem across all CMS systems??No CMS intended for the consumer market offers this that I know of, though some Pro tools might.

Tom,

"Nonetheless, saturation should measure the same at lower levels of stimulus. But your point is well-taken. Measurements of color are not typically taken at low levels of stimulus precisely because color perception at low light levels is less critical."

I am not disagreeing that in a perfect world saturation shouldn't be perfect across the board as should everything else. We are not in a perfect world though.

I am just suggesting that your number of 1.5 dE on CIE94 maximum acceptable error may not be an accurate metric at lower levels. That dE number had to be derived from visual 'taste testing.' If that was based on testing at 1000 lumens then at 10 lumens what is considered maximum acceptable could, and likey would, be a different number because of our eyes loss of color sensitivity at lower levels. It could be that your measured dE numbers are well under that max acceptable error at 25 IRE , it could be that they aren't. I just doubt a simple number of <1.5dE regardless of the lumens is going to be an accurate metric.

"But since this discussion seems to have stimulated interest in this, I will measure the Sharp 20K tonight and see what I get."

Thanks, it would be good to see this. There are already posts suggesting the Radiances CMS is 'broken' because of your initial post. Based on past discussions I think you would agree that is not at all the case.

If you have the time another measurement of your initial display at 25,50,75 and 100IRE with the Radiances CMS off would be interesting to see as well to see how pre/post measurements compare.

Thanks,

Shawn

Shawn:

I think we agree. As I said, I am not sure the color error at 25% is perceptually important, regardless of how the error is measured. However, linear behavior at 75% and 100% is, I think, important. I guess I'm agnostic about 50%.

When I began this discussion my model was grayscale tracking where white should remain a constant x0.3127, y0.329 from 20-100%. The difference with color measurements is that while the human eye becomes MORE sensitive to grayscale errors at low light levels it also becomes LESS sensitive to pri/sec color errors.

OK, I finally got around to measuring the Panasonic panel with the Radiance CMS disengaged. The results are interesting. First, the CIE chart of the primaries and secondaries at 25%, 50%, 75%, and 100% stimulus.

http://home.comcast.net/%7Etlhuffman/cie_no_cms.jpg

As you can see, the panel is nearly perfectly linear with respect to the xy coordinates.

However, when looking at a CIE94 comparison between the levels and 75%, which was assumed as the reference, we see some serious problems, especially at 100%. These problems are all due to luminance errors.

http://home.comcast.net/%7Etlhuffman/de_no_cms.gif

All in all a mixed bag. On the one hand, the Radiance CMS clearly introduces some non-linear behavior, especially with respect to secondary hues. On the other hand, it can provide nearly perfect color at a single level of stimulus and resolves color decoding problems at multiple levels of stimulus that at least this panel was afflicted with.

The bottom line, though, is that its CMS dramatically improves color performance, despite these relatively small problems.

please forgive me if I looked over this and missed it guys ;)

can the Radiance do 1920x1080p/72hz output?

thanks

-Gary

The Radiance has a maximum output of 1080p60. Some displays, when given a 1080p24 input from the Radiance, will actually drive the panel at 1080p72 or 1080p96.

Best regards,
Randy Freeman

thanks Randy

-Gary

Gary the only Scalers out right now that can output 1080p@72 is the TV-One C2-225x series, they pushed it to 1080p85 for fun to see how high it could go, but this isn't really based for HT unless you want to blend, still wish Lumagen would put a blend function on the radiance.

Athanasios

thanks for the info Athan

-Gary

Any updates on the developement of the Radiance XS?

Is this the upgrade to 1.3 status? If it is, I think it is the XE. If not, I'm not sure what the XS is. I have some info on the XE if it is appropriate.

No the XS is the lower cost single output version of the Radiance with less features. The XE is the upgraded XD, dual output. Someone correct me if I am wrong?

We are looking at the RadianceXS model but, at this time, there is no schedule for the release of this product. We will post more information when it becomes available.

Best regards,
Randy Freeman

support@lumagen.com

It'll only cost $100000 instead of $50000000000000. They don't want to kill their sales, even though Bill Gates would have to save up for the thing as-is.

The Radiance has a maximum output of 1080p60. Some displays, when given a 1080p24 input from the Radiance, will actually drive the panel at 1080p72 or 1080p96.

Best regards,
Randy Freeman

Randy the Radiance should have the bandwidth to perform 1920x817@72. In the timing menu I can make all the needed changes except for refresh. It is lock only allowing to toggle between 48/50/60.
Is there anyway to dial in 72hz?

Thx.

Actually, I think the XD does not have the bandwith, but I could be wrong. Randy or Jim can confirm one way or the other, but I'm pretty sure the topic was brought up a while back in the beta forum.

Randy the Radiance should have the bandwidth to perform 1920x817@72. In the timing menu I can make all the needed changes except for refresh. It is lock only allowing to toggle between 48/50/60.
Is there anyway to dial in 72hz?

Thx.

Alan,

I adjusted this via the pixel clock - write down the pixel clock at 48 and multiply by a factor of 1.5 and you get 72 Hz. Pixel clock will be fine with this, so you can easily use that resolution and refresh.

I would of course suggest to go for 71.93 for film based stuff.

If you need exact timings just shoot me a pm :)

Oliver

I'm glad I was wrong then. :)

Alan,

I adjusted this via the pixel clock - write down the pixel clock at 48 and multiply by a factor of 1.5 and you get 72 Hz. Pixel clock will be fine with this, so you can easily use that resolution and refresh.

I would of course suggest to go for 71.93 for film based stuff.

If you need exact timings just shoot me a pm :)

Oliver

Hi Oliver, Yes 71.93 is what I want:) Thanks for this tip I never would have figured that out. I wish there was a simple selection for 71.93 after choosing a custom res below the bandwidth limit but at least there is this work around. Ill give this a run later today!!
Doing some quick math the pixel clock for 47.95 is 118681319 x 1.5 = 178021978.5
Not sure if the .5 is important or if the lumi will allow me to enter it but Ill give this a try.

Thank you!

I added this mode to the list. We will check it out and see how much work it will take to implement.

Best regards,
Randy Freeman

Hi Oliver, Yes 71.93 is what I want:) Thanks for this tip I never would have figured that out. I wish there was a simple selection for 71.93 after choosing a custom res below the bandwidth limit but at least there is this work around. Ill give this a run later today!!
Doing some quick math the pixel clock for 47.95 is 118681319 x 1.5 = 178021978.5
Not sure if the .5 is important or if the lumi will allow me to enter it but Ill give this a try.

Thank you!

Alan,

both 1978 and 1979 should work fine, you cannot and need not enter .5 values.

Oliver

I recently purchased the Pioneer Pro FP-J1 and I am now thinking of adding the Radiance XD. My Understanding Is that with the Lumagen added to the Pioneer there is more range to calibrate the pioneer to standards like Gray Scale and CMS.
So I have a few questions and I am hoping to be guided in the right direction.

The Pioneer is a clone of the JVC RS-2 will the lumagen work will with the Pioneer?
I just upgraded from a Sony G-70-CRT to the Pioneer and except for the extended color on the Pioneer I have been very happy with the unit. Will the addition of the Lumagen allow for 6500K Gray scale the entire range? Will the CMS in the Lumagen allow for adjustments of the primary and secondary colors to standards?

I recently purchased the Pioneer Pro FP-J1 and I am now thinking of adding the Radiance XD. My Understanding Is that with the Lumagen added to the Pioneer there is more range to calibrate the pioneer to standards like Gray Scale and CMS.
So I have a few questions and I am hoping to be guided in the right direction.

The Pioneer is a clone of the JVC RS-2 will the lumagen work will with the Pioneer?
I just upgraded from a Sony G-70-CRT to the Pioneer and except for the extended color on the Pioneer I have been very happy with the unit. Will the addition of the Lumagen allow for 6500K Gray scale the entire range? Will the CMS in the Lumagen allow for adjustments of the primary and secondary colors to standards?

Absolutely! It's the best system available to do what you want.

I recently purchased the Pioneer Pro FP-J1 and I am now thinking of adding the Radiance XD. My Understanding Is that with the Lumagen added to the Pioneer there is more range to calibrate the pioneer to standards like Gray Scale and CMS.
So I have a few questions and I am hoping to be guided in the right direction.

The Pioneer is a clone of the JVC RS-2 will the lumagen work will with the Pioneer?
I just upgraded from a Sony G-70-CRT to the Pioneer and except for the extended color on the Pioneer I have been very happy with the unit. Will the addition of the Lumagen allow for 6500K Gray scale the entire range? Will the CMS in the Lumagen allow for adjustments of the primary and secondary colors to standards?

the Radiance will certainly whip that Pioneer into shape ;)

-Gary

The only whipping that will be done will be by your wife when she finds out how much you spent on it :D

I am just starting to try and figure out how to use the CMS, and am going to attempt to do a grayscale calibration. Are all the test patterns that I need available through the Radiance?

I think I just need a 100 IRE white window, and then 10% steps down from that. In the quick look I did at the test patterns, I saw a Gray window. Is that white or should I be looking elsewhere?

MENU → Output → Configs → [Config] → Color Mgmt→ Color Temp → OK

A couple updates ago Color Temp was renamed Gray/Gamma so that's what you might see.

FYI: if you intend to perform a full calibration, it is recommended to do the gamut before tuning the grayscale.

Is there anyone in NY NJ that can calibrate the Lumagen on my Pioneer / RS2?

MENU → Output → Configs → [Config] → Color Mgmt→ Color Temp → OK

A couple updates ago Color Temp was renamed Gray/Gamma so that's what you might see.

FYI: if you intend to perform a full calibration, it is recommended to do the gamut before tuning the grayscale.


Thanks Dave. I don't have the Radiance in front of me right now to verify, but I assumed the menu link you gave me is where I would set the values... are you saying this is how I get to the patterns to do the measurements etc?

I thought there would be a pattern that showed a 100 IRE White window. The one I looked at last night was labeled "Gray"... just curious if Gray at 100% and White are the same thing?

The reference patterns will show you a window (yes 100% stim gray and white are the same), but you need to be in the Gray/Gamma screen to access the XD's controls for adjusting the grayscale. This screen includes the necessary patterns, and there are control to switch between full field, window, small window, and external pattern, as well as the 11 point of the grayscale (0%, 10%, ... , 100%). So to answer your question: yes - the screen I pointed out includes everything you need.

As I said above, ideally you should calibrate the color gamut before attempting the grayscale (not obvious since the workflow's definitely more complex). If not, then at the very least use your display's controls to calibrate 100% white, and for that you should use the pattern you mention: MENU → Other → Test Pattern → Reference → OK and look at the 100% gray pattern. I think '4' on the remote lets you switch between field aka full screen and window patterns, and up/down arrows let you change the % stim in 5% increments.

Long story short: calibrate 100% white point with your display's RGB controls, then gamut if you can, then grayscale/gamma.

Cool, thanks for the info. I figured that was the guy I wanted, but I'd hate to start taking measurements off the wrong thing.

I wasn't aware that the color gamut should be calibrated first, so I will give that a go. I definitely have a lot to learn.

I am trying to learn as much as I can, and using
Greyscale and Colour Calibration for Dummies (http://www.curtpalme.com/forum/viewtopic.php?t=10457) and Basic Guid to Color Calibration using a CMS (http://www.avsforum.com/avs-vb/showthread.php?t=852536) to get me going. I also just found the Gamut Calibration pdf (http://www.lumagen.com/testindex.php?module=manuals) linked from the Lumagen site.

If you have access to the beta forum there is also a thread started by VirusKiller:

http://www.convergent-av.co.uk/forum/index.php?t=msg&th=1356&prevloaded=1&rid=528&S=1f077379b9590c3b953265e901897c0f&start=0

(You won't be able to access it if you're not allowed in the beta forum.)

It describes a workflow pretty similar to the gamut calibration pdf from Lumagen, with a couple tweaks. The thread references a couple bugs in the Radiance that have long been eradicated. It needs to be brought up to date and made available to a larger audience. :)

PS: while the generic CMS 'for dummies' type threads can be useful for learning about CMS and calibration in general, I'd be careful about relying on them too much when working with the XD, because they generally assume a calibration done with the usually limited controls found within a display, not an external box.

I am still on the 111808 firmware (looks like a new one was just released today, so I will try that). But I am wondering how well the 480i60 to 1080p24 is supposed to work.

I am using an Oppo 980 outputting 480i over HDMI as a source, and a JVC RS1 as a display. I've not been able to watch an entire movie without the display getting jerky. Some movies it will be fine for several minutes, then go jerky, others I've only had it happen a couple of times. Either way, a pause of the movie and play will usually get it going again.

Is there a setting I am missing, or are the bugs still being worked out, or is this how things are just supposed to be?

What is the genlock setting supposed to be set to for this kind of configuration? It is not quite clear to me what this setting is supposed to do.

I also have a Toshiba HD-A1 set to output 1080i60. I've had better success with this, but still not perfect.

It needs to be brought up to date and made available to a larger audience. :)Tell me about it... ;)

To get correct color decoding from DVD and other SD sources....

From reading the manual, it appears that setting the INPUT matrix to Bt601/SMPTE-C and the OUTPUT to Bt709/HD will provide correct decoding of SD sources so that SD reds don't appear too orange. Is that all there is to it, or is some other setting/calibration needed?

To get correct color decoding from DVD and other SD sources....

From reading the manual, it appears that setting the INPUT matrix to Bt601/SMPTE-C and the OUTPUT to Bt709/HD will provide correct decoding of SD sources so that SD reds don't appear too orange. Is that all there is to it, or is some other setting/calibration needed?

For SD sources I think it would be pretty safe to leave the format to 'Auto' unless you've had issues with this setting.

I guess "that's all there is to it" as far as color space conversion is concerned. Now whether your reds appear not too orange will depend mostly on your display's performance, but of course that can be addressed with the CMS. I'm sure you knew that. :)

I am still on the 111808 firmware (looks like a new one was just released today, so I will try that). But I am wondering how well the 480i60 to 1080p24 is supposed to work.

I am using an Oppo 980 outputting 480i over HDMI as a source, and a JVC RS1 as a display. I've not been able to watch an entire movie without the display getting jerky. Some movies it will be fine for several minutes, then go jerky, others I've only had it happen a couple of times. Either way, a pause of the movie and play will usually get it going again.

Is there a setting I am missing, or are the bugs still being worked out, or is this how things are just supposed to be?

What is the genlock setting supposed to be set to for this kind of configuration? It is not quite clear to me what this setting is supposed to do.

I also have a Toshiba HD-A1 set to output 1080i60. I've had better success with this, but still not perfect.

I haven't used the genlock setting yet, and I don't know much about it except that it's supposed to lock in the cadence somehow so that no frames are lost or repeated... which clearly doesn't seem to be working in your case. I would email Lumagen support - attach your config file.

I am using an Oppo 980 outputting 480i over HDMI as a source, and a JVC RS1 as a display. I've not been able to watch an entire movie without the display getting jerky. Some movies it will be fine for several minutes, then go jerky, others I've only had it happen a couple of times. Either way, a pause of the movie and play will usually get it going again.I've spotted this too with the latest firmware. I also see periodic problems with 1080i from my Toshiba XA1 HD DVD player. I will be feeding this information back to Lumagen.

I've spotted this too with the latest firmware. I also see periodic problems with 1080i from my Toshiba XA1 HD DVD player. I will be feeding this information back to Lumagen.

Cool, keep us updated. Are you going to post over on the Lumagen forum, or communicating with support directly? If posting, I'll be happy chime in with data points if it will help.

The RadianceXE Beta with HDMI 1.3 is available and shipping. Anyone who wants more information about the RadianceXE Beta should contact Lumagen.

Best regards,
Randy Freeman

support@lumagen.com
503-574-2211

Cool, keep us updated. Are you going to post over on the Lumagen forum, or communicating with support directly?I emailed Lumagen with a reproducible Oppo 980 / 480i stutter that I've found with the extended edition of Lord of the Rings: FOTR. Turns out that they were already aware that this was problematic. Any additional data points will be useful to them though.

I emailed Lumagen with a reproducible Oppo 980 / 480i stutter that I've found with the extended edition of Lord of the Rings: FOTR. Turns out that they were already aware that this was problematic. Any additional data points will be useful to them though.

I watched that movie the other day, didn't notice anything. When/how does it happen?

I watched that movie the other day, didn't notice anything. When/how does it happen?Lord of the Rings: Fellowship of the Ring (R1 Extended Edition ISBN 0-7806-3893-X)
Radiance firmware: 121208
Source: Oppo 980 (YCbCr 4:2:2 480i output with PCM over HDMI)
Display: JVC HD1
Genlock: Auto (i.e. on)
Deinterlacing: Film
Output: 1080p24 4:2:2 with 10-bit dither.

Stuttering started at the beginning of Disc 2 Chapter 33 "Moria" and seemed quite reproducible. I've also seen combing instead of stuttering.

I am not using genlock, that might be why I'm not seeing it.

I'll try the same scene. I did an experiment this morning using Raiders of the Lost Ark. With genlock off, it started getting jerky in the middle of chapter 5. I skipped back to the start of the chapter, and it did not get jerky in the same spot.

I turned genlock on (usually this is less stable for me) and started the movie over. Somewhere around chapter 15 is when it started getting jerky.

So, in my case, it doesn't really happen in particular spots, it just happens.

I'll start a thread over in the Beta forum.

So am I correct in saying that stable 1080p24 output from film sources (including broadcast sources) is still a work-in-progress on the Radiance?

Also, it's recently been indicated to me in another thread that the VXP 9350 issue with the HQV HD film resolution loss test has been confirmed to be fixed on the VXP9450/Radiance? The 9350 produced vertical bands on the horizontal line boxes in all four corners of the SMPTE pattern. It seems a little strange that after all of this time there's no further mention of this after searching this thread. I'd also be interested in knowing how the Radiance performs on both of the HQV SD/HD video deinterlacing "jaggies" tests particularly from someone who has seen the same test with Reon, Realta, ABT 2010, Pio Kuro or VXP 9350.

So am I correct in saying that stable 1080p24 output from film sources (including broadcast sources) is still a work-in-progress on the Radiance?

Also, it's recently been indicated to me in another thread that the VXP 9350 issue with the HQV HD film resolution loss test has been confirmed to be fixed on the VXP9450/Radiance?

I've just watched Transformers HD DVD at 1080i/60 from the player with the Radiance performing IVTC to 1080/23.98. Can't comment on broadcast film as mine don't use 3:2.

Out of interest I also just tried the HD DVD at 1080i/60in>1080i/60 out of the Radiance to my PJ (11S2) which uses a the older Gennum 9351. The fine horizonal lines in corner squares are visible in the SMPTE video (no strobing lines) and teh film res tests (EDIT - PJ film res test does appear to have strobing lines around the fine lines unlike the same test deinterlaced via the Radiance which passes). There is also no moire on the stadium panning shot although at 60hz the motion is juddery compared to 23.98 via the Radiance.

D

I've just watched Transformers HD DVD at 1080i/60 from the player with the Radiance performing IVTC to 1080/23.98. Can't comment on broadcast film as mine don't use 3:2.

Out of interest I also just tried the HD DVD at 1080i/60in>1080i/60 out of the Radiance to my PJ (11S2) which uses a the older Gennum 9351. The fine horizonal lines in corner squares are visible in the SMPTE video and film res tests. .
D

The fine horizontal lines must be visible, but there must also be no vertical bands/strobing at the box's edges. The vertical band/strobing is a recognized "failure mode" for this test and indicates half-resolution processing. I can reproduce this failure mode exactly by placing my Realta-based Denon DVP-602ci in "video" deinterlacing mode for this test.

Am I correct in that you are saying that the Radiance and your 9351 based pj perform similarly for this test?

Am I correct in that you are saying that the Radiance and your 9351 based pj perform similarly for this test?

No. :)

Using the Radiance to deinterlace 1080i/60 HD DVD HQV there are no vertical strobing bands/lines around the fine horizonal lines that I can see in the video or film res test. The film test motion is also smoother using 1080p/23.98 rather than 1080p/60.

The PJ deinterlacing has no vertical stobing lines in the video resolution test. On closer inspection it does appear to have a strobing outside lines around the fine lines on the film res test which was masked a little by the juddery motion so my bad. There is no moire/shimmer visible on the stadium pan but motion is a bit juddery due to 60hz.

D

Hi everybody,

I finally got my Radiance hooked up to my RS1 and used it to calibrate the greyscale and gamut with CalMAN and my i1 (I really do need to step up to an i1Pro), and have been very impressed with the results so far. I'm about 75% done, and have a couple questions.

From what I read earlier in this thread, I understand that decreasing the Y of the secondaries just involves adding equal parts of the adjacent primaries - is this correct? Now for the dumb question. To reduce Y for a primary using the Matrix mode, would I simply decrease the values of R for red, B for blue, and G for green? Or am I trying to make this too simple and missing something? Playing with the Lumagen's CMS has already taught me a lot, one thing being that I have a LOT more to learn about the underlying principles that govern properly calibrating a display - but it's been a lot of fun learning with the help of AVS so far. Thanks,

HP

From what I read earlier in this thread, I understand that decreasing the Y of the secondaries just involves adding equal parts of the adjacent primaries - is this correct?

Hmmm. I don't think that's correct. In fact I think the opposite is true :).

Try this calculator:
http://www.learn.londonmet.ac.uk/student/resources/radiance/colourcalc.shtml

Enter 0.5 0.5 0.0 in the top line (RGB) - that's yellow. Observe the conversion to xyY a few lines below. Then change the RGB line to 0.7 0.7 0.0 and convert again - Y shots up dramatically.

So - I'm not Charles Poynton, but this is the way I understand it: we're manipulating RGB values while calibrating (talking about the XD here) but are trying to reach points in the xyY space. The Y (luma) linearly depends upon R, G and B, i.e. Y=a*R + b*G + c*B. I don't remember off the top of my head the values for a, b, and c, but it doesn't matter what they are, although it is useful to know that b > a > c > 0. What does it all mean? Any increase in any of the primaries will increase the luma. However an increase in green will affect it more than an increase in red, which will affect it more than an increase in blue. (I hope I got the order right, anybody feel free to jump in and correct me).

Conclusion: you can't decrease Y by increasing the primaries. You need to decrease them. It's hard because when calibrating you need your x and y to stay constant while manipulating Y.

(I think people may be confused by the fact that if you look at the xy chart in 2D, adding say R and B to the mix looks like you're decreasing G because your point is moving away from green, but you're not! It's just a different mix with the same amount of green.)

Now for the dumb question. To reduce Y for a primary using the Matrix mode, would I simply decrease the values of R for red, B for blue, and G for green? Or am I trying to make this too simple and missing something? Playing with the Lumagen's CMS has already taught me a lot, one thing being that I have a LOT more to learn about the underlying principles that govern properly calibrating a display - but it's been a lot of fun learning with the help of AVS so far. Thanks,

HP

The matrix is not meant to be used during calibration, it's just a summary of your calibration values (so you can pass them around easily for example). You want to use the color screens for that. As for reducing Y for a primary you are correct the easiest way is to reduce the value for the primary itself, but again this will move the xy values so you need to fidget with the other two primaries at the same time to keep x and y constant. That's assuming you've already reached the correct xy values for your primaries, which you're supposed to do in the calibration process before worrying about Y.

Pfew :) Hope this helps.

I am sorry if this has recently been covered,but I am wondering what my choices are for a Bluray player with Native output including 480i? It would seem a waste to not deinterlace with the Radiance?
The future Oppo Bluray will at a very reasonable price,but I am having trouble finding any others?
I would prefer not to use 2 players. I could do it,but my clients get confused enough with one player!
Thanks

I am sorry if this has recently been covered,but I am wondering what my choices are for a Bluray player with Native output including 480i? It would seem a waste to not deinterlace with the Radiance?
The future Oppo Bluray will at a very reasonable price,but I am having trouble finding any others?
I would prefer not to use 2 players. I could do it,but my clients get confused enough with one player!
Thanks

I am not sure which ones do, but I don't think it will be that common of a feature. If you don't get any answers here, you should try asking in the Blu-ray Players forum (http://www.avsforum.com/avs-vb/forumdisplay.php?s=&daysprune=&f=149).

Even if there are some that do, I would be surprised if they would automatically switch from outputting 1080p to 480i based on the source media. So your clients would have to do some setting toggling based upon what kind of disc they are using. Might be easier to just tell them to put regular DVDs in the DVD player, and blu-ray movies in the blu-ray player, and not have to fiddle with settings, or remember which button on their remote to hit (assuming you programmed one for them).

I am sorry if this has recently been covered,but I am wondering what my choices are for a Bluray player with Native output including 480i? It would seem a waste to not deinterlace with the Radiance?
The future Oppo Bluray will at a very reasonable price,but I am having trouble finding any others?
I would prefer not to use 2 players. I could do it,but my clients get confused enough with one player!
Thanks

Sony s350 and s550 both do 480i HDMI.
Pio 51 and the equivalent Elite version also do.
Older Sony s300 does as well as even older bdp-s1.
The Oppo 83 does as you mentioned.
Denon A1 (upcoming) also will.

Sony s350 and s550 both do 480i HDMI.
Pio 51 and the equivalent Elite version also do.
Older Sony s300 does as well as even older bdp-s1.
The Oppo 83 does as you mentioned.
Denon A1 (upcoming) also will.
do these players switch output automatically without the need to manually switch between modes ?
ie when a DVD is used they put out 480i .... when a bluray is used they put out 1080p24 ?

do these players switch output automatically without the need to manually switch between modes ?
ie when a DVD is used they put out 480i .... when a bluray is used they put out 1080p24 ?

The auto res selection you describe above is usually referred to as a "source direct" feature.

IIRC the Sony 350/550 don't have a source direct but earlier models S1, S300 did. The Denon 2500BT also doesn't support source direct but does offer 480i/1080i manually. I think the Pioneer model mentioned does support source direct but check you're happy with it's speed of operation.

D

Oppo 83 also has source direct.

Sony s350 and s550 both do 480i HDMI.
Pio 51 and the equivalent Elite version also do.
Older Sony s300 does as well as even older bdp-s1.
The Oppo 83 does as you mentioned.
Denon A1 (upcoming) also will.

Thanks! I am looking for an automatic "native" player. I believe the newer Sony's have to be set manually for your desired output. Too much work :)
Do the Pioneers automatically select the "native" resolution of the disc?

Thanks! I am looking for an automatic "native" player. I believe the newer Sony's have to be set manually for your desired output. Too much work :)
Do the Pioneers automatically select the "native" resolution of the disc?

My Pioneer HD1 does. I think later models do as well. Check their threads in the Blu-ray player section.

Oppo 83 also has source direct.

The Beta OPPO BDP-83 also has Source Direct which delivers the native signal, even at 24P ;)

- Rich

I am sorry if this has recently been covered,but I am wondering what my choices are for a Bluray player with Native output including 480i? It would seem a waste to not deinterlace with the Radiance?
The future Oppo Bluray will at a very reasonable price,but I am having trouble finding any others?
I would prefer not to use 2 players. I could do it,but my clients get confused enough with one player!
Thanks

You really have to look at each disk player individually for 2 things:

- A "native" or "direct" mode (Sony uses "direct" on players that have that capability)

- 480i output capability.

Some players with direct/native modes will output 480p but not 480i and some players with 480i output capability have no direct/native mode setting.

I don't know anywhere you can find that level of detail without researching every player - unless somebody just happens to have a player that works the right way and knows it. The discontinued first gen Sony BDP-S1 would work the way you want it to work, not sure about subsequent models. The Playstation 3 doesn't currently offer 480i output, nor does it have a direct/native mode (kind of annoying considering everything else it can do).

The Playstation 3 doesn't currently offer 480i output, nor does it have a direct/native mode (kind of annoying considering everything else it can do).

Any of the players mentioned above with "source direct" will automatically provide 1080p24 for BD film, 1080i for BD video, and 480i for DVD over HDMI.

The PS3, although not capable of 480i HDMI, can be configured to automatically provide 480p for DVD, 1080i for BD video, and 1080p24 for BD film over HDMI.

Any of the players mentioned above with "source direct" will automatically provide 1080p24 for BD film, 1080i for BD video, and 480i for DVD over HDMI.

The PS3, although not capable of 480i HDMI, can be configured to automatically provide 480p for DVD, 1080i for BD video, and 1080p24 for BD film over HDMI.

Thanks everyone for the great insight. I am going to try a Pioneer 51Fd with a Radiance.It seems like a good unit overall.

Pfew :) Hope this helps.

Yes, you helped immensely - thanks much.

No, R:1, G:1, B:0 is yellow.

Lowering the value of R when the starting point is R:1, G:0, B:0 (Red) will not affect the xy of red, just the luminance (as the calculator you referenced shows).

The original poster is correct. To lower the luminance of a secondary, you lower, by equal amounts, the values of the contributing primaries.

Use the calculator you reference. Begin with R:1, G:1, B:0 (Full yellow). This yields
x0.4193
y0.5052
Y0.9278 (normalized).

If reduced to R:0.8, G:0.8, B:0, then you get
x0.4193
y0.5052
Y0.5602

-TH

Hmmm. I don't think that's correct. In fact I think the opposite is true :).

Try this calculator:
http://www.learn.londonmet.ac.uk/student/resources/radiance/colourcalc.shtml

Enter 0.5 0.5 0.0 in the top line (RGB) - that's yellow. Observe the conversion to xyY a few lines below. Then change the RGB line to 0.7 0.7 0.0 and convert again - Y shots up dramatically.

So - I'm not Charles Poynton, but this is the way I understand it: we're manipulating RGB values while calibrating (talking about the XD here) but are trying to reach points in the xyY space. The Y (luma) linearly depends upon R, G and B, i.e. Y=a*R + b*G + c*B. I don't remember off the top of my head the values for a, b, and c, but it doesn't matter what they are, although it is useful to know that b > a > c > 0. What does it all mean? Any increase in any of the primaries will increase the luma. However an increase in green will affect it more than an increase in red, which will affect it more than an increase in blue. (I hope I got the order right, anybody feel free to jump in and correct me).

Conclusion: you can't decrease Y by increasing the primaries. You need to decrease them. It's hard because when calibrating you need your x and y to stay constant while manipulating Y.

(I think people may be confused by the fact that if you look at the xy chart in 2D, adding say R and B to the mix looks like you're decreasing G because your point is moving away from green, but you're not! It's just a different mix with the same amount of green.)



The matrix is not meant to be used during calibration, it's just a summary of your calibration values (so you can pass them around easily for example). You want to use the color screens for that. As for reducing Y for a primary you are correct the easiest way is to reduce the value for the primary itself, but again this will move the xy values so you need to fidget with the other two primaries at the same time to keep x and y constant. That's assuming you've already reached the correct xy values for your primaries, which you're supposed to do in the calibration process before worrying about Y.

Pfew :) Hope this helps.

No, R:1, G:1, B:0 is yellow.R:0.7 G:0.7 B:0 is yellow when transcribed to the xy chart. I see what you mean though.

Lowering the value of R when the starting point is R:1, G:0, B:0 (Red) will not affect the xy of red, just the luminance (as the calculator you referenced shows).
Of course! But I was talking about a real calibration. You're assuming an already perfectly saturated red (xy-wise). In a real calibration, lowering R will most certainly move R's xy

The original poster is correct. To lower the luminance of a secondary, you lower, by equal amounts, the values of the contributing primaries.
That's not what HogPilot said. He thought that the luminance would be lowered by increasing the adjacent primaries.

And again, you're talking about already perfect colors, in which case there'd be no need for calibration. When you get to the secondaries, lowering both adjacent primaries by the same amount will move the secondary's xy - I guarantee you that. Unless you started with an already perfect secondary. Not likely.

Dave and Tom - thanks again for the help. I sat down with my i1 and CalMAN last night and got everything dialed in - now I finally get to enjoy the HT a little before my i1Pro gets here and I tweak everything. Talk about a great learning experience - hands-on time with a full CMS really teaches you a lot about the calibration process, and how controls interact with one another.

Hey guys, I finally go a Radiance to adjust the color gamut of the RS2 (over saturated red and greens). I'm not sure if this has been addressed before, but how exactly do you tame these colors? I have everything else dialed in however I'm not sure where to start to adjust the gamut. I see color gamut settings in the radianceXD menus, but not sure how to use them... anyone has some starting points?

Cheers!
Danny

Hey guys, I finally go a Radiance to adjust the color gamut of the RS2 (over saturated red and greens). I'm not sure if this has been addressed before, but how exactly do you tame these colors? I have everything else dialed in however I'm not sure where to start to adjust the gamut. I see color gamut settings in the radianceXD menus, but not sure how to use them... anyone has some starting points?

To desaturate a primary, add equal amounts of the of the other 2 primaries.
To desaturate a secondary, add the opposing primary.

Hey guys, I finally go a Radiance to adjust the color gamut of the RS2 (over saturated red and greens). I'm not sure if this has been addressed before, but how exactly do you tame these colors? I have everything else dialed in however I'm not sure where to start to adjust the gamut. I see color gamut settings in the radianceXD menus, but not sure how to use them... anyone has some starting points?

Cheers!
Danny

I'm not sure what you're asking for; a primer on gamut calibration, or a primer on using the XD for gamut calibration?

If the latter: http://www.lumagen.com/docs/Tip0002_GamutCalibration.pdf

If the former - I recommend Calman's help pages. The Calibration forum here on AVS is also a good starting point. :)

I'm not sure what you're asking for; a primer on gamut calibration, or a primer on using the XD for gamut calibration?

If the latter: http://www.lumagen.com/docs/Tip0002_GamutCalibration.pdf

If the former - I recommend Calman's help pages. The Calibration forum here on AVS is also a good starting point. :)

Guess that's the problem, not sure what I need to do to bring these colors into spec:

http://www.mavromedia.com/tofrom/rs2/color.jpg

So you know how to take measurements and use calibration software?

So you know how to take measurements and use calibration software?

Of course! Just not sure about how to do gamut adjustments to bring the over sat. colors down.

Of course! Just not sure about how to do gamut adjustments to bring the over sat. colors down.

The document I linked should help then. Any particular step you need help with?

The document I linked should help then. Any particular step you need help with?

I'll read that document and let you know... thanks!

Of course! Just not sure about how to do gamut adjustments to bring the over sat. colors down.


It's no different than making grayscale steps accurate... you have target coordinates for each color (your calibration software may supply them, if not, you can find the target values online fairly easily with a search).

The grayscale points are 3-dimensional if you use a target gamma to set an accurate gamma using the 11-step grayscale/gamma adjustments.

Color coordinates are also 3-dimensional so if you have xyY reference coordinates, you will be adjusting the color controls to make all 3 coordinates as accurate as you can.

You can also use the AccuPel Calibration calculator (free to download at www.accupel.com (http://www.accupel.com) in the firmware download section) if your calibration software isn't very helpful with 3-dimensional calibration (some software is very 2-dimensional... they only deal with xy which won't help you get gamma and you won't have the info/feedback you need to make the colors accurate.

If you are doing gamma with grayscale... you enter the gamma value you would like to have like 2.2 or 2.3 and after making a measurement of a grayscale point like white, the softwar will calculate luminance targets for each grayscale step. If you set the luminance correctly for each step, you'll achieve the gamma you wanted (2.2 or 2.3 or whatever).

Similarly, to do colors, you need a white measurement first. From the white measurement, the luminance values for each color are calculated and you use the reference xy coordinates for each color to set each color point as close to perfect as you can... xy = 2-dimensional, xyY=3-dimensional. CalMAN, for example, does this all automatically for you. Some other software doesn't.

Also... you have to decide the stimulus level you are going to use and sticl with it. I prefer 75% windows... you really need about that level for plasmas and for other displays, 75% puts your measurement point away from the peak white level so you have an opportunity to make your one set-point for each color's luminance a nice "average" setting over the whole range of luminance (colors have a luminance range for each color just like the grayscale luminance changes for each measured step - but we only have one point where you can ADJUST color luminance so it's better that you not pick a very low or very high white luminance point. But whatever you do, if you pick 82% for white, you also have to be certain you are using 82% for each color also or you will be all messed up (I just picked 82% at random... I use 75% and think it's a generally 'safe' value for any type of display).

I have a question concerning the HDMI 1.3 version of the Radiance. If I send a 16-Bit signal to the processor what will happen to the data during the "processing-pipeline"? The narrowest part has a width of 10-Bit (or 12-Bit?) - will the least significant bits be discarded?

I'm asking this because I have an offer for the HDMI 1.1 version that I will take. I don't care about deep color and xvYCC but I want to avoid any loss of tonal value during processing of a 8-Bit signal. Is the "old" Radiance able to output in YCbCr 4:2:2 10-Bit and is YCbCr 4:2:2 the internal processing format?

Best regards

Denis

Is anyone here planning on having their XD sent in to be upgraded to the XE? I already talked to Lumagen and got the pricing from them, and according to the lady I talked to the only difference is that the XE is HDMI 1.3 She did caveat that by saying there could be other differences that she wasn't aware of.

Aside from being able to pass all audio codecs in bitstream form to a compatible receiver, can anyone here think of any real reason to upgrade to the XE?

"Is anyone here planning on having their XD sent in to be upgraded to the XE?"
Not exactly. I have received the board and will be installing it myself.

'I already talked to Lumagen and got the pricing from them."
For reference - the default price is cheaper if the black faceplate is not included.

"the lady I talked to"
That would be Jim's wife, Kim. She's wonderful.

"the only difference is that the XE is HDMI 1.3"
That is correct. And that has not changed.

"She did caveat that by saying there could be other differences that she wasn't aware of."
She is being very careful and making sure there is no chance of the customer feeling misled down the road. Of course there is always the possibility of SOME difference at some point, but it would almost certainly be insignificant. If it were anything else I'd be very surprised it Lumagen wouldn't make any significant update available. Note that this 1.3 update is at the cost, only, to the company. Given the fact that the exchanged boards will only be used in otherwise-new units sold as "demos" it is very likely that Jim (and Kim) end up losing on the swap.

"Aside from being able to pass all audio codecs in bitstream form to a compatible receiver, can anyone here think of any real reason to upgrade to the XE?"
Maybe real, but definitely NOT justifiable or even rational. :)
Seriously. Lumagen can't even provide such a reason. But again, they are an honorable group!

I have a question concerning the HDMI 1.3 version of the Radiance. If I send a 16-Bit signal to the processor what will happen to the data during the "processing-pipeline"? The narrowest part has a width of 10-Bit (or 12-Bit?) - will the least significant bits be discarded?

I'm asking this because I have an offer for the HDMI 1.1 version that I will take. I don't care about deep color and xvYCC but I want to avoid any loss of tonal value during processing of a 8-Bit signal. Is the "old" Radiance able to output in YCbCr 4:2:2 10-Bit and is YCbCr 4:2:2 the internal processing format?

Best regards

Denis

I think the XD can output YCbCr 4:2:2 10 and 12 bits.

The CMS calculations are done in RGB.

Aside from being able to pass all audio codecs in bitstream form to a compatible receiver, can anyone here think of any real reason to upgrade to the XE?

No.

Being able to pass all audio codecs in bitstream format is not valueless. There may be (I'm pretty sure there already are) media players that can deliver those streams over HDMI but can't decode them.

At least for a time, this capability may be important. Slightly.

But this can be solved by placing the receiver that decodes the codecs ahead of the XD in the chain.

Not if I want to use the Radiance as the switching hub, which I do. My configuration also allows me to send the switched output of the Radiance to another room without having to use the receiver or projector in the main viewing room.

"Being able to pass all audio codecs in bitstream format is not valueless."
I am not aware that anyone has said that. The question was, "Aside from being able to pass all audio codecs ..." (emphasis added).

This leaves the video aspect. This is debatable even at best.
:)

People here have tended to naysay the idea of the HDMI 1.3 upgrade pretty strongly. I just wanted to reiterate that there's at least one non-buzzword reason for wanting it.

Sorry if the purpose of my question was interpreted as more XE upgrade naysaying - that was not my intent as I think it's great Lumagen is offering the upgrade for those who really want/need it.

Ultimately I just wanted to know what else one would get in addition to the HDMI version upgrade, and my question has been answered - so sorry for stirring up any ill sentiments.

Hey - don't apologize for ... nothing wrong!!!
Besides ... if you go back through this thread (I think it's this one) even Jim Peterson keeps maintaining that 1.3 and its update is useful only for audio in this component.

Thanks everyone for the great insight. I am going to try a Pioneer 51Fd with a Radiance.It seems like a good unit overall.

I tried the Pioneer for the first time today.On an existing customer's system.Lumagen HDP in this case. As advertised in "source direct" mode the 51FD outputs 480i for DVD and 108p24 for Bluray via HDMI. Load time is reasonable as well.
Black levels seemed low,so I had to bring up the input black level to compensate. Interestingly, the HDP has separate memories for both resolutions on the same input. I used Digital Video Essential disc to set the levels.
Very nice picture quality! Seems like a good choice for Radiance as well
Thanks!
Dallas

Color coordinates are also 3-dimensional so if you have xyY reference coordinates, you will be adjusting the color controls to make all 3 coordinates as accurate as you can

So I've been playing around with the RadianceXD and Color Gamut calibration... I'm stuck on one part... I can get the xy levels to reference, however, the Y seems higher... how do I adjust Y?

Danny

So I've been playing around with the RadianceXD and Color Gamut calibration... I'm stuck on one part... I can get the xy levels to reference, however, the Y seems higher... how do I adjust Y?

Danny

Have you read Lumagen's own FAQ on gamut calibration? Available from their support page:

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

Cheers,

Mark

So I've been playing around with the RadianceXD and Color Gamut calibration... I'm stuck on one part... I can get the xy levels to reference, however, the Y seems higher... how do I adjust Y?

Danny

The gamut calibration guide is a little vague (for a beginner, anyways) when it comes to explaining exactly HOW to change Y for the primaries and secondaries. For the primaries (take red for example) you'd decrease Y simply by decreasing the value of R after you've adjusted G and B to get red's x/y to be as accurate as possible. For the secondaries (again, using cyan as an example), you decrease Y by decreasing G and B by equal values (again, after editing R, G, and B to get its x/y coordinates accurate).

Remember that, as the guide explains, this is an iterative process since changing one color affects the others, and changing Y will most likely change your xy values for that color. So you'll have to run through the whole process several times. Hope this helps.

The gamut calibration guide is a little vague (for a beginner, anyways) when it comes to explaining exactly HOW to change Y for the primaries and secondaries. For the primaries (take red for example) you'd decrease Y simply by decreasing the value of R after you've adjusted G and B to get red's x/y to be as accurate as possible. For the secondaries (again, using cyan as an example), you decrease Y by decreasing G and B by equal values (again, after editing R, G, and B to get its x/y coordinates accurate).

Remember that, as the guide explains, this is an iterative process since changing one color affects the others, and changing Y will most likely change your xy values for that color. So you'll have to run through the whole process several times. Hope this helps.

Great! I'll try that! So far, it's really weird looking at a somewhat (still need to adjust the "Y"'s) properly color gamut corrected display. I'm so used to the over saturated colors that it feels dull...

Should I should set the gamma factor in the radiance before or after adjusting the gamut?

I convinced myself that it should be set before, but the Calman help files mentioned that you should be using raw color values, and not gamma adjusted ones. I draw a different conclusion each time I read the Lumagen doc.

thanks,
aaron

Should I should set the gamma factor in the radiance before or after adjusting the gamut?

I convinced myself that it should be set before, but the Calman help files mentioned that you should be using raw color values, and not gamma adjusted ones. I draw a different conclusion each time I read the Lumagen doc.

thanks,
aaron

From the Lumagen calibration document:
"The Radiance series uses a 3D RGB color-palette to allow both the primary and secondary color points to be independently calibrated for (x,y) and Y values. The 3D palette works in linear RGB color space. The Radiance processes the incoming values to remove the gamma pre-emphasis applied at the source."

So you can do gamma before doing color gamut. The gamut calibration will still be done with 'raw' colors as recommended by Calman.

So I've been playing around with the RadianceXD and Color Gamut calibration... I'm stuck on one part... I can get the xy levels to reference, however, the Y seems higher... how do I adjust Y?

Danny

I'm not exactly sure what you are getting stuck on - I'll try this and see if it gets you over the hump...

Each primary color has one "Sub" setting in the Color Gamut > By Color area. Red has SubR, Green has SubG, Blue has SubB. SubR = Subtract Red = make Red Luminance Lower.

So you make an initial white measurement - lets say you measure 75% white. Your calibration software or external calculator like the AccuPel calculator takes the white measurement data, and calculates Y targets for each color automatically. You then display 75% red, 75% green, etc. to make your other measurements and Y (SubR, SubG, or SubB) settings. If you measure 100% white, you have to use 100% colors also. Because plasmas run out of gas at high saturations, it's best to just stick with 75% window patterns for every display - that also puts you in the middle of the luminance range for each color (and white) rather than being all the way out at the peak luminance for each color and white.

Secondaries are made of 2 primaries so there's no luminance adjustment - you get what you get. But you can move the secondary towards the primary color opposite the secondary (for example, the AddG control for Magenta will move the Magenta point towards the Green point).

If you use the Color Gamut > By Color adjustment area, setting Y is fairly straight forward. Just keep in mind, if you do anything that changes the luminance of your reference white measurement you start the process with, then you need to re-measure white and re-calculate the Y targets for each color.

Do xyY for the colors first, then do grayscale - that seems to minimize the amount of iteration needed to get color and grayscale accurate at the same time. Don't forget there's a Y adjustment for each grayscale step also - your calibration software should let you specify a target Gamma and that should give you your target luminance settings for each grayscale step. If your software doesn't have that capability, you have to manually enter data into something like the free AccuPel calculator (see their firmware download page to download the calculator).

Except they're rather busy now on genlock and 60i to 24p. ;)

Sorry for this some kind of offtopic.

Is it really possible to have a NTSC source @60Hz converted to 1080p24 without judder? I suppose it does. But how? Can anyone explain please?

Is it really possible to have a NTSC source @60Hz converted to 1080p24 without judder? I suppose it does. But how? Can anyone explain please?
If the NTSC source was based on 24fps material (film) then it's actually quite easy. After deinterlacing the incoming fields in filmmode you end up with 60 frames per second, but only 24 genuine ones. The other 36 are redundant frames and the processor can easily delete those from the stream.

My TV for example does this very nice (1080p60 to 1080p24). In a videoprocessor the deletion of redundant frames would probably happen right after the deinterlacing and before a full frame 60p stream is created.

With the original 24 frames on hand (actually 23.97 per second) the processor can easily output a 1:1 cadence (1080p24), 2:2 (1080p48) or 3:3 (1080p72).

In theory, yes. In practice, few processors can do it consistently and without artifacting with broadcast material.

Thank's for the head up.

So if I understand this correctly. A NTSC DVD, almost filmaterial would benefit to be outputed in 1080p24/48 @ a native 1080p display?

Is this true even for PAL film material? (PAL DVD)

I'm about to receive my new XD in a couple of days. I will try this out, if its supported correctly as cpcat stated?

About brodcast material. The difficulty here lies within the bad flaging of the material right? And it can alter between film and videomaterial?

So if I understand this correctly. A NTSC DVD, almost filmaterial would benefit to be outputed in 1080p24/48 @ a native 1080p display?Yes.

Is this true even for PAL film material? (PAL DVD)No. PAL DVDs have the 24fps speeded up to 25fps which are then interlaced with a 2:2 pulldown process. De-interlacing (correctly) will get you 25 "original" frames per second which should be displayed at 50Hz.

About brodcast material. The difficulty here lies within the bad flaging of the material right? And it can alter between film and videomaterial?Oh yes.

Bad flagging is usually not an issue, since the Radiance detects film/video based on the cadence. The flags are not even present in the HDMI data stream. The biggest problem is mixed video/film material (commercials), and compression artifacts making field matching difficult.

Question: not sure if its been answered in this massive thread, does each output from the radiance have its own gamma adjustment? I was thinking of using it ahead of my TV-One C2-2250A's for gamma control for each Crt projector in the blend.

I wish they would have added a edge Blending feature to these or the Pro model.

Athanasios

Yes. This is addressed in the FAQ, see link in my sig.

Edit: hmm, I take that back. Not for how you want to use it. There multiple output configurations, each with its own set of settings, but you can only use one at a time for an output, and then the second output must be a copy of the first or turned off.

Yes. This is addressed in the FAQ, see link in my sig.

Edit: hmm, I take that back. Not for how you want to use it. There multiple output configurations, each with its own set of settings, but you can only use one at a time for an output, and then the second output must be a copy of the first or turned off.

But both outputs can be driven together therfore driving two displays at once correct?

if that is true then it would make sence for them to have separate calibration for each output since two different displays would be driven. I think this is something Lumagen should add to the next firmware.

Athanasios

If someone connects two displays to the Radiance outputs, they nearly always just watch one display at a time. You can store different display calibrations in two different output configurations and select one of the calibrations by a user memory button on the remote. So it's very easy to switch between displays.

Best regards,
Randy Freeman
support@lumagen.com

But both outputs can be driven together therfore driving two displays at once correct?

Yes, but the second output will be a copy of the first.

if that is true then it would make sence for them to have separate calibration for each output since two different displays would be driven. I think this is something Lumagen should add to the next firmware.

Athanasios

That is not possible. There is only one picture pipeline in the XD, what you want would require two. That is why on the second output you can have only one of two things; an exact copy of the first output, or no video at all (i.e., a 1080i blank signal). I think Lumagen will add (if it's not already there) a direct passthrough for the second output. But that's about it.

Should I should set the gamma factor in the radiance before or after adjusting the gamut?

I convinced myself that it should be set before, but the Calman help files mentioned that you should be using raw color values, and not gamma adjusted ones. I draw a different conclusion each time I read the Lumagen doc.

thanks,
aaron

Gamma Factor defaults to 1.00 - this is where it should be when you start your calibration. If you do that, after you are ALL DONE, you will use the Gamma Factor control to explore different gammas with your display.

With Gamma Factor set to 1.00 during calibration, after you are all done with calibration, you will have 3 lower Gammas to experiment with (.85, .90, .95) and 6 higher gammas to experiment with (1.05, 1.10, 1.15, 1.20, 1.25, 1.30).

You SET gamma for your display in the Grayscale adjustment area (Item0-10, 0% white-100% white). Let's say you want your default Gamma to be 2.2. You either will have software that will tell you what the luminance should be for each grayscale step based on your first grayscale measurement, or you will have to use a free calibration calculator like AccuPel's calculator you can download for free from their firmware download page. In either your calibration software or the calculator, you enter your target Gamma (2.2 in this example) and make a grayscale measurement... 100% white will work, for example. Once the calibration software or calculator knows the 100% white luminance, it can tell you what the luminance should be for each grayscale step. So as you calculate grayscale steps, you will also set the luminance control so that step matches the Target Luminance calculated by your calibration software or the calibration calculator (not all calibration software will calculate target luminance values for you - hence the need for an "external" tool like the AccuPel calculator).

With Gamma Factor set to 1.00 and your grayscale steps all set to the correct target luminance values for 2.2 gamma, your video display will have a 2.2 gamma whenever Gamma Factor is set to 1.00. Using the lower settings will produce lower gammas (may be useful for daytime viewing if there's light in the room) and using higher Gamma Factors will produce higher gamma values. Adjusting Gamma Factor does not mess up other settings (or if it does, they don't change much).

So... Gamma Factor itself is not a calibration control, per se. It is something you use AFTER calibration to get whatever gamma you'd like to have in the event you feel that your target gamma (2.2 in this example) seems to not be quite right - either for everything or just for certain things.

Higher gammas will make midtones darker and keep shadows darker longer... the black point does not change, but 5% white at a "high" gamma will be darker than 5% white at a "low" gamma. Too much of this can be a bad thing - you may not completely lose shadow detail, but the shadows can be too dark for comfortable viewing when the gamma is too high.

Randy

I'm not sure I follow the "75% or 100%". Does that mean at the moment there are additional issues using external 75% stim colour patterns as opposed to 100% stim ?

I can't speak for others but one of the primary reasons for my purchase of the Radiance XD was full CMS. If this is "flawed" in real world execution I would expect it to be fixed and not require the puchase a new "pro version" to deliver what was expected in the XD.

To avoid any confusion can we have statement from Lumagen re what should be expected of the Radiance XD CMS ? Jim commented on the Vision HDx range offering gamut "mitigation" but not full CMS does this now also extend to the Radiance in real world use ? :(

D

I second that.

The main reason for me to buy the RadianceXD is the fully working 3D CMS. Which doesn't seem to be able to adjust non linerity intensities on faulty working displays.

Unfortunally I did found this thread too late, as my VP will be delivered today, allready paid for. I do hope Lumagen support will adress this matter. It feels like the marketing foold me plenty.

About the relative good linerity for the RGB, but worse for CMY wouldn't this be a faulty matrix? I havn't done any real meassurements my self, and I'm so initsiated as a calibrator exept my own gear. But I can say there are sure some significant matters in PQ with a skewish linerity for illuminans.

Can Lumagen comment on my concerns and thought?

I hope my writing are understandable, english in writing aint my strong side.

What levels are we looking at with these two patterns?

Contrast1 - has what looks like two black bars in the big black rectangle. Is one -2 and the other +2 or something like that? I assume I should get the darker one to where it is not visible, and the other to where it is barely visible, but I am not sure how visible.

Contrast2 - The gamut calibration pdf mentions that this has a -4 and a +4 bar, but I am not really sure what I am looking at when I view this pattern... and how things should look when I dial it in properly.

I probably should have understood those two completely before tackling my gamut and greyscale calibration, but now I want to really make sure I know what is going on, before I start mucking with my gamma as I think I am losing some shadow detail.

Contrast1 with -2 vs. +2 should be dialed so the +2 is visable.

Contrast2 with -2 vs. +2 should be dialed so the +4 is visable.

Not much of a different you can think. Well, it's not!

Well, this is not all the thruth I belive. I think Lumagen has put these 2 different patterns to evaluate and dial in brightness controls with different gammas. Where different setups, enviorments call/allows higher/lower gamma values, things should be set slight different. And the low IRE may, may not be visable all the time.

eg. in +2 IRE in a specific room may not be ideal to dial in the brightness setting due to to much ambient light.

Another usage of these 2 patterns should be evaluating gamma setting. To see if your gamma is set to high you can evaluate differences between these 2 patterns. There should be a noticable difference between +2 and +4 but with a to high gamma it wont.

Hello,

I've already posted this problem in the Lumagen forum but thougt that it would not hurt to post it also in this thread to reach more PAL users.

Received my RadianceXD yesterday (had an iScan VP50Pro before). I'm very content with the performance but noticed a problem in handling of 50Hz 2:2 material ("PAL-Film").

I can see something like occasional "mild" combing - especially in slow scenes (and not in all moving parts of the image) . The only explanation I can think of is a change to Video-Mode (but why does this lead to slight combing in this case?) and back to Film-Mode.

It makes no difference if I use the "Film" setting in the deinterlacing menue.

To illustrate the problem I have recorded one scene. I used a SD-DV Cam so the quality is not very good, but the problem can be seen ~@ second 2 for a very short moment in the video.
www.bkh-von-der-mark.de/test.avi

To exclude a problem with my Radiance I have also created a short DVD with the original material. It would be very nice if you could test it and give me a feedback if the described problem also occurs on other machines (look at the mouth of the guy with the cartridge belt ~@ second 15):
http://www.bkh-von-der-mark.de/test.ISO

I can also see some combing in 50Hz video material (sometimes it even missdetects a 2:2 cadence in scens with no/ slow motion which leads to heavy combing for a few moments until it switches back to video mode). A fixed video mode would be helpful

I know that the correct detection of 50Hz material is not easy. My iScan had a "Forced 2:2" setting that forced weaving of two adjacent fields (the above scene runs flawless with this option on my iScan). Could something like this be implemented in the Radiance?

Maybe there would be the possibility to add a mode in that the deinterlacer permanently locks to a once detected cadence?

Best regards

Denis

I'v played that clip on my TViX feeding the RadianceXD with both 1080i and 576p, unfortunally it can't send native res interlaced.

I cant rule out that the tvix doesn't fiddle with the interlace signal @1080i but there is no error in deinterlacing in the XD.

I can certainly see improvements in xd deinterlace over the tvix.

So for those that are interested color gamut correction of a JVC RS2 with a RadianceXD, I can say that the results are pretty impressive. It's really noticeable with red tones, primarily with faces. Watching Mama Mia, you can really notice the people with tans don't look sun burned -- they actually look tan. Now, the correction isn't night and day, but brings it to more natural levels.

I cant rule out that the tvix doesn't fiddle with the interlace signal @1080i but there is no error in deinterlacing in the XD.
Thanks for the info. The scene runs fine if I switch to 1080i50 on my Oppo too. But the pure 576i50 handling shows the "regional combing". Not only in this movie but in principle with all 50Hz 2:2 material from time to time.

I've only provided a link to 50Hz 2:2 material in my first post. Here are two iso files with DVD structure and 50Hz video material. On my machine there is combing in many scenes so I won't post the time indices. Maybe these files help to see if there is a special problem with my unit (or underline the importance of a forced video mode; in addition to stable 2:2 handling).

www.bkh-von-der-mark.de/pal1.iso (http://www.bkh-von-der-mark.de/pal1.iso)
www.bkh-von-der-mark.de/pal2.iso (http://www.bkh-von-der-mark.de/pal2.iso)

Best regards

Denis

I will try these iso's both from my TViX (won't handle 480/576) upscaled internally to 1080i. And also from my Oppo970, in native resolution.
Which Oppo do you use? The 980?

This is in fact very interesting!

PS. I still hope for a answer to my question about the CMS.

The Radiance CMS system allows you to calibrate the three primary color points the three secondary color points and the white point. This allows you to calibrate one of the new displays that has a green color point that is outside the color triangle. Many customers have reported that the CMS greatly improved the color on such displays.

We have talked about the possibility of expanding the CMS system on a RadiancePro model to add calibration points for more IRE levels. With so many points to calibrate you, most likely, would need to use some type of automated calibration software.

You would also need to make sure that the colorimeter that you use to make the measurement is linear. Otherwise you might just be using the CMS system to compensate for meter reading errors.

For most people, the present Radiance CMS is exactly what they need to achieve great results.

Randy Freeman
support@lumagen.com

Randy,

I agree 100%. There is concern regarding non-linearity in displays and the lack of multiple points for the 3D gamut correction, but often the error is coming from the metering equipment. Unless you are willing to spend a substantial amount of money for a good spectroradiometer (i.e not an i1), you will not get accurate primary/secondary readings.

It's really not an issue. I calibrated serveral displays with the Radiance, and the CMS makes a huge difference (for the better). A slight saturation error in the low IREs due to non-linear color decoding is not issue. (YMMV of course).

Thank's for the calirification. Many CMS out there work well in x,y axis but only mess up things in the Y, resulting in a worser picture than before the calibration, especially when adjusting secondaries. The CMS it self doesn't work linear.

I suppose the radianceXD does?

From RadianceXD manual:
Trigger Output (12V at 100mA inductive, or 250mA non-inductive, load)

Is it possible to run a computer chassie fan from this output? I suppose if it use less than 1,2W (100mA) it would be fine?

To answer any curious readers, it work flawlessly!

my RadianceXD is for sale in AVS classifieds in case anyone was looking for a used one. i am upgrading to an XE - this XD works flawlessly. PM me if interested.

Why not just upgrade the XD? I thought that was possible...

Why not just upgrade the XD? I thought that was possible...

I was thinkin' the same thing exactly!!!

i have some dealer credit but it's only good for equipment ..... .. between the used XD sale and the dealer credit i hope to come out even on an XE from him.