Yeah there was alot more results but I didn’t want to post them all, but most were .01, so less than a frame of delay of average according to poonage. Not bad for a 980 buck tv.
Yes indeed, for the input lag in term of frames assuming that we have 60images/seconds the calculation is: FrameLag = 60 * InputLag (msec) / 1000
In your case and if you are always around 10msec then it means that you get 0.6 frame lag which is very good (the average gets a bit worse if you take into account your 40msec though).
EDIT: I believe that many people should thank you for your test because it’s never been done before by anyone (at least not on AVforum or AVSforum or in the french forums I watch). If i have had this information before to buy my TV I might have bought yours 
!
I have the 40W4000 and I can’t for the life of me see a game mode option.
Yes sorry, only the W4500 has one, the W4000 does approximately as good as the W4500 without the game mode, I’ll edit my previous post. More details on hdtvtest.co.uk, I guess the game mode of the W4500 disable the 100hz and some other stuffs your W4000 hasn’t.
EDIT: Ok I have added a new edit to my previous post, note that if you can test yourself your TV with a dual screen timer like what is done here and elsewhere that would be great and confirm (or not) the information provided by hdtvtest.co.uk
Will do. Will have a look at things tomorrow and see how things fare. Will let you know soon.
Ok I have done some test on my PZ86 with a dual screen, however it’s not a “real” clone mode since my monitor LCD is only 1680*1050 compatible so my plasma has to rework the image => I don’t know how much it affects the results but I don’t own any 24" monitor LCD so…
As a sidenote after doing these tests I believe that the reading of the stopwatch is subject to discussion. Let me explain, the stopwatch is precise to the millisecond and the refresh rate is only 60Hz (limited by my monitor), when taking a picture of both screens then multiple numbers are printed on top of each others. Let say 1/2/4 can be seen for the same digit.
I have used the following logic to try to figure out what was really the value printed on scree when multiple numbers are shown:
- if for instance 1/2/3 is seeen for the same digit => then it’s always the last digit the right digit, here 3
- if for instance 0/1/9 is seen: then the last number is 1
- however if 0/5 is shown for instance then you’re screw…
Note that some numbers might be more difficult to read because the TV is not in native mode…
Applying this to my panasonic plasma PZ86 reported to be 10msec or less lag from hdtvtest.co.uk I have found the following values:
LAG TV MONITOR
UNKNoWN 52 ??? 52 ???
12 2 535 2 547
20 8 859 8 879
11 14 4 14 15
15 20 703 20 718
4 25 484 25 488
UNKNOWN 30 ??? 30 ???
20 35 984 36 0
5 42 98 42 103
22 47 822 47 843
I get an average of 13msec so 0,8 frame lag which is very small.
I’d be very much interested in doing the same test with a 1080P monitor LCD.
Note: the test was done with IFC off, IFC on increases the lag by 40/50 msec.
I’ll try to post my pics later
EDIT: You can find the pics here http://picasaweb.google.com/lanfeust9/IFCOFF# ; if you have any idea why it’s so difficult so get a good reading on my setup I’d be glad to know
EDIT2: Now that I think about it, maybe my pics are not easy to read because of the camera settings, it’s a panasonic tz3, if anyone knows how to set it up to capture 1/1000sec correctly that’d be great !
I got my response after some additional testing, here is all you need for such a test (no need to say that the list is impressive):
- Camera capable of taking a picture at 1/1000sec or faster (my Panasonic TZ3 can’t…)
- LCD monitor with a 2ms response time (basically as low as possible, since if it is too high the pics will gets blury…)
- TV with a very good response time as well so that you don’t get any blury pics
So because I have a 8ms monitor and a slow cameras the pics from my viewsonic are not precise enought.
Now I am able to get very good pics from the plasma with 1250 ISO (can’t do any better) and the plasma in “standard” mode instead of “cinema” to get more light.
I’ll get one of my friends’ camera later next week and I’ll buy a new LCD anyway so I’ll retest once I’ve got everything.
well i tested my sony kdl-w4100 today and the results were bleh! =(
tested with rockband 2 guitar sensor method
all results with game mode on and everything that could possibly be turned off set to off
connected through hdmi - 60-64 ms
connected through vga - 44-48 ms
not acceptable imo and it sucks because i really like this tv…but i guess my next stop on the path to getting lagless games is getting the model poonage (fubar im assuming) recomended =/
Oh noes…
I just got the exact same model this week. I’m out of town at the moment. That much lag seems a bit absurd.
First I’ve got to ask if you’re playing on a 360 or a PS3.
Secondly, do you have the system settings on the output to match the resolution so that the TV isn’t having to convert it again?
I’ll be testing it when I get home but that’s days away so I’m hoping we can see if you can eliminate some of that lag by using any of those steps or if anyone has ideas for reducing it further.
From what I’m understanding I’m still not sure what’s a better setup for a 360 running on that model. Without running a full barrage of tests what’s supposed to work the best? VGA input or a particular system video output using either standard AV or component cables?
P0cketz
i recommend vga cables for 360 users and if u can find a vga converter box for ps3 users it will make a world of diff. and will provide no noticable screen delay
Can the 360 achieve proper 1080p over component cables? I’ve seen some claims that “real” 1080p can only be achieved through HDMI, but there seem to be conflicting reports on that. I’ve got an old (pre-Elite) 360 so I’m trying to see what my options are here. I heard there’s some converter box you can use to hook an HDMI cable up to an old 360, but that sounds like they type of thing that could introduce lag.
Before I was thinking of just going the 720p route, but then I measured my living room and realized I could fit a 50-incher in there! So now I’m wondering if I can manage a proper 1080p setup…
^^^^Hmm, could this be my answer???
The best option for the 360 is HDMI and the output configured as 1080P, as there is an integrated scaler inside the box all 720P native games will be upscale by the 360 reducing the risk of having the TV doing it (with potential lag is poor/slow scaler inside the TV).
As for the VGA connection is concerned, I do not know if it is 1080P compatible, maybe the information you read was related to an old 360 software because at the beginning the 360 was only 1080i capable. Then after an update it became 1080P compatible.
People on AVS forum as using the VGA connection and they don’t seem to report any 1080i issue but to be confirmed.
The problem with the rockband test is that it requires the end user to detect the lag with the potential issue that the same end user might lag :wgrin: or be too quick. I’m not saying you lag FloE don’t take me wrong :).
The only scientific test needs a dual screen output to your TV set and a 0 lag monitor (CRT if possible, or a known monitor LCD with as less as possible) and then compare the timing between the screens.
Now the V4100 isn’t the high end Sony TV so … does it have a game mode ?
I tested my Sharp LC42D64U yesterday, with RB2(edit: the automatic one.)
I don’t have all the details, not that it should matter, but:
In "game mode"
HDMI
1080p on the console, RB running at 720p (ie, the TV recognizing 720p)
28ms lag.
I’ve heard countless people talk about this TV being basically lag due to the “game mode” actually working properly… but 28ms is hardly lag-free.
Does anyone have any comparison tests? If I can somehow get it lower than 28ms (via firmware, other settings, whatever) I’ll gladly do anything!
i didnt use the input test…i used the one where the tv blinks and the rockband guitar sensor picks up the lag…not the one where i hit a button at a certain time
Ah ok, you need the guitar for this automatic test right ?
I guess it’s pretty much reliable then.
Just a heads up, make sure you leave your xbox 360 on “auto” for display settings, otherwise your shit will lag even if you run hdmi because it still has to upscale the image.
That seems to run counter to everything the OP said. The output on the 360 doesn’t get any feedback from the TV it’s connected to. Doesn’t auto just send a signal that forces the TV to format it and do exactly what we’re trying to avoid?
I was playing around with my TV some more, and my results are even more messed up now.
More than half the time, I can’t even get the guitar to get a reading… and sometimes you’ll see it “miss” a blink(ie, the bar doesn’t move after a flash) which ups your averages like crazy.
On top of that, I ended up with some 30ms results… in other modes, that have all the processing stuff turned on.
I’m even more confused now, the test is pretty crappy tonight. My average results are in the 40ms range, but it’s very clearly not picking up several flashes each time. =(