lol its 1080p
Considering getting a new monitor. Any good lag free models with built in speakers/headphone jack/HDMI input and decent image quality without breaking the bank?
I actually don’t know how to test for the frame data, but if you could point me in the right direction I’d be more than happy to test.
Hehe thanks! Just wanted to make sure. :chat:
I just bought an LH236H from newegg reading good things about it and lag time, and it being the “Evo” monitor for fighting games. I’ve read other good things about other minimal lag LCD monitors, but can’t find them for sale from a place I’d care to order from online.
However, I never stumbled few numbers on it in my google searches. And, even if I had, I figured I’d test it in case I got some new revision that was slower or something. Who knows.
As I’ve gone through the trouble, I figured I’d post my results, as I appreciated those who posted similar results on other models I felt I should.
Now I’m no expert, but I do know the difference between “response time” and “input lag.” That said, I am not certain of every sigle process that affect the image from CPU to GPU core to Monitor. I tried to minimize variables, but I wouldn’t be surprised if I missed any. I AM NOT STATING THESE RESULTS ARE 100% ACCURATE OR CONCLUSIVE.
I used the Timer at http://tft.vanity.dk/inputlag.html and took pictures of both screens.
Some results showed a split number or two superimposed numbers that were sometimes–though rarerly–2 frames apart. The Flash application timer was set to use hardware acceleration. I have no idea if it applies or not, but my prerendered frame number was set to 3. Perhaps I should set this to zero or 1 for further testing? There may be something else applicable I don’t know about either. Feel free to post any ideas or constructive criticism.
When it comes reviewing the results of all these pictures, I don’t like to throw results out, because throwing them out I feel can skew the results a bit. What if all of the mixed numbers for whatever reason tended to favor one monitor over the other for some reason? Every picture did represent a moment in time by both monitors display. I’d rather throw in approximated values as approx representations. -->>You may feel otherwise: If so, there raw data is there. Process it as you see fit<<–
Later I’ll try to post the pictures. If I get around to it, Maybe results through DVI out -> DVI to HDMI adaptor -> HDMI in on VH236H. I’d also like to see if “Scenery mode” has a significant reduction, because if it does not, I’d prefer to use it as it has darker blacks, that I couldn’t get from tweaknig brightness/contrast settings in game mode. Don’t hold your breath though.
Two sets of results follow, 720p and 1080p. image exposure time = 1/250 second. By “frame” below I mean ~16ms
LEGEND/KEY: (you don’t need to read the key unless you want to interpret my data or see how I arrived at my numbers)
+= +1 = Clearly VH236H 1 frame ahead of CRT
n= +0.5 = LCD shows two numbers, both same (CRT) frame, and next frame CRT show solid number
r= +0.5 = LCD shows 1 number, CRT shows 0 and 1 frame behind LCD
0= 0 = clearly same time shown
y= 0 = LCD and CRT both show split numbers, but split numbers are the same
z= 0 = LCD shows blurred both one frame behind, and one frame ahead of CRT solid number
a= 0 = LCD shows solid number, CRT shows prev and next frame relative to LCD
g= -.25 = grey area between h and 0. Same frame number is much more dominant than prev on LCD
h= -0.5 = LCD shows two numbers, half same frame, half previous frame CRT shows solid number
j= -0.5 = LCD shows 1 number, CRT shows same number and one frame ahead of LCD
c= -0.5 = LCD shows 1 and two frames behind CRTs faster score, CRT shows fastest time split with 2 frames prior
1= -1 = Clearly VH236H 1 frame behind CRT
w= -1 = LCD shows blurred 2 frames behind AND equal with CRT (LCD 094/125, CRT 125)
x= -1 = LCD shows both 1 and 2 frames behind CRT, CRT shows both 0 and 1 frame behind CRT)
b= -1.5 = LCD solid number, 1 and 2 frames behind CRT mixed number
2= -2 = Clearly 2 frames behind CRT
3 to 9 = continues on same pattern from 0 to 3
i= ignored from averages = unclear time shown/unclear double-number
_*= an inbetween or “ish” image, number skewed to represent (usually, both numbers overlapped, one MUCH brighter)
Hitachi V700 (CRT) [DVI out to dvi->dsub adaptor->dsub in] /
Versus
Asus VH236H / [DVI out->DVI in] / GAME MODE / Overdrive=100 / not 1:1, image streched to fill screen /
1280x720 60hz (720P)- Nvidia Clone Mode Nv 9600GT (dual DVI out)
(result || aproximated lag value || running total of results || result number)
0 0 0 1
1 -1 -1 2
n +.5 -.5 3
0 0 -.5 4
0 0 -.5 5
1 -1 -1.5 6
x -1 -2.5 7
y 0 -2.5 8
0 0 -2.5 9
1 -1 -3.5 10
0 0 -3.5 11
+ +1 -2.5 12
0 0 -2.5 13
h -.5 -3 14
0 0 -3 15
w -1 -4 16
1 -1 -5 17
z 0 -5 18
b -1.5 -6.5 19
0 0 -6.5 20
0 0 -6.5 21
0 0 -6.5 22
0 0 -6.5 23
0 0 -6.5 24
h -.5 -7 25
g -.25 -7.25 26
0 0 27
0 0 28
0 0 29
0 0 30
y 0 31
0 0 32
w -1 -8.25 33
n +.5 -7.75 34
0 0 35
r +.5 -7.25 36
z 0 37
0 0 38
j -.5 -7.75 39
h -.5 -8.25 40
j -.5 -8.75 41
z 0 42
1 -1 -9.75 43
h -.5 -10.25 44
h -.5 -10.75 45
0 0 46
0 0 47
0 0 48
0 0 49
j -.5 -11.25 50
1 -1 -12.25 51
-12.25/51 instances = -0.24 = average 0.24 frames behind CRT
0.24*16.67ms = 4ms average behind CRT
ASSUMING (as in, I’m not saying this is the case) Signal to both monitors averages out to to be synchronous and 1 frame=16.67ms, then VH236H (in DVI input in game mode with max overdrive) vs CRT in 1280x720 60hz (720p) in my tests showed an average of 4ms ‘input lag’. Not too shabby, especially as this isn’t its native resolution. :rock:
Hitachi V700 (CRT) [DVI out to dvi->dsub adaptor->dsub in] MONITOR DIDN’T SUPPORT 1080p, it ran in 1280x1024 with a scrolling desktop that was 1080p total resolution, I have NO IDEA how much this “throws” the results off or if it has no effect /
Versus
Asus VH236H / [DVI out->DVI in] / GAME MODE / Overdrive=100 /
1920x1080 60hz (1080p)- Nvidia Clone Mode Nv 9600GT (dual DVI out)
0 0 1
1 -1 -1 2
0 3
0 4
1 -1 -2 5
0 6
0 7
1 -1 -3 8
h -.5 -3.5 9
0 10
0 11
j -.5 -4 12
1 -1 -5 13
1 -1 -6 14
0 15
0 16
0 17
0 18
0 19
0 20
j -.5 -6.5 21
g -.25 -6.75 22
1 -1 -7.75 23
w -1 -8.75 24
w -1 -9.75 25
n +.5 -9.25 26
j -.5 -9.75 27
0 28
h* -.75 -10.5 29
b -1.5 -12 30
0 31
h -.5 -12.5 32
1 -1 -13.5 33
1 -1 -14.5 34
y 0 35
0 36
z 0 37
n +.5 -14 38
a 0 39
0 40
n* +.75 -13.25 41
j -.5 -13.75 42
1 -1 -14.75 43
h* -.75 -15.5 44
0 45
y 0 46
h -.5 -16 47
0 48
g .25 -16.25 49
0 50
0 51
0 52
0 53
w -1 -17.25 54
0 55
0 56
0 57
0 58
g -.25 -17.75 59
1 -1 -18.75 60
1 -1 -19.75 61
r +.5 -19.25 62
0 63
w* -1.5 -20.75 64
0 65
1 -1 -21.75 66
0 67
+ +1 -20.75 68
c -.5 -21.25 69
0 70
+ +1 -20.25 71
0 72
1 -1 -21.25 73
1 -1 -22.25 74
h -.5 -22.75 75
1 -1 -23.75 76
j -.5 -24.25 77
n +.5 -23.75 78
0 79
i n/a n/a n/a
-23.75/79 instances =-0.30 average/instance = 0.30 frames behind CRT
0.30*16.67ms = 5ms average behind CRT
ASSUMING (as in, I’m not saying this is the case) Signal to both monitors averages out to to be synchronous and 1 frame=16.67ms, then VH236H (in DVI input in game mode with max overdrive) vs CRT with 1920x1080 desktop running in 1280x1024 60hz (720p) in my tests showed an average of 5ms ‘input lag’. :rock:
Now, the results aren’t of course totally 100% accurate. I might have glossed over a detail or two. I think sharpening was set to 37 or something. Don’t know if that matters. I might have missed something else. I wasn’t able to eliminate all variables, or do an extremely large result set. The native resolution should show less lag as there’s no re-sample, so something seems off. 
Nonetheless, from results I’ve seen from other monitors, I’m impressed with the results so far, unless perhaps until someone might point out a huge flaw. (?)
I gave up waiting for an extremely low lag HDTV, after reading about the inherent slow response in grey-grey without the monitor knowing what frames is coming next to know how much to overdrive each pixel… as the monitor can’t predict the future, this means a buffer of a frame or two, and this means lag and/or ghosting. As I know of no cheap yet decent TN HDTV screens, I went with this.
I hope it gives similar results through HDMI, so that it will be good for gaming on my PS3 as well as my PC. So far it makes Bionic Commando Re-armed and the MegaMan 9 demo much easier to play than with the lag I was getting in game mode on a large DLP HDTV.
Assume that I meant DVI then. They are practically the same thing. 
All of the info is on the front page. Alternatively RB2 automatic sensor test done 5+ times would be great too.
Penfold: Thanks heaps for that, it looks really good. I have to admit the key and your results are too difficult for me to follow, but your summary speaks well enough for it.
I don’t know how to explain the better results with 720p either. One thing I will say though is that when RB sensor testing my friends VH222H, when he threw it in game mode, it would actually ADD around 2 milliseconds onto the results consistently. My theory is all the game mode does is adjust the image somehow to look more ‘appealing.’ It’s worth setting the monitor to defaults and seeing if the Evo (sister) monitor does the same thing.
great job Penfold. I might barrow some of your idea when I test my monitor.
This is the first time somebody tested the “evo monitor” and capturing 1 frame faster than a CRT monitor, is that even possible?
I have read that using 2 different ports and using clone mode is not accurate due to the port might not be synchronize.
Maybe try switching the ports and see if the average will still be the same. There’s usually a frame of lag between the primary port to the secondary port and also windows will always assign a primary and a secondary monitor.
got a problem. i bought a sony kdl40v4100 and the rockband 2 test is 41ms lag! my 2 part question is will the dvdo edge eliminate this problem and how hard are they to set up? anybody else gone this route and how happy were you with the results?
The dvdo will just upscale the signal, it won’t really help with lag and for the price your better of buying another tv that has less lag.
is there anything besides the hd fury2(tried one didnt seem to help) in the 650 to 700 dollar range. ireally like everything about the set exept this very big issue! i guess i could set up an evo monitor next to my tv,but i really like playing on the big screen. also does anybody think capcom can or will fix this in ssf4
This isn’t an issue with the game, it’s an issue with the TV. All of your games will lag, it’s just the timing sensitive ones that you’ll notice it.
Here’s a sneak peak on the ULTIMATE SF4 MONITOR TEST that I will be doing.
Should be able to complete the test next week if Im not to busy with work.
Image is crop to make the 2 screen closer. Just a sneak peak don’t jump to conclusion…
well i guess that settles it pulled the trigger on a evo monitor. if i only had a crystal ball to tell me how expensive it was going to be to get back into sf lol
Holy shit. not even 1ms off?
That’s fucking nice, I’d like to see a test like this done with other monitors and HDTVs.
^^;
shinshoryuken you trouble maker
I guess you did say don’t jump to conclusions.
Basically all of the TN panels I’ve ever tested I could get a single photo like claim they are lagless. 
Thanks!
Ya, the tests seem to confirm that, they’re not synchronous with a rare result of the LCD one frame ahead. I was hoping they’d average out over time, though.
I’ll try to get around to this. Who knows.
I hope that isn’t 100% accurate, because if it is, than that means for NO LAG time, the LCD would have to average skewed results of +16ms and -16ms relative to the CRT. As it’s averaged 4ms in the test that would give it the advantage, then that would mean it would average ~36ms lag were it the secondary, and thus have an actual lag of 20ms.
I’ll try to get to the vice-versa test one of these days.
That timer cannot show a one millisecond difference. Taking it down to a rating to the millisecond can only be done by averages.
It updates 60 times per second, so though it shows to the thousandths of a second, it might as well actually show a timer that displays:
H:MM:SS: Fraction
0:00:00:01/60
0:00:00:02/60
0:00:00:03/60
...
0:00:00:59/60
0:00:01:00/60
0:00:01:01/60
etc
Matter of fact, I wish it DID! (as that is the smallest increment it can actually display on a monitor running at 60 Hz.
Multiple timers… now why didn’t I think of that? Good idea. :tup:
I think 3 would be better than two… As for FPS games, the lag in the center of the screen (where the crosshair is) is the most important factor.
I think I’ll try to do that in my next test.
I’m not trying to cause trouble:wink:
Like I said, I’ll be playing the role of myth buster…which I believe will cause alot controversy!! As far as SF4 is concern, I think my test will make people shit bricks! so best take your daily fiber intake!
I was going to use 3 timers but it wouldn’t fit the screen lol.
What monitor is that being tested?