I'm not hoping, I'm not particularly fussed about NR. They could issue a software update if they didn't have time to update the boxes they're prepping for the initial ship.

I just think it's worth considering the options why they're being so mysterious about it/

Gotcha. I am sure they could just address it in a quick firmware update...

It seems reasonable to assume they have something in mind...
What?
Who knows?
Until the info comes from on high, I won't rule anything out... ;)

Sept 6th? What is significant about it? Wasn't the VP50 pushed back until the end of this month?

you are correct - nothing anymore. Just that today WAS the original ship date - something that disappeared from official sites and posts in a manner that would make George Orwell proud. A date that disappeared right around the time we did our preorders....

but as I said earlier, since I already have a VP30, I can wait until they get around to shipping the VP50. Especially since DVDO is awaiting a visit from Orkin, prior to shipping. :D

I had the VP30 (sold it) then just sold my Lumagen HDQ. Luckily I have a Crystalio2 coming this week to tide me over until the VP50 hits. It will be fun to compare. Man, I hate this hobby! :eek:

Man, I hate this hobby! :eek:
Stop lying!!!

:D

Sept 6th? What is significant about it? Wasn't the VP50 pushed back until the end of this month?

One can wish, can't he?

I guess. I just thought everyone already knew it was pushed back so there was no point in expecting it on the 6th (today)...

you guys are too funny :D

-Gary

I am not quite sure how we "slipped" to the end of the month, when the release date of the VP50 has always been September 06.

I am not quite sure how we "slipped" to the end of the month, when the release date of the VP50 has always been September 06.
Eh...??

Ah... :D

Well, that answers that. :)

I think the September (20)06 was misinterpreted to mean the 6th of September.

Well Josh, it is going to be September 10th on Sunday. So next week is considered mid September. So after next week we are going on the end of September. So that is why people are guessing the end of this month. Are you saying they will be out before then? :)

Well Josh, it is going to be September 10th on Sunday. So next week is considered mid September. So after next week we are going on the end of September. So that is why people are guessing the end of this month. Are you saying they will be out before then? :)

Actually we have "late" September before we go on to "end" of September.

Or did I hear it was going to be "Summer" 06, which puts us up to around the 22nd.

Come on, Josh. You've got to throw us a bone here. ;)

I think the September (20)06 was misinterpreted to mean the 6th of September.

When it really means the 20th of September '06.

- Collin

It will be released when it's released.

It will be released when it's released.
Are you sure? Do you have any details? Can you be more specific?... :p :p :D :D

Thanks aaronwt we did not know that. :D

Well Josh, it is going to be September 10th on Sunday. So next week is considered mid September. So after next week we are going on the end of September. So that is why people are guessing the end of this month. Are you saying they will be out before then? :)

Actually, by my book "early September" is from the 1st to the 29th, "mid September" is from the 2nd to the 29th, and "late September" is the 30th to the 45th. :D

I just picked up the new HR 20 DVR DirecTV and am very happy it has native resolution as a setting. Should go good with the VP50... Soon? :)

Actually, by my book "early September" is from the 1st to the 29th, "mid September" is from the 2nd to the 29th, and "late September" is the 30th to the 45th. :D
Spoken like a true marketing guy! :D :D

Actually, by my book "early September" is from the 1st to the 29th, "mid September" is from the 2nd to the 29th, and "late September" is the 30th to the 45th. :D

Josh--you've got too much time on your hands! :D

[But I have mine on order from AVS-- so the sooner the better!]

OK Josh....so on Sept. 46th, can I write a bitchy post about DVDO here at AVS? :D

"Oh do I remember the days in September when ...."

Josh. Quit dancing. :)

Do you remember that song?

I just picked up the new HR 20 DVR DirecTV and am very happy it has native resolution as a setting. Should go good with the VP50... Soon? :)

That is great news. I was getting so sick of having to change resolution on the HR 10 every time I switched from SD to HD. I will be getting the HR 20 as soon as it is available in my market. Hopefully I will get it around the same time as my VP50. Gettin both on the same day would be bliss.

Josh--you've got too much time on your hands! :D

[But I have mine on order from AVS-- so the sooner the better!]


Of course he has to much time on his hands, his months are 45 days long.
"45 days has September, April, June, and November" :eek:

Josh. Can you answer the question re any future plans for seperately adjustable sat and hue for each of R, G, and B. Also any plans for improving the already good scaling? Also 11 step gray scale adjustment?


All the rest have 46 except for February which has 43 (leap years 44). The real question is how many hours in each day?

By adjustable hue do you mean that in the final processing stages the red content should be able to be bled slightly into the green and blue outputs, the green content should be able to be bled slightly into the red and blue outputs, etc?

For example if objects that are pure red signal looks orange-ish on the screen you want to be able to tweak things (the tint) towards purple to compensate? And tweak each color independently of each other color?

I suppose this could be done but it would require several additional menu settings which translate to dozens if not hundreds of permutations that an ISF expert would have to tweak using trial and error.

You would need something like 79 days in every month to get through a calibration.

As far as I know, separate red, green, and blue saturation controls are the existing red gain, green gain, and blue gain controls.

Josh. Can you answer the question re any future plans for seperately adjustable sat and hue for each of R, G, and B. Also any plans for improving the already good scaling? Also 11 step gray scale adjustment?

What you really want is the ability to control R,G,B gains at multiple points across the grayscale. This will allow you to converge the guns to create a sensible grayscale. Traditional sets give you RGB (sometimes just RB) gains (hi) & cuts (lo) and this is sufficient to balance a grayscale, but the more points the better the calibration.

Then what you want is the ability to restrict the color gamut to a standard within the native color gamut of your set. The primary choices of projector and display manufacturers is often quite divergent from the standards. Samsung DLPs (my current display of choice) and a few others allow you to enter the xyY coordinates for the set's native primaries and then the coordinates for your desired gamut (709 HD or 601 SD for instance). The set will then create psuedo primaries that exactly match the standards. It's awesome. With a good calibration you can achieve perfect colorimetry. It would be great to see an iScan do something like this.

Similarly one thing the DVDO scalars do not yet do is compensate for primary differences in the sources. SD and HD have different color gamuts and it would be super cool if the VP50 could twist SD to the HD gamut. It sounds like the Lumagen boxes can do this. It's a very minor difference admittedly, but eventually it will be demanded by us videophile sorts, particularly as more things get mastered in the HD gamut (much HD is still mastered in the SD gamut).

Finally, I am all for more calibration options in the video processors, but if the processors are asked to do too much we end up harming the image due to only having 8 to 10 bits per channel going to the set. Say you need to lower the B component of an 8 bit RGB signal by half, you've reduced the possible steps in that channel from 255 to 128. This is likely to produce unavoidable false contoring affects and other problems. Folks are speculating that it won't be until we get 14 bits or greater per channel that we'll really be able to do serious calibration in an outboard device.

- Collin

Great post. Thanks

To illustrate what I'm talking about with primary calibration, I dug up a ColorFacts CIE graph from one of the times I calibrated my set. The outer white triangle with the colored points shows the native color gamut of my Samsung HLP. The inner black triangle is the ideal 601 SD gamut. The grayish points on the 601 triangle are the measured primaries and secondaries of my set after calibration using Samsung's pseudo primary technology (CCA).

You will notice, that the native red and blue are off a bit from the standard (pretty close to the HD standard actually), but look at how far out the native green is from the standard. This makes grass on a football field look cool and sells TVs, but it's hardly accurate.

Also notice how close to the standard the calibrated gray/white dots are. This sort of gamut constriction technology would rock in a processor since most people's sets cannot be calibrated to this level.

- Collin

Similarly one thing the DVDO scalars do not yet do is compensate for primary differences in the sources. SD and HD have different color gamuts and it would be super cool if the VP50 could twist SD to the HD gamut. It sounds like the Lumagen boxes can do this. It's a very minor difference admittedly, but eventually it will be demanded by us videophile sorts, particularly as more things get mastered in the HD gamut (much HD is still mastered in the SD gamut).

I may be wrong about the Lumagen's ability to do this. If I understand their materials correctly (they're not 100% clear) they do not twist between SD & HD primaries, but they do support assigning the analog YPbPr input to be either 601 or 709 encoded. This is another difference in the HD & SD specs that involves how RGB is converted to YPbPr. This is important because some upconverting sources output HD resolutions incorrectly encoded as SD YPbPr. If this signal is then interpreted as HD YPbPr the colors can get slightly but noticeably messed up (green depression mainly). Correcting for this would be another nice feature, thought probably not that important since most people would be using the iScan for upconversion and not need to correct for buggy upconverted sources.

- Collin

What you really want is the ability to control R,G,B gains at multiple points across the grayscale. This will allow you to converge the guns to create a sensible grayscale. Traditional sets give you RGB (sometimes just RB) gains (hi) & cuts (lo) and this is sufficient to balance a grayscale, but the more points the better the calibration.
The Lumagens do this over an 11-point scale -- and have a neat trick where they report the IRE of the center of the screen, for when you have the DVE DVD's black screen with grayscale boxes in the center. Each one of their 11 steps is from 0 IRE through 100 IRE, and you adjust RGB so that IRE on screen is same as source.

Crystalio 2 does the same thing over a 20-point scale.

Another neat color feature on Lumagens is hue & saturation correction for red & green push -- and the simple way it works by simply removing color channels when configuring, removes the need for colored filters to peep through.

DVDO are moving in this direction with their independent gammas for Red, Green and Blue -- but multiple point parametric grayscale and color push fix would be awesome addition to functionality. :)

Saying that, DVDO's approach is probably the best for 99% of users -- and sounds really easy to configure.

StooMonster

Ordered my VP50 Today.....Whooya!

I was told the VP50's will ship at the end of September, so we all have to sit tight. Anybody ordering SDI??

Jaime

DVDO are moving in this direction with their independent gammas for Red, Green and Blue -- but multiple point parametric grayscale and color push fix would be awesome addition to functionality. :)

Saying that, DVDO's approach is probably the best for 99% of users -- and sounds really easy to configure.

StooMonster

DVDO, is also including 1024 point (10 bit) RGB correction in the VP50

Output Controls -
Output Resolution, Sync Type, Colorspace (RGB or YPbPr/YCbCr), Frame Lock, Gamma (R/G/B and up to 1024 point Gamma individual RGB correction), Display Profiles (10)

Anybody ordering SDI??
I guess most with SDI sources already had an older iScan with SDI. So we can rip the SDI card out of the old iScan and move it over to the VP50.

Yes for SDI. I also have an SDI to HDMI converter.

I just hope the end of September does not turn into mid October...

I just hope the end of September does not turn into mid October...

or November...

Or December... Didn't this happen with the VP30? Atleast I have a C2 coming today to keep me company unitl the VP50 hits... :)

I just hope the end of September does not turn into mid October...

That would tough time :rolleyes:

or November...

I don't even want to hear about this :eek:

DVDO, is also including 1024 point (10 bit) RGB correction in the VP50

...1024 point Gamma individual RGB correction
It's nothing like the 11-point or 20-point parametric grayscale & gamma correction of Lumagen and Crystalio.

From the VP50 manual you can see that VP50's Gamma Correction is pretty simple, which is why I said "DVDO's approach is probably the best for 99% of users -- and sounds really easy to configure."

VP50's default gamma curve is linear (1.0); but you can individually adjust gamma of red, green, and blue in range 0.5 to 2.5 -- these curves are obviously 10-bit processing. However, you can adjust each color once.

On the Lumagen you can do this same thing 11 times, and on the Crystalio II 20 times. So rather than a gamma curve the red, green, and blue can be higher or lower in the bright, dark, or mid ranges. Very nice, but only really of use to ISF calibrators not to 99% of owners.

However, it would be nice to have the option in some future version of firmware. :)

StooMonster

Well, but personally, I wouldn't know how to adjust the gamma stuff at all. So is it useful *at all* without ISF calibration? I always thought these gamma correction stuff is only useful for ISF calibration. And I thought the VP50 supported that now, just like the Lumagens and the Crystalio II do?

Me being confused... :confused:

Well, but personally, I wouldn't know how to adjust the gamma stuff at all. So is it useful *at all* without ISF calibration? I always thought these gamma correction stuff is only useful for ISF calibration. And I thought the VP50 supported that now, just like the Lumagens and the Crystalio II do?

Me being confused... :confused:

Yeah it's a bit tough to adjust without tools. You could probably adjust it with a $150 light meter, Excel, and a lot of patience, but it's only a few minutes work with ColorFacts.

The approach StooMonster is talking about on the Lumagen would give the calibrator a lot of control. Having multiple points along a curve for each primary allows a calibrator to tune grayscale as well as gamma. The DVDO approach allows one to correct displays which already have smooth curves, but are too deep or shallow. Allowing separate control of this for RGB doesn't seem that useful too me. Displays are rarely if ever miscalibrated in such a way that each primary has a different gamma curve. Note that flipping through the VP50 manual it is not clear that it will work this way. It sounds like the manual is describing the VP30 style single curve correction control. The "individual R/G/B Adjustments (0.5-2.5)" feature doesn't appear to be documented. It wouldn't shock me if this were one of those features delivered later in a software update.

- Collin

Josh,

Just thinking ahead a little bit...how will HDMI audio like Dolby TrueHD be handled in the audio conversion process in future scalers? Since DTS-HD Master Audio (for example) will primarily be output via HDMI v1.3...will there be an HDMI audio output as well that will pass through those signals for Blu-ray/HD-DVD? I guess analog outputs can be used, but that would complicate the set-up a little bit and only allow for 5.1 audio whereas 7.1 is in the works for the lossless formats.

I currently own a VP30 and maybe I'm off base for asking, but I like to have some sort of upgrade path in mind to minimize the upgrade costs where possible. Thanks!

No chance of the release date becoming sometime in October if September is 45 days long or so. The issue I would guess is bug removal etc disclosed by the beta testers. You do not hear from them because they are essentially under nondisclosure agreements. While I don't expect delays, far better to delay things a bit for elimination of more bugs and more inititial features (to be announced at Cedia, I expect). Less bugs and additional features mean many more happy end users. And Josh LOVES happy end users.

No chance if September is 45 days long or so. The issue I would guess is bug removal etc disclosed by the beta testers. You do not hear from them because they are essentially under nondisclosure agreements. While I don't expect delays, far better to delay things a bit for elimination of more bugs and more intitial features (to be announced at Cedia, I expect). Less bugs and additional features means many more happy end users. And Josh LOVES happy end users.

I agree 100% ! Better to not be disapointed, and pulling your hair out when it comes... This is my first DVDO product, and I want to feel it was worth the wait...

Besides just got the PDP 5070 Pio last Month, and the 971 Oppo with SDI last Week... So I have plenty to keep me busy!

Royd

Just thinking ahead a little bit...how will HDMI audio like Dolby TrueHD be handled in the audio conversion process in future scalers? Since DTS-HD Master Audio (for example) will primarily be output via HDMI v1.3...will there be an HDMI audio output as well that will pass through those signals for Blu-ray/HD-DVD? I guess analog outputs can be used, but that would complicate the set-up a little bit and only allow for 5.1 audio whereas 7.1 is in the works for the lossless formats.



VP30 is HDMI 1.1 and this means it will handle multi channle PCM which means it is fully compatible with HD DVD player, as this is designed to decode in the players and output PCM on HDMI 1.1. On BD players you will limited to 1.1 (even if the machine is 1.3) which means you will be limited to the internal decoding on their players to output HDMI 1.1 PCM multi channel. Unfortunately BD players are MUCH more limited in this respect than their HD DVD cousins. A BD player will not be able to pass say lossless TrueHD to a processor via the VP30.

I do not want to throw gasoline (petrol) on the fire but ...

And for those that thought HDMI 1.3 was farther away, I guess not, at least in Europe anyway.

Does this announcement from Toshiba (Europe) mean the VP50 HDMI I/O will be/are obsolete? Does anyone know what Toshiba means by " Deep Color Technology"?

Perhaps the HDMI I/O connection and chipset, are on a changeable/upgradeable board in the VP50?

From "Audioholics"

Toshiba Announces HDMI 1.3 HD DVD Player for Euro Market

Today Toshiba announced today the introduction on the market of the European-wide first commercially available HD DVD Players. The HD-E1, similar to their current US model, is due to market in the middle of November. At the beginning of December will follow the HD-XE1, which is a bit more technologically advanced and sports higher-quality audio and video performance. The Player comes in a flat black Slimline Design and is equipped with the newest HD DVD technology, and of course the ability to upconvert standard definition DVDs to 720p and 1080i (HD-E1) and even 1080p (HD-XE1). The HD-XE1 supports HDMI 1.3 and Deep Color technology (nonexistent compatible display required of course) which appears to be a first.

Even if I get an HD DVD player with HDMI 1.3 I would still be routing it through my receiver first(like I'm doing with my current HD DVD player) before routing it to the VP30/VP50. HDMi 1.3 won't change a thing for me with the VP50. I would just need to get an HDMi 1.3 receiver that can decode the new codecs. As far as deep color, the TV would have to be able to handle the higher color depth as well as the recorded content and neither does right now.

Does anyone know what Toshiba means by " Deep Color Technology"?


That's in the HDMI 1.3 specs. It's the support of color depths up to 48-bit.

http://www.hdmi.org/press/pr/pr_20060622.asp

That's in the HDMI 1.3 specs. It's the support of color depths up to 48-bit.

http://www.hdmi.org/press/pr/pr_20060622.asp
Not really a big deal as the media doesn't support it and maybe one display currently on the market does, and it's only 32-bit.

Even if I get an HD DVD player with HDMI 1.3 I would still be routing it through my receiver first(like I'm doing with my current HD DVD player) before routing it to the VP30/VP50. HDMi 1.3 won't change a thing for me with the VP50. I would just need to get an HDMi 1.3 receiver that can decode the new codecs. As far as deep color, the TV would have to be able to handle the higher color depth as well as the recorded content and neither does right now.
And that's even if the players will pass the new codecs, and if the media itself is authored to do that.

From my point of view, HDMI 1.3 is pretty much a non-event as I really don't see any practical added value.

That's in the HDMI 1.3 specs. It's the support of color depths up to 48-bit.

Most of which are beyond the range of human vision.

Most of which are beyond the range of human vision.

Much like audio frequency ranges when specified 20-20kHz, also where the extremes are not within aural ranges.

However, these items (from the HDMI website) do sound interesting.

> Lets HDTVs and other displays go from millions of colors to billions of colors

> Eliminates on-screen color banding, for smooth tonal transitions and subtle gradations between colors

> Enables increased contrast ratio

> Can represent many times more shades of gray between black and white. At 30-bit pixel depth, four times more shades of gray would be the minimum, and the typical improvement would be eight times or more

It could be beneficial to total picture quality if a DVDO processor could "process" this information. The better the information to process the better the output, provided the processing is capable.

Those are all related to the color bit-depth, and as far as I know, there is only one display on the market, a Brillian(?) that can process 32 bit color. That's not to say there won't be more in the future, although I haven't really seen much talk about.

Plus, as noted by others with more knowledge, I believe the media itself has to be encoded with that bit-depth to make use of that feature, and to date, I don't believe any of the codecs being used for HD-DVD or BD do, or are planned to. I could be wrong on this part though, I'm sure Josh Z has a much better handle on this aspect of it.

DeepColor does make some sense. Why? The source is encoded in YCbCr 8bit 4:2:0. The VP internally calculates in YCbCr (AFAIK), but better VPs calculate in at least 10bit, because deinterlacing, scaling, noise reduction, color correction result in values that are "finer" than 8bit. The display displays RGB. So somebody has to convert 10bit YCbCr to RGB. Usually we let the VP do this conversion, because the VP is supposed to do it better than the display. When using HDMI 1.2 (or older) HDMI only supports 8bit when transporting RGB. The conversion from 10bit YCbCr to 8bit RGB is lossy. A conversion to 10bit RGB would be better, cause it could avoid some banding/dithering.

Ok, but the problem is that HDMI 1.3 chipsets are not available today. The big Toshiba with HDMI 1.3 is announced for the beginning of December. Do you want to wait that long for your VP50? The VP50 is announced to ship in September already. That's simply too early for HDMI 1.3.

I'm sure the next VP from DVDO (maybe September 2007?) will support HDMI 1.3. And by the time the next VP from DVDO will be ready for release, maybe we'll then also have some displays which can make use of DeepColor. Today we don't.

Yea. And at CES 2008 (Jan 2008), the HDMI folks will be announcing version 2.0 (the biggie :))

I know the VPx0 range has 10-bit processing, but what resolution is the chroma?

For example both Lumagen and Crystalio II are both 10-bit processors, but the Lumagen has 4:2:2 chroma resolution whilst the Crystalio II has 4:4:4 chroma resolution processing thoughout.

Are VPx0, and specifically VP50, 4:2:2 or 4:4:4 chroma resolution throughout processing?

Personally I am hoping 4:2:2 -- and that "YCbCr 422" HDMI 1.2a output is 10-bit deep, because this means I can pass 10-bit signals to my display without downsampling from 4:4:4 to 4:2:2.

StooMonster

Yea. And at CES 2008 (Jan 2008), the HDMI folks will be announcing version 2.0 (the biggie :))
Maybe, but I don't think I'll care. I'm not aware of anything they could add which would me make care. Do you have any ideas of features which could me want to wait for HDMI 2.0 over HDMI 1.3?

Personally I am hoping 4:2:2 -- and that "YCbCr 422" HDMI 1.2a output is 10-bit deep, because this means I can pass 10-bit signals to my display without downsampling from 4:4:4 to 4:2:2.
I don't agree. That would be like saying we don't want VPs to use 12bit internal processing, because we can only get 10bit out, anyway. That doesn't make any sense to me. I think the higher the bit rate and chroma resolution is during internal calculations the better. Obviously if the internal bit rate and/or chroma resolution is higher than HDMI can output, the internal data needs to be rounded/dithered down for output. But there's no reason why the end result should look any worse than if you calculate with lower internal bit rate and chroma resolution from the get go. Quite the opposite, I'd say - at least if rounding down and dithering is implemented correctly.

Man, some great info in the posts the last couple of days.
I have been very interested in this 48 bit Color stuff.
Thanx everyone. :)

Maybe, but I don't think I'll care. I'm not aware of anything they could add which would me make care. Do you have any ideas of features which could me want to wait for HDMI 2.0 over HDMI 1.3?



A more robust physical connection would be nice, although that is something that may be part of 1.3, I haven't read the complete spec in quite awhile.

A more robust physical connection would be nice
True. But let's say HDMI 2.0 came along with nothing but a better physical connector. In that case a simple adapter would do. So that wouldn't stop me at all from buying HDMI 1.3 equipment.

True. But let's say HDMI 2.0 came along with nothing but a better physical connector. In that case a simple adapter would do. So that wouldn't stop me at all from buying HDMI 1.3 equipment.
Oh, me neither, in fact, I suspect I'll be happy with 1.1 for at least a few more years.

I don't agree. That would be like saying we don't want VPs to use 12bit internal processing, because we can only get 10bit out, anyway. That doesn't make any sense to me. I think the higher the bit rate and chroma resolution is during internal calculations the better. Obviously if the internal bit rate and/or chroma resolution is higher than HDMI can output, the internal data needs to be rounded/dithered down for output. But there's no reason why the end result should look any worse than if you calculate with lower internal bit rate and chroma resolution from the get go. Quite the opposite, I'd say - at least if rounding down and dithering is implemented correctly.
Yes, I agree with you for HDMI 1.3 and the future.

However, for today's world of HDMI 1.2a (i.e. VP50, my 10-bit display) the VP working with 10-bit chroma at 4:2:2 is ideal because 10-bit can be transported at this 4:2:2 resolution over HDMI; and the display itself can upsample from 4:2:2 to 4:4:4 with no loss of information.
Result = no downsampling at any point.

Whereas if the VP is working with 10-bit chroma at 4:4:4 internally it has to be downsampled to either 10-bit 4:2:2 or 8-bit 4:4:4 for transport over HDMI.
Result=either chroma resolution downsample (4:4:4 > 4:2:2) or color depth downsample (10-bit > 8-bit).

* off the top of my head I can't remember if HDMI 1.1/1.2a allows for 12-bit at 4:2:2 in which case replace 10-bit above with 12-bit.

StooMonster

Or December... Didn't this happen with the VP30? Atleast I have a C2 coming today to keep me company unitl the VP50 hits... :)


Joe you will have to let me know what you think about this VP :)

@StooMonster, I still don't agree. If I could choose between:

(1) Calculate internally with 10bit 4:2:2 and output 10bit 4:2:2. Or
(2) Calculate internally with 10bit 4:4:4 and output 10bit 4:2:2.

I'd choose (2). Of course it doesn't make much sense going from 4:2:0 to 4:4:4 and then back to 4:2:2, if that's all that's going to change. But the VP is doing much more than changing chroma depth. All internal calculations when doing deinterlacing, color/gamma correction, noise reduction and scaling will have less rounding errors if you calculate in high calculation depth. If you then in the final step round/dither down to 4:2:2, that should look better than doing everything in 4:2:2 from the beginning. At least that's my understanding.

I know where you're coming from. I've seen your questions about whether to use 10bit YCbCr 4:2:2 or 8bit YCbCr 4:4:4 or RGB 4:4:4 as the connection between the VP and the display. But my believe is that usually you get the best quality, if you use the highest internal calculation depth you can reasonably use and then round/dither down for the output. Why else would Algolith use "up to 14bit" calculation depth in their Mosquito?

@StooMonster, I still don't agree. If I could choose between:

(1) Calculate internally with 10bit 4:2:2 and output 10bit 4:2:2. Or
(2) Calculate internally with 10bit 4:4:4 and output 10bit 4:2:2.

I'd choose (2). Of course it doesn't make much sense going from 4:2:0 to 4:4:4 and then back to 4:2:2, if that's all that's going to change. But the VP is doing much more than changing chroma depth. All internal calculations when doing deinterlacing, color/gamma correction, noise reduction and scaling will have less rounding errors if you calculate in high calculation depth. If you then in the final step round/dither down to 4:2:2, that should look better than doing everything in 4:2:2 from the beginning. At least that's my understanding.

I know where you're coming from. I've seen your questions about whether to use 10bit YCbCr 4:2:2 or 8bit YCbCr 4:4:4 or RGB 4:4:4 as the connection between the VP and the display. But my believe is that usually you get the best quality, if you use the highest internal calculation depth you can reasonably use and then round/dither down for the output. Why else would Algolith use "up to 14bit" calculation depth in their Mosquito?

Yeah the extra calculation depth internally is important. It's all about the rounding errors when moving between stages of the pipeline. The difference in PQ on the 10 bit VP30 vs the 8 bit HD+ was noticeable even on a setup with 8 bit SDI in and 8 bit RGB HDMI out.

- Collin

I was reading the specs on the DVDO web site. It says that the VP 50 scales a 1080P input. What does it scale it to and why would you want to process it at all? Also I didn't see anywhere that it has a 1080P pass-through. Can someone enlighten me?

Thanks.

Yeah the extra calculation depth internally is important. It's all about the rounding errors when moving between stages of the pipeline. The difference in PQ on the 10 bit VP30 vs the 8 bit HD+ was noticeable even on a setup with 8 bit SDI in and 8 bit RGB HDMI out.
I never said bit depth wasn't important -- and in my posts mention 10-bit and 12-bit -- my question is about chroma resolution 4:2:2 or 4:4:4.

StooMonster

I was reading the specs on the DVDO web site. it says that the VP 50 scales a 1080P input. What does it scale it to and why would you want to process it at all? Also I didn't see anywhere that it has a 1080P pass-through. Can someone enlighten me?
Say you had a 1366x768 plasma you'd want to downscale 1080p to that resolution, or a 1280x720 projector the VP50 will do that downscale much better than HD DVD / BluRay player downconverting 1080p to 720p itself (and can support resolutions other than 720p).

StooMonster

Thanks StooMonster,

How would it handle a 1080P input with a 1080P output?

I never said bit depth wasn't important -- and in my posts mention 10-bit and 12-bit -- my question is about chroma resolution 4:2:2 or 4:4:4.

StooMonster

Well more chroma resolution is more bits dedicated to chroma and therefore I think the same logic applies. The more data passed between stages of the pipeline the better, with fewer opportunities for rounding errors. I don't think there is an advantage to using a lower bit depth or chroma res just because its a better match for the output device, which I thought was the point you were making.

Like yourself, however, I would like to know more about the pixel formats used for the internal pipeline of the VP50. We know it's 10 bit, but we don't know the chroma res.

- Collin

Like yourself, however, I would like to know more about the pixel formats used for the internal pipeline of the VP50. We know it's 10 bit, but we don't know the chroma res.

And a follow up. DVDO's semiconductor page http://www.anchorbaytech.com/products/ABT102chip.html reveals that the ABT102 outputs 10bit 4:2:2. And their two scaling chips (similar but not the same as the VP30 scaling engine) accept at most 10 bit 4:2:2 in. I assume that the 20 bit refers to a 10 bit chroma bus and a 10 bit luma bus, else they'd make a big deal about the VP20/30/50 having 20 bit processing. Anyway, I think we can say with relative certainty that the VP20/VP30s are 10 bit 4:2:2 machines. I don't think it would be to outrageous to assume the VP50 is as well.

- Collin

Anyway, I think we can say with relative certainty that the VP20/VP30s are 10 bit 4:2:2 machines. I don't think it would be to outrageous to assume the VP50 is as well.

These are all safe assumptions.

Is there any change in the VP50's RGB component(bnc) input concerning allowing 480i input? The VP30 will not allow 480i to RGB(bnc) input. It would be good to have an additional input to benefit from the abt102 480i processing.

thx, Tom

And a follow up. DVDO's semiconductor page http://www.anchorbaytech.com/products/ABT102chip.html reveals that the ABT102 outputs 10bit 4:2:2. And their two scaling chips (similar but not the same as the VP30 scaling engine) accept at most 10 bit 4:2:2 in. I assume that the 20 bit refers to a 10 bit chroma bus and a 10 bit luma bus, else they'd make a big deal about the VP20/30/50 having 20 bit processing. Anyway, I think we can say with relative certainty that the VP20/VP30s are 10 bit 4:2:2 machines. I don't think it would be to outrageous to assume the VP50 is as well.

- Collin

Maybe I'm confused here but wouldn't the VP50 be using the newer ABT1018 chip?

http://www.anchorbaytech.com/products/ABT1018chip.html

Can we get a confirmation on this from Josh (either before, during, or after Cedia)?? :confused:

The VP50 does not use one our ASICs for video processing. All video processing algorithms reside on 2 very large FPGAs.

How would it handle a 1080P input with a 1080P output?
From the manual, I would guess that it processes 1080p input (brightness, contrast, gamma, etc.) and then outputs 1080p signal.

StooMonster

The VP50 does not use one our ASICs for video processing. All video processing algorithms reside on 2 very large FPGAs.

With a 3rd FPGA for UI & control?

The VP30 looks to be a 3 chip design : deinterlacer, scaler, & another FPGA.

- Collin

Anyway, I think we can say with relative certainty that the VP20/VP30s are 10 bit 4:2:2 machines. I don't think it would be to outrageous to assume the VP50 is as well.These are all safe assumptions.
Josh, thank you for the confirmation.

As the entire signal path though the VPx0 visual processors is YCbCr 4:2:2 10-bit and you can select "YCbCr 4:2:2 (10-bit)" as output, this is perfect for current HDMI versions. :)

StooMonster

The VP30 uses 2 much smaller FPGAs for Scaling and Routing and the SiI504 for deinterlacing. The ABT102 adds another relatively small FPGA for deinterlacing.

The VP30 uses 2 much smaller FPGAs for Scaling and Routing and the SiI504 for deinterlacing. The ABT102 adds another relatively small FPGA for deinterlacing.

Interesting.

Is the division of labor on the VP50 then 1 FPGA for the deinterlacer and 1 FPGA for scaling, routing, etc? Or are things more fluid than that?

- Collin

The VP50 does not use one our ASICs for video processing. All video processing algorithms reside on 2 very large FPGAs.

Doh! .......................Thanks Josh-- now I feel dumb with my question.

The ABT102 adds another relatively small FPGA for deinterlacing.

Interesting what you guys consider "small". The ABT102 uses a 2 million-gate Spartan 3, which is the 3rd largest part in that line, but I guess you did say "relatively", too. I wonder what the VP50 will be using.

--Andre

My XCard with the PMS SDI output card works fine with my Crystalio, but it didn't work at all with a Lumagen HDP I briefly owned. I am pretty sure the problem was the Lumagen's lack of flexibility in being able to switch between positive and negative SDI polarity, a feature the Crystalio has.

Now I'm reading in some old threads that past products from DVDO had this same problem with the XCard SDI source. Was this ever fixed? Anyone have an XCard sending 480i via SDI to a VP30 or HD+? If not, Josh could you comment on whether this should be a supported scenario on the VP50?

Or, alternatively, does anyone know of a device that outputs clean 480i over HDMI or can be modded for SDI and can read DVD files from a fileshare? I wouldn't mind ditching the Xcard if there was a substitute for streaming ripped DVDs.

Thanks!
-Pat

VP30 + ABT102 perfect on PAL, but crops off 3 lines at the bottom on NTSC. Without ABT102, no go on PAL, but works on NTSC. I've posted quite a bit on this topic, but can't find the thread(s) I've made most posts in. A quick search on all the posts I've made (there aren't that many :) ) will probably get you the whole picture.

P.S. I'd love to have a working X-Card (I have the PMS SDI card - no probs here). Ordered the X-Card from X-Passport, but they sent me one without the digital Output components, i.e no connector or series Resistor paks. I'm still trying to get them to respond to my emails to have them either take it back or supply me with the extra components. It appears that DRM has stopped them supplying the digitally enabled X-Cards. :(

quick note on calibration and gamma adjustments.

Gamma adjustment is important for adjusting between PAL, NTSC and HDTV (all requiring slightly different Gamma settings), but I hardly see the point of separate gamma adjustments per channel.

color calibration a processor makes sense - you can either do it on a per-point basis (i.e., 5, 11 or 20 points - although that sounds like simply too much work for me) or as a 2 point basis (bias, gain).

Keep in mind that color calibration is ideally done in the display device in order to avoid banding, since the number of bits on which the calibration is done is quite limited. On the Lumagen, I calibrate the display first and then use the 11 point calibration to fine tune the results. This should also be the correct approach here - not full blown calibration, but as a fine tuning device.

The VP50 does not use one our ASICs for video processing. All video processing algorithms reside on 2 very large FPGAs.
Thank you for the information!

Now where's the fan? Is it mounted on one of these FPGAs or is it somewhere else, just blowing air through the case?

Interesting what you guys consider "small". The ABT102 uses a 2 million-gate Spartan 3, which is the 3rd largest part in that line, but I guess you did say "relatively", too. I wonder what the VP50 will be using.

--Andre

The next step are Spartan3 with 4 and 5 million gates.

Looks like they maybe the Virtex 5's "You can use Virtex-5 FPGAs to realize cost effective video and audio interfacing, real time HD video processing, complex video and audio compression codecs,"

http://www.xilinx.com/esp/broadcast/processing.htm

http://www.xilinx.com/esp/broadcast/broadcast-bridge.htm

Ofer,

Gamma for NTSC and HD are the same, BT.1361 (or rec709).

As the entire signal path though the VPx0 visual processors is YCbCr 4:2:2 10-bit and you can select "YCbCr 4:2:2 (10-bit)" as output, this is perfect for current HDMI versions.

It is not that simple. You really need to look at a chroma multi burst pattern that shows both horizontal and vertical to decide which one to use as the display interacts. For example, on a Marantz VP11S1, you want to use RGB out. That display clips with either YCbCr input. On the Samsung HL-S5088, you want to use YCbCr 444 out to get the best chroma results.

It is not that simple. You really need to look at a chroma multi burst pattern that shows both horizontal and vertical to decide which one to use as the display interacts. For example, on a Marantz VP11S1, you want to use RGB out. That display clips with either YCbCr input. On the Samsung HL-S5088, you want to use YCbCr 444 out to get the best chroma results.
Stacey, what's a good example of a chroma multi-burst pattern showing both horizontal and vertical? How do I decide?

Perhaps you already know the best output for my display, the Pioneer 50-inch 1080p plasma (model numbers vary with regions)? ;)

StooMonster

Looks like they maybe the Virtex 5's "You can use Virtex-5 FPGAs to realize cost effective video and audio interfacing, real time HD video processing, complex video and audio compression codecs,"

http://www.xilinx.com/esp/broadcast/processing.htm

http://www.xilinx.com/esp/broadcast/broadcast-bridge.htm

It would cheaper to use the same chip structure that the one used in the VP20, VP30 and the ABT102.

I realized that I missed the latest firmware release for the VP30 (version 1.10). Since I pre-order the VP50, I don't really feel the need to go through the process of upgrading the software/recalibrate everything again and again.

It would cheaper to use the same chip structure that the one used in the VP20, VP30 and the ABT102.

Yes but they "DVDO" has said the design or the chips used in VP30 did not have the capability of processing full 1080i/p HD, so that's why the VP50. Not that it really matters how DVDO designed the VP50, just fun speculating while we wait.

I guess we will all know soon, once the first person pulls the cover to see the guts :)

Yes but they "DVDO" has said the design or the chips used in VP30 did not have the capability of processing full 1080i/p HD, so that's why the VP50. Not that it really matters how DVDO designed the VP50, just fun speculating while we wait.

I guess we will all know soon, once the first person pulls the cover to see the guts :)

Not having the device to play with, guessing parts is the only fun left. :D

VP30 + ABT102 perfect on PAL, but crops off 3 lines at the bottom on NTSC. Without ABT102, no go on PAL, but works on NTSC. I've posted quite a bit on this topic, but can't find the thread(s) I've made most posts in. A quick search on all the posts I've made (there aren't that many :) ) will probably get you the whole picture.(

Thanks escon, I think I found the threads you posted on. They all seem to refer to one of the PMS SDI mod cards for modding your own DVD player. The SDI XCard Converter product is slightly different. It has been rock-solid for me on the Crystalio and the both versions of the Holo3dgraph, but the problem I had with the Lumagen and the references to difficulties with other DVDO products worry me. Is nobody using this with a VP30 or other DVDO processor?

... The SDI XCard Converter product is slightly different. It has been rock-solid for me on the Crystalio and the both versions of the Holo3dgraph, but the problem I had with the Lumagen and the references to difficulties with other DVDO products worry me. Is nobody using this with a VP30 or other DVDO processor?
How is it different? I've looked very closely at it physically, and the two certainly appear to be identical apart from the interconnect connector. As I said, I can't fire up my X-Card as it has missing components connecting the data lines to the connector - and hence the SDI card has zero output of course

I'm on the pre-order upgrade for the VP50 and my question has to do with plasma owners as there is not even a handful of 1080P yet. I own a 1080 Sharp and primarily considered 1080P LCD's but I really like this new Samsung plasma that's coming out with an anti-glare screen - here a pic of it in an all glass building in daylight and the flash does not even pass onto the display that can be seen on the left bezel. It has 13 bit processing and won't be 1080P for about a year. My room is too sun filled for the Panny 1080P.

Seeing how PDP's are not 1080P but for two how do the other panels match up with the VP50. Can the VP50 make up for the fewer pixels or must I have a 1080P panel? Does the VP50 display 1080P on a 768 panel? Sorry guys but the plasma comparisons throw me but I would guess this product is made for them also.

Here's that new Samsung HP-S6373.

http://images.akihabaranews.com/news_pics/11756/2.jpg

Wow, what a beautiful lady! I'd like to see her in 1080p!! :eek:

westa, generally when using the VP50 you should have a display which allows 1:1 pixel mapping. 768p may be enough, depending on how far away you sit from the display and how big the display is. If you want to go really big and sit near then 1080p may be worth it. Otherwise, 768p displays are just fine. However, only few 768p displays allow 1:1 pixel mapping, while it seems that more and more 1080p models support it.

Stacey,

Specs might be the same, but I don't like to watch NTSC (I almost never do these days, but that's beside the point) beyond 2.2-2.3 gamma, but HDTV looks much better at 2.4 (sometimes even slightly higher)

PAL works great with 2.3... If memory serves, the original specs called for 2.8 gamma...

Not sure if it's my personal taste, but I would definitely love to control it this way.

Westa, that picture of that Samsung screen looks too good to be true. And I think the white reflection on the bezel is a light, not the camera flash, which wouldn't reflect back to the camera from that angle.

I'm on the pre-order upgrade for the VP50 and my question has to do with plasma owners as there is not even a handful of 1080P yet. I own a 1080 Sharp and primarily considered 1080P LCD's but I really like this new Samsung plasma that's coming out with an anti-glare screen - here a pic of it in an all glass building in daylight and the flash does not even pass onto the display that can be seen on the left bezel. It has 13 bit processing and won't be 1080P for about a year. My room is too sun filled for the Panny 1080P.

Seeing how PDP's are not 1080P but for two how do the other panels match up with the VP50. Can the VP50 make up for the fewer pixels or must I have a 1080P panel? Does the VP50 display 1080P on a 768 panel? Sorry guys but the plasma comparisons throw me but I would guess this product is made for them also.

Here's that new Samsung HP-S6373.

http://images.akihabaranews.com/news_pics/11756/2.jpg
Well I have asked myself the same question every time I look at the superb picture qaulity that I get on my Pioneer 5000EX Monitor 1920x1080p I very recently was given a tial demo of the Vantage HD and the Pioneer was better without it.There has been such a leap forward in plasma technology that I think they are no longer needed certainly on the very latest ones.Of course that is just my opinion for what its worth

Well I have asked myself the same question every time I look at the superb picture qaulity that I get on my Pioneer 5000EX Monitor 1920x1080p I very recently was given a tial demo of the Vantage HD and the Pioneer was better without it.There has been such a leap forward in plasma technology that I think they are no longer needed certainly on the very latest ones.Of course that is just my opinion for what its worth
Or was the VantageHD not very good for what you were watching?

Westa, that picture of that Samsung screen looks too good to be true. And I think the white reflection on the bezel is a light, not the camera flash, which wouldn't reflect back to the camera from that angle.

I don't know about you but I'm viewing it on dual DVI Monitors and I swear I can see what looks like a professional photographer holding a lamp flash, My DVI monitor gives me a crystal clear PQ and I'm not saying your wrong but mine is clear enough that I think I see someone holding a photographers lamp - not something we would use but a professional would.

Someone suggested it was photoshopped but this was done at a Press Release so how does one do that and not get caught? The whole reason of displaying in that glass building was to make a very valid point if it is genuine and I don't have reason to suspect otherwise I have testimonial from a Canadian Member that saw it put to the test with spot lights placed on the screen and confirms PQ and true anti-glare features work. Not trying to dispute you or anyone else but it's really a struggle for any panel besides a Panny or Pioneer to get any respect on this forum.

If the VP50 is not developed for less than 1080P than what is left for PDP Owners with 768? Isn't the NEC TheaterSync a true improvement on an NEC PDP? It certainly wasn't developed for 1080P was it? Then why wouldn't the same apply with this Samsung?:)

Wow, what a beautiful lady! I'd like to see her in 1080p!! :eek:



Japanese broads are so Hot!! http://forums.allaboutjazz.com/images/smilies/pimp.gif

-Gary

Noobie question, Is there any improvements to picture quality if you hookup your components to the VP50 and then output HDMI to the PRO-FHD1? If so, now much am I going to notice? Thanks.

Stacey, what's a good example of a chroma multi-burst pattern showing both horizontal and vertical? How do I decide?

I have generated HD versions for an upcoming disc Don and I are developing.

I have generated HD versions for an upcoming disc Don and I are developing.


Nice Tease!!!

On another topic - does anyone have any new information about when these babies will start shipping?

By September 45 :D

By September 45 :D

It is the 13th, September is almost over were is my VP50? :D

I have generated HD versions for an upcoming disc Don and I are developing.
I'll be looking forward to that, last disc for ABT was excellent.

And I you're right ... I think my 1080p Pioneer plasma may actually prefer 8-bit 4:4:4 RGB (least banding).

StooMonster

I
The whole reason of displaying in that glass building was to make a very valid point if it is genuine and I don't have reason to suspect otherwise I have testimonial from a Canadian Member that saw it put to the test with spot lights placed on the screen and confirms PQ and true anti-glare features work. Not trying to dispute you or anyone else but it's really a struggle for any panel besides a Panny or Pioneer to get any respect on this forum.



FWIW, I have a Sammie DLP that IMO has some of the best ant-glare properties I have seen.
It wouldn't surprise me if Samsung has improved their technology. ;)

Stacey I would assume you are speaking of working on both BR and HD-DVD versions ? that would be sweet :)

the SD ABT disc is top notch ;)

-Gary

I'll be looking forward to that, last disc for ABT was excellent.

And I you're right ... I think my 1080p Pioneer plasma may actually prefer 8-bit 4:4:4 RGB (least banding).

StooMonster

You have a FDH1?!? Wow... That's what I would have gotten if I could have came up with the scratch... I Love my 5070 though...

Just waiting for the VP-50 now... Sigh...

Royd

Japanese broads are so Hot!! http://forums.allaboutjazz.com/images/smilies/pimp.gif

-Gary

Yeah... Asian girls are just hot period... Japanese chicks usually don't have the back though...

My old girl was the hottest... She slayed her. She was Indonesian & Swiss... Eurasian....

Royd

Isn't Samsung Korean? :)

Yes.

you know what now that you mention it, that girl does look Korean

oops

-Gary

How is it different? I've looked very closely at it physically, and the two certainly appear to be identical apart from the interconnect connector. As I said, I can't fire up my X-Card as it has missing components connecting the data lines to the connector - and hence the SDI card has zero output of course

I have confirmation from PMS that the source of the negative polarity is the Xcard's MPEG decoder, not the PMS Xcard SDI Converter add-on card. So it looks like unless DVDO can be persuaded to support toggling of SDI polarity, the Xcard with SDI will never work as a pure digital path for DVDs stored on a PC hard drive.

I would guess if enough people stated it is a valuable feature that DVDO could add the feature. So if you are interested in this please let them know!

It is a shame as this is a really compelling scenario without any alternative that I can find. The Helios X3000 is supposed to be able to do 480i over HDMI, when it comes out, but it has pretty limited DVD file support. It only supports VOBs, not IFOs or ISOs so you lose all menu control and have to rip episodic DVDs as separate VOBs. Does anybody know of a network media/DVD player that will either do 480i over HDMI, or preferably is capable of having an SDI output mod performed?

-Pat

anyone at cedia planning on reporting back on the vp50 and any dvdo announcements?

also, when is the show open and when can we expect to hear feedback from
various demos, etc?

thanks

Does anybody know of a network media/DVD player that will either do 480i over HDMI, or preferably is capable of having an SDI output mod performed?

-Pat
Kaleidescape (http://www.kaleidescape.com/) I think lists 480i HDMI output from its network movie player, but -- having not yet won the Powerball -- I don't have personal confirmation of that.

:confused:

-John

I don't know about you but I'm viewing it on dual DVI Monitors and I swear I can see what looks like a professional photographer holding a lamp flash, My DVI monitor gives me a crystal clear PQ and I'm not saying your wrong but mine is clear enough that I think I see someone holding a photographers lamp - not something we would use but a professional would.

Someone suggested it was photoshopped but this was done at a Press Release so how does one do that and not get caught? The whole reason of displaying in that glass building was to make a very valid point if it is genuine and I don't have reason to suspect otherwise I have testimonial from a Canadian Member that saw it put to the test with spot lights placed on the screen and confirms PQ and true anti-glare features work. Not trying to dispute you or anyone else but it's really a struggle for any panel besides a Panny or Pioneer to get any respect on this forum.

If the VP50 is not developed for less than 1080P than what is left for PDP Owners with 768? Isn't the NEC TheaterSync a true improvement on an NEC PDP? It certainly wasn't developed for 1080P was it? Then why wouldn't the same apply with this Samsung?:)
Yeah if I squint the right way, I can see a photographer. The distorted perspective on the screen could mean the left edge is close to straight on now you mention it.

If that's a true representation of how the anti-glare works, it's pretty impressive, but I tend not to believe things until I've seen and twiddled them.

I don't understand your questions about resolution. Can you rephrase? Who said the vp50 isn't for people with screens of <1920x1080 ?

Kaleidescape (http://www.kaleidescape.com/) I think lists 480i HDMI output from its network movie player, but -- having not yet won the Powerball -- I don't have personal confirmation of that.

:confused:

-JohnI too am interested in an HD/SDI network player. But you're right, at $US10K, it's not quite the goer for me either. It just seems that there aren't any PC SDI graphics cards out there that are reasonable priced AND can just simply put out what the PC puts out on its normally connected monitor - most seem to need running out of Photoshop or the like programs. Being able to simply put out what shows on the PC screen (attached to its normal graphics card) and the use of a wireless remote "mouse/pointer, allows you to run ANY files on your PC, e.g Powerpoint and standalone exe slide files. Looks like an HTPC at some 2K is still the most economic way of doing it. My display and VP30 are some 15m away from my PC, so SDI is certainly the best way to connect my PC to it.

Has anyone found a better solution or more appropriate/suitable PC SDI card?

I have confirmation from PMS that the source of the negative polarity is the Xcard's MPEG decoder, not the PMS Xcard SDI Converter add-on card. So it looks like unless DVDO can be persuaded to support toggling of SDI polarity, the Xcard with SDI will never work as a pure digital path for DVDs stored on a PC hard drive.

I would guess if enough people stated it is a valuable feature that DVDO could add the feature. So if you are interested in this please let them know!

It is a shame as this is a really compelling scenario without any alternative that I can find. The Helios X3000 is supposed to be able to do 480i over HDMI, when it comes out, but it has pretty limited DVD file support. It only supports VOBs, not IFOs or ISOs so you lose all menu control and have to rip episodic DVDs as separate VOBs. Does anybody know of a network media/DVD player that will either do 480i over HDMI, or preferably is capable of having an SDI output mod performed?

-Pat
Well, I would certainly be very interested in DVDO making the SDI port accept both sync polarities. It seems strange though, that the MPEG decoder on the X-Card puts out a different sync polarity than the MPEG decoders on standalone DVD players. Not sure that I'm totally convinced of their explanation.

I don't understand your questions about resolution. Can you rephrase? Who said the vp50 isn't for people with screens of <1920x1080 ?
Must be a misunderstanding - the Samsung PDP is a 768 Panel not 1080P Native and someone sort of responded that the VP50 would not be beneficial to a non-1080P panel (this Samsung Plasma) which doesn't make sense to me since there are not even a handful of 1080P PDP's? Sorry but after owning three LCD's with large windows I cannot stand competing with washout and reflections of everything in the room that a Plasma presents which is the only reason I won't touch the Panny 65" 1080P. My hope was the Samsung 63" with anti-glare would provide the best of both but short on 1080P for now anyways.

Why else would an NEC TheaterSync be developed when the NEC is not 1080P yet. I just don't want to spend $6K on the panel and then $3K on a VP50 and see no or little improvement as $3K is worth more than nominal advances and you'd have a very small market audience if it were only advantageous to 1080P native panels, that didn't make sense to me. Thanks :)

ok so I think there were a number of others who expressed the opinion, that I share, that a 768p panel will also benefit in a big way from proper deinterlacing of 1080i.

Having said that, 63" is a very big screen and depending on viewing distance, the relatively low resolution might be a lot more noticeable.

ok so I think there were a number of others who expressed the opinion, that I share, that a 768p panel will also benefit in a big way from proper deinterlacing of 1080i.
This is very much the case. Before my 1080p plasma I used to own a 1366x768 model, and have had the opportunity (with various equipment) to test this out.

1080i deinterlaced to 1080p (1920x1080 pixels) and then downscaled to 1366x768 is far-and-away superior to upscaling individual 1080i fields (1920x540 pixels) to 1366x768.

You do not need a 1080p display to benefit from VP50, you simply need a 1080i source and VP50 will produce a better picture for you even if your display is 720p.

StooMonster

http://www.dvdo.com/new/press_09-14-2006.php

Additional unannounced pending industry certifications and VRS technologies will further cement VP50 as the premier product in the video processor category.

http://www.dvdo.com/new/press_09-14-2006.php

Indeed... but we knew this already, have to see if Cedia produces some 'unofficial' news now as the official release is old info, for us anyways. ;)

Indeed... but we knew this already
We only knew that "There will be more features/differentiators added to this list". We didn't know (although we may have guessed it) that the new features would be "pending industry certifications and VRS technologies".

I think you've stretched that sentence too far mate. It's surely just "pending industry certifications" ... whatever value that's supposed to imply.

"pending industry certifications and VRS technologies".


Please clarify your point.

Please clarify your point.
I've just quoted the DVDO press release. You should ask DVDO for further clarification, not me! :) Or am I misunderstanding you?

The sentence doesn't use the best grammar; it would read better with parenthesis, or some commas, or maybe simply rewritten!

Additional (unannounced pending) industry certifications and VRS technologies will further cement VP50 as the premier product in the video processor category.

Additional, unannounced, pending, industry certifications and VRS technologies will further cement VP50 as the premier product in the video processor category.

i.e. the industry certifications and VRS technologies are unannounced but pending.

Does that make it clearer?

StooMonster

I've just quoted the DVDO press release. You should ask DVDO for further clarification, not me! :) Or am I misunderstanding you?

Anybody who could answer. This announcement is more puzzling that clarifying.

is the vp50 being demo'd live or only discussed in a press release?

is the vp50 being demo'd live or only discussed in a press release?
It is almost certainly being demo'ed at CEDIA right now.

I like that the press release states it is shipping in mid-September 2006. That would be right about...now?

I like that the press release states it is shipping in mid-September 2006. That would be right about...now?
Wrong up to the 29th is mid September. :eek:

Actually, by my book "early September" is from the 1st to the 29th, "mid September" is from the 2nd to the 29th, and "late September" is the 30th to the 45th. :D

I like that the press release states it is shipping in mid-September 2006. That would be right about...now?

According to Josh - 45 days in September, so maybe a week? :D

Seriously, according to Jason it's supposed to be "late" September whatever that means.

I like that the press release states it is shipping in mid-September 2006. That would be right about...now?


Not according Josh extended calendar :D

At the press conference yesterday they announced pending THX certification. This would make it the first THX certified video processor.

They also announced PReP and have it running on the show floor. PReP is the algorithm that will take 480p, 576p and 1080p as input, re-interlace and then perform inverse telecine and such. It seems to work really well.

At the press conference yesterday they announced pending THX certification. This would make it the first THX certified video processor.


Cool, it will look nice right next to my TiVo S3, the first DVR to be THX certified. :p

(not that THX cert really means a whole lot to me) :D

At the press conference yesterday they announced pending THX certification. This would make it the first THX certified video processor.

They also announced PReP and have it running on the show floor. PReP is the algorithm that will take 480p, 576p and 1080p as input, re-interlace and then perform inverse telecine and such. It seems to work really well.
Not sure what benefits THX certification brings us. But PReP is definitely very nice.

Edit: Here's a link with a tiny bit more information about THX certification for video:

http://www.hdbeat.com/2006/09/14/thx-cedia-thx-certification-for-video-displays/

At the press conference yesterday they announced pending THX certification. This would make it the first THX certified video processor.

I wish the THX seal still meant as much as it used to.

They also announced PReP and have it running on the show floor. PReP is the algorithm that will take 480p, 576p and 1080p as input, re-interlace and then perform inverse telecine and such. It seems to work really well.

What are the advantages of that?

Not sure what benefits THX certification brings us.Bragging rights!
But PReP is definitely very nice.Absolutely. Especially with the minefield of HDMI.

They also announced PReP and have it running on the show floor. PReP is the algorithm that will take 480p, 576p and 1080p as input, re-interlace and then perform inverse telecine and such. It seems to work really well.
PReP will solve the 576p output problem of Sky HD in UK, and enable any HDMI DVD player to output 480i/576i over HDMI.

I am really looking forward to seeing this in action. :p

StooMonster

What are the advantages of that?
Most source devices refuse to output 480i. They insist on outputting 480p instead. In such a case a video processor is usually satisfied and just rescales that to the output resolution. The problem is that most source devices do a rather bad job with deinterlacing. Now with PReP (whatever that acronym means) the VP50 can undo the bad deinterlacing of the source device and replace it with its own much better deinterlacing.

I wish the THX seal still meant as much as it used to.



What are the advantages of that?
Some devices (e.g. SkyHD) will always output progressive over HDMI if they detect that the destination device supports it which means you're lumbered with the cadence detection & deinterlacing algorithms of the source device (Stoomonster has shown a Lumagen with edited EDID table can pretend it won't accept 576p and it then behaves itself but that's really a bit of a hack)

edit: yeah, what those guys said :)

PReP = Progressive ReProcessing.

StooMonster

Some devices (e.g. SkyHD) will always output progressive over HDMI if they detect that the destination device supports it which means you're lumbered with the cadence detection & deinterlacing algorithms of the source device (Stoomonster has shown a Lumagen with edited EDID table can pretend it won't accept 576p and it then behaves itself but that's really a bit of a hack)

edit: yeah, what those guys said :)

PReP will certainly be much appreciated, but I hope they still do the EDID "hack" which has been mentioned. Stoo - any more news from Josh on that front? Reinterlacing will be great, but not having to do it in the first place would be even better.

At the press conference yesterday they announced pending THX certification. This would make it the first THX certified video processor.

They also announced PReP and have it running on the show floor. PReP is the algorithm that will take 480p, 576p and 1080p as input, re-interlace and then perform inverse telecine and such. It seems to work really well.

I don't care about THX certification but PReP sound cool.

I believe EDID editing will come soon after PReP.

However, PReP could well be superior for two reasons:

PReP will work for all 480p/576p/1080p sources thus making likes of HD-A1 into 480i over HDMI device. :)

Extracts the original unprocessed interlaced signal :)

To transmit 480i/576i over HDMI a technique called pixel doubling is employed, how this works is it turns 720x240 into 1440x240 (720x288 into 1440x288) for transmission over HDMI.

This could go wrong at either transmission or receiving stage, each end requires processing.

Pixel doubling is supposed to be done by repeating each pixel twice, so if fields were scaled rather than horizontally bob'ed (i.e. interpolated data was created) that would be bad for transmission.

Likewise if the receiving device didn't handle receiving pixel doubled properly, e.g. it scaled 1440x288 back to 720x288 rather than stripping out pixel column, that would be bad too.

PReP doesn't require any horizontal processing, it simply identifies the original fields from a series of progressive frames and extracts them. :D

StooMonster

I believe EDID editing will come soon after PReP.

However, PReP could well be superior for two reasons:

PReP will work for all 480p/576p/1080p sources thus making likes of HD-A1 into 480i over HDMI device. :)

Extracts the original unprocessed interlaced signal :)

To transmit 480i/576i over HDMI a technique called pixel doubling is employed, how this works is it turns 720x240 into 1440x240 (720x288 into 1440x288) for transmission over HDMI.

This could go wrong at either transmission or receiving stage, each end requires processing.

Pixel doubling is supposed to be done by repeating each pixel twice, so if fields were scaled rather than horizontally bob'ed (i.e. interpolated data was created) that would be bad for transmission.

Likewise if the receiving device didn't handle receiving pixel doubled properly, e.g. it scaled 1440x288 back to 720x288 rather than stripping out pixel column, that would be bad too.

PReP doesn't require any horizontal processing, it simply identifies the original fields from a series of progressive frames and extracts them. :D

StooMonster

that is fantastic, and seals the deal for me.
i've been beating up (with no success) the makers of network media players to provide 480i over hdmi and find they have no concept that sigma designs 8620 deinterlacing and/or scaling could be 'sub-optimal'

i hadn't realized the issue was pixel doubling, so the problem was more complex than i had thought......and now DVDO has a solution. that is outstanding.

now all i (and ofer and others) need to do is get the the networked media player to output the input resolution as an output option over hdmi, i.e.

1080i => 1080i
720p => 720p
480i => 480p => PReP/dvdo
(unfortunately, all currently available network media players that i know of are designed to scale all inputs to a single selectable but fixed output resolution)

at least the size of the problem has been reduced as PReP permits a solution over a single hdmi output (i had thought i'd need to use 480i over component to the VPx0 and switch inputs as needed)

i have a demo set up next week with a local dealer so this is wonderful timing.

Sounds like this could make my 2 month old Oppo 970H obsolete long before its expected time.

Sounds like this could make my 2 month old Oppo 970H obsolete long before its expected time.

Possible, but I need to see that be demonstrated first. :)

Possible, but I need to see that be demonstrated first. :)Are you telling us that you are getting on a plane to CEDIA? Go see Josh in person? ;)

StooMonster

So is the VP50 still on schedule for release this month?

Sounds like this could make my 2 month old Oppo 970H obsolete long before its expected time.

I was just wondering about that. I would prefer an (unfortunately 480p-only) Oppo 971 for its lack of HDCP, but wouldn't its Faroudja deinterlacing alter even a 480p output enough that the original interlaced fields were pretty much unrecoverable?

I guess I am unsure what exactly PReP would do. For film material, it could just undo the reverse telecine if that was done well, but I'm not sure what the point would be, since then it would just redo it. For video or mixed video/film material, I could see recovering the original 480i fields from a 480p signal if the source device just did a simple weave or something, but if the source had tried to do motion-adaptive deinterlacing or any other serious processing (like a 971 does), then the original fields would be lost, right?

you know what now that you mention it, that girl does look Korean

oops


Hmmmm, that sounds like something PeeWee would have said in Porky's.....

I guess I am unsure what exactly PReP would do. For film material, it could just undo the reverse telecine if that was done well, but I'm not sure what the point would be, since then it would just redo it. For video or mixed video/film material, I could see recovering the original 480i fields from a 480p signal if the source device just did a simple weave or something, but if the source had tried to do motion-adaptive deinterlacing or any other serious processing (like a 971 does), then the original fields would be lost, right?

After bob, or after weave, or after motion adaptive de-interlacing you would expect that each progressive frame has either all original odd lines with even lines created/obtained from somewhere, or all original even lines with odd lines created/obtained from somewhere. Provided that the PReP processor can identify the original lines each time (helped by identifying a cadence that goes: odd even odd even) surprisingly good extraction of original interlaced fields can be accomplished.

More sophisticated processing including diagonal processing during the first de-interlacing can result in "original" lines being altered and then extracting back the original interlaced fields in a subsequent PReP process will not be entirely successful.

Inverse telecine'd de-interlaced material, if perfectly done from source with perfect 3-2 cadence, will re-interlace perfectly. If there was a glitch such as a bad edit, resulting in the first de-interlacing doing fallback on motion adaptiveness or something else, then the need for PReP to identify original versus created scan lines in the received (first de-interlaced) material surfaces again.

Are you telling us that you are getting on a plane to CEDIA? Go see Josh in person? ;)



Sorry, but no...
What I meant was the possiblity of re-interlacing 480p by the VP50 and therefore making 480i HDMI players obsolete.
Great...if it can be done in a way that makes it indistinguishable from 480i. :)

In the announcement here on the VP50 at CEDIA, there's a picture of it:

http://www.avsforum.com/avs-vb/showthread.php?t=724331

What's the smaller box on top of it and is that a different remote?

TIA.

- Gerry

is that a different remote?

- Gerry

I saw that too and got pretty excited. Went to the DVO website and checked the new owners manual for the VP50. Alas, it's showing the same remote as the VP30.

Looking at the picture again it looks like they may have been using a universal remote.

Mike

That box is likely the video source for the unit and the remote is probably for that.

All I can say at this point is that HD-DL is all that was promised and more...

Wonder if the THX logo will be on the front panel, and on the silver bezel?

StooMonster

All I can say at this point is that HD-DL is all that was promised and more...
No more combing in your torture tests? :)

The smaller box is a HDMI switcher DVDO is selling which most likely is made by Radiient.

In the announcement here on the VP50 at CEDIA, there's a picture of it:

http://www.avsforum.com/avs-vb/showthread.php?t=724331

What's the smaller box on top of it and is that a different remote?

TIA.

- Gerry
That remote looks identical to the Oppo 970 remote, maybe they were using an Oppo hooked up to the VP50.

All I can say at this point is that HD-DL is all that was promised and more...


that is a understatement of epic porportions :)

-Gary

That remote looks identical to the Oppo 970 remote, maybe they were using an Oppo hooked up to the VP50.

Thats their new 4 hdmi switch box.

Daniel.

Thats their new 4 hdmi switch box.

Daniel.
I know, I was referring to the remote off to the right in the pic GerryWaz asked about.

A 4 port HDMI box would hardly need a remote with that many buttons. :)

Just looked again, maybe he was referring to the little flat piece sitting on top of the switcher, doesn't look like a remote unless it's laying face down.

that is a understatement of epic porportions :)

-Gary

Care to elaborate... :)

He can't, DVDO has the NSA watching all his communications very closely. :p

I know, I was referring to the remote off to the right in the pic GerryWaz asked about.

A 4 port HDMI box would hardly need a remote with that many buttons. :)

Just looked again, maybe he was referring to the little flat piece sitting on top of the switcher, doesn't look like a remote unless it's laying face down.

Thats the place that holds the remote for it :)

http://gallery.avsforum.com/data/509/CIMG2412.JPG

Daniel.

Thats the place that holds the remote for it :)

http://gallery.avsforum.com/data/509/CIMG2412.JPG

Daniel.
Okay, that's why it looked strange, but why have a remote holder on the very device you'd want to control, presumably at a distance, right on the device, wouldn't you just use the actual buttons on the device itself?

Maybe there are no buttons, ala the Zektor component switcher I've used.

Okay, that's why it looked strange, but why have a remote holder on the very device you'd want to control, presumably at a distance, right on the device, wouldn't you just use the actual buttons on the device itself?

Maybe there are no buttons, ala the Zektor component switcher I've used.

It comes with two remotes (pro and consumer). You can use one and store one on the unit. It looks like the remotes still function when docked, allowing you to customize the buttons on the surface of the unit. Not necessary, but kind of neat.

- Collin

Okay, so it's like a removable interface, that is kinda cool.

no I don't care to :)

I have been using the VP50 for a while now in my personal reference HT setup, I can only call it reference because of the VP50 now, and that is no exageration :)

let me tell you something guys, you can dream of this thing all you want, but when you get your hands on it, it will surpass all thoughts or dreams you ever had

I have been viewing HD-DVD's deinterlaced 1080i to 1080p on my CRT pj for a while now and my jaw has to be lifted up to be able to recover from the most amazing picture I have ever seen

the 1080i deinterlacing is perfect and I give it that remark without hesitation or regard at all, I just finished viewing the Firewall HD-DVD, words cannot describe the image quality

my setup is as follows:

VP50 feeding a NEC 1352 CRT PJ via HDMI setup for 2.40:1 on a 2.40:1 Draper Cineperm screen, my output resolution is 1920x800p, eactly the resolution that 2.40:1 HDTV movies have :) no scaling involved for 2.40:1 1080i HD movies from any source, the VP50 is handling all my sources including SDI DVD, HD-DVD, D-VHS and BR upcoming

something else that needs to be mentioned, the SDI DVD performance is improved over the VP30 with 102 card, and it pleases me very much, DVD which became enjoyable with the VP30 just went a step further on the VP50

I have been wanting to blather on about this bad boy forever, I could write for hours upon hours

I must simply say this:

Thanks a million to DVDO, I would NOT have my current setup and picture quality without the VP50 in my system, no question about it, it is the centerpiece that ties everything together giving me the most amazing film reproduction that I have ever seen, true 1080i to 1080p is not something to be taken lightly, those who are picky like me demand the best and the VP50 is it

I can answer questions

Thanks a bunch guys

-Gary

Goodluck with that post 11a... :rolleyes:

no I don't care to :)

those who are picky like me demand the best and the VP50 is it

I can answer questions

Thanks a bunch guys

-Gary

So I gather that the VP50 should start shipping this Week?

Thanks,

Royd

I doubt that is the kind of question he can answer.


I would like to know how it would improve an HD A1 feeding a SimC3x.

warren

yes I would say it might improve 1080i material with the Sim, I am not sure what quality processing the Sim has though

the VP50 is the best 1080i to 1080p scaled to XXX(in your case 720p) that is available

Royd you would need to ask DVDO that :)

-Gary

...
the SDI DVD performance is improved over the VP30 with 102 card, and it pleases me very much, DVD which became enjoyable with the VP30 just went a step further on the VP50

-Gary

Gary,

Thanks for the preview. Do you know if 480i over HDMI input showed similar improvements?

I'm eager to see the new HD deinterlacer for my 1080i sources but any improvement to DVD will also be very welcome - NR? - if you can say?

Thanks,
Mike

Mike

sorry, I don't use 480i over HDMI only SDI, but it won't make a difference in that regard, 480i via anything was slightly improved

also there is a negative sharpness control which works wonders in sucking EE out of DVD's and HD for that matter (which is a very rare case)

-Gary

Gary, I don't see any mention of what framerate you are sending from the VP50.

I have been using the VP50 with the HD-A1 and the Marantz VP11S1. I currently feed 1080p48 from VP50 to VP11S1.

Stacey I am a 60hz guy, I have tried to use 48hz/72hz many times before but could never get used to how it looked and preferred 60hz, my output is 1920x800p/60hz and thanks to the user specified input and output res, I can dial in 2.40:1 across the board :)

I haven't enjoyed HT this much in years, it truely is a new game with the Toshiba HD-D1 and the VP50, don't need to forget D-Theater either ;)

this scaler is worth every penny and twice that

-Gary

the simc3x uses dcdi deinterlacing and pixelworks scaling.

Personally I dont think that the Sim does a good job scaling/deinterlacing because I see no difference with the 1080i output or the 720 output of the HD A1. This is the case both before and after version2 firmware.

In addtion, I find that most of my HD is not that much better than the best of SD.

Warren

no I don't care to :)

I have been using the VP50 for a while now in my personal reference HT setup, I can only call it reference because of the VP50 now, and that is no exageration :)...
Gary,

Did the NDA and any extensions expire on 9-17 (looks like you may have broke it by a few seconds:D )? If so it would stand to reason that shipments would start next week.

Also did DVDO update the firmware while you had it and if so what was added and will be the release version?

I am very curious to A/B this unit with my CrystalioII. I love the user interface of the C2 and the fact that it does have two HDMI outs but of course picture Q is my main priority. If the VP50 does as good or better than there is no reason to spend that much more on the C2. Of course doing as good or better will be no small feat. Also I like the fact that I just need to use an USB flash drive to do firmware upgrades. That is 10 times easier than having to disconnect everything and take my unit to my office upstairs. I wish other companies would start this feature. I got my C2 for a great price, as I will the VP50 so it really does come down to picture Q and user flexibility... I am excited to try a VP50 and if Gary is this excited it must be good! :)

Hai,

Can the beta testers comment on the PReP system ?, How far does it go ? does it have settings you can tune ?.

For example can it be used for htpc use ? say take a mini mac make it output 'bad' 1080p say using fieldscaling or bob and recover the 1080i and redo things ? or it its aimed at 480i/576i. also seems to me that its harder to de-deinterlace ntsc than pal is this correct and how well is pal supported in this setup. give us more more more more info ! :)

Daniel.

He can't, DVDO has the NSA watching all his communications very closely. :p

No we dont.... :cool:

Stacey I am a 60hz guy, I have tried to use 48hz/72hz many times before but could never get used to how it looked and preferred 60hz, my output is 1920x800p/60hz and thanks to the user specified input and output res, I can dial in 2.40:1 across the board :)

I haven't enjoyed HT this much in years, it truely is a new game with the Toshiba HD-D1 and the VP50, don't need to forget D-Theater either ;)

this scaler is worth every penny and twice that

-Gary

Gary,

I'm new at this for sure, but isn't all film (meaning what's on the DVD), based material 24fps, so doesn't even blu-ray upscale? Not being true 1080p 60?

Thanks,

Royd

Can the beta testers comment on the PReP system ?, How far does it go ? does it have settings you can tune ?.

To Infinity and beyond... :) PReP is currently not in the 1.0 SW. It should be in the first upate. At CEDIA, it was a user option under Input Adjust. It sits below the deinterlacing option. In the UI it is PReP and the options are on and off.

or it its aimed at 480i/576i.

480p, 576p and 1080p.

I have not seen much discussion, but you can also send in 720p60 and have it output 24, 48 and 72 Hz.

BTW - To get 24p, you actually need to set it in the frame rate menu, not the format menu, even though it is listed there.

To Infinity and beyond... :) PReP is currently not in the 1.0 SW. It should be in the first upate. At CEDIA, it was a user option under Input Adjust. It sits below the deinterlacing option. In the UI it is PReP and the options are on and off.



480p, 576p and 1080p.

I have not seen much discussion, but you can also send in 720p60 and have it output 24, 48 and 72 Hz.

BTW - To get 24p, you actually need to set it in the frame rate menu, not the format menu, even though it is listed there.

well if its To Infinity and beyond... then as a pal user i don't want to feed it 720P60 but 720P50 and 1080P50 and have them turn it into 24,48,72hz. I mean how hard can it be if it has no limits :), really only thing they need is a buffer to store say for a 2hour movie 4% of the total size right ? then resync the video and audio .... now THAT would be a killer feature the last frontier for us pal users......

But really how possible would this PReP be useful for HTPC users ? output using a simple deinterlacer from the PC and then have the vp50 redo it?


Daniel.

I have not seen much discussion, but you can also send in 720p60 and have it output 24, 48 and 72 Hz.

BTW - To get 24p, you actually need to set it in the frame rate menu, not the format menu, even though it is listed there.
I believe DVDO call it Progressive Source Input Cadence Detection and it has been kept a bit quiet, I think it's pretty impressive myself.

24p is the 1080p output I have from my VP50, my Pio 50-inch 1080p panel supports it, but I do find all the VP50's 1080p menus versus refresh rates slightly confusing.

StooMonster

Can the beta testers comment on the PReP system ?
Been waiting to see it ever since I first suggested it, when it turns out that Dale already had it on his to-do list for a couple of years at least.

We didn't have it before CEDIA but I've already sent Josh a bunch of emails asking when I can have it -- sorry Josh.

StooMonster

Been waiting to see it ever since I first suggested it.

We didn't have it before CEDIA but I've already sent Josh a bunch of emails asking when I can have it -- sorry Josh.

StooMonster

Yeah i guess now i know its a upcoming feature i can't blame you beta testers for not having too much feedback yet. Also i hate the fact that its patent pending since some if not all of the ideas for this came from feedback by users. They should use copyright and trademarks for this the patent system stops innovation by its silly range and long expire times ... sigh ...

note to dvdo, you don't need the patent system it seems your users are a loyal bunch and esp. external scaler makers need more innovation not less by having 1 maker block a whole area of features.

Daniel.

Also i hate the fact that its patent pending since some if not all of the ideas for this came from feedback by users.
That's simply not the case. I developed the basic algorithms (and the name) for PReP about 3-1/2 years ago, and the basic concept was not suggested by a 'user' but was developed internally at ABT. The implementation in the VP50 uses virtually the same algorithms developed back then. This has been on ABT's list of potentially patentable IP for about 2-1/2 years (our IP counsel has had this list for at least that long), and has been on the 'to be implemented' list of features for the ABT deinterlacer since its conception over 2 years ago.

- Dale Adams

Thanks for the clarification Dale. As usual way ahead of the curve, and submissions from enthusiastic amateurs. :) Please accept my apology for any confusion caused.

I'm just glad that we are getting PReP in the near future.

StooMonster

That's simply not the case. I developed the basic algorithms (and the name) for PReP about 3-1/2 years ago, and the basic concept was not suggested by a 'user' but was developed internally at ABT. The implementation in the VP50 uses virtually the same algorithms developed back then. This has been on ABT's list of potentially patentable IP for about 2-1/2 years (our IP counsel has had this list for at least that long), and has been on the 'to be implemented' list of features for the ABT deinterlacer since its conception over 2 years ago.

- Dale Adams

Nice to see you back Dale maybe i should not have aimed it like this, but i still think using patents is a bad way to protect your work like this. Im all for protecting works but there are better methods then patents in software. They are bad unless they are very limited in scope but i guess we will see how limited it is in this case.

The patent system on a whole is not good for users or most of the companies involved.

Again sorry Dale if you take this personally you are one of the inovators in this area we all know that and i have the highest respect for your work i just feel the need to point this out whenever i can that software patents are a huge danger and a mess.

Daniel.

PS: atleast i get a bonus point for getting you back on the forum :)

NDA was lifted, I can now talk about the VP50 too :)

For SD, the difference between the VP50 and the VP30/ABT102 are not huge, there's a small improvement in deinterlacing. My torture test passes with almost perfect results. There are very infrequent combings on the test - roughly the same as with the original HQV tests I ran with the DF. Real world material will rarely, if ever, comb.

locking time is fantastic and 2:2 works great even with very tough sources.

HD-DL does the same "ducky lucky" magic as the SD-DL. 1080i locks great and works in both film and video modes. Video mode is a new one and there's simply no combing on 1080i that i can see. Period. I don't have "tough" 1080i sources, but it simply works...

1080i50 and 1080i60 works fantastically well. conversion is a piece of cake and the results are two steps up from my display's own deinterlacer. My 1080 test sequence (5th Element, Lilu breakout scene at 50Hz) passes with fantastic clarity, and amazing detail.

Tomorrow I'm scheduled to receive a 1080P display for testing (PDP5000EX) and look forward to hooking up the VP50 to it and seeing how well it works vs. the display's own processor.

The image is slightly darker than I'm used to (still trying to figure out why), seems like a calibration issue that I need to test.

Gamma feature is great. There are 3 gamma curves for R, G & B. I personally don't see why one would need separate gamma coefficients. What we need is full gamma curve adjustments (PC controlled, or multipoint settings). I think this is more of a proof of concept and it definitely works...

In short, ABT have made dramatic progress and I think they have a definitey winner on their hands.

Im all for protecting works but there are better methods then patents in software.
It's not a software patent (application).

- Dale Adams

Thank you Beta Testers for your comments. :)

Can't wait to hook up mine when it ships...

To Infinity and beyond... :)
I have not seen much discussion, but you can also send in 720p60 and have it output 24, 48 and 72 Hz.

BTW - To get 24p, you actually need to set it in the frame rate menu, not the format menu, even though it is listed there.

I have a couple of questions for you experts based on everything I'm reading. :)

Among other inputs, I'm going to have a Tosh HDDVD player with 1080i output (HDMI) set and a Pioneer Elite 59avi DVD player with 480i output (HDMI) running into the VP50. Since these will be film sources, does that mean I should output 1080p24 into my television. I have an HP 6580N which over HDMI will accept 1080p at 24,30, and 60 fp???

And if I upgrade the Tosh in December to an XA2 which will output 1080p60 what will I want to do in the VP50 for output?

Thanks!

Now that the beta tester's can talk...

Have they fixed the Star Wars bug? The one where the deinterlacer drops in and out of lock on the opening crawl? Episode III is probably the worst. The original theatrical editions are pretty bad at causing this too.

Another one that appears to really confuse the ABT102 is the Pinky and the Brain Season 1 DVD set. Anybody try that one?

- Collin

HD-DL does the same "ducky lucky" magic as the SD-DL. 1080i locks great and works in both film and video modes. Video mode is a new one and there's simply no combing on 1080i that i can see. Period. I don't have "tough" 1080i sources, but it simply works...

1080i50 and 1080i60 works fantastically well. conversion is a piece of cake and the results are two steps up from my display's own deinterlacer. My 1080 test sequence (5th Element, Lilu breakout scene at 50Hz) passes with fantastic clarity, and amazing detail.

In my Review of DVDO iScan VP50 (http://www.avsforum.com/avs-vb/showthread.php?t=724723) I mentioned my 1080i50 video torture test is Discovery HD and National Geographic HD in UK, these frequently defeat my Lumagen Vision Pro HDP's deinterlacing but the VP50 has been completely comb/error free. :)

My Toshiba HD-XA1 HD DVD player looks excellent outputting 1080p24 via VP50.

Tomorrow I'm scheduled to receive a 1080P display for testing (PDP5000EX) and look forward to hooking up the VP50 to it and seeing how well it works vs. the display's own processor.
I've had a Pioneer PDP-5000EX (EMEA code name for 50-inch 1080p plasma) for a few months now, and have been feeding it via VP50 of late. The displays own processing is actually quite good -- it doesn't comb on 1080i50 video like the Lumagen does -- but I think the VP50 makes it even better.

I had to adjust the tint on the 5000EX using VP50's test patterns and green filter because the default image was way too red with VP50. Also, it's quite hard to turn off all the processing in this panel -- it does a lot!

The image is slightly darker than I'm used to (still trying to figure out why), seems like a calibration issue that I need to test.

Gamma feature is great. There are 3 gamma curves for R, G & B. I personally don't see why one would need separate gamma coefficients. What we need is full gamma curve adjustments (PC controlled, or multipoint settings). I think this is more of a proof of concept and it definitely works...
I agree the image is darker, but was easily calibrated with 5000EX.

I also agree that Gamma Control feels like a 'work in progress', but alternatively I think its simple interface could work well for most users.

StooMonster

Have they fixed the Star Wars bug? The one where the deinterlacer drops in and out of lock on the opening crawl? Episode III is probably the worst. The original theatrical editions are pretty bad at causing this too.
I specifically mention the answer to this is my Review of DVDO iScan VP50 (http://www.avsforum.com/avs-vb/showthread.php?t=724723) ;)

StooMonster

I wanted to add also that the VP50 is much much cooler than the VP30, because of the fan, the fan is 100% inaudible for those worried about that ;)

the VP30 case gets very hot, not so with the VP50 :)

you will want to send 1080i only from the HD-DVD players, the 1080p from BR and HD both is not the true 1080p from the disc, it is 1080i converted to 1080p by the players, 1080p is pointless from these players, send the VP50 the 1080i and sit back and enjoy!! :)

-Gary

Among other inputs, I'm going to have a Tosh HDDVD player with 1080i output (HDMI) set and a Pioneer Elite 59avi DVD player with 480i output (HDMI) running into the VP50. Since these will be film sources, does that mean I should output 1080p24 into my television. I have an HP 6580N which over HDMI will accept 1080p at 24,30, and 60 fp???
Likewise my DVD player and HD DVD player are film only sources, therefore I have VP50 set for 24p output. My display handles this very well, motion is smooth an natural.

The menus and DVD extras are generally in video and can look a terrible if you try to play them. But I don't watch extras, and can still use the menus even if they look rubbish, because to me it's worth it for judder free content.

And if I upgrade the Tosh in December to an XA2 which will output 1080p60 what will I want to do in the VP50 for output?
The VP50 can convert 1080p60 to 1080p24 with ease.

StooMonster

I specifically mention the answer to this is my Review of DVDO iScan VP50 (http://www.avsforum.com/avs-vb/showthread.php?t=724723) ;)

StooMonster

Great review.

Bummer they haven't fixed the Star Wars bug yet. Dale said at one point that he had some sort of fix for it. I'm sure we'll see it someday.

- Collin

Great review.

Bummer they haven't fixed the Star Wars bug yet. Dale said at one point that he had some sort of fix for it. I'm sure we'll see it someday.

- Collin

The good news is that now it can be fixed and updates using firmware.

Daniel.

From what I know the bug is actually in the source material encoding.

I watched EPS2 and 3 in 1080i and they look amazing on the VP50.

Have they fixed the Star Wars bug? The one where the deinterlacer drops in and out of lock on the opening crawl? Episode III is probably the worst. The original theatrical editions are pretty bad at causing this too.
The 'film bias' setting on the VP50 will not drop out of lock on the Star Wars opening credits (at least not with a decent quality digital source - I haven't tested too many analog sources). Most of the versions of beta software had a bug where the hardware was not being programmed correctly to fix this, so you may well see some reports from beta users which indicate that this was not fixed. It will still drop out of lock if you use the 'auto' mode, though, as that has all the error-checks for problem sources enabled.

- Dale Adams

Likewise my DVD player and HD DVD player are film only sources, therefore I have VP50 set for 24p output. My display handles this very well, motion is smooth an natural.

The menus and DVD extras are generally in video and can look a terrible if you try to play them. But I don't watch extras, and can still use the menus even if they look rubbish, because to me it's worth it for judder free content.


The VP50 can convert 1080p60 to 1080p24 with ease.

StooMonster

Hi-- Do you have any idea what a DLP like the HP will do with a 1080p24 signal? I'm wondering if its going to interlace it to 1080i24 then do a pulldown to 1080i30 and then deinterlace it to 1080p60???? Any suggestions on how to find out--and the folks on the HP phone line go blank when I ask them this question.

Thanks.

Lore

That reminds me that I usually have SDI input set to 'Film Bias' as that's all I play on DVD player, but atm it's set to 'Auto'; I'll do the same to my HD DVD player HDMI input.

It's a highly recommended tweak, worked a treat on previous iScan models.

StooMonster

Thanks for the clarification Dale. As usual way ahead of the curve, and submissions from enthusiastic amateurs. :) Please accept my apology for any confusion caused.
No apology necessary (except perhaps by me for not saying something earlier).

I think it's only fair to point out that Stoo did independently come up with the idea of extracting the original fields from a deinterlaced signal. He just did it about 3 years after ABT. ;) (And for the record, I don't know if anyone else has come with this idea before ABT. If so, I'm unaware of it.) He had a great motivation for this - a Sky HD box that wouldn't output 576i from HDMI and would only send out a deinterlaced signal. He also came up with a good workaround using a Lumagen processor's ability to create a custom EDID and force the Sky HD box to actually output 576i.

- Dale Adams

VP50 feeding a NEC 1352 CRT PJ via HDMI setup for 2.40:1 on a 2.40:1 Draper Cineperm screen, my output resolution is 1920x800p, eactly the resolution that 2.40:1 HDTV movies have :)

Gary, to set output resolution to 1920x800p, are you simply changing vertical size from 1080 to 800?? If so, how does this work? The VP50 doesnt squash the pic but just removes the black bars when watching 2.40? I'm confused.

The image is slightly darker than I'm used to (still trying to figure out why), seems like a calibration issue that I need to test.
I agree the image is darker, but was easily calibrated with 5000EX.

Is the image "slightly darker" on all the displays tested?

Paul

This might head off a bunch of posts every day asking "when are you shipping?"

Is the image "slightly darker" on all the displays tested?

My calibration settings are identical between the VP30 and VP50 on my Mitsubishi DLP projector.

The 'film bias' setting on the VP50 will not drop out of lock on the Star Wars opening credits (at least not with a decent quality digital source - I haven't tested too many analog sources). Most of the versions of beta software had a bug where the hardware was not being programmed correctly to fix this, so you may well see some reports from beta users which indicate that this was not fixed. It will still drop out of lock if you use the 'auto' mode, though, as that has all the error-checks for problem sources enabled.

I'll have to try it with Film Bias the next time I think of it. I was disappointed to see it still dropping in and out of lock using Auto. The Auto also continues to have problems with a few PAL DVDs, which I reported as Beta feedback.

Is the image "slightly darker" on all the displays tested?


Did anyone also notice this when using the VP30??

Gino, if your source input resolution is 2.40:1 1920x1080i (like for example Tomb Raider HD-DVD or I Robot DVHS) then yes it would be displaying only the 2.40:1 portion(800 lines) as 1920x800p, no scaling , at this time I am just running everything at 1920x800p, including all HD content, DVD's of all aspect ratios, so everything is being scaled except 2.40:1 HD movies, the VP50 is so nice that I don't feel the need to optimize yet for every aspect ratio because that would be "CRT setup hell" on my NEC, I would need at least 5 different setups/memories or more, the VP50 again is so good that I am just running it at 1920x800p 2.40:1

slightly darker, didn't really notice that, I have my VP50 set for 1.10 Gamma R/G/B which gives the best blacks and details on my CRT PJ :)

-Gary

NDA was lifted, I can now talk about the VP50 too :)

For SD, the difference between the VP50 and the VP30/ABT102 are not huge, there's a small improvement in deinterlacing. My torture test passes with almost perfect results. There are very infrequent combings on the test - roughly the same as with the original HQV tests I ran with the DF. Real world material will rarely, if ever, comb.

locking time is fantastic and 2:2 works great even with very tough sources.

HD-DL does the same "ducky lucky" magic as the SD-DL. 1080i locks great and works in both film and video modes. Video mode is a new one and there's simply no combing on 1080i that i can see. Period. I don't have "tough" 1080i sources, but it simply works...

1080i50 and 1080i60 works fantastically well. conversion is a piece of cake and the results are two steps up from my display's own deinterlacer. My 1080 test sequence (5th Element, Lilu breakout scene at 50Hz) passes with fantastic clarity, and amazing detail.

Tomorrow I'm scheduled to receive a 1080P display for testing (PDP5000EX) and look forward to hooking up the VP50 to it and seeing how well it works vs. the display's own processor.

The image is slightly darker than I'm used to (still trying to figure out why), seems like a calibration issue that I need to test.

Gamma feature is great. There are 3 gamma curves for R, G & B. I personally don't see why one would need separate gamma coefficients. What we need is full gamma curve adjustments (PC controlled, or multipoint settings). I think this is more of a proof of concept and it definitely works...

In short, ABT have made dramatic progress and I think they have a definitey winner on their hands.

Ofer thanks for your review.

I wanted to add also that the VP50 is much much cooler than the VP30, because of the fan, the fan is 100% inaudible for those worried about that ;)

the VP30 case gets very hot, not so with the VP50 :)

you will want to send 1080i only from the HD-DVD players, the 1080p from BR and HD both is not the true 1080p from the disc, it is 1080i converted to 1080p by the players, 1080p is pointless from these players, send the VP50 the 1080i and sit back and enjoy!! :)

-Gary

Good news I was worried with this fan.

How look the VP50 processing of 1080i from cable box?

No apology necessary (except perhaps by me for not saying something earlier).

I think it's only fair to point out that Stoo did independently come up with the idea of extracting the original fields from a deinterlaced signal. He just did it about 3 years after ABT. ;) (And for the record, I don't know if anyone else has come with this idea before ABT. If so, I'm unaware of it.) He had a great motivation for this - a Sky HD box that wouldn't output 576i from HDMI and would only send out a deinterlaced signal. He also came up with a good workaround using a Lumagen processor's ability to create a custom EDID and force the Sky HD box to actually output 576i.

- Dale Adams
Thanks for that Dale, you're a gent.

StooMonster

How look the VP50 processing of 1080i from cable box?

In my case it looks fantastic. However I have a fiber optic feed from my cable co. which might make a big difference in the quality of the input signal.

The difference is more spectacular on 480i from cable on SD channels as it was with the VP-30 with ABT102 card. I feel it is slightly better using the VP-50.

Josh, what is my tracking number? :D

In my case it looks fantastic. However I have a fiber optic feed from my cable co. which might make a big difference in the quality of the input signal.

The difference is more spectacular on 480i from cable on SD channels as it was with the VP-30 with ABT102 card. I feel it is slightly better using the VP-50.

Good news.

there is a certain darkness to the unit, it is fixed during calibration. Not really sure what it originates, could be that it's limited to my beta unit only.

Dale,

Could you elaborate more on what you can accomplish by reinterlacing? Does this mean that if something was deinterlaced adequately, the deinterlacer basically kicks in and fixes just those areas that suffer from problems?

Dale,

Could you elaborate more on what you can accomplish by reinterlacing? Does this mean that if something was deinterlaced adequately, the deinterlacer basically kicks in and fixes just those areas that suffer from problems?
No, that's not the way it works. PReP identifies and extracts the original field from each incoming progressive frame, and then runs the resulting sequence of fields (i.e., an interlaced signal) through the VP50's deinterlacer. In effect, it un-does the original deinterlacing and re-deinterlaces the signal.

- Dale Adams

does that mean that lip sync lag is greater when PReP is enabled?

Hi Dale,

Thanks for the clarification, do you simply extract the fields as-is (i.e., odd & even lines)?

got my pre-order in. :)

i'm can't wait to use this with my Zensonic networked media players (z500).
i use these to network recorded video around over a gigabit lan.

in addition to OTA HDTV, i have a lot of NTSC recorded MPG files (recorded at
12mbps) that the Zensonic currently interlaces to 480p over hdmi, so the re-interlacing will be great to have.

can someone explain the EDID change in the vpx0 processor that is supposed
to let/trick the source (z500 in my case) output hdmi at 480i?

got my pre-order in.
Gah :eek: , you made my heart race a little before I realized you meant that you placed your order, not recieved your order. Don't do that to me. I'm all too eager to believe that there's an iScan getting dropped off at my doorstep even as I type. And the 45th won't be for, what, almost a week? :)

can someone explain the EDID change in the vpx0 processor that is supposed to let/trick the source (z500 in my case) output hdmi at 480i?
Many sources output 480p over HDMI and have no software settings to force them to output 480i.

This can be overcome for a number of sources, if EDID is set to...
480i = YES
480p = NO
...the the sources output 480i and not 480p.

* Same applies for 576p and 576i over HDMI.

Although PReP may be a better answer as sometimes sources can be confused by EDID, however, it would be nice to have the choice.

StooMonster

Stoo,
Where do you edit the changes to the EDID file? On the VP50 I presume? And how does one do it? Or it can only be done by the DVDO boys?

Is the VP50 shipping yeeetttt???? Josh u there?

Cheers!

I just orderd mine also....too much of a good deal to pass up with the trade in. I have 4 HDMI sources now so the HD+ just doesn't "fit" well in my system. Life is good.

Where do you edit the changes to the EDID file? On the VP50 I presume? And how does one do it? Or it can only be done by the DVDO boys?
It can't be done on VPx0 range (yet).

To demonstrate that this works I used a Lumagen scaler that has options to EDID settings.

DVDO VPx0 owners in Europe (and UK in particular) have been waiting for this option for quite a while, as European HDMI set-top-boxes all output progressive SD signals and to make use of ABT102 you need an interlaced SD signal.

Although VP50 has PReP, which extracts interlaced signals from progressive sources.

StooMonster

Anyone with a Sony XBR1 considering a VP50? I realize the XBR1 can't accept a 1080p signal, but I'm on the fence about upgrading from my VP30. I'd like to hear from someone who's made the jump and can offer some thoughts.

Thanks!

Is the VP50 shipping yeeetttt???? Josh u there?



Yes and Yes!

Starting today the VP50 started shipping :D

woohoooooooooo!!! :D

Will we be sent tracking/shipping information? Especially us international customers?

Tracking information is sent the moment your order is processed (if you ordered directly from DVDO/ABT). If you purchased your unit elsewhere you will need to contact them to get shipping information.

Sounds like a well executed launch. Congrats to the product manager... :)