or Connect
AVS › AVS Forum › Display Devices › Flat Panels General and OLED Technology › LCD motion blur: Eye-tracking now dominant cause of motion blur (not pixel transition/GtG)
New Posts  All Forums:Forum Nav:

LCD motion blur: Eye-tracking now dominant cause of motion blur (not pixel transition/GtG) - Page 6

post #151 of 184
Thread Starter 
If it helps, giving FireFox a little higher priority in Task Manager does seem to help a little bit.

I actually was the one who convinced the FireFox people to support 120Hz monitors -- see FireFox Bugzilla #856427: Add vsync support to Windows. You can see from my comments in the bugs that I eventually convinced them to allow animations to run at 120fps on 120Hz monitors. They are now working on other bugs that is causing the juddering/stuttering, they are already familiar with it. So FireFox will do the animation as good as Chrome, eventually.

Meanwhile, Opera 15 also runs the animations pretty well too at 120fps. IE10 works great but has a bug that limits VSYNC animations to 105Hz, I've reported the bug to Microsoft already in their bug tracking system (and acknowledged by Microsoft). But of course, they're a lot harder to reach than the Chrome/FF teams. And for example, there are situations where I can run at 120fps but still gets lots of stutters in FireFox. This pattern this stutters a lot in FireFox in full screen mode on 120Hz monitors (not necessarily on 60Hz) because it can't consume more than about ~2% of CPU before it begins to stutter.

It is a challenge to make all web browsers run perfectly at fps=Hz but this recently became possible in 2012 onwards, with the advent of GPU-accelerated web browsers that is gaining the ability to run animations synchronized to VSYNC in pure HTML5 with zero plugins. (TestUFO does not use Flash! It even works on iPad which even also synchronized to VSYNC too.)

For now I need to give people the boilerplate advice to use Chrome, because just about any version of Chrome (not running multimonitor) will usually run this test reliably. Opera 15 now uses the Webkit engine (Chrome based) and manages to sync to VSYNC now too. Rest assured, I'm coaxing FireFox and Microsoft to fix their respective issues, so I can also recommend them too, so that all four major browsers all have the capability to animate to VSYNC at 120fps @ 120Hz. I am calling out an existing new 2012 W3C standard draft that recommends animations running to refresh rate, just that not all browser makers have yet have had a widespread test tool as good as TestUFO, to test their web browsers with, and also to obtain a 120Hz monitor (Mozilla even purhcased a 120Hz monitor based on my advice in Bug #856427).

That said -- TestUFO (www.testufo.com) is the world's first precision motion test website -- and even I think it is still impressive this can be done at all, even in some web browsers, considering the nature of web browsers. So give all browser makers some slack, all of them will eventually catch up. It may take a year, but they will all catch up. They have to -- just see Unreal Engine: Epic Citadel for HTML5. Play this 3D game engine demo in FireFox. You see where web browsers are going -- ability to run better-than-console-quality 3D video games using 100% standards (HTML5, WebGL, JavaScript). In five years from now, I expect TestUFO will be reliable on the vast majority of systems. TestUFO doesn't even work on Internet Explorer 8 mind you (Not possible -- it displays an error message: HTML5 required.)
Edited by Mark Rejhon - 7/21/13 at 8:25pm

Gear mentioned in this thread:

post #152 of 184
Your web app is useful (very interesting), but why implement it using web stuff at all? IMO the purpose would be much better served by a standalone application (similar to PerPixAn).
post #153 of 184
Quote:
Originally Posted by rurifan View Post

Your web app is useful (very interesting), but why implement it using web stuff at all? IMO the purpose would be much better served by a standalone application (similar to PerPixAn).

 

The purpose would be to reach (and educate) as many people as possible as to what this is all about.

 

The number of people willing to download an app is miniscule in comparison to the number of folks willing to click on a link.

post #154 of 184
Thread Starter 
Quote:
Originally Posted by tgm1024 View Post

The number of people willing to download an app is miniscule in comparison to the number of folks willing to click on a link.
Ding-dong -- this is the winner -- you just hit the jackpot.
Quote:
Originally Posted by rurifan View Post

Your web app is useful (very interesting), but why implement it using web stuff at all? IMO the purpose would be much better served by a standalone application (similar to PerPixAn).
I can reach more than 100 times as many people with just simple link clicks.

And it works on Mac, works on iPad, works on PC, as long as you're running a supported browser with GPU acceleration. Reliable on many recent GPU-accelerated computers and some tablets (newer iPads, Windows tablets, Windows phone, even BlackBerry 10 supports VSYNC). TestUFO links are highly shareable -- Facebook, Forum, comments, email, etc.

Examples:

3 UFO's at 1440 pixels/sec comparing 15fps vs 30fps vs 60fps
... www.testufo.com/#test=framerates&count=3&pps=1440

Eiffel Tower at 480 pixels/sec
... www.testufo.com/#test=photo&photo=photo-eiffel.jpg&pps=480

Moving line, light yellow on light blue; showing ghosting effects on many LCD's:
... www.testufo.com/#test=blurtrail&foreground=FFFF80&background=8080FF&thickness=8&height=-1&ppf=8

Can you launch custom configured motion tests just by clicking a link?
TestUFO is the world's first motion test that lets you do that.
Just go to TestUFO, choose and configure your favourite test, and then share the link!
At some point, I will also support < EMBED /> so bloggers can embed motion examples in their webpages.
New TestUFO tests will be added on a gradual basis (e.g. moving test patterns)...

Yes, browsers are not yet always reliable at synchronizing to VSYNC. However, look at this: www.testufo.com/browser.html -- more and more browsers are now supporting synchronization to VSYNC, making TestUFO possible. Even a lowly iPad (2nd gen or later) or an iPhone 4+ runs this website perfectly full screen 60fps, for example! Eventually in five years, compatibility will be a non-issue as long as you're running a computer built this decade (except for netbooks). In some test groups (videogamers using high end systems), I've seen over 90%+ of people run TestUFO perfectly without a single framedrop after the first 3-seconds (loading/synchronizing). The bigger reach and platform-independence, more than compensates for the inability to run on some 3-year-old entry-level computers that don't have a GPU fast enough for full-screen browser animations at 60fps... Some computers run this test instantly & perfectly -- some computers need a bit help (Task Manager priority raise) -- and some computers can't keep up. But this problem will decline over the coming years. The current cellphone you are using probably runs TestUFO better than a 5-year-old laptop (except for Android devices, though).

I have not made much publicity about TestUFO yet, because this is just a beta.
Edited by Mark Rejhon - 7/25/13 at 5:31pm
post #155 of 184
You might be able to reach even more people if you just encode your test patterns as 60fps video files. That way anyone can just put them on USB flash drive and plug into any TV at Best Buy. Virtually every TV now has a built-in media player. Much easier than dragging a laptop around with you. That is the method I use when evaluating motion performance of TVs at the store.
post #156 of 184
Thread Starter 
Quote:
Originally Posted by Wizziwig View Post

You might be able to reach even more people if you just encode your test patterns as 60fps video files. That way anyone can just put them on USB flash drive and plug into any TV at Best Buy. Virtually every TV now has a built-in media player. Much easier than dragging a laptop around with you. That is the method I use when evaluating motion performance of TVs at the store.
That's another technique too, and a legitimate one, and possibly may be added eventually (e.g. video recordings of TestUFO animations) as "Download Video" links in each TestUFO page, possibly. A low priority addon, but it sounds like something I could add as demand warrants....

That said, video also introduces new variables:
- Colorspace issues (e.g. RGB vs YUV)
- Inability to measure 1-pixel motion blur increments, due to natural blur in video compression (as well as 4:2:2 chroma)
- Fixed framerate. Inability to play animations at 120fps
- Some video players are unable to play 1080p/60fps smoothly, necessary for accurate motion blur tests on 1080p displays
- Non-native resolutions. (if you're not using a 1080p display)

With 960 pixels/second motion, 1 pixel is approximately equal to 1 millisecond of motion blur, so such motion tests are quite easy to control with computer-based 1:1 pixel mapping.

My blog (www.blurbusters.com) is initially specialized towards elimination of motion blur during computer/gaming, so it requires computer graphics quality animations, rather than video animations. The blog 120Hz native refresh rate, and requires motion tests that are capable of synchronizing to true 120fps@120Hz and true 144fps@144Hz. Presently, few displays has the ability to play 120fps files, and YouTube does not have an easy method of playing 60fps or 120fps without frame decimation. The web browser is currently the easiest mass-market method of showing gaming-relevant framerate animations to the public. TestUFO on laptop in Best Buy is a secondary'/tertiary purpose -- the main target audience is being able to post TestUFO links in places like HardForum and Overclock.net, and expanding beyond that over time. Also, about 10% of BlurBusters users are actually 2560x1440 monitor users. Video at 1:1 pixel mapping on those displays would be hard to do.

In modern web browsers, web-based animations are now able to automatically synchronize to the refresh rate (e.g. your 60Hz laptop you're viewing this webpage on, your 75Hz desktop, your brother's 120Hz gaming computer, your HDTV, your iPad). Playing 60fps@75Hz looks very different than 60fps@60Hz, so framerate synchronization with refresh rate is extremely important; because more than 50% of BlurBuster users are 120Hz monitor users... I can't cater to that with just video files alone (but that's still a good idea as a supplement!)
post #157 of 184
My comment regarding video test files was aimed strictly at evaluating televisions. The vast majority of them only handle 60hz inout anyway and often do internal processing at 4:2:2. So you would not be losing much with a 4:2:0 encoded video file. They also are nowhere near 1-pixel of blur so we don't need that kind of accuracy yet. Due to the high amount of blur present, even a scaled 720p file might be enough to differentiate them. All the higher-end 2013 ones I tested a few months ago accepted 1080p 60hz h.264 files either in .ts or .mkv format - probably for camcorder compatibility.
post #158 of 184
Thread Starter 
Yes, you're right. 1 pixel motion blur accuracy is probably overkill for HDTV's, also because of viewing distance -- e.g. sitting from sofa, etc.
The video files thing is a good idea.

That said, Blur Busters keeps in touch with FPS game players (e.g. CallSignVega of OCN) who swear by LightBoost=10% -- for that tiny further improvement in motion blur (a mere 1ms), and TestUFO has already started to suggest more than 90% of average population (random people I pull in and show to my screen blindly, "which looks clearer?") likely can see 1ms differences in motion blur under controlled conditions. (It apparently translates to the clarity seen during fast strafing and fast 180 degree turns in FPS gaming whenever framerate=Hz) This is an interesteing future area for scientists to study, because high-definition 120fps computer game motion with zero source-based limitations (no source blur, no source softening, no nothing), apparently starts to reveal 1ms differences in motion blur by the human eye whenever framerate=Hz (120fps@120Hz) at standard 1:1 computer monitor view distances at 1080p (where the individual pixels resolve). This was uncharted territory until recently.
Edited by Mark Rejhon - 7/26/13 at 10:37am
post #159 of 184

I love the testufo site.  I spent some time with the chasing squares....I like it.

 

However, I'm finding myself wishing that all of the tests had pause button.

 

Can you make it so that clicking anywhere in the test image (or providing a button) will pause the stuff on and off so I can better see what the test image is dead still at particular points?

post #160 of 184
Thread Starter 
Quote:
Originally Posted by tgm1024 View Post

However, I'm finding myself wishing that all of the tests had pause button.
I've added this to the list of requests. Thanks for the suggestion!
post #161 of 184
Sorry if this is not the right place for it, but I was wondering if there was any way to get 120Hz working in my setup.

I have a Sony HX900 LCD, which unfortunately has backlight scanning and full array local dimming, combined with a fast UV2A LCD panel, but it lacks the Impulse mode which enables backlight scanning without interpolation - that was introduced with the HX920.
According to your MPRT test, it has a persistence of 12.4ms (p.s. it would be nice if you had the option to reverse direction of these patterns so my head isn't spinning when I finish)
It's a 3D capable display so the TV is capable of receiving and displaying a 120Hz input.

However, I have been unsuccessful in creating 120Hz resolutions from my PC, at either 720p or 1080p. Sony seems to be very strict on what it will or will not accept. (I can get 48.5Hz for HFR film playback for example, but not 48.0Hz as it then falls below the range it will sync to for a "50Hz" signal)

It supports 720p60 via Nvidia's 3DTV Play - which is really just sending a 720p120 signal to the display.
Is there any way to use this mode to display 2D content at 120Hz, similar to the "hack" that enables Lightboost with 2D gaming?
post #162 of 184
Thread Starter 
Quote:
Originally Posted by Chronoptimist View Post

I have a Sony HX900 LCD, which unfortunately has backlight scanning and full array local dimming, combined with a fast UV2A LCD panel, but it lacks the Impulse mode which enables backlight scanning without interpolation - that was introduced with the HX920.
According to your MPRT test, it has a persistence of 12.4ms (p.s. it would be nice if you had the option to reverse direction of these patterns so my head isn't spinning when I finish)
Thanks for the suggestion, it will be taken into consideration! Also, there is a massive improvement to the TestUFO MPRT test (www.testufo.com/#test=mprt) coming up in the future. Most LCD's have an MPRT of about 14-16ms at 60Hz, due to the sample-and-hold effect, which creates the eye tracking motion blur (e.g. www.testufo.com/#test=eyetracking )
Quote:
It supports 720p60 via Nvidia's 3DTV Play - which is really just sending a 720p120 signal to the display.
I'm not 100% sure what the 3DTV play format is. I know there are 3D formats that pack two frames into one frame, and let the display decode them into two separate frames. These were essentially standardization workarounds to allow 3D to work over existing 2D signal standards without needing to double the signal rate. There are probably some ways to embed 2D into this 3D signal (e.g. side-by-side) that gets converted to 120Hz internally in the display. For these, it may be possible to masquerade a 120Hz 2D signal by embedding two refreshes (one refresh and then the next refresh) into one frame, and letting the display show them sequentially. This would require a hardware device to do the appropriate encoding, although there might be a way to do it using a computer, by hacking the drivers. As of right now, Blur Busters knows of no such hacks for side-by-side, and other 3D formats -- except for forcing 120Hz 2D output via televisions supporting frame-sequential 3D 60Hz/60Hz. But frame-sequential is different from framepacked formats such as side-by-side.
post #163 of 184
The easiest method to test black-frame insertion on any 3D capable set would be to encode a side-by-side video, where one side is always black. When played back at 120hz, it should produce the desired effect. I think only 1280x720p is supported for 120hz playback on most devices due to HDMI bandwidth limitations.
post #164 of 184
Quote:
Originally Posted by Wizziwig View Post

The easiest method to test black-frame insertion on any 3D capable set would be to encode a side-by-side video, where one side is always black. When played back at 120hz, it should produce the desired effect. I think only 1280x720p is supported for 120hz playback on most devices due to HDMI bandwidth limitations.

Interesting idea.

Seems lke motion test software is an ideal candidate to render itself in 1080p60 SBS with a black frame. Unfortunately you need to own the whole screen to do this effectively, so testufo's browser-based presentation may not be able to produce suitable SBS frames.
post #165 of 184
Thread Starter 
Quote:
Originally Posted by rurifan View Post

Seems lke motion test software is an ideal candidate to render itself in 1080p60 SBS with a black frame. Unfortunately you need to own the whole screen to do this effectively, so testufo's browser-based presentation may not be able to produce suitable SBS frames.
Correct, browsers don't natively do 1080p60 SBS, so that's not an option.

However, theoretically a special driver modification could convert 120Hz into 1080p60 SBS to a set that converts it to frame-sequential at the display level (e.g. shutter glasses), which would end up looking like 2D 120Hz on such a display. So anything (games, TestUFO, anything else) would work through that conduit.
post #166 of 184
Quote:
Originally Posted by Mark Rejhon View Post

I'm not 100% sure what the 3DTV play format is. I know there are 3D formats that pack two frames into one frame, and let the display decode them into two separate frames. These were essentially standardization workarounds to allow 3D to work over existing 2D signal standards without needing to double the signal rate. There are probably some ways to embed 2D into this 3D signal (e.g. side-by-side) that gets converted to 120Hz internally in the display. For these, it may be possible to masquerade a 120Hz 2D signal by embedding two refreshes (one refresh and then the next refresh) into one frame, and letting the display show them sequentially. This would require a hardware device to do the appropriate encoding, although there might be a way to do it using a computer, by hacking the drivers. As of right now, Blur Busters knows of no such hacks for side-by-side, and other 3D formats -- except for forcing 120Hz 2D output via televisions supporting frame-sequential 3D 60Hz/60Hz. But frame-sequential is different from framepacked formats such as side-by-side.
Now that you mention it, I think they are using frame packing for this. So to do 120Hz 2D gaming, you would need to have an extra frame of latency. It would probably be worth it though. Side-by-side might also be an option for 960x1080 rather than 1280x720.

It's too bad that nothing currently supports this, because I'm really happy with my TV apart from gaming, where latency could be better (it's about 30ms) and motion handling could be better. (with ~12ms persistence)
post #167 of 184
Hey guys, I'm late to the party. But I am also interested in a lot of this discussion. I just upgraded my computer at home after being prompted by some of Mark's research specifically into the Seiki monitor.

Carry on, but just wanted to say hello.

I wrote the software that did some of these original tests, and it just drew circles instead of UFO's. I had that software linked from my Wiki page on the subject for a while, but I think Mark's continued work in this area is fantastic. He's really been helping a lot of people!
post #168 of 184
Thread Starter 
I was just given heads up by Chronoptometrist that John Carmack probably indirectly referred to my ribbon tests in his YouTube video beginning at 9:08
http://www.youtube.com/watch?v=93GwwNLEBFg#t=9m08s

His talk about motion blur issues caused by persistence (which creates eye tracking motion blur). It is very nicely explained at 12:30, using a tablet computer, begin watching 30 seconds prior:
http://www.youtube.com/watch?v=93GwwNLEBFg#t=11m55s

This is exactly the same type of subjects Blur Busters Blog talks about.
Also a great link to Valve Software's talk about motion blur:
http://blogs.valvesoftware.com/abrash/down-the-vr-rabbit-hole-fixing-judder

And also keep an eye on this new thread:
Why We Need 1000fps @ 1000hz This Century
Edited by Mark Rejhon - 8/4/13 at 12:40pm
post #169 of 184
Yep. He is a real Chronoptometristoptimist tongue.gif
post #170 of 184

haha.  I was going to mention that he's not a time traveling eye doctor...
 

post #171 of 184
Quote:
Originally Posted by tgm1024 View Post

haha.  I was going to mention that he's not a time traveling eye doctor...
No, I just always under-estimate how long it takes to get things done. It's a stupid name really, but I guess I'm stuck with it.
post #172 of 184
Thread Starter 
FYI, I've posted my homebrew ribbon test research at Electronics Hacking: Creating a Strobe Backlight. It also includes research combined in, from an European researcher's who did reverse engineering for the purpose of scientific / vision researcher usage, but it has been expanded further based on my previous Arduino Scanning Backlight experiment which I put on the shelf after LightBoost was discovered to do essentially the same thing.

This is of interest to people:
--> who want to know how LightBoost strobe backlights works
..........Surprising discoveries includes:
................Blanking interval lengthening techniques (to create more time for LCD pixels to settle before strobe)
................Strobe flash slightly overlaps beginning of next refresh (0.5ms) since pixels haven't visibly started transitioning
................Response-time acceleration algorithm that take into account of Y axis of pixel for freshness of pixel prior to strobe
--> how to modify existing LightBoost strobe backlights
--> how to create a strobe backlight experiment from scratch, even using an LCD panel's own existing edgelight.

This is of interest to geeks, programmers, engineers, display manufacturers, electronics hobbyists, etc.

For LightBoost displays, such an impressively fine-tuned response time acceleration algorithm aims to try to make the pixels as perfect/complete as possible at the exact moment of the strobe. Since the LCD scans top-to-bottom in total darkness before the strobe, the pixels at the bottom edge has had less time to settle. Thus the response time acceleration algorithm used in LightBoost LCD's actually include the Y axis of the pixel within its response time acceleration. Rounding errors sometimes occur, which yields horizontal bands during the TestUFO Flicker Test (at www.testufo.com) configured to Yellow/Blue, or the TestUFO Blur Trail test (8 pixel thick, configured to Yellow/Blue), also immediately revealing human-visible proof of the Y-axis component in the response time acceleration algorithm.

There is a way for an electronics hobbyist to add an experimential strobe backlight to an existing monitor, albiet not as high quality as LightBoost. Someone who buys a cheap 1ms or 2ms 60Hz monitor can hack its existing LED backlight to flash during the blanking interval. ToastyX CRU or PowerStrip can be used to lengthen the blanking interval from the computer side instead of within the display, to create a longer blanking interval that fits a 1ms or 2ms LCD pixel persistence. A simple electronics circuit can flash the panel's existing LED backlight (with a slight boost pulse) during the blanking interval. A phasing adjustment to adjust the timing and length of the strobe, including allowing the strobe to occur early or late, even overlapping the next refresh, since that actually apparently produces better results by giving more time for previous refresh pixels to settle, while the next refresh pixels haven't visibily started transitioning yet in the first fraction of a millisecond (e.g. 0.5ms). The TestUFO Blur Trail (configured to various colors, with an 8-pixel bar @ 8pps or 16-pixel bar @ 16pps) is perfect for calibrating (by human eye) the strobe length and strobe phase, until the minimum artifacts occur along the center band of the monitor, with slight extra ghosting artifacts along the top/bottom edge of the monitor (top due to initial transitions of next refresh, bottom due to remnant transitions from previous refresh).

From a hobbyist's perspective, you will typically not be able to modify the overdrive algorithm in an existing monitor beyond a limited extent, but at 60Hz this will be less of an issue where you have more time for LCD pixels to settle. But advanced software developers and future enterprising manufacturers may be interested to know that software-based response time acceleration is possible, as is shown by ATI Radeon's "LCD Overdrive" feature in graphics drivers. This concept is useful for prototyping a custom overdrive algorithm before comitting it to a monitor's firmware (if you're a display manufacturer, or a hacker lucky enough to be able to reprogram a monitor's firmware).
Edited by Mark Rejhon - 8/7/13 at 6:48am
post #173 of 184
forgive me if i'm overlooking something obvious here, but would it not be possible to wear a pair of stroboscopic glasses and achieve the same effect as lightboost/'motion impulse'?
post #174 of 184
I've never gamed on a pc before. I've actually never owned a pc before, only laptops.

I am thinking about building a pc. Of course I need everything, on down to a monitor, mouse and keyboard, so when I spec one out the costs seem to get up there. I'm looking to accomplish a few things, not just gaming on this pc, but it would be nice to have very low input lag for BF.

What would be the least expensive setup (meaning gpu and a monitor) that can do lightboost for a game like BF3 / BF4? I was originally assuming that I would much prefer a 27" monitor.

Are there any "Korean" lightboost monitors on ebay similar to the type of 'Korean" IPS deals out there?

EDIT- I actually went to Microcenter today and purchase a lot of parts for my build. When I was there I looked at a 24" side by side with a 27" (they were both turned off). I think the 24" will be plenty big for a beginner (me). It actually seemed quite big to me, and the 27" almost seemed too big. So I am happy that I will get to save some money there, and still feel like it is still a pretty big monitor.
Edited by mphfrom77 - 8/11/13 at 7:25pm
post #175 of 184
Thread Starter 
Quote:
Originally Posted by cochaak View Post

forgive me if i'm overlooking something obvious here, but would it not be possible to wear a pair of stroboscopic glasses and achieve the same effect as lightboost/'motion impulse'?
Yes, it's possible. However, it would not be as good quality as using an optimized strobe backlight monitor. See Electronics Hacking: Creating a Strobe Backlight to just touch the surface of understanding the detail that goes into engineering a strobe backlight for a modern LCD display. Viewing the high speed videos of a 2007 LCD and a 2012 LCD, shows the top-to-bottom sweep scan pattern, a Switzerland researcher's LightBoost reverse engineering findings, findings that LightBoost displays do several special things simultaneously (blanking interval lengthening, Y-axis compensated response time acceleration algorithms, and other very interesting behaviours). Ordinary 3D shutter glasses modified to strobe between LCD refreshes, would help to reduce motion blur, but you will obviously be hampered by the LCD transition speed limitations of the shutter glasses itself, so it won't be as good as "pure" strobes as a high-speed LED backlight strobe.
post #176 of 184
In case anyone missed it, there's now an alternative to lightboost for reducing sample-and-hold 2D blur:

http://www.eizo.com/global/products/foris/fg2421/index.html
http://gaming.eizo.com/wp-content/uploads/file/turbo240_whitepaper.pdf

At least this one uses a PVA panel of some sort with much better contrast, colors, and viewing angles than the existing TN lightboost monitors. Input lag is a bit worse.
post #177 of 184
Quote:
Originally Posted by Wizziwig View Post

In case anyone missed it, there's now an alternative to lightboost for reducing sample-and-hold 2D blur:

http://www.eizo.com/global/products/foris/fg2421/index.html
http://gaming.eizo.com/wp-content/uploads/file/turbo240_whitepaper.pdf

At least this one uses a PVA panel of some sort with much better contrast, colors, and viewing angles than the existing TN lightboost monitors. Input lag is a bit worse.

 

I just read through both of those links and while the white paper is a good and concise description of the problem, I see absolutely no difference between it and the current state of interpolation+pulse style display.

post #178 of 184
It's not just the eye though.

The heart of it lies in the content. Much of the content has that blur (Especially motion blur) to begin with.

Freeze frame during any film or TV with the camera in motion.
post #179 of 184
Quote:
Originally Posted by MrBonk View Post

It's not just the eye though.
The heart of it lies in the content. Much of the content has that blur (Especially motion blur) to begin with.
Freeze frame during any film or TV with the camera in motion.
That's part of the reason motion handling is a bigger issue for gamers than people watching films or other TV content.
PC gamers especially - even if a game does offer motion blur, most PC gamers prefer to game at high framerates with the motion blur disabled.
Motion blur in games is largely designed/tuned to try and smooth out the sub-30fps framerates of console games.
post #180 of 184
Quote:
Originally Posted by Chronoptimist View Post
 
Quote:
Originally Posted by MrBonk View Post

It's not just the eye though.
The heart of it lies in the content. Much of the content has that blur (Especially motion blur) to begin with.
Freeze frame during any film or TV with the camera in motion.
That's part of the reason motion handling is a bigger issue for gamers than people watching films or other TV content.
PC gamers especially - even if a game does offer motion blur, most PC gamers prefer to game at high framerates with the motion blur disabled.
Motion blur in games is largely designed/tuned to try and smooth out the sub-30fps framerates of console games.

 

Yeah, there is a lot of travel (in screen space) for games, especially first-person-shooters.

New Posts  All Forums:Forum Nav:
  Return Home

Gear mentioned in this thread:

AVS › AVS Forum › Display Devices › Flat Panels General and OLED Technology › LCD motion blur: Eye-tracking now dominant cause of motion blur (not pixel transition/GtG)