cool.. staff. so it seems that once you hit 32" it is like 3 times as much lag than 27" monitor at Displaylag.com
@twistedsymphony - when this thing says 4ms.. are we saying 4ms input lag when gsync is on with a compatible gsync card? or is it 4ms input lag with or without gsync like if you connect a console.
http://www.newegg.com/Product/Product.aspx?Item=9SIA0AJ3YZ9174&cm_re=gsync-_-24-009-893-_-Product
it says 4ms "response time"... "response time" is NOT "lag", it's a completely different measurement, and it's only a very small contributing factor when considering the total lag you experience while playing.
When you're talking about lag you have
1. Input lag - is how long it takes for your button push to be registered by the game code, generally this is effected by the controller design (how long it takes to encode the button push into a data packet for wireless transmittal or a USB signal etc. as well as how long it takes the internal architecture of the PC or console to interpret that signal and actually register the button push in the game code. Really bad keyboard encoders on MAME setups (including early 1st gen JPACs) actually had bad lag here, most wired setups have very small amounts of lag even when being encoded for USB or JVS.
2. networking lag - if you're playing online you may need to wait for a signal to be sent or received before the game code can make a decision on what to do next. If you are playing online this is the largest lag contributor, typically effected by "ping time" which is how long it takes to send a receive a response from the server you're connected to. Typically it has little to do with how fast your connection is and everything to do with how far away from the server (or other players) you're located as well as how many "hops" from internet server to internet to server it takes to get to the actual game server and back.
3. video processing lag - which is how long it takes for your console or your PC to generate the next frame after it makes the decision about what to do next. If you enable v-sync or buffering to improve the graphics this can add as much as 16ms ( or 1 whole frame) of lag because you're telling the PC to wait that long before releasing the frame. Also if you have a really under powered PC this the step where the lag comes from. At the end of this step you have a signal leaving the console or PC in the form of RGB, VGA or HDMI.
----- everything before this is determined by your console or PC, everything after this is determined by your monitor -----
4. display input lag - not to be confused with the controller input lag from step 1, this is the amount of time it takes your display to decode the signal that was sent to it from your PC or console, In general it's assumed that VGA is better than HDMI since the signal is being streamed to the display pixel by pixel with a VGA signal while HDMI generally needs the whole frame of data to be sent before it can start interpreting it. Also many devices encrypt the digital signal which means that your display will also have to decrypt it before it can use it. ALL digital display technologies will have some amount of lag due to this step, even Digital CRTs. However, this is the step that the G-Sync monitors claim will be greatly reduced. This step is what REALLY separates the low lag displays from the high lag displays.
5. scaling lag - LCDs and other modern displays are "fixed pixel" displays, which means that they can only ever display images at one resolution, their "native resolution", if the signal you're sending the display is NOT the native resolution then your display will have to scale that image to match it's native resolution, this takes time and adds lag. if you run your console or PC at the native resolution of your display then you can eliminate this step (though it may result in lag being added at step 3). It's important to note that most "720P" displays typically have a "1366x768" native resolution, while the 720P spec calls for "1280x720" that means that sending a 720P signal to a "720P" display might still require some scaling. Higher end displays are very good at scaling, they make hardware scaling chips that have a sub 1ms processing time, however most lower end displays perform software based scaling which is cheaper, but more time consuming.
6. post processing lag- many modern HDTVs perform other image adjustments in order to improve the image quality, these might make movies and TV look better but it generally adds a lot of lag, if your TV has a "game mode" this essentially disables this step. Though some TVs don't completely disable all post processing in game mode, and other displays don't even offer a game mode to disable any of it. Technically scaling is part of this step but I'm listing it separate since they can be controlled separately. Computer monitors and industrial monitors (like what they use in arcades) are generally better at this than TVs because they don't bother with any of the post processing crap at all.
7. response time - once the video signal has been decoded and any scaling or processing are done. the electronics then tell the actual LCD part of the LCD to change from one color pattern to the next. Response time is how long it takes for a single pixel to change from one shade of gray to another shade of gray. Technically this isn't considered lag at all, but obviously it's still some amount of time before the change in color is perceivable so when you're talking about "real world" lag then it's definitely part of the overall picture. It's important to note that technically only a portion of the response time contributes to lag because the lag testing device will perceive the change before the display has completed the transition from one color to the next. This is also part of the reason why you can't compare the reading from two different lag testing methods even if they "work the same way", the difference in the hardware design can lead to vastly different lag reading depending on how much or how little it is effected by the display's response time.
----- everything before this has to do with your display ------
8. human reaction time - technically has nothing to do with the console or pc or display but it does take some time for you to see an image, process what happened and then push a button in response, for most humans this takes about 20ms just to put everything in perspective and close the loop on the whole lag cycle.
When we're talking about the Guitar hero testing method that's measuring the lag from step 1 up through step 7, everything but the human part of the equation. When we're talking about the Leo Bodnar lag test that's testing from step 4 through step 7, everything having to do with just the display (which is perfect for comparing displays against each other). When you see a display claiming "4ms response time" then that's ONLY looking at step 7 and nothing else.
