I am calibrating my sets with a PS3 and Video Essentials BluRay disk. I have tried the PS3 with superwhite on and off, and with limited range on and off. I cannot get the test pattern to show the below black and above white. What am I doing wrong? I am using HDMI on a 45 inch Aquos and Samsung 550 plasma.

To get BTB and WTW:
Limited = on
Superwhite = on
DVD/Blu-Ray = YPrPb

If you turn limited off, BTB/WTW are clipped and video levels 16-235 are stretched to PC levels 0-254. If you turn superwhite off then BTB/WTW are clipped and only 16-235 are sent to the display (0-15 and 236-254 contain no data).

Go to Blu Ray/DVD settings - turn HDMI colorspace from RGB to ycbcr, then below black will show on the pluge patterns of DVE.

RGB full/limited should be left to Limited, but this setting will only affect games/xmb

If you are setting the display output for Blu Rays to YPb (which is correct), then wouldn't you want to set RGB to Full so that PC level material (such as games and XMB) will be displayed correctly? (or perhaps I should say optimally, with no level re-mapping and color banding)?

It seems like all of the suggestions I see that say set RGB as Limited are telling you the setting it should be for video material. But Blu Ray's should be setup for YPb anyways, so why would you also setup RGB that way?

I used to blindly follow the common RGB limited setting, until some testing by a member noticed some color banding issues with some game trailers using that setting. Of course this assumes that your TV can actually accept a PC level 0-255 signal, which I think many new TV's are certainly capable of. My TV, a Samsung HL61A750 LED based DLP has settings for HDMI black level Normal (goes with RGB full) and Low (goes with RGB limited) that seems to account for this.

For reference, here are some threads in our owners forum where we are discussing this issue right now. I'd really like to hear what other people think on this matter, as I don't understand why you would setup the RGB settings for video, if Blu Rays don't even use that setting?

http://www.avsforum.com/avs-vb/showthread.php?p=13985616#post13985616
http://www.avsforum.com/avs-vb/showthread.php?p=13992461#post13992461

If you are setting the display output for Blu Rays to YPb (which is correct), then wouldn't you want to set RGB to Full so that PC level material (such as games and XMB) will be displayed correctly? (or perhaps I should say optimally, with no level re-mapping and color banding)?

It seems like all of the suggestions I see that say set RGB as Limited are telling you the setting it should be for video material. But Blu Ray's should be setup for YPb anyways, so why would you also setup RGB that way?

I used to blindly follow the common RGB limited setting, until some testing by a member noticed some color banding issues with some game trailers using that setting. Of course this assumes that your TV can actually accept a PC level 0-255 signal, which I think many new TV's are certainly capable of. My TV, a Samsung HL61A750 LED based DLP has settings for HDMI black level Normal (goes with RGB full) and Low (goes with RGB limited) that seems to account for this.

For reference, here are some threads in our owners forum where we are discussing this issue right now. I'd really like to hear what other people think on this matter, as I don't understand why you would setup the RGB settings for video, if Blu Rays don't even use that setting?

http://www.avsforum.com/avs-vb/showthread.php?p=13985616#post13985616
http://www.avsforum.com/avs-vb/showthread.php?p=13992461#post13992461


Are you going to change the TV's level depending on if you are playing games or watching Blu-ray?

Are you going to change the TV's level depending on if you are playing games or watching Blu-ray?

No, the setting on the TV only applies when viewing RGB sources. So, when watching Blu rays it is grayed out, and has no effect. It only becomes changeable in games and XMB dashboard. So, it seems to autosense on some level, which is nice. Maybe this is a new feature, or not common to other non Samsung sets? Has this been a limitation, and is it why so many people have recommended RGB limited?

No, the setting on the TV only applies when viewing RGB sources. So, when watching Blu rays it is grayed out, and has no effect. It only becomes changeable in games and XMB dashboard. So, it seems to autosense on some level, which is nice. Maybe this is a new feature, or not common to other non Samsung sets?

Some TV's try to autodetect it, but with a varying degree of success. My Sony A3000 can't do it from the PS3. I honestly don't know how a TV would know unless there is a provision to signal it over HDMI.

The reason it is grayed out for YCbCr signals on your set is probably that virtually every video source (which are all encoded as YCbCr if they are on optical media) should use 16-235, so the TV just uses 16-235. When it sees an RGB signal, well that could actually be a PC signal or otherwise where you are trying to use the TV as a PC monitor, in which case it gives you a choice. If you were using it as a PC monitor, you would want 0 - 255.

Of course, all of this is speculation since I don't own your set, but it would make sense to do it that way.

Has this been a limitation, and is it why so many people have recommended RGB limited?

No. The reason RGB limited is suggested is quite simple: it is the correct setting for video sources. You can debate about games, but when it comes down to it games don't always adhere to any standard anyway. It seems to be getting better recently, and every game I have seen looks great with RGB limited. With games, the graphics are generally being generated on the fly, and it is (should) be about as easy to output 16-235 as it is 0-255. IMO games designed for consoles should output 16-235 to be consistent with video standards.

I downloaded the Wipeout content that was referenced in the links you provided. What you are seeing is not banding, but an "effect" they have done. Pause the video when one of the gray "slides" first comes into view. You will see several other slides stacked behind it, each one of them being transparent. What you are seeing is the other slides behind the first one, not banding. To further verify this, look at the shadow underneath the slide. Any banding there? Any banding in the actual game material?

A lot of people mistake these kinds of things all the time. There are people who swear up and down that RGB Full gives "better blacks," when all they are seeing is black crush. The black isn't any deeper, there is just more of it since all the near black shades of gray are being crushed into black.

Good info. In the wipeout demo, I know the "effect" you are talking about, and the "banding" we are talking about is definitely different. It is noticeable on our sets. Here is by far the best description I've seen on the matter:
http://www.avsforum.com/avs-vb/showthread.php?p=12581788#post12581788

Seems like with MY set, I need to decide if I want optimum (non-banding) on games or video.

I downloaded the Wipeout content that was referenced in the links you provided. What you are seeing is not banding, but an "effect" they have done. Pause the video when one of the gray "slides" first comes into view. You will see several other slides stacked behind it, each one of them being transparent. What you are seeing is the other slides behind the first one, not banding. To further verify this, look at the shadow underneath the slide. Any banding there? Any banding in the actual game material?

A lot of people mistake these kinds of things all the time. There are people who swear up and down that RGB Full gives "better blacks," when all they are seeing is black crush. The black isn't any deeper, there is just more of it since all the near black shades of gray are being crushed into black.

Good info. In the wipeout demo, I know the "effect" you are talking about, and the "banding" we are talking about is definitely different. It is noticeable on our sets. Here is by far the best description I've seen on the matter:
http://www.avsforum.com/avs-vb/showthread.php?p=12581788#post12581788

Seems like with MY set, I need to decide if I want optimum (non-banding) on games or video.

Where exactly are you seeing banding?

Hmmm, makes me question what i believe on this matter...


Isn't the point of BTB just to be one more reference for black level calibration?

I run on full RGB both on ps3 and TV and still manage to calibrate black level based on the 2% white bar (good old make it disappear then go back 1 notch).





Well while i was thinking what to post i realized this must be a preference thing, either to calibrate your tv for ybp + WTW or for full rgb.

I guess the absolute best would be simply write down picture (ideally black level only, right?) settings for each mode and adjust the TV depending on the content.

My TV has 5 different settings per video input but sadly all but cinema seems to be setup to oversaturate red :( so i can't try that until i get a hold of a colorimeter and mess with the colors from the service mode...


So i guess i would choose RGB full as i play games much more than movies, to each his own.

Here's some more fuel for the fire. Unless you know for sure that you have a display that has both a minimum 10 bit panel and minimum 10 bit processing, there is no possible way to avoid banding when using the warm (D65) color temperature. Due to the fact that digital displays require different amounts of headroom for each primary to achieve D65, on a 8 bit display atleast 1 of the primary colors will have far less then 255 steps (and possibly less then even 219 steps which video levels require). Additionally it's unlikely that an 8 bit calibrated display at any color temperature has exact 1:1 step mapping for any of the primaries. Each of the RGB colors will be individually be mapped to whatever range is available. You may have a situation where red has 250 steps, green has 200 steps and blue has 220 steps. No matter if you choose PC or video levels you will have either skipped or repeated steps for each primary. All of that isn't even taking gamma into account. After your set remaps the colors for gamma, it's just a huge mess as far as level mapping on an 8 bit display. Only when you have a minimum 10 bit display/processing does levels of the actual source input really come into play.

Also with video game systems there is no set standards for output. As was stated by Mr.D in another thread, a sure sign that a game isn't using any kind of set standard is the inclusion of gamma/brightness controls.

http://www.avsforum.com/avs-vb/showthread.php?t=967106

Generally speaking I've found the higher profile 360 games ( Halo3 , Mass Effect) seem to know about video levels: Ubisoft seem to generally screw this up Far Cry on the consoles is painfully using PC levels. Oblivion looks to be using PC levels too but at least they have included a brightness slider ( which they wouldn't need if they knew what they were doing , when's the last time you saw a dvd or BD/HDDVD with a brigtness slider on the menu screen?).

I actually expect games artists to know literally nothing about this sort of stuff these days. I've even had one or two suggest its not an issue and standards don't matter: they soon get to grips with it when I thrown their CG back at them a few times because they can't tell me what colourspace it is. I'm not going to waste my time figuring out what they intended it to look like when they can't even tell themselves.

Calibrate for video levels and there are games that use PC levels. Calibrate for PC levels and there are games that use video levels. In short, you can't win. Just pick one and learn to live with it.

Where exactly are you seeing banding?

There is the "effect" that you mentioned, the square border type things around the image. Then, the banding I see is in the body of the gray box. There are round contour banding lines where the gray transitions from lighter near the center of the box, down to the bottom darker gray edge of the box. I can try and snap a picture to illustrate it, if you don't follow my description. When I switch from RGB Full/HDMI Black Normal the banding goes away, and it appears like a smooth gradient. Switch to RGB Limited/HDMI Low, and everything looks nearly identical, (Contrast, brightness, etc stay the same), except there is then slight banding in what used to me a smooth gray gradient.

I think sperron summed it up pretty well, (thanks for your great posts by the way, very helpful!). It may not be possible to eliminate all banding on these digital displays. I know under certain conditions, I can see some minor banding on various test patterns, gray scale ramps, etc.

I think you just pick one setup that will work for the majority of your intended use, and live with any of the minor consequences, (like some banding if the content was encoded the other way, or adjusting a game brightness bar).

I guess I'm fortunate as it is easy to switch my set back and forth with the RGB range and HDMI black level setting, so if I really care I can counteract some of the minor annoyances. At least I don't need to adjust the contrast or brightness level, as that would get very tedious having to change that everytime as well.

Here's some more fuel for the fire. Unless you know for sure that you have a display that has both a minimum 10 bit panel and minimum 10 bit processing, there is no possible way to avoid banding when using the warm (D65) color temperature. Due to the fact that digital displays require different amounts of headroom for each primary to achieve D65, on a 8 bit display atleast 1 of the primary colors will have far less then 255 steps (and possibly less then even 219 steps which video levels require). Additionally it's unlikely that an 8 bit calibrated display at any color temperature has exact 1:1 step mapping for any of the primaries. Each of the RGB colors will be individually be mapped to whatever range is available. You may have a situation where red has 250 steps, green has 200 steps and blue has 220 steps. No matter if you choose PC or video levels you will have either skipped or repeated steps for each primary. All of that isn't even taking gamma into account. After your set remaps the colors for gamma, it's just a huge mess as far as level mapping on an 8 bit display. Only when you have a minimum 10 bit display/processing does levels of the actual source input really come into play.

Also with video game systems there is no set standards for output. As was stated by Mr.D in another thread, a sure sign that a game isn't using any kind of set standard is the inclusion of gamma/brightness controls.

http://www.avsforum.com/avs-vb/showthread.php?t=967106



Calibrate for video levels and there are games that use PC levels. Calibrate for PC levels and there are games that use video levels. In short, you can't win. Just pick one and learn to live with it.

I'm trying to process what you are saying here and make sure I have it straight. If I have it correct, then what you are saying is that RGB Bias/Gain controls work by decreasing/increasing the available bit depth?

For example, suppose Red and Green Gain is perfect with respect to D65, but Blue is too high. When I turn the Blue Gain control down, the set starts limiting the available blue bit depth so that the highest it can display is something less than 255?

There is the "effect" that you mentioned, the square border type things around the image. Then, the banding I see is in the body of the gray box. There are round contour banding lines where the gray transitions from lighter near the center of the box, down to the bottom darker gray edge of the box. I can try and snap a picture to illustrate it, if you don't follow my description. When I switch from RGB Full/HDMI Black Normal the banding goes away, and it appears like a smooth gradient. Switch to RGB Limited/HDMI Low, and everything looks nearly identical, (Contrast, brightness, etc stay the same), except there is then slight banding in what used to me a smooth gray gradient.

I think sperron summed it up pretty well, (thanks for your great posts by the way, very helpful!). It may not be possible to eliminate all banding on these digital displays. I know under certain conditions, I can see some minor banding on various test patterns, gray scale ramps, etc.

I think you just pick one setup that will work for the majority of your intended use, and live with any of the minor consequences, (like some banding if the content was encoded the other way, or adjusting a game brightness bar).

I guess I'm fortunate as it is easy to switch my set back and forth with the RGB range and HDMI black level setting, so if I really care I can counteract some of the minor annoyances. At least I don't need to adjust the contrast or brightness level, as that would get very tedious having to change that everytime as well.

I went back and looked at that trailer again, and I think I see what you are seeing. It looks like a circular banding pattern on the slides.

I'm still not convinced it is a PC vs. Video level problem. I tried it both ways on my TV (Sony A3000) and I saw the banding either way. I also tried mismatching it so that the TV expected PC levels and the PS3 output Video levels, and vice versa, but I still saw the banding.

I'm not saying 100% that it isn't a level problem, but I would suspect that this particular case is just in the source. I have never seen banding on my set when viewing gray ramps.

I'm trying to process what you are saying here and make sure I have it straight. If I have it correct, then what you are saying is that RGB Bias/Gain controls work by decreasing/increasing the available bit depth?

For example, suppose Red and Green Gain is perfect with respect to D65, but Blue is too high. When I turn the Blue Gain control down, the set starts limiting the available blue bit depth so that the highest it can display is something less than 255?

Exactly. An 8 bit panel only has 255 steps per color. If you have head room left on any of your gains, that means there are digital steps that are not being used. At the other end of the spectrum you can have a "run out" on one of you colors (generally red) which means that all 255 of it's steps are being used and it doesn't have enough steps to match the other colors at higher levels of stimulation (100% white for instance). When you balance your grayscale you are matching the digital level of each color which is needed to achieve D65 at any particular stimulation percentage. Then there is also gamma which again remaps your digital levels for each stimulation percentage. In effect the your gains dictate the upper limit on digital levels per color while your bias and gamma controls stretch the levels like a rubber band to try and match your calibration targets. This whole process also happens on a 10 bit panel with 10 bit processing, but hopefully there are enough steps that banding shouldn't be an issue.

I went back and looked at that trailer again, and I think I see what you are seeing. It looks like a circular banding pattern on the slides.

It's useless to use a compressed trailer to judge banding. It's especially useless to do so when the trailer was most likely automatically compressed with no hand tuning. It's more likely that any banding you see is a result of the video compression. The game may or may not reflect the same banding. Check out the pirates of the carribean at world's end trailer available on PSN. There is plenty of banding in the dark areas and fog that is not on the actual Blu-Ray. Also with a video game trailer are they using PC levels in and video levels out? If so then banding may be introduced when being converted from 0-255 to 16-235 before the trailer was even encoded.

Exactly. An 8 bit panel only has 255 steps per color. If you have head room left on any of your gains, that means there are digital steps that are not being used. At the other end of the spectrum you can have a "run out" on one of you colors (generally red) which means that all 255 of it's steps are being used and it doesn't have enough steps to match the other colors at higher levels of stimulation (100% white for instance). When you balance your grayscale you are matching the digital level of each color which is needed to achieve D65 at any particular stimulation percentage. Then there is also gamma which again remaps your digital levels for each stimulation percentage. In effect the your gains dictate the upper limit on digital levels per color while your bias and gamma controls stretch the levels like a rubber band to try and match your calibration targets. This whole process also happens on a 10 bit panel with 10 bit processing, but hopefully there are enough steps that banding shouldn't be an issue.

Is there some place where this is discussed? I'm not saying you are wrong, but it seems very odd. Have you seen banding increase on blue/red/green ramps as the gains are decreased?



It's useless to use a compressed trailer to judge banding. It's especially useless to do so when the trailer was most likely automatically compressed with no hand tuning. It's more likely that any banding you see is a result of the video compression. The game may or may not reflect the same banding. Check out the pirates of the carribean at world's end trailer available on PSN. There is plenty of banding in the dark areas and fog that is not on the actual Blu-Ray. Also with a video game trailer are they using PC levels in and video levels out? If so then banding may be introduced when being converted from 0-255 to 16-235 before the trailer was even encoded.

Agreed, which is why I commented that it was probably in the source.

It's not quite as straightforward as I have presented it since the manufacturer should have implemented some form of dithering to minimize banding/posterizaton. Dithering is what makes a 6 bit TN computer LCD panel look acceptable even though they are not actually an 8 bit panel. Here's (http://www.behardware.com/articles/606-4/test-the-influence-of-electronic-components-on-lcd-colors.html) a page on dithering in relation to a 6 bit TN panel that I just dug up. Here's (http://en.wikipedia.org/wiki/Dithering) the official wikipedia article on dithering. Check out the dithering examples using the cat picture in the wikipedia article. If the 256 color picture can look that good, you can imagine that an 8 bit display can look far better even if it has less then the full 8 bit range to work with. This an example of why quality processing in a TV is so important. Excellent processing can make up for many shortcomings in the actual panel and possibly the source material.

Anyways, it's yet another reason why people assuming that they are really missing anything by going with limited might be wrong. If the raw RGB renders that the PS3's video card does are properly dithered from 0-255 to 16-235, then there will most likely be no banding and very little subjective difference between the 2 on a properly calibrated display. Of course we don't know how the PS3 accomplishes PC levels to video levels. It could be that developers are rendering video levels inside the standard PC levels and sending that out (in this situation you could get banding using full RGB since it would have to stretch 16-235 to 0-254 for display). We really need an experienced PS3 dev to straighten this all out for us.

I could not get it to work. I changed BD output from RGB to the other. I change superwhite to on, and range to limited. No below black through HDMI.

Not that any firmware that I'm aware of didn't pass BTB, but have you tried updating the firmware in your PS3? Also, are you running straight from the PS3 to your TV? It's unlikely, but something like a receiver could be stripping the BTB info from the signal. Have you had another source pass BTB to your display over HDMI?

Is there some place where this is discussed? I'm not saying you are wrong, but it seems very odd. Have you seen banding increase on blue/red/green ramps as the gains are decreased?
This is certainly the case with LCDs, and very likely most other digital displays. It's one of the reasons why, even though virtually all sources are 8-bit, even 10-bit processing/panels is not enough to avoid posterisation in the image.

I've yet to find any digital display that can produce smooth gradients. Pioneer's Plasmas have been the best I've found so far from limited amounts of testing, but they have a rather noisy/dithered image, presumably to try and mask it.

It's one of the reasons why—with LCDs at least—you also don't want to touch the contrast/brightness controls unless you have to.

I'd love to see what high end monitors such as Eizo's LCDs with 16-bit LUTs look like, but even if they can produce gradients on par with a CRT and have good colour reproduction etc., the black levels won't be anywhere near good enough for me and motion handling won't be any good. (and obviously being monitors they'll be quite small)

Most colour management systems also increase the amount of posterisation in the image.
I could not get it to work. I changed BD output from RGB to the other. I change superwhite to on, and range to limited. No below black through HDMI.
The PS3 will only show BTB/WTW when BD/DVD playback is set to YCC (not RGB or Auto) and you have ‘Super White’ enabled. If you still can't see it when Super White is enabled, then your display—or possibly something else in your display chain—is clipping BTB/WTW.

Exactly. An 8 bit panel only has 255 steps per color.

This is only part of the picture, and not really correct. A theoretical 8-bit panel would have only 256 steps, but a digital panel is a linear device, and requires a great deal more than 8-bit past the LUT. If the panel is only 8-bit, you will get very severe banding always with 8-bit content (which is gamma corrected). But I'm not aware of any 8-bit displays out there now, but maybe there are such piss-poor displays. If the gains/cuts operation are done in the 8-bit domain before the LUT (unlikely since that's a terrible design), then yes it is likely that adjusting them will introduce banding issues as the ranges are remapped in an 8-bit domain. I've not encountered a digital display that does this. Some processing methods on PC software did, and obviously did severe damage to the image.

If you have head room left on any of your gains, that means there are digital steps that are not being used. At the other end of the spectrum you can have a "run out" on one of you colors (generally red) which means that all 255 of it's steps are being used and it doesn't have enough steps to match the other colors at higher levels of stimulation (100% white for instance).

The the thing is that most any digital display will have a great deal more than just 255 steps. There may be one out there that literally is only 8-bit, but it's a display you'd never want to use regardless. Any possible damage by adjusting cuts/gains would be absolutely DWARFED by the severe damage done to the image by the limits imposed by the fact that it's 8-bit to start with.

When you balance your grayscale you are matching the digital level of each color which is needed to achieve D65 at any particular stimulation percentage. Then there is also gamma which again remaps your digital levels for each stimulation percentage. In effect the your gains dictate the upper limit on digital levels per color while your bias and gamma controls stretch the levels like a rubber band to try and match your calibration targets. This whole process also happens on a 10 bit panel with 10 bit processing, but hopefully there are enough steps that banding shouldn't be an issue.

10 bit isn't even enough to handle 8-bit properly. It's why you will see some minor banding effects as you adjust most digital displays, but this will often appear even in the default setting. There sometimes is one spot or a range where banding issues may be slightly more reduced or more pronounced, but unless you've got 12 bits or more through the LUT, you don't really have enough to de-gamma 8-bit content cleanly. And to test this, you need to use the banding free deep ramps patterns on Avia II or Avia PRO. Most other ramp patterns on test discs have inherent banding to some degree.


It's useless to use a compressed trailer to judge banding. It's especially useless to do so when the trailer was most likely automatically compressed with no hand tuning. It's more likely that any banding you see is a result of the video compression. The game may or may not reflect the same banding. Check out the pirates of the carribean at world's end trailer available on PSN. There is plenty of banding in the dark areas and fog that is not on the actual Blu-Ray. Also with a video game trailer are they using PC levels in and video levels out? If so then banding may be introduced when being converted from 0-255 to 16-235 before the trailer was even encoded.

This also, is very likely. Banding is a common problem in many systems, but it is also a common problem on a lot of content out there. Many people attribute banding issues to either systems or content without having a reference to know which it really is (and it could be both!).

Not that any firmware that I'm aware of didn't pass BTB, but have you tried updating the firmware in your PS3? Also, are you running straight from the PS3 to your TV? It's unlikely, but something like a receiver could be stripping the BTB info from the signal. Have you had another source pass BTB to your display over HDMI?

I am running through a Sony ES5300 receiver.

It is a new PS3 with an internet update. I assume that updates the firmware.

I recently updated my PS3 and my BTB is gone. It used to pass just fine through a Denon 2807 -> Lumagen HDP -> RS-1. My A35 passes BTB and the PS3 used to. I wish I could go back to whatever firmware I had before, but I wanted to get the DTS HD MA decoding...

I have a slightly related question:


I calibrated an old LCD TV for the PS3 over HDMI using DVE basics BluRay, before doing that I needed to know if the TV supported full RGB, so I turned it on in the PS3 display menu and looked at this image: http://nicolaspeople.com/ch3rokeesblog/wp-content/uploads/2008/01/fullrgb_test.jpg

I could see the first 16 squares so after "verifying" that the TV supported full RGB I proceeded to calibrate using DVE BR, I set my BR/DVD output to RGB too, since I mostly play games and don't mind losing a bit of accuracy for movies.


I couldn't see any black crush on the ramps while calibrating, mind you, I wasn't looking for it either, still I didn't notice anything weird.


So yesterday I finally had some time to get and burn an AVS HD 709 disc and went straight to the flashing black lines pattern... I couldn't see a thing below 17! Switch it to RGB Limited and back just in case... nothing! Loaded up my RGB test picture and I can still see all the squares...


So... at this point I'm just perplexed and need your advice guys.



I didn't have much time but I think my test with the RGB squares image was like this:

PS3 set at full RGB (TV has no related options), load RGB squares, can see them all.
Go to video settings and set PS3 to RGB limited, first 15 squares disappear (verified this by increasing brightness to high levels).



Thanks in advance.

I could see the first 16 squares so after "verifying" that the TV supported full RGB I proceeded to calibrate using DVE BR, I set my BR/DVD output to RGB too, since I mostly play games and don't mind losing a bit of accuracy for movies.


I am confused, you are unclear here. You seem to be saying that you're using another BR/DVD player as a source other than the PS3? Is this correct?

If so, that device may be clipping, unlike the PS3.

Nope, sorry, I am saying I looked at the picture using the PS3's picture browser. And by 'BR/DVD" I meant I went into the BR and DVD settings menu in the PS3 and set the movie output format to RGB (the full/auto/limited option is actually in the general display settings menu).


:)



Sorry, I assumed everyone would be familiar with the PS3 settings menu. There is only 1 TV and the PS3 involved in all this.

Can you see below black on the PS3 using the pluge patterns on the DVE HD Basics disc?

I will try in a couple of hours when I get home.

As far as I know if I set movie output to ycbcr and superwhite on the PS3 will pass BTB right?


Dang that's obviously the best test for BTB support on the display.



I'll post my findings ASAP.



BTW is there some sort of database where I can report my findings on capabilities of screens or players?

Here is what I found:

YPrPb + superwhite = BTB OK
YPrPb + superwhiteOFF = NO BTB

RGB full or limited = NO bars under 17 on AVCHD



All the cases above: The RGB square test always shows all the squares...





I guess that solves it.

1- You can NOT test if a TV supports full range RGB by looking at pictures on the PS3.
2- The TV does not support full RGB.



Thanks guys.


The TV is a polaroid TLMB2626 by the way.

wow, good info here.. thanks guys

Here is what I found:
YPrPb + superwhite = BTB OK
YPrPb + superwhiteOFF = NO BTB

RGB full or limited = NO bars under 17 on AVCHD
That's correct, as I said above, the PS3 only shows BTB/WTW information when set to YCC with Super White on. With anything else, BTB/WTW is clipped.


1- You can NOT test if a TV supports full range RGB by looking at pictures on the PS3.
2- The TV does not support full RGB.
That's not correct, you can test to see whether or not a display supports Full RGB with test images in the photo viewer/web browser.

With DVD/BD it always clips BTB/WTW with RGB.
With games/other functions, it does not.

When set to Limited RGB, it renders everything with an 0-255 range (well, it may actually use higher than 8-bit precision internally, I'm not sure, but to keep things simple, lets call it 0-255) and then compresses that to 16-235. So you lose some precision/steps of gradation but do not lose BTB/WTW information.

When set to Full RGB, it renders it all as 0-255, and outputs 0-255. If your display does not support Full RGB, BTB/WTW will be clipped. (BTB in particular, WTW information will sometimes be shown, but will not look correct)