Please help to edify ProspectorPete on RGB ranges and proper video setups *spinoff of spinoff*

[Shifty Geezer]

In REC 709 color gamut, black is not defined as r:0 g:0 b:0, however, on a blue ray it is.

Where are you getting that info?

Besides which, it's moot. You said cinema doesn't design for true black. When the creators of any of those space movies you reference set out to represent space, they didn't say to themselves, "let's make space a dark grey instead of black." Space is supposed to be black in those films. Ergo, if it's not black on the display equipment its being shown on, the display equipment is inadequate. In most cases that's because the tech (projector, LCD) can't represent the black properly. In other cases it's because the equipment isn't set up correctly, taking the brightness levels of the encode and failing to map them correctly to the display.

In your images above, the black of space is about 4% brightness. This is below the minimum threshold for blacks in movie standards. Therefore the display apparatus should be set so that everything at 4% brightness (actually 6.25% brightness) is black. There are numerous links in the chain that can mess with this, from the display setup to the output of the playing device to the mapping of the video software.

Viewed on PC using the full RGB range, a histogram like this would be used to map the video image to the output with true blacks and white...

 

[ProspectorPete]

With your (interpretation?) of true black, you just demonstrated the (pre 2017) OLED effect;

compare the red boxes:

In this review they have something similar: http://www.hdtvtest.co.uk/news/kd65zd9-201610164372.htm

Where the ZD9 edged ahead of the OLED was in the area just above black. As Sandra Bullock was hurling through space in Gravity, her silhouette appeared more distinct against the night sky, even if the LG OLED rendered the stars brighter and the blacks deeper. To reveal similar amounts of shadow detail, the LG’s [Brightness] and gamma would have to be adjusted to such an extent that the 0 cd/m2 blacks were ruined, or near-black artefacts were harshly exposed.

 

[ProspectorPete]

Also space is not dark grey, but whatever it is, when the LG B6 for example renders space as true black (as intended by the director right), then there is loss of detail in shadows.

 

[xz321zx]

In such an AMOLED display device, however, TFT characteristics such as driving TFT threshold voltage Vth and process tolerance factors (e.g., mobility, parasitic capacitance, and channel width/length) are non-uniform among pixels due to process tolerances. For this reason, non-uniformity of luminance may occur in the AMOLED display device. To solve this problem, a data compensation method is employed. In accordance with this data compensation method, the characteristic parameters of the driving TFT in each pixel driving circuit are measured, and input data is adjusted, based on the result of the sensing.

Publication date: Mar 24, 2015
Original Assignee: Lg Display Co., Ltd.

Link:

http://www.google.com/patents/US8988329

Only plasma and OLED TV's have the potential to show black levels approaching 0 when other content is being displayed. This is due to their ability to turn off light emittance at the pixel level.

Someone someday should arm her/himself with an actual photon counter and decide that, theory says neither OLED nor PDP nor CRT can do it (bc. glow), until then, valuing or devaluing displays based on this criteria is either OCD-ism or hot air.

 

[xz321zx]

(...)

...so LG's earlier "solution" for per pixel threshold voltage variation is to clamp near black values to "zero" .

Besides, if OLEDs modulation characteristics are so damn good , the LI-FI guys should be all over it , in piles, actually (except they aren't , and it's mostly black because of the black matrix surrounding the subpixels ) .

 

[ProspectorPete]

Talk about thread backfire lol

 

[Silent_Buddha]

Someone someday should arm her/himself with an actual photon counter and decide that, theory says neither OLED nor PDP nor CRT can do it (bc. glow), until then, valuing or devaluing displays based on this criteria is either OCD-ism or hot air.

Hence my wording.

black levels approaching 0

Nowhere did I claim that any of the technologies could actually show 0 while simultaneously being able to be switched on at any moment to display something brighter than 0. That said, OLED and Plasma can get closer to 0 than any existing LCD technology at similar brightness levels.

Regards,
SB

 

[Shifty Geezer]

Measuring photons at 'zero brightness' via a device is immaterial when the thing that matters is how it appears to the human eye, and anyone who's seen an OLED screen in the dark knows the blacks are black and it's awesome!

 

[digitalwanderer]

Slightly, (ok, completely), off-topic; what is your official tip for 2017 Shifty?

 

[xz321zx]

Measuring photons at 'zero brightness' via a device is immaterial when the thing that matters is how it appears to the human eye, and anyone who's seen an OLED screen in the dark knows the blacks are black and it's awesome!

Then if you have uniformly awesome black, you'd also need uniform near black, and they either sidestep that with clamping or it's non-uniform (or it's only uniform for the duration they are letting reviewers do their thing).

Besides ,if stacked LCD can do over >1:200.000 on-off contrast , and peak brightness in >10% screen area without ABL, how emissive tech has edge then, once both has to be calibrated at the pixel level for uniformity ?

 

[xz321zx]

OLED and Plasma can get closer to 0 than any existing LCD technology at similar brightness levels.

OLED panels have always been known to have higher contrast ratios than LCD panels, but that may be about to change with Panasonic's recently announced LCD IPS display. The display boasts a 1,000,000:1 contrast ratio,

Sharp made this around 2004-2006 already, and there are countless university research prototypes because of HDR . Looks like 30-40 watt extra consumption is on the high side, once they can get a 100 nit 40" something TV to work with that "much" . Maybe a pre-polarized light source can help that, not to mention submillisecond response time color sequential.

http://www.avsforum.com/forum/40-ol...9500-oled-owners-thread-555.html#post42605154

http://www.avsforum.com/forum/40-ol...9500-oled-owners-thread-555.html#post42605562

http://www.avsforum.com/forum/40-ol...9500-oled-owners-thread-555.html#post42606282

 

[ProspectorPete]

Besides ,if stacked LCD can do over >1:200.000 on-off contrast , and peak brightness in >10% screen area without ABL, how emissive tech has edge then, once both has to be calibrated at the pixel level for uniformity ?

Fans will now focus on extreme viewing angles, or the fact that OLED can be 2mm thinner?

Btw, you are really knowledgable about this subject, I noticed severe stutter in all LG OLED models; if you turn off all motion smoothing options it becomes stutter town. A colleague showed a test image of a white screen with a black vertical bar running across from right to left. Recording slowmotion footage and analysing you could see the vertical bar was 'angled'; with the top or bottom part coming in earlier.

The panel was not refreshing 'progressive', but from top to bottom. Does that make sense?

 

[xz321zx]

Fans will now focus on extreme viewing angles, or the fact that OLED can be 2mm thinner?

Btw, you are really knowledgable about this subject, I noticed severe stutter in all LG OLED models; if you turn off all motion smoothing options it becomes stutter town. A colleague showed a test image of a white screen with a black vertical bar running across from right to left. Recording slowmotion footage and analysing you could see the vertical bar was 'angled'; with the top or bottom part coming in earlier.

The panel was not refreshing 'progressive', but from top to bottom. Does that make sense?

Newer submillisecond response rate liquid crystals have no problem with angles ( better response time is because of low viscosity and that's because these are only microns thick) .
https://www.reddit.com/r/science/comments/5rhzw9/new_liquid_crystal_could_make_tvs_three_times/

Consumer OLED's motion is sample&hold , both plasma and blinking backlit LCD have edge there. Some Sony mastering grade OLED has blinking , but that's built for 100 nits, so they can make that trade off. Against bright ambient light, with HDR, consumer OLEDs not so much, and consumption is already too high.
Traditionally top-to-bottom updating IS progressive. Usually only displays with binary grayscale try to refresh the whole screen all at once, or so fast , that top to bottom can't be apparent .

 

[goonergaz]

Obviously this discussion has gone way over my head, sorry if my posts were 'beneath' you but I was only trying to help as 99% of the time your suggested issue is fixed tby calibration!

Bit OT, but is there any hope that OLED will get a uniformed colour at various viewing angles? Main reason I still have a plasma, I hate watching football where half the pitch looks lush and the other scorched! Obviously it's not just football, but that's the easiest way to spot the issue and describe the effect to people.

 

[iroboto]

Obviously this discussion has gone way over my head, sorry if my posts were 'beneath' you but I was only trying to help as 99% of the time your suggested issue is fixed tby calibration!

Bit OT, but is there any hope that OLED will get a uniformed colour at various viewing angles? Main reason I still have a plasma, I hate watching football where half the pitch looks lush and the other scorched! Obviously it's not just football, but that's the easiest way to spot the issue and describe the effect to people.

Going to second goonergaz. I just found this thread and I'm going to need a summary of this black level 16-255 bit. I calibrated my Panny plasma but I think I'm still using 16-255 and I don't get pure black. But it's close, it's certainly not snow, but it's certainly not fully off either. I don't get black crush either though. When I switch to 0-255 black crush and stuff start happening.

I followed my calibration numbers from this site:
http://www.plasmatvbuyingguide.com/plasmatv/panasonic-picture-settings.html

 

[ProspectorPete]

Newer submillisecond response rate liquid crystals have no problem with angles ( better response time is because of low viscosity and that's because these are only microns thick) .
https://www.reddit.com/r/science/comments/5rhzw9/new_liquid_crystal_could_make_tvs_three_times/

Consumer OLED's motion is sample&hold , both plasma and blinking backlit LCD have edge there. Some Sony mastering grade OLED has blinking , but that's built for 100 nits, so they can make that trade off. Against bright ambient light, with HDR, consumer OLEDs not so much, and consumption is already too high.
Traditionally top-to-bottom updating IS progressive. Usually only displays with binary grayscale try to refresh the whole screen all at once, or so fast , that top to bottom can't be apparent .

Thanks for your explanation, the artefact which I noticed was most probably caused by sample&hold; you could see a definite gradient vertically which makes sense now that I think of it.

Going to second goonergaz. I just found this thread and I'm going to need a summary of this black level 16-255 bit. I calibrated my Panny plasma but I think I'm still using 16-255 and I don't get pure black. But it's close, it's certainly not snow, but it's certainly not fully off either. I don't get black crush either though. When I switch to 0-255 black crush and stuff start happening.

I followed my calibration numbers from this site:
http://www.plasmatvbuyingguide.com/plasmatv/panasonic-picture-settings.html

It could be better to calibrate it yourself, even without professional tools you should be able to get the correct black level settings.
http://www.avsforum.com/forum/139-display-calibration/948496-avs-hd-709-blu-ray-mp4-calibration.html

I do have to say if you use Xbox then all bets are off because me or friends were never able to get the full picture out of them

 

[xz321zx]

Thanks for your explanation, the artefact which I noticed was most probably caused by sample&hold; you could see a definite gradient vertically which makes sense now that I think of it.

Actually I addressed the choppy motion part , s&h is responsible both for choppy movement and smeared image .

Took me a while to remember this paper , but I think the other artifact you saw is illustrated here in fig. 6 ? :
http://www.poynton.com/PDFs/Motion_portrayal.pdf

 

[ProspectorPete]

Actually I addressed the choppy motion part , s&h is responsible both for choppy movement and smeared image .

Took me a while to remember this paper , but I think the other artifact you saw is illustrated here in fig. 6 ? :
http://www.poynton.com/PDFs/Motion_portrayal.pdf

Amazing paper. Can't believe it was written almost 20 years ago! I don't think the artefact has to do with eye tracking, as it was also visible on camera. I will record it again, but this time also with the camera turned 90 180 and 270 degrees, just to exclude that the artefact is by a combination of the screen refresh and camera sampling. I will upload it here later this week.

Also again thanks for your time and contributions

 

[Silent_Buddha]

Sharp made this around 2004-2006 already, and there are countless university research prototypes because of HDR . Looks like 30-40 watt extra consumption is on the high side, once they can get a 100 nit 40" something TV to work with that "much" . Maybe a pre-polarized light source can help that, not to mention submillisecond response time color sequential.

http://www.avsforum.com/forum/40-ol...9500-oled-owners-thread-555.html#post42605154

http://www.avsforum.com/forum/40-ol...9500-oled-owners-thread-555.html#post42605562

http://www.avsforum.com/forum/40-ol...9500-oled-owners-thread-555.html#post42606282

Yes, everything in discussion had been of technology that is in the consumer market currently. The greyscale banding has been a prominent issue for LG's OLED panels prior to this year. That's reportedly been addressed for this years panels, but the jury is out as to how well and to what extent it has been addressed. I'm waiting to see more data from review sites. One site tested it with expensive lab equipment and said it's much improved, but provide no data on their measurements nor any screenshots of the screen in a darkened room.

I'll be interested to see how well Panasonics light modulating cell will be at controlling light emittance and more importantly how well it'll do at blocking light emittance once it gets to the consumer market. Hopefully it does well and addresses some of the most glaring faults of LCD tech (bad black levels that contribute to halos, whitewash, grey-ish blacks, etc.) which are greatly exacerbated in darkened rooms when watching movies. Unfortunately, I'm guessing it is expensive to manufacture panels using that as the consumer market (even the high end of the consumer market) aren't being targeted for adoption of this.

Suitable applications:
High-end monitors for broadcasting, video production, medical, automotive, and other fields

Hopefully someone decides to put one into the high end consumer TV market even if it'll be 20k USD or more.

Regards,
SB

 

[xz321zx]

Unfortunately, I'm guessing it is expensive to manufacture panels using that as the consumer market (even the high end of the consumer market) aren't being targeted for adoption of this.

High-end monitors for broadcasting, video production, medical, automotive, and other fields

Hopefully someone decides to put one into the high end consumer TV market even if it'll be 20k USD or more.

It's functionally equivalent to an LC layer stack, there is not much cost prohibitive about that.

However if you wanted to make an OLED, with stable emitted wavelengths near threshold voltage , that might be cost prohibitive ... IMHO they probably refer to the wavelengh stability of spatial light modulators (LCD etc. ) with "high-end" .

One site tested it with expensive lab equipment and said it's much improved, but provide no data on their measurements nor any screenshots of the screen in a darkened room.

Saw that link here, that was the review with refusal to send out the TV , so you won't know a thing about long term stability.