Linus Tech Tips Video Production

[technix]

No, this is totally false and in the vein of what you are saying is actually backwards. You may or may not have better clarity of vision than average. That is a different measure. Having 20/15 vision just denotes that you are a bit farsighted. This makes you worse at appreciating the resolution of a cell phone or computer screen than someone with 20/20 vision.

I have never took the American-style test, and maybe I made a wrong translation. The Chinese standard visual acuity tests are designed to test angular resolution directly, and a higher angular resolution corresponds directly to a higher score on the test. My raw score is 1.5 on both eyes, and for reference the usual "good" being 0.8-1.2, and 1.5+ being considered "excellent."

In practice, jitter is caused by unexpected delay in scene generation, typically because of storage I/O.  That scene will be eventually displayed, but much delayed from the point in time it was supposed to represent.  The study I linked to shows that even at 50ms (one twentieth of a second), the smallest jitter they tested, such a jitter affects the player results.

Some game developer actually do handle that, but by eating system memory. When I load up a scene in Cities:Skylines on my workstation, I usually expect ~80GB memory usage out of my 128 since the game is programmed to use up to a certain percentage of system memory, and since I have that much memory it just dumps a whole bunch of decompressed stuff into the main memory to combat jitter. Too bad most games are built to allow running on 16GB or even 8GB system memory, and jitter became unavoidable when storage can not catch up.

This can be tested though by comparing a slightly loaded mechanical hard drive (a lot of seek and load time) versus an idle NVMe SSD (virtually no seek time and extremely fast load,) which can be carried out on the same computer even.

No it's not a limit, it's average vision. It was something I was told a long time ago and it was about the ability to distinguish the dots not neccessarily focus on the dots. I suppose the other way of expressing it is the shortest break in a line that one can clearly see. so at 4 PPmm I think we are fairly safe to say that that is a good resolution and much more is just an overuse of technology.

Then explain to me why I can tell two dots 0.08mm apart from 15cm away, with perfect focus and all? If focus is not needed the dots can be even tighter?

I used to develop iOS apps, and it was common for me to work at a pixel level on a Retina screen. Even in those conditions I can still see the pixels perfectly. Call me a trained eye if you want to, but since the cutting edge display technologies are for trained eyes anyway, there is a reason to go above and beyond. You may say that type of monitor is excessive, but for me it can be something bare minimum to work with so I can see the app UI 100% pixel perfect and still get about the same physical size.

[Simon]

Yes you can gawp at a single pixel if you like. But if you are reading text 4K is plenty to make up enough smoothness to trick your eye as you are not dwelling on individual pixels but the overall image. For moving imagery you stand no chance of pinpointing a single pixes even at full HD.

[technix]

Yes you can gawp at a single pixel if you like. But if you are reading text 4K is plenty to make up enough smoothness to trick your eye as you are not dwelling on individual pixels but the overall image. For moving imagery you stand no chance of pinpointing a single pixes even at full HD.

Both the claims here implies a regular non-professional consumer.

* For people that is in the digital publication and mobile app development business it is necessary to be able to pick out single-pixel errors even at Retina density.
* For game developers and CG artists they even need to pick out single pixel errors in high resolution fluid animations.
* As of pro gamers being able to pick out single pixel differences can immediately turn into an advantage in a tight competition.
Those above are the kind of trained eyes consumer grade display technology just won't cut it for them. While for you it can take a while before you start taking up 8K 144Hz panels, those people are already reaching for things like 12K 240Hz.

Even my own daily driver monitor, a Dell P2415Q, a 4K monitor at 24 inch, is prosumer grade gear, since I still occasionally do iOS. Oh I can still pick out pixels on that from 50cm away, and that pixel that DOA'd still irks me.

[Simon]

What are you on about? what pixel errors? so you go through every possible frame of a video game to check for "bad pixels", give it a rest.

[technix]

What are you on about? what pixel errors? so you go through every possible frame of a video game to check for "bad pixels", give it a rest.

For game developers it is mainly picking out texture and hitbox errors. Play through the game, and check if every texture is applied correctly, and if collisions make sense. Both errors can be just a few pixels off in a single playthrough, but it uncovers underlying issues that can lead to jarring frames given certain gameplay.

As of CG artists every frame is supposed to be pixel perfect to begin with.

[Nominal Animal]

Dead or always lit pixels on a display are noticeable/irksome because they cause a continuity error, something that the perception part of our brain is wired to auto-highlight.

Similarly, it is a different thing to be able to detect single-pixel continuity errors, than actually perceive things at that resolution.  Different parts of the brain involved.

[SiliconWizard]

Dead or always lit pixels on a display are noticeable/irksome because they cause a continuity error, something that the perception part of our brain is wired to auto-highlight.

Similarly, it is a different thing to be able to detect single-pixel continuity errors, than actually perceive things at that resolution.  Different parts of the brain involved.

Yes. That seems hard to grasp for some though.

The fact that we're dealing with a discrete canvas can yield a variety of artefacts that we're able to perceive even though we are unable to "see" each "pixel" individually at high enough a resolution. Our ability to discriminate each pixel individually is ony a small part of the story - even a rather minor one at high resolutions.

[thm_w]

but one mouse has a 20ms input lag

We're not talking input lag (and I chopped some discussion of why not from here - happy to post it if you're bored ). But if we were, I think you've missed that these would be separate machines and, hence, not synced to each other. As we saw from the previously posted paper, a typically useful (for testing) game has a 10ms tick, and a 60Hz refresh gives less jitter than that (I assume - anyone disagree?). That's ignoring stuff like propagation delay and round-trip times, etc.

Its effectively similar: https://en.wikipedia.org/wiki/Input_lag
"In video games, input lag is either the delay between the television or monitor receiving a signal and it being displayed on the screen (see display lag below), or the delay between pressing a button and seeing the game react."

I'm trying to make the comparison as simple as possible. A 10Hz display, on average, will take longer to display the appropriate action on screen compared to a 100Hz display, no?

[PlainName]

A 10Hz display, on average, will take longer to display the appropriate action on screen compared to a 100Hz display, no?

Of course.

Probably

I'm trying to make the comparison as simple as possible.

Don't make it so simple that important factors are left out. For instance, the action tick is assumed to be much faster than the display refresh, but that's not necessarily the case. If you can use a silly 10Hz value for display refresh, we can have a silly 100ms value for action tick to match. In that case, the 100Hz monitor doesn't give you much (kind of - action at the start of a scan might appear to occur sooner than a parallel action at the end, but that also assumes a linear display scan that lasts for the refresh period).

[thm_w]

Don't make it so simple that important factors are left out. For instance, the action tick is assumed to be much faster than the display refresh, but that's not necessarily the case. If you can use a silly 10Hz value for display refresh, we can have a silly 100ms value for action tick to match. In that case, the 100Hz monitor doesn't give you much (kind of - action at the start of a scan might appear to occur sooner than a parallel action at the end, but that also assumes a linear display scan that lasts for the refresh period).

But even with 100ms tick rate, the action will on average still show up faster on the 100Hz monitor (on average 50ms earlier right? assuming display refresh and tick rate are not synchronized to the same clock). Lets stick with vsync, gsync or freesync being on.
Whether it is not much or a lot is not the concern, the question is just, will you see something faster.

[BrianHG]

How did this thread on Linus Tech Tips Video Production youtube quality turn into a thread on monitor refresh rates and delays?

[Simon]

How did this thread on Linus Tech Tips Video Production youtube quality turn into a thread on monitor refresh rates and delays?

Because some people would not admit that HD was plenty for youtube videos and that 4K really is luxury and to justify 8K started banging on about games which as you point out is not the original topic.

[EEVblog]

How did this thread on Linus Tech Tips Video Production youtube quality turn into a thread on monitor refresh rates and delays?

You must be new to forums 

[PlainName]

will on average still show up faster on the 100Hz monitor

Sure.

And even faster on a 1000Hz monitor, so why are we not all lusting after those? There is a point where it doesn't actually matter any more. Clearly, the silly 10Hz is not there and presumably your 100Hz is. Where is the point between those  two markers where the gain isn't worth better kit?

[BrianHG]

How did this thread on Linus Tech Tips Video Production youtube quality turn into a thread on monitor refresh rates and delays?

the writing/management for production is an art/social-science (and maybe psychology)
the equiping is tech and science
the viewing is biology and psychology (I tried a bit of this part)
the business and marketing ...
how youtube processes the video is another pot of ...

plenty to go around
 

All and everything said above in the past few posts, ok.  But, when some of the commenters get around to discussing monitor pixel speed refresh rates, latencies pros and cons in a video production thread, (especially when there is a monitor thread about all this running in parallel in the 'General Computing' subforum)  just escapes the fundamentals of this thread.
But, ok, it wont kill me...

[Nominal Animal]

For what it is worth, BrianHG, I've read every message in this thread from the "what is significant in video/display" angle.  Everything seems to flow pretty well if considering each message from that viewpoint, even if there has been considerable disagreement.

[Simon]

For what it is worth, BrianHG, I've read every message in this thread from the "what is significant in video/display" angle.  Everything seems to flow pretty well if considering each message from that viewpoint, even if there has been considerable disagreement.

And that is what most of this has been about. View angle, that is why no matter haw big your screen there is a maximum resolution required to look at the whole screen.

[thm_w]

will on average still show up faster on the 100Hz monitor

Sure.

And even faster on a 1000Hz monitor, so why are we not all lusting after those? There is a point where it doesn't actually matter any more. Clearly, the silly 10Hz is not there and presumably your 100Hz is. Where is the point between those  two markers where the gain isn't worth better kit?

Because 1000Hz monitors don't commercially exist yet.
When they do at a reasonable price, people will buy them: https://www.eevblog.com/forum/chat/1khz-refereshrate-for-the-microled-screen/

[PlainName]

When they do at a reasonable price, people will buy them

Undoubtedly. People will buy anything if it had a bigger number than the old model. But are you seriously saying that a 1000Hz monitor will improve your experience over, let's say, a 500Hz monitor?

[BrianHG]

When they do at a reasonable price, people will buy them

Undoubtedly. People will buy anything if it had a bigger number than the old model. But are you seriously saying that a 1000Hz monitor will improve your experience over, let's say, a 500Hz monitor?

Maybe not in a frame rate sense, but, it the display has a limited number of shades, or if it is only 4-6 bit per color, at 1000hz, a frame to frame dithering pattern will allow a reproduction of any missing luminance values.  This is particularly useful with superbright LED displays as the darker shades are just too huge a step since the output is a linear step while our eyes work logarithmically.  This is usually needed for huge LED wall type displays where you want an above 240Hz refresh so that outdoor signs don't flutter in direct sunlight as people drive by, but, that refresh speed lowers the amount of available PWM shades.

[PlainName]

at 1000hz, a frame to frame dithering pattern

That's an interesting (not to say legitimate) take on it. Kind of sidesteps the point being argued, though

[BrianHG]

at 1000hz, a frame to frame dithering pattern

That's an interesting (not to say legitimate) take on it. Kind of sidesteps the point being argued, though

To get so many nits of luminance on such a small chip without generating too much heat means the display's pixels are probably driven by 1 bit PWM similar to their video wall counterparts.  And with no phosphor whatsoever, even 180hz flutters to the eye like the first generation DLP projectors.  When I used to work on such large video walls, the refresh rate went from 150hz to 4000hz.  At 150hz, I got 16 million colors, but with a nasty flutter in daylight.  At 4000hz, I got a steady picture, but only 512 colors.  Through some Z-dithering trickery, I manages to get a beautiful 240Hz or 360Hz with close to a billion colors allowing the screens to be run indoors at 1/4 brightness without faces being shown with ugly red saturated contours since the colors were being drawn in the first 16 to 25 shades for each RGB color.

With such a bright microdisplay, with 0 phosphor, they need at least a good Z-dithering pattern faster than the video source for when they drive the display at lower humane contrast levels.

Otherwise, that device operates in the linear domain and will junk a lot of wasted light due to resistance and aging of the display's active matrix will become apparent fast as well as with temperature.  (My educated guess if their chip is a huge set of linear operating fets at each pixel)

The really old linear drive LED walls had pixel consistency problems and weren't too bright due to the heat each 32x16 module would generate.  The newer PWM designs got just as hot, however, they were 4x brighter and pixel consistency became as good as a normal LCD display.

[Simon]

When they do at a reasonable price, people will buy them

Undoubtedly. People will buy anything if it had a bigger number than the old model. But are you seriously saying that a 1000Hz monitor will improve your experience over, let's say, a 500Hz monitor?

Maybe not in a frame rate sense, but, it the display has a limited number of shades, or if it is only 4-6 bit per color, at 1000hz, a frame to frame dithering pattern will allow a reproduction of any missing luminance values.  This is particularly useful with superbright LED displays as the darker shades are just too huge a step since the output is a linear step while our eyes work logarithmically.  This is usually needed for huge LED wall type displays where you want an above 240Hz refresh so that outdoor signs don't flutter in direct sunlight as people drive by, but, that refresh speed lowers the amount of available PWM shades.

You should be a politician.

So we already have monitors that do 8 bit per colour, as I have already explained my 4K IPS monitor is like looking at photopaper prints. So why would a monitor have less colour ranges? if it's a limitation imposed by your theoretical 1000Hz technology and requires "over sampling" to fix you have just lost the effective frame rate. So what was gained? other than being ablo to claim a higher frame rate.

What does direct sunlight or driving have to do with the refresh rate?

[KL27x]

Linus uses these things because that's what his channel is about. The latest and greatest and most expensive stuff that money can buy. His channel is sponsored by people selling this stuff and this idea.

Linus uses it. Linus is successful. If you want to have a successful Youtube channel, you should use this stuff, too.

How you gonna sell $5,000 video cards, if you get by just fine with a $100.00 one? How would you come off as genuinely excited about the latest and greatest [w/e], when you don't use the latest and greatest, yourself?
 

[Simon]

who said anything about expensive video cards?