| Forum Index > BW General |
|
Cyro
United Kingdom20285 Posts
120hz did succeed, the main reason that they're not highly popular is cost and the vast majority of consumers not caring or not caring enough about input lag, motion quality etc. 60hz is a relatively low refresh rate that's on the bottom end of acceptable, yet not abhorrent enough to inspire mass adoption of another standard | ||
|
kka34629095
20 Posts
On March 25 2016 01:35 Cyro wrote: I don't understand what lines you're talking about there 120hz did succeed, the main reason that they're not highly popular is cost and the vast majority of consumers not caring or not caring enough about input lag, motion quality etc. 60hz is a relatively low refresh rate that's on the bottom end of acceptable, yet not abhorrent enough to inspire mass adoption of another standard so, on a framebuffer display like LCD , without eye-tracking-blur countermeasure, the blur results in "resolution overlap",multiple rows of pixels turn into single lines in effect 1920x1080x 30fps @ 60hz = slide show 1920x1080x 60fps @ 60hz = 300 lines of motion resolution 1920x1080x 60fps @ 120hz = 300 lines of motion resolution (same) 1920x1080x 120fps @ 120hz = 600 lines of motion resolution 120hz not succeeds because you can't reliably hit 120fps, and without that it's the same. And even if you have constant 120fps (no chance really ), it will look only as good as 1280*720 , because that amount of blur makes 1080 vertical resolution redundant. Or you can "empty" the framebuffer between frames to emulate the CRT ( I know blanking is the better word but I wanted to illustrate with 'emptying' ). On a display without framebuffer (high intensity in short time slice), you have all the motion resolution you'd ever want, however , without v-sync you get tearing, and with v-sync there will be hiccups if the framerate is too high - 120hz is way too high. 120hz failed , because pixel count always gets the priority , and 4K60hz will be the next interesting format to most people since they still seem to enjoy whatever bad temporal behaviour, even slide shows, and they'll buy incrementally better hardware to hit steady 60 frame. | ||
|
Cyro
United Kingdom20285 Posts
blurry 1080p is blurry 1080p, it's still 1080p. The blur doesn't effectively lower the resolution, it just makes the image look relatively worse. I think this part it will look only as good as 1280*720 , because that amount of blur makes 1080 vertical resolution redundant is outright false or at least poorly understood--- because pixel count always gets the priority That's extremely subjective. A LOT of people use >60hz monitors. It's a minority, but so what? Almost everyone uses crap. 1080p monitors are about 400x more popular than 4k monitors on the Steam hardware survey, another minority for you. VR headsets are also going performance absolutely over pixel counts with 90-120hz displays across the board and promises of further improvement past gen 1. | ||
|
corn322
United States16 Posts
Your result is messed up in some ways Indeed they are. When unable to control shutter speed of your camera, well, it sucks. Do note that I was testing for input lag and not response time in these images. I used the stopwatch method as described by tftcentral. Further down that page the explain their new methods of testing which are more accurate than photographing webpages. And, on further investigation of my monitor, tftcentral found the Asus to have an 8.3ms input lag time. Now response time, on the other hand, was found to be between 4.6 and 12.6 ms, depending on which greys you were transitioning between and what monitor setting you have chosen. -- I don't understand what lines you're talking about there 1920x1080x 60fps @ 60hz = 300 lines of motion resolution 1920x1080x 60fps @ 120hz = 300 lines of motion resolution (same) 1920x1080x 120fps @ 120hz = 600 lines of motion resolution Yeah I'm lost too. Maybe do you mean that an image moving across a 60Hz screen at 60FPS will move 8 pixles/frame and a 120Hz screen at 120 FPS will move 4 pixles/frame? because pixel count always gets the priority I play SC2 on the CRT at 1024x768 @ 85Hz, just so I could experience (if I could detect it) the increased frame rate and lower frame time compared to the Asus at 1920x1200 @ 60HZ 4K60hz will be the next interesting format I read a rumor that higher color gamut and contrast is the "next big thing." | ||
|
iloveav
Poland1478 Posts
2. LCD wiill always be slower than CRT. This is simply because of the technology: CRT creates images at the speed of light (cathode tube), LCD crates images by partially covering light sources (has moving parts, not as fast as light). 3. Response time can be measured in many ways: Black to white, White to black, Black to White to Black and finally Grey to grey. Grey to grey tells you nothing. The reason is because you dont know what shade of grey (no pun intended) is used, and it is important to know. Black to white is not that useful because you dont know white decay (how fast white vanishes after you switch to black again), meaning you can have ghosting or similar (It also does not give you much since different monitors will have different brightness settings, and you dont know how it was measured). White to black tells you how good the brightness control and decay is, and it is a decent way to identify the response time of a monitor. Black to white to black (or white to black to white) is the only solid form of testing response time as it has to do the full cycle. Anything 5ms and under is perfect. In other words, Id take a 5ms b-w-b before a 1 ms GTG. Finally, please dont mistake response time with input lag (from the monitor) and total input lag (from your system). Total input lag is quite higher than just input lag, usually about an additional 300-400 ms. http://www.humanbenchmark.com/tests/reactiontime Test here for knowing your response time (mine with lag free plasma tv is at 230 best, 280 worst). In terms of Hz, the more, the more fps you can get. Good in shooters, not so much in rts. In theory the more hz, the more frames, the less time (delay) between frames. But the delay is very low even in a monitor of 60hz (if i remember correctly its like 16,6666 ms of delay between each frame). Other factors just take priority over hz in monitors. | ||
|
kka34629095
20 Posts
On March 25 2016 04:39 Cyro wrote: 120hz is marginally better than 60hz even if you're running @60fps blurry 1080p is blurry 1080p, it's still 1080p. The blur doesn't effectively lower the resolution, it just makes the image look relatively worse. I think this part is outright false or at least poorly understood it does, "effectively lower the resolution" ,it's called 'smearing' , it means high frequency details get turned into some blobby mess. Assuming ideal anti-aliasing, 1080p at 120hz with framebuffer blur is only better than 720p (and framebuffer blur) in slow motion , close-up scenes. It's debatable if the usual RTS ingame camera position constitutes a good scenario for this "tilt" for spatial resolution fidelity because you can't have too much movement if viewed from that far. 60fps at 120hz is the very same thing as 60hz except for lag. Oh, if something is even more quixotic than 120hz it's the AR/VR stuff, especially with the close to eye scenes (haha) , instead of solutions I see ostentatious stuff . | ||
|
Piste
6175 Posts
| ||
|
Cyro
United Kingdom20285 Posts
Total input lag is quite higher than just input lag, usually about an additional 300-400 ms. http://www.humanbenchmark.com/tests/reactiontime Test here for knowing your response time (mine with lag free plasma tv is at 230 best, 280 worst). The input lag - as in actual time between giving an action and having it performed on screen - is about 20ms on an optimized system (>60hz LCD, CRT, vsync off etc) My reaction times are about 170ms visual, 130ms auditory but it's very important to know that reaction time is not directly comparable to input lag. You can have a 170ms reaction time and 50ms input lag - or a 210ms reaction time and 10ms input lag and the input will happen at the same time if you're reacting to something on the screen, but it will feel very different to the users. The difference between a 20ms and 30ms lag - or especially between a 20ms and 40ms lag is easily visible and feelable in a blind test. This video should show some of that: ------- In theory the more hz, the more frames, the less time (delay) between frames. But the delay is very low even in a monitor of 60hz (if i remember correctly its like 16,6666 ms of delay between each frame). With a typical optimized system right now, you can have about 15ms latency from everything outside of the display refresh. A 144hz refresh makes the lag time about 15.00 - 21.94ms on the screen A 60hz refresh makes the lag time about 15.00 - 31.67ms on the screen. The average lag is substantially higher, the peak lag is way higher and more importantly, the variance in input lag is much bigger as well. A screen refresh of 60hz can often be over half of the total peak input lag of a whole system. | ||
|
DiMiaN
Finland299 Posts
![[image loading]](http://i.imgur.com/XDPiL5m.jpg) IMO in RTS games minor response or/and input lag doesnt affect gameplay unlike FPS games... | ||
|
Cyro
United Kingdom20285 Posts
| ||
|
WinterViewbot420
345 Posts
On March 27 2016 00:54 Cyro wrote: This thread is a confusing mess of people comparing display latency to display response time to human reaction times etcetcetc I literally work with this stuff for a living and people were telling me I'm wrong. I gave up and now I just bask in the stupidity of others. | ||
|
AleXoundOS
Georgia457 Posts
In practice these flat screen high refresh rate monitors cannot perform the full refresh to the desired value of colour on fast switching scenes. Pixels are just not fast enough to switch the colour at desired refresh rate. And what we get is not the real requested color, but partially changed color on the way to the requested. | ||
|
Cyro
United Kingdom20285 Posts
The issue is not just about monitor refresh rate. Refresh rate can be even 1000Hz. But does it really matter if it refreshes after a significant delay? For input lag, yes. The monitor refresh time adds a delay, other stuff just adds yet more delay. Depending on the system, the refresh can be 50%+ of the lag or only a small fraction like 10% of it. That video does not seem to have an issue with display lag; what's highlighted is the motion blur on the LCD. That kind of blur happens even with near-instant pixel switching because of the sample-and-hold effect described here - http://www.blurbusters.com/faq/oled-motion-blur/ That effect is the difference between this: ![[image loading]](http://www.blurbusters.com/wp-content/uploads/2013/05/CROPPED_60Hz.jpg) and this: ![[image loading]](http://www.blurbusters.com/wp-content/uploads/2013/05/CROPPED_LightBoost50.jpg) on a fast moving object, even on the same monitor with the same pixel transition times. It shows up differently depending on how the camera is set up - cameras act quite differently to human eyes, and these two images were taken with a pursuit camera^ aimed to simulate human eye tracking. Either way, you can see that something is screwy with the motion performance in your video. First generation LCD's were pretty awful for pixel change times, but we've gotten to the point where they are of little relevance on the fastest monitors compared to motion-ruining effects like sample+hold. ![[image loading]](http://www.blurbusters.com/wp-content/uploads/2013/06/motion-blur-graph.png) CRT is in the ~1ms region. Half a millisecond isn't very noticable in practice, but motion blur increasing by a factor of 15 is painfully obvious. | ||
|
JieXian
Malaysia4677 Posts
On March 24 2016 03:43 Piste wrote: Well at least in the posted vid there are long part following a pattern, so you don't actually have to memorize each three thousand individual key presses. How can a musician play an instrument without notes? Piste, I watched the entire video with my mouth wide open. It was very impressive. And yes I am a musician who can play my instruments without notes. It's a different thing. For starters, in an instrument like a piano, each key sounds the same, whereas in the game there's no such thing, so the same type of memorisation is not possible. | ||
|
LaStScan
Korea (South)1289 Posts
Take a look at this and compare to Flash's FPVOD's game response time. Flash's FPVOD Edit: OH NICE. My icon changed to Marine. x) | ||
|
Cyro
United Kingdom20285 Posts
| ||
|
mca64Launcher_
Poland629 Posts
On March 28 2016 20:46 LaStScan wrote: Since this conversation goes to anywhere, I'll be keeping to use another program to make my game response time faster... Take a look at this and compare to Flash's FPVOD's game response time. https://www.youtube.com/watch?v=OlbKyloOQLc Flash's FPVOD Edit: OH NICE. My icon changed to Marine. x) go play real LAN games like Flash on that video | ||
|
yB.TeH
Germany414 Posts
On March 28 2016 20:46 LaStScan wrote: Since this conversation goes to anywhere, I'll be keeping to use another program to make my game response time faster... Take a look at this and compare to Flash's FPVOD's game response time. might as well use maphack | ||
|
Cyro
United Kingdom20285 Posts
Their comments of a quarter to half second of lag not being noticed are quite amusing, but they're talking about more average users. Today in SC2 and i assume brood war with settings that multiplayer is played on, the actual delay is more around 100 milliseconds i think. | ||
|
LaStScan
Korea (South)1289 Posts
| ||
| ||
StarCraft 2 StarCraft: Brood War Stormgate Dota 2 League of Legends Counter-Strike Heroes of the Storm Other Games |
|
Sparkling Tuna Cup
Online Event
Esports World Cup
ByuN vs Astrea
Lambo vs HeRoMaRinE
Clem vs TBD
Solar vs Zoun
SHIN vs Reynor
Maru vs TriGGeR
herO vs Lancer
Cure vs ShoWTimE
Esports World Cup
Esports World Cup
Esports World Cup
CranKy Ducklings
BSL20 Non-Korean Champi…
BSL20 Non-Korean Champi…
Bonyth vs Sziky
Dewalt vs Hawk
Hawk vs QiaoGege
Sziky vs Dewalt
Mihu vs Bonyth
Zhanhun vs QiaoGege
QiaoGege vs Fengzi
|
|
|
EPT
