HDTVTEST (Vincent Teoh) Discussion Thread

[PUsokrJosh305]

Originally Posted by Auditor55

Well, how do you supposed to take advantage of the wide color gamut that Dolby Vision and HDR are supposed to offer with a display with limited brightness? Isn't that the reason display manufacturers are trying to produce displays with higher and higher peak brightness, which is mainly to maximize HDR formats?

It's interesting that this is being discussed because I'm now having 2nd thoughts about using LG's DTM on Dolby Vision Titles over just turning Dolby Vision "On" on my Panny 820.

I popped in Spider-Man: No Way Home the other day just to check it out and noticed the Dolby Vision version on the disc had a bit more "color pop" than just using LG's DTM. In fact, it seemed as the the regular HDR10 version had a bit more gray/dullness to it, even with DTM "On." Plus, my Panny 820 is going down to 800 nits as the minimum instead of 860 or something like that that Now Way Home is graded for.

Because of this, I'm now starting to think that I may need to enable Dolby Vision for those titles and when I play HDR10 discs, I will still use the DTM.

[chip75]

The thing that grinds my gears is that Samsung has four other picture modes to ruin with most of it's TVs, but leave FILMMAKER MODE alone, red flags were them doubling it's brightness to 200 nits in SDR for starters.

Not that they'll ever do anything (or even read this), but:

Dear Samsung,

Do your funky stuff with Dynamic, Standard, Natural, and Movie modes, but try and have FILMMAKER MODE as accurate as possible, it doesn't hurt any of your customers (you may even gain some). You're forced to put Energy warnings whenever you change something that affects power output, but you can voluntarily add, "FILMMAKER MODE represents an image closest to industry standards, images may appear less vibrant that other picture modes."

Or Filmmaker Mode could actually mean something to the UHD Alliance that licence the trademark to manufacturers and they could call them out when they literally disregard creator's intent and the spirit of the whole thing.

[Geoff D]

Originally Posted by Auditor55

Well, how do you supposed to take advantage of the wide color gamut that Dolby Vision and HDR are supposed to offer with a display with limited brightness? Isn't that the reason display manufacturers are trying to produce displays with higher and higher peak brightness, which is mainly to maximize HDR formats?

The need for moar nits isn't to blast your eyeballs to death every time but to have it in reserve should you need it, to get towards a future when tone mapping isn't actually needed because PQ-based HDR is an absolute luminance system i.e. the nits it's mastered to are the nits it should be displayed at, and if it can't be displayed at those nits then it needs to be tone mapped to 'fit' x display. This is Dolby Vision's core USP, that the tone mapping is derived from a suite of metadata generated by taking the master HDR grade and carefully adjusting it to fit a certain level of performance like 100 nits, 600 nits, 1000 nits and so on.

So saying "Dolby Vision is useless if it hasn't got the brightness" is completely missing the point of what DV is and does. If you personally feel that a TV isn't "bright enough" for DV then the exact same thing applies to HDR10 because it literally uses the same transfer function as DV, but you didn't mention HDR10 in that post.

[Staying Salty]

Originally Posted by chip75

The thing that grinds my gears is that Samsung has four other picture modes to ruin with most of it's TVs, but leave FILMMAKER MODE alone....

Or Filmmaker Mode could actually mean something to the UHD Alliance that licence the trademark to manufacturers and they could call them out when they literally disregard creator's intent and the spirit of the whole thing.

Originally Posted by Staying Salty

For Samsung, that mode will be known as "Creator's Intent".

[GiuseppeM]

https://www.4kfilme.de/erschummelt-s...-oled-tv-s95b/

[I DO BLU]

PS5 VRR Goes Live Without System Update! But Sony Forgot to Do This. - YouTube

[Auditor55]

Originally Posted by Geoff D

The need for moar nits isn't to blast your eyeballs to death every time but to have it in reserve should you need it, to get towards a future when tone mapping isn't actually needed because PQ-based HDR is an absolute luminance system i.e. the nits it's mastered to are the nits it should be displayed at, and if it can't be displayed at those nits then it needs to be tone mapped to 'fit' x display. This is Dolby Vision's core USP, that the tone mapping is derived from a suite of metadata generated by taking the master HDR grade and carefully adjusting it to fit a certain level of performance like 100 nits, 600 nits, 1000 nits and so on.

So saying "Dolby Vision is useless if it hasn't got the brightness" is completely missing the point of what DV is and does. If you personally feel that a TV isn't "bright enough" for DV then the exact same thing applies to HDR10 because it literally uses the same transfer function as DV, but you didn't mention HDR10 in that post.

That's a pretty good explanation. With that being said, wouldn't a higher nit capable display better maximize what DV does?

[Auditor55]

Originally Posted by I DO BLU

Nobody Ever Believed S95B Was Accurate. QD OLED Is Still Great. - YouTube

Why would any true videophile try to defend inaccurate displays. Inaccurate displays are totally distracting.

[BrownianMotion]

Originally Posted by Auditor55

That's a pretty good explanation. With that being said, wouldn't a higher nit capable display better maximize what DV does?

A higher nit display would, of course, be better for HDR in general, because it would have to rely less on tone mapping. In fact, that's why you'd see less of a difference between regular HDR10 and DV on a high nit display. In the most extreme case (a 10,000 nit display) you wouldn't need tone mapping at all, so the dynamic metadata of DV would be irrelevant.

DV does have benefits other than dynamic metadata that are unrelated to tone mapping, though.

[Auditor55]

Originally Posted by BrownianMotion

A higher nit display would, of course, be better for HDR in general, because it would have to rely less on tone mapping. In fact, that's why you'd see less of a difference between regular HDR10 and DV on a high nit display. In the most extreme case (a 10,000 nit display) you wouldn't need tone mapping at all, so the dynamic metadata of DV would be irrelevant.

DV does have benefits other than dynamic metadata that are unrelated to tone mapping, though.

I see less of a difference already on my C8 between HDR and DV. My C8 was measured at 910 nits on a 10% window.

The S95B, according to FOMO, was measured at 1500 nits on a 10% window, which is much brighter than any OLED to date, yet some folks complain it won't do DV so they reject it.

[BrownianMotion]

Originally Posted by Auditor55

I see less of a difference already on my C8 between HDR and DV. My C8 was measured at 910 nits on a 10% window.

The S95B, according to FOMO, was measured at 1500 nits on a 10% window, which is much brighter than any OLED to date, yet some folks complain it won't do DV so they reject it.

I wouldn't trust "FOMO's" measurements. Was the set in filmmaker mode and properly calibrated, or did he measure 1500 nits in torch mode?

Anyways, I think most people who reject the S95B will do so on the basis of its accuracy, or lack thereof.

[teddyballgame]

Originally Posted by Auditor55

I see less of a difference already on my C8 between HDR and DV. My C8 was measured at 910 nits on a 10% window.

The S95B, according to FOMO, was measured at 1500 nits on a 10% window, which is much brighter than any OLED to date, yet some folks complain it won't do DV so they reject it.

The 1500 nit readings were bursts that could not be sustained. No idea if that transfers to real content either.

Samsung has lost credibility when it comes to test windows though. They have some undefeatable black box of processing working behind the scenes at all times. I'd be more interested in measurements derived from real content that a calibrator is intimately famliar with and has a good understanding for the luminosity and color values for specific parts of different scenes. You'd need a reference monitor though.

[chip75]

Originally Posted by teddyballgame

You'd need a reference monitor though.

If you say say "Reference Monitor" three times into a glossy OLED panel Vincent Teoh appears with a copy of Power Rangers on UHD.

[Auditor55]

Originally Posted by teddyballgame

The 1500 nit readings were bursts that could not be sustained. No idea if that transfers to real content either.

Samsung has lost credibility when it comes to test windows though. They have some undefeatable black box of processing working behind the scenes at all times. I'd be more interested in measurements derived from real content that a calibrator is intimately famliar with and has a good understanding for the luminosity and color values for specific parts of different scenes. You'd need a reference monitor though.

Isn't that how peak brightness is measured, not real content brightness, which is usually on a 10% percent window?

[Auditor55]

Originally Posted by BrownianMotion

I wouldn't trust "FOMO's" measurements. Was the set in filmmaker mode and properly calibrated, or did he measure 1500 nits in torch mode?

Anyways, I think most people who reject the S95B will do so on the basis of its accuracy, or lack thereof.

Yes, but I've read comments where people where saying things like no DV no purchase. So the lack of DV on the s95B, just like the QLED displays, cause some to want pass on the S95B.

I would not like a display that's inaccurate because it's distracting to me.

[avs commenter]

Originally Posted by BrownianMotion

A higher nit display would, of course, be better for HDR in general, because it would have to rely less on tone mapping. In fact, that's why you'd see less of a difference between regular HDR10 and DV on a high nit display. In the most extreme case (a 10,000 nit display) you wouldn't need tone mapping at all, so the dynamic metadata of DV would be irrelevant.

DV does have benefits other than dynamic metadata that are unrelated to tone mapping, though.

But wouldn’t some scenes be brighter than intended without DV?

[MisterXDTV]

Samsung did this specifically to make QD-OLED stand out more compared to WOLED...

If you use reference settings, you will soon understand that MOST content CAN'T use QD-OLED color gamut/volume as Vincent showed...

Movies are mastered to DCI P3 standard, and just like Vincent showed, the image for the most part doesn't even get past Rec.709, a TV standard since 2006

If people would get this, then the HUGE QD-OLED advantages in color quality become marginal... And that's bad for marketing...

[Fendergopher]

Originally Posted by avs commenter

But wouldn’t some scenes be brighter than intended without DV?

Unless the display is pushing highlights further than the content itself is mastered at, i.e. Samsung's approach, then no. With proper PQ EOTF tracking and all else being equal, then HDR10 and DV should play back at the same exact brightness.

[BrownianMotion]

Originally Posted by avs commenter

But wouldn’t some scenes be brighter than intended without DV?

In terms of full screen brightness, yes, so there could be instances where the HDR10 image would look more "washed out" or over exposed as a result, and you would lose some highlight detail.

But the point still stands. The brighter the display (and the better the tone mapping) the less difference there will be between the two.

Originally Posted by Fendergopher

Unless the display is pushing highlights further than the content itself is mastered at, i.e. Samsung's approach, then no. With proper PQ EOTF tracking and all else being equal, then HDR10 and DV should play back at the same exact brightness.

It's actually possible, as seen at 18:19 of this video:

Dolby Vision vs HDR10 4K Blu-ray HDR Comparison - YouTube

[teddyballgame]

Originally Posted by Auditor55

Isn't that how peak brightness is measured, not real content brightness, which is usually on a 10% percent window?

I don't know how long peak brightness is generally expected to be maintained when they do 1% readings. Vincent seems to think Samsung is cheating and that the 1500 highlight would not translate to real content. I guess that's something to find out with the full review.

Here's his comment about it from last week.