VESA's AdaptiveSync Display Certification Is Not Good Enough

[@seamon9732]

Hey VESA, smart move is hiring this guy.

[@guyfawkes8873]

VESA is a marketing organisation, that’s why stuff like HDR 400 exists. They cater to the tech firms that sponsor them, not to the customer buying the products.

[@N0N0111]

@@guyfawkes8873 Yeah people don't know why VESA is there, clearly they are plotting how to get crafty certifications to go through for their partners in crime.

[@faiyazhassan9240]

Welcome back. I would love to hear your thoughts on XG2431, or the new mini-leds

[@Rosaslav]

Or OLEDs

[@perdana5786]

Excellent video. VESA should really watch this video. You should make your own Aperture Grille Display Performance Certification Standard and make display makers adopt it if you can.

[@ApertureGrille]

Heh! I hope I get invited to their next meeting :) But it's tough; Blur Busters is way more well known that I am, but manufacturers seem reluctant to adopt the Blur Busters certification. I can see it from their perspective: slightly deceptive marketing will sell a ton of monitors, whereas adhering to a stringent quality standard will probably have them actually selling fewer panels.

[@4headgaming448]

Fantastic video! It was really easy to follow along with and your arguments are very well explained. Thanks for what you do :)

[@N0N0111]

The mouse pointer help me a lot, it's little overwhelming where to look to follow the narration :)

[@ApertureGrille]

@@N0N0111 Glad that helped! I've been looking over some of my old videos and trying to improve on how I present the charts. I think the giant mouse cursor helps. :)

[@Quick_Sa_Fugim]

These certifications used to mean something when I was young and foolish but is pretty clear that are just a tool of promoting products instead of keeping us aware of what we buy. Anyway, I wouldn't buy any monitor without a deep review. I am always waiting your videos and they are coming at a too slower pace to my liking. I still hope you'll start reviewing monitors at some point.

[@musicxxa6678]

Suprisingly, anti consumer Nvidia is actually honest with this stuff although you have to pay premium.

[@whismerhillgaming]

did they really ever mean something ? I mean back when internet wasn't that widespread then certifications were something we looked at for lack of other things

[@vgamedude12]

Wish they gave you more monitors to review. Definitely get how people are jaded by the display market seeing things like this.

[@dynastes4938]

Excellent work. It is a shame, you do not post more often. Hardware Unboxed and tftcentral do good work, but you might be the most knowledgeable and thorough tester out there to be able to call out this stuff. Keep it up, man, you rock. PS: I also bought my Y27q-20 for your recommendation. Have not regretted it ;D

[@freakystyle2407]

good and informative video. thanks good to see you back.

[@astreakaito5625]

We missed you! ;_; Killer video again, very informative!

[@ApertureGrille]

Thanks! I actually look for your comments on other monitor reviews to see what you think.

[@HelvetinRoopeAnkka]

Glad to see you back

[@ApertureGrille]

Thank you! AdaptiveSync Display was just too confusing for me not to make a video about it.

[@eniff2925]

Really good video, very interesting! Thank you for shedding light on topics not many are familiar with!

[@awemowe2830]

Hope this video blows up.. More people really have to know how fucked certifications are.

[@OCPyrit]

Great video and underrated channel.

[@yar23396550]

True! Too many flaws in the CTS. Can't agree with you more!

[@KeinZantezuken]

03:55 - you definitively should make one dedicated video explaining what test one can run "at home" and how to run them properly and how to gauge the result. This probably will be most useful display video in decade.

[@jamescampbell8482]

Welcome back, awesome video as always. What do you think about the XG2431 and the latest Adus 500hz? Sounds to me like blue phase mode LCD's will be the only ones to meet true 1 ms response time in RGB step changes. That blurb about not testing over/undershoot across the refresh rate range struck me as the ultimate admission that most manufacturers don't want to spend the money required for the electronics/ really fine power regulation, nor do they want to bin LCDs more than they do. I've always wished they could take the technology from a MEMS projector (a raster scanning laser beam) and use blue lasers to light-up phosphors or Quantum Dots on glass instead of for projection. Imagine Prysm's Laser Phosphor Display, but MEMS scanning based. In theory it would let you have a modern flat panel that mimics some of the best aspects of CRTs without the high power usage, vacuum, the lead, etc. Would be really nice to have a display exist again that could be low persistence even at low framerate, and also have a display that doesn't have a fixed native resolution.

[@perdana5786]

Why aren't blue phase LCDs possible yet

[@brett20000000009]

After seeing what a joke their hdr certs are not surprising really.

[@astreakaito5625]

I wonder if they're being lax about this side of the cert because the technology simply can't keep up with tighter requirement? The only monitors able to get a strict Aperture Grille approved ASD cert would be OLEDs haha Also, vrr flickering is surprisingly common and almost no reviewer seems to notice it. It's subtle but it happens even on the best Oleds

[@Quick_Sa_Fugim]

They can make different grades of the same certification just like power supply have.

[@ApertureGrille]

I didn't fully look over my old data from the VG279QM or even the Lenovo Y27q-20, but I suspect with a little tuning, they could be made to pass, the VG279QM especially. I found a few transitions that would fail, and those were almost always overshoots that were just a bit too high, but undershoot wouldn't be a problem since the VESA spec doesn't really care about that. Both companies would probably have to re-issue new versions, though. And to my great shame, I've never actually been able to test a modern, high refreshrate VA panel. VA seems to have the most trouble with flickering at lower framerates. Lots of people are reddit were complaining about those Samsung Odyssey monitors.

[@Lem_On_Lime]

@@ApertureGrille My G7 Faker Edition restarts itself really often, and has a scanline like issue when on screen image is certain color & brightness values. (Like @ night in Raft. It looks like the image is interlaced, and half the data is dropped, which makes the dark scene even darker.)

[@MrXenon003]

Great Video as always. Really appreciate the deep dive into why the Certification is too lax

[@pacman8689]

21:07 - Thanks for bringing out this point. I think it is extremely grave. Besides, g2g results in adaptive-sync mode at different refresh rates are very useful for predicting the quality of backlight strobing, because they are equivalent to using that fixed refresh rate with the largest possible Vertical Total.

[@Vegemeister1]

It seems like overdrive for strobed backlight modes should be handled completely differently. The brightness in linear space is the integral of the product of the pixel response with the backlight pulse. Convert that to perceptual space, and optimize the overdrive to minimize the error. I expect the result will be a 4D lookup table of (start_color, end_color, row_number, strobe_percent). Maybe you can get more accurate result with smaller lookup table by representing the last two as (row_backlight_on, row_backlight_off). In any case, I think it's likely a good strobed backlight implementation could reduce overshoot/undershoot ghosting, because the backlight flash occurs in the later part of the LCD transition, when the transmittance is closer to the target value and changing more slowly.

[@eddi3x3]

I think you are talking about blur buster approved monitors :) those ARE the best monitors out there.

[@jpss8659]

will you be testing the upcoming 500hz monitor?

[@DarkLOLable]

He probably would love to! But buying monitors costs a lot of money haha

[@jpss8659]

@@DarkLOLable yea i suppose the channel is not big enough to get free stuff for reviewing, tho these smaller channels are always better and informative than the bigger ones, there is no obnoxious bloat...

[@Naveronasis]

I uh... have a Viewsonic LCD monitor from 1998 that supports 77hz.... if you want to have some fun. Viewsonic VP150 I believe it was the first consumer LCD mass produced and sold to the public for desktop use. Naturally laptops had full color ones since the early 90's but you can't run modern software on them without disassembly and connecting the monitor to something else... The most interesting part is the higher than 60hz refresh rate. Anyway, let me know.

[@fandoom9261]

You did a good job

[@jambazz]

Thank you for drilling in to this material and cooking it down to something digestible for us viewers. Could you dive into the best >200hz monitors that excel in variable refresh rate across both amd and Nvidia GPUs? I have a monitor with a g-sync processor, i love it, but it's time to get a new, and i would love to find one that won't limit me to either amd or Nvidia.. I understand how many freesync monitors are g-sync compatible, but how about vice versa? Any true g-sync monitors that support amd freesync?

[@skitch6663]

Is it early Christmas or what

[@josipreponj1928]

I think the main focus with this certification was flickering, frame drops and jutter. But still, what is the point when everything is overshooting. I wonder how would they deal with vrr + strobing since everything of the above gets amplified. 🤔

[@ApertureGrille]

I think so too, but it's weird that they would even bother with the response time mandates at all. I guess just the flicker tests alone wouldn't be enough to justify all the press releases they sent out. :) Btw, VESA says that "Beta versions of the VESA AdaptiveSync Test Tool are available from multiple GPU vendors," so I wonder if I can ask NVIDIA or AMD to send me the executables. If not, I can probably produce something similar in UE4.

[@ApertureGrille]

And hey, if you get this, I've been on a different PC for a while now, so I've missed a bunch of emails. I'll send a proper reply pretty soon!

[@josipreponj1928]

@@ApertureGrille Yeah, they could of just skipped overdrive part. I see why they did it, but they didn't do it right. Unacceptable for someone specialized in that area. It would be interesting to see some of those .exe tests. Who knows maybe they're flawed too hh. No worries, see ya on mails.

[@th4ros96]

Great video!

[@eniff2925]

What do you think about the new VESA ClearMR standard?

[@lulkLogan]

Ugh, what a mess. It's still pretty much the wild west with game-focused features in the display world. Not until more people start reading and watching Aperture Grille and Blur Busters.

[@JuanGarcia-lh1gv]

I'm convinced VESA is trying to make these kinds of certifications as easy to pass as possible to get as many companies on board. Then with each revision of the certifications, make them slightly harder to pass. It seems as if they are trying to gradually improve monitor quality over years as opposed to making a very strict set of tests that very few monitors could pass now.

[@theedstv]

more monitor videos pls?

[@N0N0111]

24:52 Monitor marketing nonsense is just so insane. Now they have 0.1ms on all the boxes when we clearly know they can hardly hit 1ms on IPS/VA And the whole HDR extravaganza hoax for PC gaming monitors is so insane! Finally we are getting OLED and QD - That will change a lot!

[@ItIsMik]

Great video : )

[@LunarLaker]

smart guy with american accent ASMR? subbed

[@kaddasixseven3581]

terrible verdict of everyone involved in VESA. shame on you VESA..

[@Nicolas-qc3jf]

You should make a video on retrotink 5x vs the OSSC.

[@musicxxa6678]

It's not surprising if we look at HDR400 nonsense. Vesa is meaningless. We need to have proper reviews to have an idea otherwise they will find a way to scam consumers.

[@ShadyFury]

Brother can you recommend me a monitor for league compitition ? my performance is much better when I moved from 240 to 280 vg279qm, do you think going to 360hz ASUS ROG Swift 360Hz PG259QN would give me advantage or no I've heard it got lots of ghosting, what do you think?

[@saiprasad8078]

Did you rename your channel?

[@HelvetinRoopeAnkka]

He did

[@ApertureGrille]

Yep! I don't really know what I'm doing on KZbin, but I'm trying! Having the channel name "a5hun" was pretty dumb of me. :)

[@Lem_On_Lime]

We need more teeth!

[@movax20h]

When they say

[@ApertureGrille]

Both calculations (over and undershoot) use the width of the transition in linear terms as the denominator, so for shorter transitions, like the one at 20:56 in the video, a relatively modest RGB overshoot does represent a pretty large portion of the full transition width. Agreed about the integral, that's what my CAD score represents, and I use it exclusively instead of response times to "grade" monitors. It really should be called PANCAD, for Perceptually Adjusted Normalized Cumulative Absolute Deviation, but it essentially takes the area away from a perfect square wave transition (the integral), applies an overshoot penalty, and normalizes the result to the transition width. I'm seriously considering dropping the normalization part, though, since errors in smaller transitions (0 to 32 for instance) don't visually mean as much as errors in large transitions (0 to 255), but that would make G2G CAD heatmaps essentially unreadable.

[@Lem_On_Lime]

@@ApertureGrille And readability is the ultimate goal, or the testing doesn't mean much.

[@darkl3ad3r]

From my experience, Freesync panels are just garbage in general. They can't single cycle refresh below even as high as 72 fps many times, choosing to rather double refresh the same image at 144hz instead. It's cheap trash that I will never buy again.

[@ghos7bear]

9 views 3 Jun 2022