Announcement

Collapse
No announcement yet.

A Low-Latency Kernel For Linux Gaming

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • aht0
    replied
    Windows does not do everything best, but it does enough things simply better or does offer more options. More enough so that majority of users are not willing to put up with design and functionality issues inherent to "free Linux software". Open software tends to be pile of individual small or bigger programs. Some work reasonably well, some should have devs shot because of UI/functionality design. MS software tends to be far more deeply integrated to work with each other to the degree that there is often enough no real alternatives to using it. And windows software in general often has no analogues among "Linux software"

    Be it office, gaming or daily use.

    Leave a comment:


  • ParisStaltic
    replied
    ping Paradox Uncreated - I don't understand why windows users who already have over 90% of the Desktop market need to delude themselves that Windows does everything best. Maybe it does Desktops best for Average Joe non-technical user but it should be obvious it doesn't do everything best or Linux wouldn't have more support for more platforms and be used on everything from embeddeds, thru smartphones, to Supercomputers. I am so sick of hearing Windows users buy into the MS Spin that Windows, especially Win 10, revamped the whole latency issue and can now achieve down to 1 msec latency. Baloney! That is a damned lie when one considers the system as a whole because there is no way MS is going to make a kernel that so many of it's clients can't run because their hardware is run-of-the-mill crap.

    BOTTOM LINE - Windows' kernel has not changed and won't change in this area.. All that has changed is priority scheduling and a suggestion to turn off services....period! The prime obstacle through which all software must operate to reach hardware is the kernel and MS doesn't even offer an optional low-latency kernel. BSODs caused by allowing software direct access to hardware is a thing of the past ONLY because that mistake was rectified after a scanner crashed Win 98 SE when it was plugged in on Live TV. It took a few years but direct access was discontinued and the embarrassment still haunts them. Everything goes through the kernel now and since Win 2K and that kernel is conservatively designed for 16-20 msec.

    Specific to gaming, Windows still has some advantages but latency isn't one of them.

    Leave a comment:


  • Paradox Ethereal
    replied
    When jitter is getting low, hardware latencies matter aswell ofcourse. http://wiki.linuxcnc.org/cgi-bin/wiki.pl?Latency-Test

    I think in the end, low jitter is what everybody wants from large many cpu machines, to multicpu home computers. It really makes an OS feel high-end, good and responsive, with very low latencies possible, and smooth graphics. So in that, the standard linux config really should be low-jitter

    Best Regards.
    Last edited by Paradox Ethereal; 06-23-2017, 02:51 PM.

    Leave a comment:


  • aht0
    replied
    Originally posted by RCL_ View Post
    Modern games stabilized around two numbers: 30 FPS (mostly console games) and 60 FPS (PC games). 30 FPS is the lowest acceptable FPS, and it's chosen out of necessity, because it turns out that some people prefer (with their dollars) richer graphics (possible with 33 ms per frame) over smoother experience.

    As for network games, network latency (ping to server) is rather orthogonal aspect to visual latency because of two factors:

    1) ALL network games [try hard to] predict server response in advance (by essentially running the same code - on possibly stale/incomplete data - on clients) and then correct/compensate/lerp after getting authoritative server data. Games would be unplayable if you there was a delay of 20-50 ms between you pressing forward and you actually moving forward.

    2) Game tick (i.e. when game entities are updated, when they "think" - which in network client-server based games only happens on servers) already happens at different (often lower) rate than drawing. For user experience, it's much more important to update animations (which often are local to clients), non-game-affecting physics (particles, smoke, etc), and other "visual" things in a network game, which happens client-side.


    P.S. When talking about games, it's misleading to use FPS. It's better to compare frame time in milliseconds, and for latency tests, it's better to compare standard deviation of this value instead of an "average FPS".
    Used to be 60 for PC. It's tilting towards 120Hz and 144Hz more and more. Bunch of such specialized monitors on gaming peripherals market.

    Leave a comment:


  • Paradox Ethereal
    replied
    As for benchmarking jitter/latency, more tools exist for this now it seems:

    https://lwn.net/Articles/725238/

    So the linux kernel might be getting better at this all the time! And my impression is that many linux engineers care more about this, than windows, so.. However if windows fixed their jitter aswell, it would indeed be a good OS, and with its software support, be a winner. People still complain about lost frames and untight timing etc, on windows tough. And I have never heard of anyone doing close 0.33ms audio on it. Some technical papers on windows still talk about 10ms as a good latency target, indeed (LOL).

    Best Regards.

    Leave a comment:


  • Paradox Ethereal
    replied
    Talking of resolution I later found an undocumented maximal antialiasing mode I used on 4K, and ran the old game Unreal at 30 fps, and it looked and played great. This on windows though, but it probably is possible somehow to do this on linux aswell(?)



    One frame = 25 MB Lol..

    As I wrote on this: The new mode is demonstrated here on Unreal "Skaarj Generator" in 4k + the undocumented 3x3 Supersampling + 4x SGSS -no MSAA- Nvidia mode. They complement eachother, aligning to the highest possible antialiasing. Furthermore anisotropic optimizations must be off, and filtering set to 2x, and LOD Bias -0.5. This is near the very optimal of psychovisual "high resolution", so an ultimate setting! Not documented, but possible.

    You may need a compatibility bit if you are going to try it on windows, I found it on a german site, also.






    Last edited by Paradox Ethereal; 06-23-2017, 10:04 AM.

    Leave a comment:


  • Paradox Ethereal
    replied
    And update on what I said on 72/73 hz refresh/FPS, I think 30 fps is quite sufficient on 4K, so the resolution and detaillevel you run definately has something to say on that.

    Leave a comment:


  • ParisStaltic
    replied
    Originally posted by Paradox Uncreated View Post
    Latency

    And frames above 73hz really isn`t useful anyway.
    .
    Just FTR that statement may only have meat on the bones if your refresh rate is at or under 73Hz. Back when games allowed extreme user control but still in effect today, setting FPS to match Refresh made for extreme smoothness. For example capping Quake 3 Arena at 120FPS and setting Refresh to 120 Hz makes tricks possible that are extremely fifficult or even impossible at different settings. It is my understanding that this creates a hard sync in timing and i can assure you, whatever the cause, you can feel it. This of course assumes your monitor is capable of 120Hz.

    Note - Yeah I know Q3 is a 20+ year old game and even those that still play it are mostly on Quake Live, but the effect is the same and afaik independent of the game being a natural timing issue.

    Leave a comment:


  • Paradox Ethereal
    replied
    Xorg performed quite well, definately got better performance than windows on this jitter sensitive game. So for "ultrasmnooth" graphics, one can already do that.

    I quit making this kernel though, because GNU fundamentalism seems to be a thing that haunts the linux environment, and ultimately oneself is there, flaming as possessed oneself. But my last tests of running 0.3 ms audio was really nice. Nothing else has caught up to it yet, I think. And windows keep on being jittery from what I hear. - Great fun to be able to get as deep into configuration as doing something like this, but I agree that GNU needs to be toned down, or mature, or something along the line, for it to be mainstream acceptable.. "Be Free Hacker" says RMS, but the culticness of it was more like a sect.

    Peace.

    Leave a comment:


  • ParisStaltic
    replied
    It's too bad that some so disparage Xorg because as is common problems coexist with power and while Xorg may seem an overlapping disorganized mess it is nonetheless wide open and full of config power and is still being developed and refined. Wayland may be an improvement but it must also come at a cost. So whether you feel powerful or stuck is no matter because mouse movement/response is eminently controllable in Xorg. I used to use Xset but apparently it is no longer enough on it's own but increased power comes with the new rules. If you want to own your mouse especially on an rt-lowlatency kernel check this out http://tinyurl.com/nzahdtm

    Leave a comment:

Working...
X