CS:GO input lag test - multicore rendering

Everything about latency. Tips, testing methods, mouse lag, display lag, game engine lag, network lag, whole input lag chain, VSYNC OFF vs VSYNC ON, and more! Input Lag Articles on Blur Busters.
devi
Posts: 2
Joined: 29 Sep 2014, 15:44

Re: CS:GO input lag test - multicore rendering

Post by devi » 04 Oct 2014, 14:38

TheLeetFly wrote:devi, could you please repeat your test with a fps lock at 128 frames [maybe 64 AND 128 tick with this framerate] ?!
In my opinion the drawtimes of the frames result in originate in the difference between lock and tickrate.

Thank you in advance.

EDIT: Oh sorry, I overlooked your second image. Would you agree with me, that the variance in this picture Image
originates of the Interpolation of the tickrate itself? In one frame the real TICK is applied and renderd and in the second frame the interpolation of the tickrate occurs. One of these two steps is the longer frame and the other one is always the shorter one.
Hi, FYI I can't read your private message because of message board limitations for new users. But yes, I would attribute the difference of the two peaks to the time it takes to do physics simulation.

flood
Posts: 929
Joined: 21 Dec 2013, 01:25

Re: CS:GO input lag test - multicore rendering

Post by flood » 04 Oct 2014, 15:28

RealNC wrote:
flood wrote:it's unfortunate that csgo can't run at a standardized fps like quake live (250) or cs1.6 (100)
You can run it at 100tick if you want. You'll obviously need to install the dedidated server for this, and configure the server for 100tick:

http://css.gamebanana.com/tuts/5638
i mean the in-game fps isn't fixed.

and csgo's allowable tickrates are different from css.. in csgo there is no 100, only 64,128,and 102.4 (<<ya nice numbers volvo wtf???)

badugib
Posts: 7
Joined: 02 Jun 2014, 10:15

Re: CS:GO input lag test - multicore rendering

Post by badugib » 05 Dec 2014, 06:21

Does input lag matter when multicore rendering is enabled even if you use a crt monitor? Which is basically 0 input lag on monitor?

User avatar
RealNC
Site Admin
Posts: 3743
Joined: 24 Dec 2013, 18:32
Contact:

Re: CS:GO input lag test - multicore rendering

Post by RealNC » 05 Dec 2014, 14:11

badugib wrote:Does input lag matter when multicore rendering is enabled even if you use a crt monitor? Which is basically 0 input lag on monitor?
It doesn't change. Why should it?
SteamGitHubStack Overflow
The views and opinions expressed in my posts are my own and do not necessarily reflect the official policy or position of Blur Busters.

User avatar
Chief Blur Buster
Site Admin
Posts: 11648
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: CS:GO input lag test - multicore rendering

Post by Chief Blur Buster » 05 Dec 2014, 20:26

The lag of the monitor itself does not change, but the lag of the game engine MAY change.

Very minor increase in input lag, only if you are GPU-bottlenecked, for multicore rendering.
Forum member sharknice has a nice webpage with a graph that shows multicore lag increase if your framerate is capped low.

CS:GO multicore input lag
He used my high speed video input lag testing method, to determine this...

Image
sharknice wrote:Conclusion

If you are limiting the frame rate and have enough power to reach that limit only using one core you should definitely disable multicore rendering. If your CPU is bottlenecking your frame rate you may want to enable multicore rendering to get a higher frame rate, but if you are GPU bottlenecked you should leave it off.

Theory

Instead of every core being combined to render a single frame faster each core produces its own frame. Since the processing all happens at once it can't grab input that hasn't happened yet. The extra frames rendered from the extra cores are based on interpolation from previous data instead of new data.
In the chart, the lag difference was about 5ms.

Practically, for the average gamer, this difference does not matter.

However, 5ms matters a huge deal for the pro gamer attending a tournament with money prizes. I can confirm that 5ms lag _differentials_ are noticeable by highly attuned individuals in certain situations. Also, mathematically, consider a fast 180 degree turn which pans the screen horizontally at about 4000 pixels/sec -- it means you will lag by 20 pixels behind in the motion vector (5/1000ths of 4000 equals 20). You can overshoot your aiming of crosshairs more often during fast-aiming situations, if the latency is different from what you trained for. Or your enemy sees you an average of 5ms sooner than you do, and there's lots of shoot-eachother-at-sametime situations (here, the lowest lag often wins). There's also an additional factor where lag helps In the Olympics-style equivalent of gaming competition, milliseconds matter. In real life, consider the starting-gun situation; a lot of olympic racers start running less than 5ms apart from each other after they hear the starting pistol (that's for real -- 2012 London olympics, for example). In competitive gameplay, all gamers may hear somebody else's clumsiness (somebody falls off a ledge by accident) they all turn around and see the same clumsy gamer, the person who frags that clumsy gamer get the frag point first. Likewise, there are a LOT of "starting-pistol" style situations in pro competitive play in certain games like that. So now you know multiple reasons why 5ms matters! So it can over time, a 5ms advantage, can theoretically all add up very slightly, even for the pro gamers who can't feel the 5ms differentials. Another example is background gunshots and bomb explosions, and reacting fast enough before you're fragged. Of course, the audio can be lagged for different individuals -- but the audio will typically have less lag than the display itself. But as you can see, when you're trying to win professionally, it helps laypeople to understand why a 5ms advantage can help. I don't play professionally, but I understand multiple reasons why 5ms matters to a lot of them!

Also use a very high fps_max when you're doing multicore rendering. Also, this probably means if you're using GSYNC and hitting the GSYNC limit often, you probably want to turn off multicore rendering.
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on Twitter

Image
Forum Rules wrote:  1. Rule #1: Be Nice. This is published forum rule #1. Even To Newbies & People You Disagree With!
  2. Please report rule violations If you see a post that violates forum rules, then report the post.
  3. ALWAYS respect indie testers here. See how indies are bootstrapping Blur Busters research!

flood
Posts: 929
Joined: 21 Dec 2013, 01:25

Re: CS:GO input lag test - multicore rendering

Post by flood » 06 Dec 2014, 07:36

without histograms of the actual data, you can't really say that the differences in those plots are statistically significant

from a technical perspective 5ms is quite a lot though, and if there is actually something that adds that much delay, the game dev really failed there.

User avatar
RealNC
Site Admin
Posts: 3743
Joined: 24 Dec 2013, 18:32
Contact:

Re: CS:GO input lag test - multicore rendering

Post by RealNC » 06 Dec 2014, 11:19

I'd like to add that this game isn't playable without multicore rendering. It's not really a choice. You have to use it, otherwise you're getting way too little FPS. And this game NEEDS 150+ FPS for you to be any good in it.
SteamGitHubStack Overflow
The views and opinions expressed in my posts are my own and do not necessarily reflect the official policy or position of Blur Busters.

flood
Posts: 929
Joined: 21 Dec 2013, 01:25

Re: CS:GO input lag test - multicore rendering

Post by flood » 11 Dec 2014, 21:25

RealNC wrote:I'd like to add that this game isn't playable without multicore rendering. It's not really a choice. You have to use it, otherwise you're getting way too little FPS. And this game NEEDS 150+ FPS for you to be any good in it.
wtf does it seriously affect fps?

User avatar
RealNC
Site Admin
Posts: 3743
Joined: 24 Dec 2013, 18:32
Contact:

Re: CS:GO input lag test - multicore rendering

Post by RealNC » 12 Dec 2014, 01:02

flood wrote:wtf does it seriously affect fps?
It doubles it on my quad i5.
SteamGitHubStack Overflow
The views and opinions expressed in my posts are my own and do not necessarily reflect the official policy or position of Blur Busters.

User avatar
Chief Blur Buster
Site Admin
Posts: 11648
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: CS:GO input lag test - multicore rendering

Post by Chief Blur Buster » 12 Dec 2014, 14:16

RealNC wrote:I'd like to add that this game isn't playable without multicore rendering. It's not really a choice. You have to use it, otherwise you're getting way too little FPS. And this game NEEDS 150+ FPS for you to be any good in it.
On the other hand -- If you are playing on a GTX Titan, and just 4X AA (or less), you can easily get 150fps+ with single core rendering. Once you're in the stratospheric framerates, the single-GPU latency difference of 300fps and 500fps is less than 1.5ms difference (1/300 minus 1/500), smaller than the latency cost of parallelizing rendering via various forms (e.g. SLI or multicore rendering).

SLI adds slight lag too, so there's an interesting question about whether to use SLI+singlecore or singlecard+multicore (pick your poison). Or worst case, doing both.

Assuming you can reduce the latency cost of multicore rendering to darn near zero, it can then become preferable, and you might win more games with it. But there are systems out there, that wins more with singlecore rendering, too. Depending on where your bottleneck is.
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on Twitter

Image
Forum Rules wrote:  1. Rule #1: Be Nice. This is published forum rule #1. Even To Newbies & People You Disagree With!
  2. Please report rule violations If you see a post that violates forum rules, then report the post.
  3. ALWAYS respect indie testers here. See how indies are bootstrapping Blur Busters research!

Post Reply