Hah, cool montage. 
I just ordered an Asus monitor myself since my TV just took a dump.
And what a coincidence! I had the same question.
Luckily I do have a HD cable box. The thing is there is only one HDMI port on the Asus monitor which my 360 will occupy.
So my other options are finding something that converts RGB to VGA or something. Plus audio is important too. Watching TV is no good without sound!
The 360 supports VGA natively, you just need the cable.
Not at all. The goal is to measure how much input lag your TV has, not how well your camera can pick up frames on Ryuâs standing jab. Your camera does the job, which is good. Thatâs not the issue. What is an issue is you introducing human timing as a variable, making things less accurate. Whatâs a bigger issue is you trying to accurately work out input lag in a span of 9 frames, 12 frame, 20 frames, or some other time window. Itâs just not going to happen. What you call â0 lagâ in that situation could be several frames in reality, and your camera wouldnât be able to tell the difference.
BTW, the word is âinherent.â âInheritâ means something else entirely. It took me a minute to figure out what you were talking about.
The camera method records one frame for every frame the game outputs. Both work at 60 fps and you can tell that each frame the camera records is exactly 1 frame on the CRT. Thus you can count frames and find out how long it takes to do a move and if itâs longer than the established frame count, your tv lags. I think youâre failing to understand this method.
Human timing is a variable in that it can possibly make a 1 frame error if your finger contacts the button between frames, making it hard to tell. If you install an LED like NKI did, you basically remove human timing from the equation. But thatâs not necessary, you just have to find a clean button press sequence.
Youâre not making any sense here. Let me re-explain how this works.
Camera records at 60fps. Game outputs at 60fps. Each frame you see on the camera is one frame of the game.
The camera does NOT randomly skip frames. It does NOT randomly decide to speed up or slow down above or below 60fps. How do you know this? Because the scan-line on a CRT recording at 60fps never moves. If it were anything but 60fps*, youâd see it moving. Thatâs why when you video tape a CRT at 24fps, you see it moving down the screen. * at 30 fps you also wouldnât see it moving, but it would be obvious that it was at 30 fps.
So we now know that the camera records at the same exact speed the game runs. You now want to measure input lag. Well, what is input lag exactly? Itâs the time between a user inputting a signal and the results happening on-screen. We can assume it takes negligible time for the signal to travel from the controller to the system, and most likely negligible time from the system to the tv. What we want to find out is if the tv takes longer to output an image once it receives information.
When you point your camera at the tv, you need to somehow indicate exactly when you create an input. Letâs say that is LP. Now assuming it always takes the exact same amount of time for a LP to come out (a good assumption in most games), you can count how many frames it takes from creating the input to the result or any fixed occurrence in the animation, such as the start of the animation in sprite-based fighters (as opposed to 3d models, which are more complex and harder to spot the first frame of animation). This count is the exact amount of time it takes your attack to hit.
So now you know Ryuâs c.LP takes 9 frames to hit, according to the test. Wait, 9 frames? Frame data says 3 frames! SF4 has 6 frames of inherent lag, which means even a move thatâs supposedly âinstantâ takes 6 frames, including moving and jumping. But if you buffer two c.LPs together, the inherent lag will be gone and the second c.LP will come out 3 frames after the first c.LP ends, thus showing the actual frame data to be 3 frames, not 9.
But what if you try on a laggy LCD and find that Ryuâs c.LP takes 17 frames to hit? Well, you know that the move takes 3 frames and thereâs 6 frames of inherent lag. 17-3-6=8 frames of lag on your LCD. You can test multiple times with multiple attacks for which you know the frame data (or find it yourself), but the results will always come up the same, unlike the stopwatch method.
You can test with ANY move, be it a 20 frame attack or a 3 frame attack, youâll get the same results in the end that tell you how much lag your TV has. Itâs just easier and faster to count to 9 than it is to count to 26.
There are all kinds of flaws in this logic, not the least of which is that input lag doesnât always happen strictly in even frame ratios.
So what do you think is the more accurate way to measure input lag then?
KoshTheKoala already had a good explanation why the stopwatch test wasnât the ideal test. I also posted an article that gave a more scientific reason why the stop watch test wasnât accurate.
http://www.prad.de/new/monitore/specials/inputlag/inputlag.html
Iâve actually done numerous stopwatch test with my monitors, and after taking hundreds of pictures and various camera setting, I couldnât get a consistent result. I will get numbers I can barely read, get 0 input lag on most pictures then all of a sudden jump to 2/3/4 frames of lag on a few, the result was all over the place.
Ok, I see. I just have a standard box right now, but Iâm upgrading and will be getting an HD box anyway. Makes sense that it would have HDMI, so I should be all set. I was just worried when I looked at my current box and all it had was a coax.
I wasnât dead-set on the Asus, but now that I know I will be able to use it as a TV, why not? Plus, I was planning on getting this model because it has component (for recording via PVR). Do you have any recommendations over this Asus or any reason why I should look into something else?
Heh, yeah. I canât justify buying a monitor for only taking to tourneys and recording games on. But, if it doubles as a TV and can replace my old boob-tube in the bedroom, they hell yes!
Like I said, the Asus Iâm getting has components, but even if you get one without, you can just hook it up via VGA. You just have to buy the cable; thereâs an official XB360 one. Thatâs how I have my main TV hooked up to it in hopes of reducing the lag that I know must be there (42" LCD).
The only reason I would look into something else is cause that model is 16:10, not 16:9.
You will either have black bars or have the image stretched.
Well I have the 360 Elite model which has HDMI so I can easily plug it directly to the Monitor.
My Cable box was the problem since thereâs only one HDMI port on the Asus monitor which my 360 will take up. And my HD Cable box has no VGA.
However, I just bought a 3-way HDMI switcher so I think I just solved that problem!
Do you have an HD Cable box? Or does it have just component, AV hookups? You should be able to use it to watch TV since thereâs a bunch of cable and acessories out there that can convert one connector to another. Plus Cable companies already made the jump to digital.
What you can also do is get a HDMI to DVI-D converter, since HDMI and DVI-D are exactly the same except that HDMI carries audio as well. Since most monitors have one HDMI and one DVI-D, that could work if you use external speakers for your setups.
Hello everybody.
I play sfIV on PC and on my X-Box 360
-My Pc is hooked up to an acer 222W LCD monitor (which is supposed to be lagfree)
-My X-Box is hooked up to a 42PFL7662D LCD TV by coax cable.
I have read numerous times that Philips displays have a big input lag. But I could realy never tell any difference between the two screens. Either I am to much of a fightingame noob or my LCD TV has no lag. To find that out, i decided to do the Guitar Hero 5 Videolatency Check.
Acoording to the Guitar Hero 5 Test my LCD TV also has 0 ms. Now I wonder if I did the Test right. I went to options and selected latency calibration. There was an option which said âCalibrateâ. When I selected it, I had to hit one Note several times.
Afterwards it said 0ms on the top of the screen. I ran the test again and hit the note late on purpose, to see if the results are different. And it really showed me an Inputlag of 45ms. So the test seems to work.
So did I do the right test? And if yes, can it be, that an 2 years old Philips LCD display has an 0ms Inputlag?
Greets Marc
No, I was saying you could hook up your 360 to the monitor through VGA⌠leaving the HDMI for the cable box. But hey, if that switcher works, there ya go!
I have been told, and have seen it mentioned numerous times in this thread, that a GH or RB test is not totally accurate.
Ok, I thought that it is not 100% accurate. But it showas 0ms so the inputlag should be relativly low then, or not?
That test is the most reliant on human error. If you can hit notes exactly at the 1-frame timing level, then Iâd say you might be able to rely on it. If you canât, like most people probably canât, you wonât get a reliable reading.
Youâre better off completely ignoring the results of the test rather than trying to use it for support that your tv has/doesnothave lag.
Here is another test that I did using the 60fps camera method.
PC SF4 everything on high setting running 60fps
-> HP LP3065 ->DVI-D dual link -> max res 2560 x 1600 @ 60 Hz
best 2nd looking SF4 Iâve seen.
its an S-IPS panelâŚso thereâs lag.
as you can see compared to the CRT test I posted earlierâŚRyu didnât hit till the 12th frame.
So the second time the test tells you it has 45ms of lag (about 3 frames) and you ignore it?
Yep, as Kosh said the GH results are completely worthless. Initially I thought that they might be okay, but I see so many dodgy results in this thread like yours. The only game test worthwhile is the RB2 test, and only after multiple times to find an average, and preferably after comparing it to a screen that has been tested using a more reliable method.
Geez I wish I had access to a 60fps camera though. Does anybody know a still camera that does video and can do this? Preferably one where I can playback frame by frame.
Anyone have any recommendations for a 24" monitor with both a DVI and HDMI input? Component/VGA would be a plus too if available. Thanks for any help.
Ah, I see.
But isnât HDMI better quality than VGA?
I rather my 360 have the better picture.
some monitors/TV disable post processing when using VGAâŚbut yeah some not all.