Quantcast
[ 3 / biz / cgl / ck / diy / fa / g / ic / jp / lit / sci / tg / vr / vt ] [ index / top / reports / report a bug ] [ 4plebs / archived.moe / rbt ]

Due to resource constraints, /g/ and /tg/ will no longer be archived or available. Other archivers continue to archive these boards.Become a Patron!

/g/ - Technology


View post   

[ Toggle deleted replies ]
File: 137 KB, 1228x729, maxresdefault.jpg [View same] [iqdb] [saucenao] [google] [report]
64608801 No.64608801 [Reply] [Original] [archived.moe] [rbt]

How many of your games run at native Windows speeds on Linux?

>> No.64608875

none huh

>> No.64608904
File: 37 KB, 680x444, d95.jpg [View same] [iqdb] [saucenao] [google] [report]
64608904

>>64608801
>used to play this game as a kid
>Midtown Madness 2
>reinstall for nostalgia
>Windows 8 and 10 fucking butcher the performance of the game
>never goes above 15fps at 640×480 low
>freeze whenever my car hit another
>installed game on Wine in Linux
>flawlessly maxes out at 1080p60 with widescreen patch
>no issues or lags, even with heavy mods
>mfw

>> No.64608949

>>64608904
https://pcgamingwiki.com/wiki/Midtown_Madness_2
>use dgvoodoo (turns directdraw, Glide to dx11)
>still works better than wine

>> No.64608965

>>64608801
none

>> No.64608973

>>64608801
all of them run better on linux, somehow
>xonotic
>urban terror
>quakeworld
>quake 3 cpma
>counter-strike

>> No.64609007

>>64608904
>>64608973
are we all pretending ultra old games are relevant?

>> No.64609009

>>64609007
OP said "your games", these are the only games I play

>> No.64609019
File: 67 KB, 1440x1440, pn1dz02.jpg [View same] [iqdb] [saucenao] [google] [report]
64609019

>>64609007
Just like OP pretends that this thread belongs to /g/ and not /v/

>> No.64609050

>>64609019
Nobody on /v/ uses linux and that's a testament to its usefullness.
>>64608973
>run better
Bah, anyone can already get FPS higher than their refresh rates on those games.

>> No.64609078

>>64609050
>Bah, anyone can already get FPS higher than their refresh rates on those games.
Entirely depends on your hardware, and there are noticable differences from getting your refresh rate and twice your refresh rate, especially in some games. The time between frames, that the last frame updates from when your screen refreshes. If it is newer, you will feel less latency, so 300fps feels a lot better than 180 fps, or something. Regardless, most of those games capped fps.

>> No.64609100

>>64609078
Any scientific evidence behind this?
>not just vsynch-ing

>> No.64609119

>>64609100
It's just how rendering works, if you aren't playing high speed games you might not notice, but vsync is shit since you get the maximum latency on your frames. Try cap your counter-strike fps at whatever your refresh rate is, then try play at 400fps and say you can't feel a difference. You can probably even see the difference with a video.

>> No.64609123
File: 106 KB, 480x529, 1506899860644.png [View same] [iqdb] [saucenao] [google] [report]
64609123

>>64609050
>Nobody on /v/ uses linux and that's a testament to its usefullness.
>/v/ kiddo logic

>> No.64609130

>>64609100
It's just the time it takes to render a frame. Why would you want "scientific evidence"? For the same reasons you want responsive monitors, you also want to render a frame faster.

>> No.64609135

>>64609119
that's just because source engine is fucking garbage and needs at least twice your refresh rate to not look like shit

>> No.64609142

>>64609135
Why do you think you need twice your refresh rate? But it's true that q3 at 125fps looks smooth with 180hz

>> No.64609156

>>64609130
Literally what's it matter how many frames the GPU has ready when the monitor grabs whatever's in the buffer every 60th of a second? Wasted power.
>>64609119
Been there, done that, no difference. I actually run with 1 max pre-rendered frame on nvidia control panel and it's no different from 400fps.

>> No.64609163

>>64609156
>Been there, done that, no difference. I actually run with 1 max pre-rendered frame on nvidia control panel and it's no different from 400fps.
I guess you have another bottleneck, maybe your monitor is bad, or your mouse, or some other softwares creating latency.

>> No.64609175

>>64609163
Tried it all the way back on my CRT, champ. Then again several years ago on a 60hz TN LCD, later OC'd to 77.
Resident Evil 5 looks dope on 77 + enhanced vsynch tho. Zero latency and no tearing.
What are your 'render ahead' aka 'max pre-rendered frame' counts?

>> No.64609178

>>64608801
>How many of your games run at native windows speeds on linux?
How should I know? How many of your games run at native linux speeds on windows?

>> No.64609185

>>64609156
>Wasted power.
For you. Notice all those 1337Ub3r_n0sc0pe_r3bU7331 FPS players who have 1080Tis and play at the lowest settings. Setting prerendered frames to 0 and minimizing frametime has great effects on gameplay, especially games where twitch reflexes come into play. For an exaggeration of this, try playing something like DX9 skyrim with an extremely intensive ENB. You can really feel how sluggish and unresponsive it is when frametime rises over 16ms, and how "light" and responsive the game feels when the frametime drops again.

>> No.64609196

>>64609175
vsync has a buffer, no? by definition it would increase latency
I don't have these settings, since I have open source driver and intel graphics

>> No.64609258

>>64609185
>over 16ms
I get a stable 16ms at 60fps in all my games, which sometimes dips to 15.x and moving my mouse feels responsive, like an extention of my arm.
Last game I tried was STALKER just several hours ago.
What does feel sloppy is CSGO on de_nuke primmarily. What a shitshow.

>> No.64609330
File: 44 KB, 378x453, 1lgr72.jpg [View same] [iqdb] [saucenao] [google] [report]
64609330

>>64608801
>Linux
>Gayming
Pick one.

>> No.64609358
File: 177 KB, 500x625, 1445940602436.jpg [View same] [iqdb] [saucenao] [google] [report]
64609358

>>64609258
It was, as I said, an exaggeration. The same applies when going from 16 to say, 8ms, albeit it is less noticable. This is another reason why it is important, actually - having spikes around a 5ms average are far less noticable and less likely to ruin your game than those around the 16ms average.

>> No.64609385

>>64609358
I don't understand what you're getting at, you only see the results of your input every 60th of a second. How is that different if your game runs at 60 or 600fps?

>> No.64609400
File: 505 KB, 1125x972, Linux AMD.png [View same] [iqdb] [saucenao] [google] [report]
64609400

>>64608801
None because AMD is a joke

>> No.64609415

So many efforts here and there to bring gayming to linux but none have werked, it's sad

>> No.64609454
File: 30 KB, 600x394, unigine_heaven-amd-windows_official-vs-linux_mesa-apples_to_apples.png [View same] [iqdb] [saucenao] [google] [report]
64609454

>>64609400
Second-gen GCN like the Bonaire was famously problematic with older versions of the kernel and Mesa.
The open source drivers are generally faster than anything AMD has ever provided even in apples to apples tests vs Windows.

>> No.64609481
File: 66 KB, 528x792, 1516771685520.jpg [View same] [iqdb] [saucenao] [google] [report]
64609481

>>64608801
>gayming
>loonix

>> No.64609496

>>64608904
chain menu heh

>> No.64609566

>>64609385
It's not about the FPS itself, lower frametime merely grants higher FPS as a side effect. You can limit FPS with certain tools (not vsync or similar) to archieve low frametime with 60 (or whatever your monitor outputs) FPS.
>you only see
See is the keyword here, there is much more than meets the eye. Lower input lag together with higher consistency due to lower spikes will improve a twitch, reflex based game greatly. There's a difference between sending a bullet 8ms or 16ms after recognizing your target. I urge you to try something like playing a multiplayer FPS at the best possible graphics and dropping it to the lowest. Maybe force lower clocks, even.

>> No.64609590

>>64609566
Sure but I still fail to see how whatever happens behind the scenes is important when you're getting feedback at a set frequency.

>> No.64609667

>All versions of Windows will now have an 'S' mode where they will only run UWP apps from the Windows Store.
>86% of Windows users who use S for a week stay in S mode.
>All new OEM systems will ship in S mode.

GNU/Linux is the future of traditional PC gaming regardless of a minor performance penalty from translating D3D->GL because those games won't run at all on future versions of Windows.

>> No.64609698

>>64609667
>minor
See >>64609400. 5x worse performance.

>> No.64609748
File: 83 KB, 600x752, tomb_raider-linux-mesa_vs_nvidia-official.png [View same] [iqdb] [saucenao] [google] [report]
64609748

>>64609698
See
>>64609454

Second-gen GCN was basically defective and didn't match the documentation. Even now I'm not sure if the devs have completely fixed it.
Benchmarks like this were pretty common for a long time. Every card is faster than the second gen card in testing.

>> No.64609769

>>64609748
Wouldn't that mean it should suffer on windows too?

>> No.64609785

>>64609769
The proprietary driver team didn't have to use the same documentation released to the open source community.
I think even the firmware files that got distributed with the two stacks are slightly different at times.

>> No.64609822

>>64609566
>There's a difference between sending a bullet 8ms or 16ms after recognizing your target.
There is no difference because doing so is physically impossible for humans.

>> No.64610022

>>64609822
Doesn't matter, any twitch gamer can feel the latency from vsync on

>> No.64610045

>>64609007
>says urban terror isn't relevant in current year
typical normie

>> No.64610445
File: 10 KB, 224x224, 2018-02-05-10-52-41-.png [View same] [iqdb] [saucenao] [google] [report]
64610445

>>64608801
This runs about 2.5 times *faster*
I didn't even know an average of 2677 fps at 4k on an NVIDIA GTX 1050 Mobile was possible in anything.
T-tt-thanks.. linux?

>> No.64610817
File: 86 KB, 1024x768, pc-52857-31515262525.jpg [View same] [iqdb] [saucenao] [google] [report]
64610817

>>64608904
Some time ago I tried to run Blair Witch Project games on wine but I had some CATASTROPHICAL issues with interface drawing performance. Basically I had to wait about 3 minutes or more to see in-game menu.
Fortunately I never had such problems with other games.

>> No.64610872

>>64608801
I got Hyperdimension Neptunia Re;Birth 1 to run at 30fps on Linux, whereas it would would be very choppy on Windows, around 15-20fps.
Granted, this was on a toaster of a computer.

>> No.64611992

>>64610872
Any tutorials on this anon? I bought all the Neptunia games before making the jump and I haven't had the chance to play them...

>> No.64612093

>>64608801
>>>/v/

>> No.64612144

>>64610445
I tried to install osu multiple times, but each time it just says "Your game client is beeing updated" and then restarts over and over again. Do you have any idea what could be causing that? It seems like some winfags are experiencing that too, but I didn't manage to find any solutions.

>> No.64612149

ITT:
>I play shitty games and it's all Linux's fault that they're bad!

>> No.64612166

All of 'em. My uncle works at Linux, so I get the fastest ports only employees get to use.

>> No.64612204

>>64612149
It's Linux's fault there's no DX support, therefore no official videogame support.

>> No.64612275
File: 5 KB, 141x141, 1507328979786.jpg [View same] [iqdb] [saucenao] [google] [report]
64612275

>>64612204
>It's Linux's fault there's no DX support

>> No.64612296

>>64612204
LMAO you can't be serious mate

>> No.64612312

>>64612275
>>64612296
Yeah it's not windows, therefore no DX + it's shit.

>> No.64612322

>>64612204
You're trying too hard

>> No.64612519

>>64609667
what will really happen is that PC gaming will completely die

>> No.64612554

>>64611992
Use a 32-bit wine prefix, or else you can't install wmp to watch the movies. Then use winetricks to install xact_jun2010 devenum quartz wmp9 windowscodecs vcrun2012 vcrun2013. Finally, install the game.

>>
Name (leave empty)
Comment (leave empty)
Name
E-mail
Subject
Comment
Password [?]Password used for file deletion.
Captcha
Action