u-man wrote:Dont wanna sound like a asshole, but your statement regarding input-lag and the Filthypant test is simply wrong in all terms.
?
Both articles state otherwise.
u-man wrote:What you all need to pay most attention to, is the end of the article. This test was not done good, in terms of quality and valuable info
Which is why I used terms like 'preliminary' and 'not properly documented yet', maybe I should have said 'bullshit' ?
u-man wrote:There are shaders that produced less latency, than the default with no shader at all
And some also considerably more, even though the test isn't reliable enough there's no reason to ignore that part.
u-man wrote:byuu wrote a very good and recent article about input-lag:
https://byuu.org/articles/latency/
He is just a little wrong with the numbers for audio, but otherwise it is pretty much this.
Well, again;
byuu wrote:Before a frame can be pushed from the video card to the display, the frame has to be rendered and rasterized by the video card itself. For 2D, this usually just means getting the pixels into video RAM. For 3D, it means rendering all of the 3D draw commands. But even 2D can gain a frightening increase in latency by the use of user-programmable pixel shaders to perform software-based processing effects like CRT simulation, scanline simulation, image sharpening, etc.
u-man wrote:The max. input-lag from a shader *could* be 16ms or 1 frame so to say any higher lag you would expect only from a non optimized shader.
Even if it ends up being proved that it doesn't go higher than a frame at worst, that's still a lot considering an average of 2 frames is already about the acceptable ceiling for many people, add another frame on top when using pixel shaders and we're in the 3 frames area, which easily gets noticeable and annoying.
In any case you can do the test yourself, personally I definitely feel HLSL adds enough delay on top to be annoying, Royale too last time I checked (quite a while ago though)
Don't know in regards to Geom or others since I haven't played a lot using them.
Nobody's trying to set a general rule yet, bot nobody's ruling out the possibility that pixel shaders can add delay either, quite the opposite even, when you read both articles (if this is not at all what they meant then they really need to consider re-reading their own statements and re-write), which can imho be a real problem in cases where your environment and setting already add their own.
Now if anyone has plans to make more detailed and clear research on the topic in the future, that'd be great.
... not the same settings...
...
. A shader that produce lag, can only have two reasons: it tax your GPU way too much or it is written very badly.
