Slashdot Mirror
Windows vs. Linux On 3D Performance
Linux Games have posted this article about Windows VS Linux on 3D performace. They tested Quake III with Matrox G400, NVidia GeForce 256 DDR, and 3DFX Voodoo 3 3000 -- all with their latest drivers (both Linux drivers and Windows drivers). There are some interesting results, and even a few surprises. What do you think about the results?
Does anyone use statistical theory when benchmarking. I ask all the authors when I read their articles and they all don't bother using it. Instead they come up with ridiculous experiments and the wrong conclusions. i.e. Toms Hardware Guide
into the ground, before this'll make a splash.
.sig. It isn't very big.
As someone else stated, this proves why he stays with Windows. Even though it is very close, it's not there yet.
Now, this isn't a flame of the driver writers, because (as was also previously stated), the windows people have had years to fine-tune their drivers for games under Windows. But with Linux being such a superior OS (we all know it), it should CRUSH Windows. IMHO, only then will it make a difference, not even being equal will match, because then people will still stick with Windows, because that's what they know.
There has to be a reason for the switch, to the hard-core gaming crowd, that slightly higher FPS could be the reason. The next question is, what about for the rest of the world? Do we want everyone and their grandmother to use Linux?
This is my
This is my
--An Oldie, but a Goodie!
Games completely take over your system, when they run. Linux, being a true multitasking system, does not allow the game to have a completely free reign over system resources,
You can minimize the effects of this on a test by stopping all other programs. I.e. kill inetd, mail daemons and so on. True multitasking is not a problem when no other tasks run. This frees up some memory too.
From the article:
In our communication with NVIDIA, they've told us that at lower resolutions they're seeing Linux leading Windows.
Now, I'm not a linux guru, but it seems to me that Linux shouldn't be doing this, but rather the opposite, as its got less "crap" loaded/taking up memory/etc than Windows. ie) Shouldn't Linux go faster than windows when a heavier load is applied?
Or is this a direct result of the drivers (and their level of optimization)? Is it merely the drivers which are holding linux back in the performance field? Or are there other issues?
-=++=- Lord, grant me the serenity to accept the things I cannot change, the courage to change the things that I can, and the wisdom to hide the bodies of those people I had to kill because they pissed me off.
I used to have a cool sig.
3D graphics is exceptional, since direct interaction with the hardware is required for acceptable speed. Both Win32 and the above X drivers provide just such a direct-rendering capability.
(essentially the X drivers let OpenGL functions act like system calls, rather than going through the X IPC mechanism like 2D drawing commands. On Windows *all* drawing is done this way) </I>
A self-contradiction. Using OpenGL system calls is not direct messing with hardware. It bypasses X for sure, so no networked display. But it still uses a system interface instead of accessing hardware.
This means OpenGL will work with any hardware as long as an OpenGL driver is provided for it. Using hardware directly would give you a game that works with one 3D-card only.
I have NT 4.0 and have had it for at least four years. I regularly do the following things, most of them simultaneously:
- Web browse in several windows at once
- Read email (with a bloaty client no less (OE))
- ICQ (with desktop floating enabled)
- Task manager
- SSH 2.0 client (often x2 or x3)
- GUI Chess client
- GUI IRC client
- Winamp
Sometimes, in addition to these, I:
- Play Quake 2 over Internet
- Compile in C++Builder (which uses 50mb of ram itself)
I used to, in addition to this, do:
- SETI@home (before I decided it was lame)
- GetRight (before I noticed how poorly coded it was and how much it hung)
I have a Celeron 300 and 96Mb of ram. I note that the total ram usage is sometimes well over 100Mb, but tasks which are idle get their display memory swapped out, and it is only slow when i switch back to a task that I have had minimised for some time (even if it was minimised and still doing stuff).
My current uptime is 11wks 4days 36mins 9secs.
I'm sure there may have been a few different things similar to what I'm speaking in development, but none I know that have a great push behind them.
Also another issues that must be brought to attention... if companies are going to create these games, people must buy them. I don't condone paying for sometihng your not going to use, but I'd hate to see games being pirated from these companies that are starting to port games and create games from Linux.
Maybe I'm just a retard (I'm sure I'll have atleast 10 AC's reply that I am) but we desperately need development of a more game friendly API that would seperate itself from the OS in the fashion that DirectX does...
or maybe Linux needs something like MFC! (sarcasm!)
------- What exactly is real?
This makes the assumption that the X bashers know what they're talking about.
With DRI the 3d pipeline bypasses X. There is some resource usage by X for font/pixmap caches but it is negligible and wouldn't have caused the slowdowns seen here.
People just want a scapegoat and X happens to be the handiest thing to point a finger at. The real problem is that very few people understand X and even fewer people contribute to the XFree86 team.
There are people in XFree86 who are concerned with speed, and experiments done in the past have proven that XFree86 drives the cards as fast as they can possibly go. Performance problems are in most cases caused by lack of documentation about the cards acceleration features, not because X is getting in the way.
Remember, at it's heart XFree86 is an async-io synchronisation mechanism. Any windowing system needs a similar mechanism, be it locks or mutexes or message passing. Changing from one mechanism to another will not change performance, it simply moves the "bottleneck" somewhere else.
Um, on my system I tested Q3 with and without DRI, the DRI rendering is almost 20% faster than without. I suspect that if these "tests" that they did used Xfree4's DRI, they would be much faster.
Spring is here. Don't believe me, look outside!
Linux was not invented.
It is reimplementation of old ideas in , sometimes better and new , style.
Check out the extensive documentation for the utah-glx project. They explain how memory/dma works on the cards (except the tnt due to no specifications being released).
Lars -
And they do perform better on Win98 ( at least as compared to W2000)
I remember reading somewhere some stats about it...
Nope, it is Linux that sucks in that particular role. In the other words, games on Linux sucks because .. ( makes no difference for most people)
Hardcore gamers will almost always stick with Win 98/ME.
Actually, many are switching to Windows 2000 - especially when there's a 5-10% speed improvement in Quake3.
overhead that Linux or any multiuser OS with a good security model has.
Win2k's certainly better than 98 in that respect, yet it's quicker on less mature drivers.
--
qube http://www.quake3arena.co.uk
To be fair they should have tested on an Alpha as well.
They should have also tested the Linux 3D with XFree86 4.0 -- what were they thinking...
Lars -
Um, Active Desktop isn't much. It just makes the desktop an OLE container. Just about every application in windows now days has windows which are ole containers.
FYI I have my active desktop on all the time, and use web view in explorer....
And also, Microsoft never forced it on you. IE4 setup asks you, IE5 doesn't even include it. And you can always turn it off...
That was FUNNY ...
"Patience is a virtue, afforded those with nothing better to do." - I don't remember
"Patience is a virtue, afforded those with nothing better to do." - I don't remember
Well, he does use Win2000.
I don't think there is a person alive who would say they prefer win9x/nt4 active desktop to be on.
I did it once so that I could have different pics of Britney showing all the time, and be able to move them easily, but after it fucked up every 5 minutes I got sick of it.
As much as I hate to say it, TPC is a load of shit. Their top two boxen(*), running MS-SQL, are the latest greatest from Compaq that won't even be on the public market for 6 months. The competition are Alpha boxen from 1998, running Oracle 8i. It's surprising that the Compaq machinen(**) only finished about 30% behind.
(*) I think this is, grammatically speaking, crap, but am trying it out to get a feel for it.
(**) The next new innovation?
Only because they were comparing it to Win 3.11, the lamest piece of crap in a long time.
Win2k comparisons are being done against Unix and Linux and all the rest that is out there.
This flamebait will hopefully be marked down if moderation has any meaning at all, but...
Unix/Linux advocates better start realizing that Windows/2000 is pretty damn good. There are some compatability headaches, but it's been rock solid for me.
--
Sometimes it's best to just let stupid people be stupid.
Aureal management:I'm not dead *yet*!
Oh, but you will be soon.
Aureal management:I'm feeling better!
Anomalous: inconsistent with or deviating from what is usual, normal, or expected
Anomalous: deviating from what is usual, normal, or expected
Canard: a false or unfounded repor
i don't know why i'm so pulled into this, but here's my stats on how often things seem to crash. btw: at my apartment i am currently running Win2k, Win98, Suse Linux and FreeBSD 3.4
OS's
Win2k:
At work1: twice, due to some soundcard driver fucking up.
work2: all the freakin time, it's some hardware issue.
Win98: a lot, i'm running a netshow server on one. about every 2 days i get a message about my soundcard driver bailing out. system is really slow. but nothing helps until a reboot.
Linux: i think once, but i can't remeber why.
FreeBSD: i few times while i was fucking around with trying to get some CD Burner working:
Microsoft Apps:
Office apps: nothing i can remember, but i don't use them much.
Visual Studio: InterDev, VB and VC++ die all the time. about twice a week
Explorer: (this is the shell for windows people). about once a week explorer dies. this doens't lose any data, i just have to restart it so it's not a biggy.
IE: i dono, 4 times a week. but i use it about 12 hours a day.
some reasons people may think NT/Win2k is unstable. NT can get into a very bad state when an app has gone south. this doesn't happen much, but you will know when windows are drawing and responding veeeery sloooowly. if you are patient enough to kill the bad app, things will go back to normal.
Also explorer does die a lot. this is the shell (i.e. the desktop, and startbar). this is no bigy as you can bring up the task manager and restart explorer. think of it as X crashing.
do i think *nix's are more stable then NT's? hell yes. do i think they are unstable usless OS's? hell no. my work enviorment is Win2k and FreeBSD. i use Win2k as my desktop and telnet into my *nix machines to do anything. I've used KDE, Gnome, Blackbox, etc.. and IMHO they are all slow, unreliable and inmature in comparsion to the Windows GUI.
-Jon
this is my sig.
That's why all the relevant libraries are LGPL (or equivalent). Closed-source applications can use them.
All Linux users please be patient. Linux will soon surpass Windows in gaming. But Linux's advancement has priorities. Before Linux goes into gaming, it needs to work on some other aspects of the OS. I don't even use windows anymore ofr gaming. I get my fix with the quake series, Starcarft on Linux, and Unreal tournament. The graphics aren't really that bad considering how long windoze has had video drivers.
(Yes, that's a good question: how does Linux handle DMA transfers from userland to PCI busmaster devices? Temporary buffers?)
No, why in the world would you want to use a temp buffer when the PCI bus-master can see the entre 32-bit memory range?
Drivers for PCI bus masters generally tell the PCI device to transfer to/from some physical address range. The userland app gives the driver a pointer to virtual memory. The driver convert this to a physical address by looking it up in the page tables - which is dead easy.
The driver may also have to cope with how a contigous range of virtual pages are broken up into scattered physical pages. This may, in the worst possible case, break IO requests into 4k chunks. It is usually not that bad though, and modern hardware have ways to get around it anyway.
Note that none of this gives linux any disadvantage whatsoever, as windows too uses the page tables. Well, unless you are running windows 3.0 in "standard mode" that is...
I know it's not *impossible* to write open source in Linux... I'm asking if those wonderful tools that you were talking about mandated that the game become open-source. If so, it may be useful in a Machivellian sense to sacrifice that bit of code to make it easier (and more attractive) to write native or ported Linux games. Did Loki use any GPL'ed tools or develop their own? Many vendors would prefer not to reinvent the wheel.
Someone go out, profile, find the critical path in x86 Mesa code and optimize it by handwritten assembly (using MMX, 3DNow! and SSE). See if you can decrease branching factor, make better use of modern L1 and L2 caches or reduce memory loads altogether and preload data from memory before actually using it.
Granted, portability is gone after that, but it's performance that matters, eh?
Hey, don't look at me! :)
He's obviously talking about win98, dumbass.
/. posters are for linux, and it is no good. You give the rest of us tro^H^H^H windows users a bad name.
It's well and good that you are a windows fan, but I think you should apply some sense and moderation to your posts. You are just as much a FUD zealot for windows as most
I have to agree with ya, Linux does suck running games. In linux, i can run Q2 at about 25 frames, at 640x480......and in win98, i get ~120 frames at 640 (TNT2).
~~~~~~~~~~
Martee
Makes sense.
Dreamcast + keyboard + mouse + Quake III -- All demonstrated at E3, and it looked dern pretty!
4 player split screen on one machine, and 8 players online. (assuming you're hosting it on a DC, and not connecting elsewhere... in that case it's limited by the server)
---
This sig has been temporarily disconnected or is no longer in service
He is not talking about direct access to hardware in DOS sense. What he is asking for is thin, unified layer on top of hardware.
Makes sense to me.
Not really - all the tested X drivers use direct-to-hardware rendering, same as Windows. If the apps were run through the X socket, you'd see linux at more like 5% of windows performance instead of 90% =).
That said, X is really showing its age. IMHO network transparency is neat but not necessary at all for 99.9% of users. I'm sick and tired of crappy, jittery 2D on linux - perhaps the solution is pure client-only rendering. Those that need apps over the net can use VNC.
DOS 6.0 actually could run in partially protected mode 32bit code using DPMI and EMM386.EXE. The code ran in virtual 86 mode(similar to how DOSEMU works BTW), and the kernel, if you could call pre Win95/98 system files that, also ran in real mode.
DOS 7.0 and above provide system calls to ease the entrance into 32 bit protected mode, and then the graphical user interface kernel takes over. After entering this Mode DOS 7 STILL provides a couple of system services. (Take a look at some of Microsoft's books on System Device Drivers programming if you don't believe me, or Caldera's online files involving their recent case.)
In Win95/98 you get a strange mix of protected mode and virtual 86 mode drivers, which don't always like to play nice, especially dealling with older or flakey hardware that doesn't like newer drivers. (I remember having to use DOS mode drivers for my 2x cdrom, when Win95 just came out.)
It was actually possible to run 32 bit programs in Win 3.1, using Win32s, (if anyone ever remembers doing such a thing, I used to use PovRAY like that).
Win 95/98 don't provide the same level of protection that Linux provides while running in protected mode. So even though the processor is in protected mode, you are running programs with VERY high levels of hardware access, compared to their linux conterparts.) Which is probably what the original poster was talking about. An OpenGL app in linux has to go through something like:
Linux Kernel < XFree86 < OpenGL < Program
NT4 and W2K have a completely different infratructure than Win 95/98, which is the reason why NT4 had little to no DirectX support. Microsoft couldn't make drivers designed to live so close to the hardware, without severely sacrificing in other areas.) OpenGL and NT4 work very nicely together as I recall, and OpenGL cards will thrive in NT4. (Half-Life is a good example.)
So, it would be much fairer to compare W2k or NT4 to Linux than Win95/98 to Linux.
------ 24.5% slashdot pure
"I just want to see Windows run ICQ, an IRC client, an MP3 player, web browser, HTML editor, web server, telnet daemon, distributed.net client, retrieve my e-mail automatically, maybe run a game or two without gasping and choking. Maybe do some word processing or image editing once in a while."
:P), graphics utils such as Colorific, keyboard and mouse software.
By the time that I've logged into my Windows (NT) box, I have nearly 50 processes running including:
MP3.com Beam-It, Creative SB software, McAffee, PPPoE client services, RAS software, SETI@home, WinFax, Yahoo Messenger, PGP, Norton CrashGuard (!!!
When I start working, I will add multiple MS VC++'s, JBuilder 3 Enterprise, some Java processes, MS VSS, SQL Server 7, Query Analuzer, IE, Netscape, Outlook, bash, VPN connections, IIS, ISAPI extensions using DCOM, Word, Excel, pcAnywhere (hosting and connections to multiple servers), MSDN, NT Explorer, Emacs (including gnuclient), Rational Rose, Real Player, DVD player software, etc - all of the apps that I need whilst working for a DVD/internet company.
If I want to take a break, do I close them all down? No - I just go and play Quake 3 on top of all that. Do I reboot a lot? No. Does it crash a lot? No. Do I think that a lot of Linux advocates who claim NT crashes a lot know what they're talking about? No. Can Linux be unstable or in a state requires rebooting? Of course!... I played around with the Sound Blaster live stuff a while ago. I managed to hang the computer a few times. I was forced to reboot many times as sometimes I couldn't unload the module, and other times I couldn't insmod it again.
Don't tell me that Windows is unstable and that you cannot do a lot of things simultaneous without effecting the OS. It's just not true. I've been developing software under NT for some time and I can leave the machine up and running for long periods of time without rebooting.
In some respects it already does. However, some of the documentation on how to use the high speed interfaces in XFree86 and Linux are a little sparse so it is possible that some people aren't aware of what's available.
For example, DGA 2.0 in XFree86 4.0 provides similar services to DirectX 2D. You can have framebuffers mapped directly into your client. No X encoding/decoding overhead for images. In effect a direct rendering infrastructure for 2D.
Then there's XVideo which provides services for registering video sources. This will assist applications working on video playback.
DRI now provides a 3D pipeline which matches (perhaps exceeds) Direct3D for the quality of the pipeline, and does it with clean OpenGL.
Some of these services are only new, or have changed since the XFree86 3.3 days, but the base is there to work with, so go use it.
People should keep in mind that Linux wasn't all that fast when it came out either. Do people remember the awful IDE performance? Or that 1.0 sucked wind through a straw for networking? Even SCSI performance was pretty bad until 2.0.
XFree86 needs people to work on it and make it better too. There are lots of areas that can be improved, or better documented, and applications need to be modified to exploit the new faster code in XFree86 4.0.
What isn't needed is a lot of complainers, or people who think that starting again is the best way to fix things (and then repeating all the same old mistakes that XFree86 has already been through and solved).
Ok, who the heck wants to have a comparison between crappy Win'98 and Linux? Give us a Win2000 vs Linux!! That would be a really good comparison as they approach each other in function and security... Let's see if the all new, all good NT kernel is better.
Reading A New Rendering Model For X it appears, that without bashing X the feature set could be improved. However the bypassing of X for the 3d pipeline was the main feature I was interested in.
tangent - art and creation are a higher purpose
postmoderncore - art and creation are a higher purpose
I'm curious, did you buy Windows NT 2000? _every_ person I have spoken to either on irc or in person has _pirated_ Windows NT 2000 when they claim they "switched"
Not that I can blame them, that bloated motherfucker is nearly $400 a copy here.
Lars -
If you want good performance under X you need to:
1) Buy yourself a matrox G400 card or something as powerful. You can't easily do X on an S3 Trio 3d or similar chipsets. 32bit colour in high resolutions are very nice on the Matrox G400.
2) Get a PS/2 mouse and use it instead of a serial mouse. Serial mice don't seem to get enough updates under X and cause window moving and scrollbar fiddling to be slow and clunky. This is the most important part, Do Not Use A Serial Mouse In X.
3) If you want an extra blast of performace, install and use Xfree 4.0. Xfree 4.0 is much faster than Xfree 3.3.X, but isn't as compatible with strange hardware as Xfree 3.3.X
So there you have it, 3 steps to decent X performace. It won't be as good as windows, but it isn't far off.
The differences between XF3.3x+Mesa+Glide and XF4.0+DRI are:
www.eFax.com are spammers
ya know, this presents an interesting solution to the MS trial. open up directx. even if the powers that be dont want to force MS to open up the whole os, they could at least go for directx, microsoft would have nothing to lose from that(it can be argued that they have nothing to lose and everything to gain to from opening the whole os..but thats another debate). then a directx compatable api suite could be written for linux. im sure theres plenty of people that would love to work on a open source multimedia api suite. it would make it easier to port games over, close the performance gap between the two os's, and thus satisfy the government by putting a dent in MS's monopoly by breaking down the last reason people have to stick to windows. kill three birds with one stone, whoop!
Troll. The reason windows has these often used audio api's is marketing nothing but. The Gravis ultrasound from way back in the early 90's had 32hardware channels and was supported for windows and linux equally well. As far as needing " ....a good way to get huge amounts of data and commands to the sound cards" well you certainly shouldn't be looking at soundblaster cards. Again the GUS did this right in the early 90's where you would load the card with the samples you needed and then trigger the samples when you wanted to play them. An interrupt would be triggered when a channel was done playing. ( You could also have samples loop, decay, etc) and all this had full support in windows AND Linux. In short go away troll; or at least read up before you speak.
you're more of a fucking idiot if you keep windows installed on your computer. i wiped that shit off years ago. i bet a cockless bitch like you still uses it.
-- You see, there would be these conclusions that you could jump to
This I entirely agree with. The font handling in X11 is an absolute joke; a relic of ideas and limitations of the 80s which certainly shouldn't be holding us back these days.
It's a great thing Keith is working on making fonts better for XFree86. He is a very smart guy and if there's anyone that can fix the font mess in X, he's the one to do it.
But keep in mind that limitations in X are in almost all cases fixed through extensions. Good examples are XInput, or DGA, or GLX, or DPS. Not a single one of these features was planned for when X was written, but now you can seamlessly support all these features in your programs.
I'm particularly impressed with DPS which now gives XFree86 users the equivalent of Macintosh's QuickDraw. Print quality rendering on the desktop and at amazingly high speed too.
Or GLX which gives us an OpenGL implementation that is simultaneously a direct renderer (when it is possible), an indirect renderer (if you don't have permissions to directly access the hardware) and a network renderer (providing X's extremely useful network transparency), all automatically sorted out at runtime.
Or DGA which gives you direct access straight to the framebuffer, but with the benefit of still being able to use XAA for hardware accelerated blits and fills and the like. This makes DGA much better than SVGALIB for 2D graphics.
Keith may eventually produce something like an XFree86-Font extension, or XFree86-AntiAlias. It is an exciting time to be using XFree86. It's all getting better, and faster, and wonderful things are being done. It's always sad to see people who harp on the few bad things in X rather than look at all the great improvements that are being made and developed.
XFree86 problems can be solved, but not if you sit on your laurels and complain.
I just want to see Windows run ICQ, an IRC client, an MP3 player, web browser, HTML editor, web server, telnet daemon, distributed.net client, retrieve my e-mail automatically, maybe run a game or two without gasping and choking. Maybe do some word processing or image editing once in a while.
That's exactly what my system is! My main development machine has tonnes of stuff installed and running at any one time, multiple instances of VC++ and VJ++, always has a IIS running (obviously), telnet servers, and various office applications (word, outlook) loaded. ICQ is on all the time, WinVNC, GetRight, FlySwat, Corel & Microsoft Indexers, Winamp etc...and I have no problems. (all these programs are open at once most of the time).
Really, I don't understand this idea people have that windows crumples under some load like that....
Maybe you just need to treat your system as stable...I find when people expect their systems to fail, they often find a way to make it fail. I treat my systems as rock solid stable systems, I do put them thru their paces, and they still run. I don't go and reboot them just for higiene either.
I mean, if you think the kind of stuff you say you'd liked to see windows doing without choking is like some kind of list for making windows choke I really really can't understand it. I EXPECT my machines to do exactly that, and I use my machines for exactly that...and more...
But I am running Windows 2000
I mean, perhaps you should try NT or Windows 2000. The kind of feelings of instability you've built up regarding windows is quite common around here....not sure how realistic it is now days.
I've found Xfree crashes one too many times for me to do any GUI based work with Linux.
Basically now, if i have a Windows 2000 box that BSOD, I get shocked and really try to figure out what's going on. I'm a windows user, but I don't find any crash acceptable anymore.
However, people are becoming frustrated with NT, and longing for the stability/flexibility they had with their IRIX based systems. Major animation houses are now moving to Linux, at least for rendering. Maya has its renderer ported to Linux already, and houses that write their own software are porting to Linux also. (I've spoken to people from two major houses who said, "we're giving up on NT.") But, on the actual interactive end, where 3D performance counts, the major packages have not yet been ported, (though the houses may be porting their in-house packages.) So... much of the discussion of Linux for 3D animation users is, "Will, or when will, enough packages port to Linux so that we can use it instead of NT?" Also, "Will we have the level of OpenGL performance we need for these applications?" It's good to see the gaming article, because, attention to 3D that helps the gamers winds up helping the animators too, but I would like to see more articles addressing these aspects of 3D Linux performance.
you dumb motherfucker. Active Desktop is a braindead piece of shit, much like yourself. who the fuck wants all that bullshit going on a fucking desktop. put the fuckin icons out and leave me the fuck alone. you cockless microsoft groupies deserved the living hell you've created for yourselves.
-- You see, there would be these conclusions that you could jump to
Here's the deal...
XFree 4.0 which _has_ been released, supports DRI (Direct Rendering Infrastructure) which lets you have direct access to the graphics hardware giving a _big_ improvement over XFree 3.X's older archecture.
Now, it is still a bit of a pain to set up, and you can't run Quake or Quake II (only Quake III) under it, but it gives _awesome_ performance. Have they bothered to try this, they would have found Linux KICKS WINDOWS ASS.
Check 3dfx's linux website, source forge, and XFree for more info...
Take a look at the benchmark results on http://www.tpc.org/ and you'll see some Windows doing some *real* work faster and cheaper than many UNIX based solutions.
</flamebait>
The Windows gaming crowd has a lot of things going for it:
First, lots of companies have thrown lots of really smart programmers at the problem of maximising FPS, the holy grail of gaming. Yes, there's a lot of Linux programmers out there that are just as smart, if not smarter, but the Windows game programmer is following a beaten path and has a head start.
Second, the hardware drivers are all developed Windows first, everything else second. You can bet your booty that only a tiny fraction of programmer hours have gone into the Linux drivers, compared to the Windows drivers.
Thirdly, gaming tends to grab the entire system and take it over completely. Timesharing? Yeah, right. The kernel, however, still has the overhead of paying attention to a variety of different processes. I'd like to know, for instance, what else was running on that test box while it was up. Maybe rerun the tests on the Windows box with ICQ, some kind of mail notifier, MSN Messenger, and AOL IM all running too, to simulate the backend load of Linux. :)
Fourthly, as some people have mentioned, there's the overhead of the X system... although I really wouldn't expect that to be handling the pixels. I always thought that a "fullscreen" driver kicked in, and shouldered X out of the way, while a game was running.
Like the Mindcraft benchmarks, though, this gives the Linux community something to aim for, and possible places to look at, as well as a kick in the seat of the pants to get further on it. I can't wait to see what we have by Christmas, personally!
RollingThunder
Warlord, Starslayer Tribe
(FPS gaming group for the non-gamers)
PS: What about Linux's performance in online gaming? Does it eke out a reduction in the other half of the gaming grail, ping time?
I am surprised that this became a Linux vs Windows kind of thing. Windows is not even involved in this story. When a game developer writes on Windows, he would be stupid to use any more of Win32 than needed to create the window and delete the window. Most Windows games today are programmed in DirectX, which offers almost all the services needed by a game. Once one learns how to set up a windows and do some threads, almost everything else a game needs is done through calls to DirectX instead of OS calls. In addition to that, most games spend most of their time inside their own code anyway, so the OS has a negligible effect on performance, (assuming of course that the app has direct or near direct access to hardware). Som in reality the competition is Linux vs. DirectX. Quite an impressive showing on Linux's part, at least for now. DirectX (true to its name) has much less overhead that any OS ever could. I wonder, however, if this success is short-lived. Quake isn't exactly a service intensive game. Sound is pretty basic, as is input. They only thing that really matters in this case is the speed of 3D subsystem. Game, however, will eventually evolve, in particular using more sound. Some games (thief in particular) already do this. I really wonder if Linux can handle yet another subsystem that needs direct access to hardware. As 3D sound controllers become more complex (like Aureal's controllers) the OS will need a good way to get huge amounts of data and commands to the sound cards. In these types of games, Windows will keep trumping Linux. The main problem is that Linux doesn't have DirectX. In absense of that, it has no consistant API that pushes the OS out of the way and allows the game to take over the system. Take a look at the rest of the APIs. DirectSound, DirectInput, DirectDraw, and to some extent DirectPlay (for its flexibility) whoops anything availible on Linux. (Any Linux driver-hardware combo that allows 32 HW accelerated 3D sound streams in addition to 96 normal streams? Windows had that in the A3D 2.0 days.) What Linux really needs at this point is a DRI for the rest of the sub systems. Only then, will Linux become better than Windows for games. Some will say the current situation is enough. Linux is almost as good as windows, it performs almost as well. Almost will not cause anybody to switch, however. If Linux really is a technically superior system, it should be able to trounce windows, not merely equal it.
A deep unwavering belief is a sure sign you're missing something...
I never said a word.
--
It's a
-- Danny Vermin
Dual P2 450 w/ 256Mb RAM + 512Mb swap.
Memory footprint is typically below 350Mb unless I have JBuilder running. NT swaps aggressively, so unless have JBuilder in memory, I typically have 50-85Mb of RAM free, leaving the machine feeling very responsive.
I did upgrade it from a celery 366 w/ 128 Mb... that was unusable, especially when using JBuilder, which of course likes to use 100-150Mb, with at least 70Mb swapped into memory at all times.
Dual P2 450 w/ 256Mb RAM.
;) It'll be interesting to see what happens at your end when Q3 starts.
If you're going to accuse people of lying, make sure you can put your money where your mouth is.
Download pcAnywhere trial and send me a personal email. I'll let see for yourself. Then you'll own me an apology
"Windows is still somewhat based on DOS"
Err, no. What do you think the "NT" stands for in "Windows NT"?
Win2K is based on NT and its performance is great. Quake3 on my dual NT workstation outperforms Win98 hands down.
I believe the point he's trying to make is that for those people who have ever considered running Linux, more and more bases being covered by Linux and Windows begins to look less and less appealing due to BSoDs and other stupid Windows tricks.
Have you killed Morgoth?
Because the bandwidth of OpenGL is significant. Really huge in fact. We're talking 100s of megabytes a second over the bus to the video card. In the near future we will be looking at 1000s of megabytes a second.
You can far more easily bog down a 2D chipset without the bus breaking a sweat.
But there is no alternative! Read what I wrote up above. All windowing systems have to provide some synchronisation method. Some of them do it with message passing. Some require mutexes. Some require async-io.
Windows does it with message passing. And you still need to assemble "packets" and you still need to call an IPC method to send the packet. It isn't any different to X when you get right down to the nitty gritty.
X's synchronisation method is amazingly cheap on UNIX systems, and has the natural benefit of also providing network transparency with little extra cost (just some byte reordering as packets are built).
I don't understand why people can't see that all windowing systems have the exact same problem here. Why do you think Microsoft was forced (by games developers) to include DirectX? It wasn't because Windows is blindingly fast.
But people like Jamie will bash specific bad points in X and with good reasons. I'm not seeing any good reasons here on slashdot: just a lot of vitriol and finger pointing and cliched remarks like "X is slow" and "X is bloated", without any actual thought about what the real problems might be. Remember that DRI is bypassing X, so it is a big leap to blame X for the problems here.
That said, X is really showing its age. IMHO network transparency is neat but not necessary at all for 99.9% of users.
Yeah, it's really high-bandwidth (at least if you're on a modem). But as more and more homes get DSL/Cable and local LANs, the more useful GUI-over-the-network will be.
perhaps the solution is pure client-only rendering.
At the cost of reimplementing every GUI application? Ports of GTK+ and QT would handle most of it, but Netscape and other proprietary apps aren't going to deal well with X going away. Not to mention interoperability... I like being able to run emacs off a Solaris machine on the other side of the network. Also it'll be a lot of work. The Berlin project has been working on their stuff for, what, years now, and it'll be a while yet before anything real is running on it. Replacing X is a very long-term proposal (though Berlin looks really cool).
It's all X's fault. That said, having a networked client/server GUI beats the shit out of a single-user, single desktop GUI anytime.
It's a shame you can't mod and post in the same story; I'd like to be able to both negate the "insightful" rating and explain why it's BS.
Take a look at the drivers they used. Not a one of them sends data over the X pipe. The X server basically is there to say, "yeah, you can bang directly on the hardware" and then get out of the way.
If they were sending data over the X pipe, you'd definitely know it. 3D hardware acceleration is often bandwidth limited; you could get up to a 50% drop in framerate without direct rendering. Smart design would reduce this problem, but I still suspect you'd see 70% optimal framerate, max... and in the LinuxGames tests, 70% was the worst case, not the best.
What was the best case? 99% framerate. This suggests to me that it's not idle processes or the kernel hogging CPU, it's not any weirdness from X or kswapd... it's just that some drivers are better than others. And right now, it looks like Windows drivers are 5% to 40% better than Linux drivers. Frankly, since Windows sales are 500% to 40000% better than Linux sales, I'm not complaining about driver quality.
I am surprised to see the 3Dfx drivers do so poorly, though. Isn't anyone helping out Daryll Strauss now that we've got source code available?
But of course, just how much can you trust the benchmarks? They ran it on one game, using a particular configuration, for a specific kernel.
Well, aside from the usual difficulties, there was one special case; the Matrox testing was done using full OpenGL drivers under Linux and a specific Quake 3 "TurboGL" driver under Windows; TurboGL drivers are Matrox's OpenGL subsets designed to run one game per DLL. In fact, the TurboGL driver postdates the Linux Mesa drivers; at one time (and probably still) the Linux implementation was significantly faster than the full Matrox OpenGL.
It's not 'insightful'. The poster jumped to conclusions.
I'll ask the same question delmoi asked :
This is so completely wrong, how the hell did it get modded up?
Your claim that a $300 console machine will be more powerful than a PC's graphics system is laughable. If you take any particular console system, and look at the graphics accelerators currently on the market on its release-- MAYBE, I mean MAYBE the console will stand up fairly well in comparison. But given a few months the PC world has rapidly passed it by and fillrate, and features of the PC gfx cards will blow the console out of the water.
Personally I'd rather look at games on my TNT2 Ultra, than on a PlayStation or Dreamcast or N64. PS2 looks nice, but again, NV 20 will blow it out of the water, and X-Box's NV 25 blows both out of the water. (in pixels/sec terms, and in features-- being able to do crazy amounts of multitexturing in fewer and fewer clock cycles, and accelerating math functions used in 3d programming (see PS2 coverage on Ars Technica)) All of which will be surpassed by the fast paced 3d accelerator world.
I also like the depth of PC games more than that of console games, on average. While I can't deny my urges to play fighting games on the console systems (something PC games lack are decent standard gamepads-- but then again the DC controller sucks for Capcom fighting games), I can't say there is much for an RPG fan to be played on the console systems. Console loyalists may be yelling "Final Fantasy" and "Dragon Quest", etc, right now but I would say that these are adventure games with RPG elements. (the same description could be applied to PlaneScape: Torment.. blah). When I mean real computer RPGs I mean things like the upcoming Elder Scrolls: Morrowind from Bethesda, and DW Bradley's Wizardry 8. Freedom to wander between tons of side quests, and avoid continuing main plotlines for extended periods of time, along with amazing flexibility to customize the character(s) you play as. Non-linear gameplay and finding best combinations of characters to beat the game are things that give these games replayability.
The console will NEVER EVER replace the PC. It might try to become REALLY similar to the PC (keyboard, mouse, etc), but the limitations caused by not being able to upgrade components, or not being able to use your own preferred ISP and method of connection (Dreamcast) will keep PC gamers on their PCs for a long while.
Don't get me wrong, I love consoles, but they are replaced every 5 years and the PC cycle is like every couple months (for gfx accelerators)
Someone was telling me today that the technology of the 3D accelerator market is rapidly outpacing the CPU market, and that NVidia said something to the effect of, "Moore's Law is for wimps".
BTW- the Voodoo 5 is just die shrunk and overclocked Voodoo 3 technology, which is die shrunk and overclocked Voodoo 2 tech, which is die shrunk and overclocked Voodoo 1 tech. (*GRIN*GRIN*) See John Carmack's comments on their fabled T-Buffer (simply an accumulation buffer that can be done on any modern gfx card). I believe their FSAA acceleration gimmick is the first real innovation they've had in years. I think game developers would prefer that they spent more money on better drivers, T&L acceleration (next year will be the year of the T&L and AGP, as games will be beginning to take advantage of it), wider bus to memory, and more efficient multitexturing. Simply put, given a competent programmer making a new game-- the GeForce 1 & 2 blow the V5 series out of the water in quantity of features that the developer can use to make his game look prettier without taking too big of FPS hits. The V5 will have to use raw fillrate and offload the rest to the CPU to accomplish the same.
Ok, this post is too long. G'night.
While it's the ugliest and "slowest" of the possible cards to compare- it is more of an apples to apples comparison.
I am not merely a "consumer" or a "taxpayer". I am a Citizen of the State of Texas
"As Linux grows because of its capabilities in other areas and its openness, it will gain market share, and the disparity between the two will decrease."
- 6.html
So if things were different they wouldn't be the same. Go figure. & Linux will improve over time. Oh my gosh, what a revelation.
You're saying that this isn't a comparison of the OS. I think the article was pretty straight forward in saying that it was just comparing Windows vs. Linux 3D Performance in Q3A. Do we really need to give Linux a hug and say "Don't worry that doesn't really matter because it will get better in the future." and make excuses every time someone points out a short coming?
Linux comes up short when comparing Windows vs. Linux 3D Performance in Q3A. Accept it and help provide suggestions in improving things or move on. Excuses doesn't make Linux any better.
I don't mean to be hyper critical here. I hear allot of "Well windows has lots of developers." or "Well windows has X or Y." and such statements when Linux comes up short in certain areas. In the end that doesn't help anything, in fact Linux does do poorly in these tests. I refer to the last line in the Linux Advocacy how to.
http://howto.tucows.com/LDP/HOWTO/mini/Advocacy
And the 400% figure comes from....?
The Xfree-v4 DRI driver from 3dfx still goes through Glide. Check out this page at 3dfx. You need the new Glide installed before you can install the DRI X-server.
The IIRC the 3.3.6 X-driver used DGA (and so was full screen only). As a result it is entirely possible that it should be as fast as the DRI solution, or faster if the DRI implementation is not yet as well optimised.
So I am quite prepared to believe that the DRI implementation could be *slower* than the older version at the moment.
Windows' failure to have a model that allows components to be replaced by enhanced/redesigned versions, like the VFS, kernel abstractions, and X Windows abstractions (XFree86 4 didn't break support for all applications because it has many new features) is what has "caused it to suck eggs". Windows should never have been divided into two seperate lines (NT and 9x), because it has caused nothing but headaches for people who desire a stable OS. (not releasing newer DirectX releases for NT users, etc) If Microsoft truly desired to use their market force in a positive manner, they should have radically redesigned Windows after 3.x and created emulators for 3.x and DOS apps on the new platform. (they did, afterall, have all the specs!).
/.) , and APIs and get a stranglehold on any developers who invest more than a small amount of time or money into developing on that platform. (eg. why did D3D developers not port their games to other OS'? because they didn't know how to do the same tricks in OpenGL).
The answer to why MS hasn't gotten a clue and written their OS like a cleverly packaged and beautified POSIX system is because they are infinitely greedy and lazy. They would rather create new slightly (making their own versions of Java?) or completely incompatible protocols (see the Kerberos article here on
I bet you said the same thing back when Super Nintendo came out..
Why do I keep typing pythong?
I would like to point out to the vindictive moderators and flamers that the article specifically talks about how it uses WINDOWS 98. Not NT. Not Win2k.
Many thanks to those who noticed this, and pointed out the details of the way windows works with DOS.
Thanks also to those who actually READ THE ARTICLE.
WARNING: there is a trojan on your
I use Windows when I want to play games or do any real work (despite StarOffice's attempts, it's nowhere near Word). When I need to program in C++ or just browse the internet, I use Linux.
In addition, I'd love to know what version of Windows you guys are running. I've been running Windows 98 SE solidly for over 4 weeks, with no need to reboot. I like the concept of having a crashless operating system, but is it really required for the average user? I don't mind a crash here and there, if it affords me extra affordiability and solid support (MS's Technet section of their website is the most comprehensive I've seen by any software developer).
And to those who say "Well, Windows 98 SE rarely crashes because it's Windows 98 Second Edition, the latest in a long string of upgrades", all I have to say is: "What about Linux?" Doesn't anyone remember the almost hourly crashes of the first few point releases (it took 2 years to get anything close to stable).
- I don't care if they globalize against free speech. All my best free thoughts are done in my head.
IMHO this is the best way to improve the quality of Linux as a whole.
People inside the comunity making benchmarks. Open Benchmarks. Showing where the deficiencies and problems are to get mindshare/resources into getting it fixed/improved.
Remember how Samba got a performance boost when a MS-sponsered benchmark (of all things) showed some problems?
These things are a good thing (tm) for Linux.
-- Aji con Todo!
Um, if netscape leaked memory, why not just shutdown netscape? You don't have to reboot you know. Or even better, don't use netscape.
:P).
...i mean 'winblows' users (man i feel so l33t now), couldn't be wrong? How many NT licenses do you think Microsoft has sold?
And your experience sounds like your hardware fucked up...
I mean I've had horror stories with Linux, where I just lost everything after a kernel panic. It's hardly a proof that Linux totally sucks is it? (there's other proofs for that
A million unsatisfied windows
All drivers under NT/windows could potentially render the system unstable
Uh huh, what was the point of that statement? The same applies to linux...you even said it yourself.
If we all just chipped in 1$ each, linux could have gotten 600% more performance:)
FRA: STFU GTFO
If you have a dual NT workstation, I'll bet it beats the pants off Win98. 95/98 is usually faster than NT on equivalent hardware (though I can't vouch for Win2K speed), but it takes NT to take advantage of a second CPU. Running Win98 on a dual workstation will leave half your clocks playing pinochle.
--The basis of all love is respect
I had first implied that you were lying assuming you had a remotely typical hardware configuration. But then I made an exception for extraordinary hardware, so no, I don't owe you an apology. :-)
When I use something, I expect it to live up to its promises. If a part says that its temperature range is -20 to 200 degrees fahrenheit, it damn well better work when it gets up to 190 degrees. If you buy a fridge, you expect it to keep your food cold. Is it acceptable if every once in a while the fridge gets warm and all your food is ruined? So why do we complacently let Windows crash and lose our work? I expect my operating system to work.
When you're doing real work (like spreadsheets, word processing, data entry, etc.) you want the operating system to do its job. I've lost too much data to Windows to trust it with anything more than my high scores.
Why do I get a (II) [GLX]: __glXInitImports not found. GLX disabled (II) [GLX]: Add the following line in your XF86Config file: (II) [GLX]: (II) [GLX]: Load "glx" (II) [GLX]: (II) [GLX]: More hints: (II) [GLX]: Make sure you are loading libglx.so, not libglx.a (II) [GLX]: Make sure you are NOT explictly loading libGLcore. glx will load it. upon starting X? where the hell is__glXInitImports??? PS. I am not explicitly loading GLcore and there is no libglx.a. Any help appreciated.
skiy. www.Smokedot.org Drug Info, Rights, Laws, and Discussion
Damn, that's pretty interesting, but I'm not sure if LINUX was ever part of that....
-------- "All I want in life's a little bit of love to take the pain away" --Spiritualized
In fact Linux doesn't do anything either. Developers do things to Linux. So the developers that want desktop Linux develop desktop improvements, the developers that want server Linux develop server improvements, and both sides of Linux improve at the rate developers work on them at.
I guess the collary here is if you want Linux development to go in a particular direction you should work on it yourself, and try and convince the Linux developer community that your direction is the best. And from that, if you are not communicating with the Linux developer community, you have no business complaining Linux isn't going the direction you want.
tangent - art and creation are a higher purpose
postmoderncore - art and creation are a higher purpose
It's NOT fast. The bandwidth is limited by the network in addition to the encoding to & from GLX protocol at each end. If you're fill limited then you probably won't see a difference once you get going. If you have significant geometry to draw in immediate mode or the X server cannot hold all your resources (display lists & textures mainly) in memory then you will see a major performance degredation over a network.
An NT BSOD is a very bad thing. It usually means something is pretty solidly hosed in either software or hardware and you'll either need to fo some serious bug-squishing, a re-install or hardware maintenance. It's about the equivalent of a kernel panic.
BSODs in NT are usually reproducable -- BSODs in Win9x are usually random crashes that people just have to live with. If you're 98 machine is blue-screening consistently you'd do well to do a simple re-install. If an NT machine doing it, it's usually worth reading Microsoft's list of STOP error codes to see why it happened and if you can fix it.
--srj/mmv
The drive letters are there for the benefit of higher-level programs (including much of Windows itself) that still rely on that notation. NT's kernel uses a different notation (take a look at the BOOT.INI file in NT:
multi(0)disk(0)rdisk(0)partition(1)\WINNT="Windows NT Workstation Version 4.00"
NT also uses UNC paths (\\SERVER\SHARE\FILE) pretty heavily too.
But you're right, it would be nice if Microsoft would scrap the drive-letter device model. I know Windows 2000 lets you mount network shares as folders on a local disk -- I don't know if this functionality has bee extended to include mounting physical disks as folders.
--srj/mmv
I meant overall stability, not specifically games. Yes, X is not entirely stable, but X is not part of the operating system. The X server is an application, not an integral part of the OS. An X server crash doesn't hose the OS in Linux/*BSD, unlike a Certain Other Operating System.
Otherwise, I agree completely with you. Windows is currently much better suited to games than X on any platform. An alternative, though, is to have Linux games not use X at all. This would be more like games back in the days of MSDOS.
The late betas of the OpenGL rewrite are extremely fast. I'm not sure if the port is just faster or if Be has really great OpenGL guys. It's probably a combo.
Curiously, when I did this in my Linux install of Q3 demo, I got .3 fps less with r_smp set to 1 than I did with it on zero. I run a dual PII 333 setup with Mandrake 7.0. Is there any trick to getting Q3 to use both processors?
Logic ... merely enables one to be wrong with authority. -- Doctor Who
You cant compare computer animation generated by todays hardware with tv or film. When you film something with a camera, the shutter is open for a certain (non-zero) time per frame. This leads to temporal anti-aliasing, in other words motion blur. Current consumer hardware can not do motion blur. Therefore, with the same fps as in a film you will see artefacts that you dont in the film.
Many people can see the difference between, say 30 and 50 fps, even if it is constant.
Additionally, the fps is not constant and also you want to have some reserve for more complex games. It is quite clear that there will be games with more polygons, with higher depth complexity and even more effects in a short while.
Here's a quick test, drop your monitor's refresh rate from whatever it is down to 60Hz then switch it back up. Big difference huh? Games running at a full 60fps have a huge visual distance. You see alot less blurring which I particularly don't enjoy (the blurring of low framerates). The problem I have with the 33fps is that the same exact hardware with a different OS has such a drastic difference in hardware performance. Linux ought to be doing better than Windows, not worse.
I'm a loner Dottie, a Rebel.
Curiously, when I did this in my Linux install of Q3 demo, I got .3 fps less with r_smp set to 1 than I did with it on zero. I run a dual PII 333 setup with Mandrake 7.0. Is there any trick to getting Q3 to use both processors?
You need to have r_smp set to 1, and either put it in your config file or reset the video after the r_smp command. On my dual 450 celeries running 2k on a TNT2, I get about about 20 more frames per second.
Play Q3 in Win98, pull up a console and type /r_smp 1 and see what happens. Ohhhh yeah, Win98 can't use multiple processors. Sheesh silly me. Tomshardware was restricted in its test because they needed to stick to a single processor. Linux, Be and MacOS all support SMP also and benefit from it.
I'm a loner Dottie, a Rebel.
DRI is not only full screen, it also supports rendering in a window. I've seen it with my own eyes in XFree4.0 with my Voodoo 3 2000.
He. I like (and share) that attitude, but you do realize that RMS (who developed the license) does not agree, right? In fact, the official GNU party line is that the Lesser GPL should only be used for non-exclusive functionality, since it doesn't coerce other developers to use free licenses. It's interesting to note that people seem to use the license for quite different kinds of libraries -- SDL, for example, provides a rather "unique capability", but is still LGPL... Well, just food for thought, or something.
main(O){10<putchar(4^--O?77-(15&5128 >>4*O):10)&&main(2+O);}
Either you have a multi-processor machine w/ absurd (>512M) amounts of memory or you are full of shit.
I'm giving you the benefit of the doubt, NOT flaming. I've been running NT4 for 5 years and it is possible to do all that at once, but not w/out some serious hardware.
My own experience - 57fps under Win98se, 62fps under Windows 2000 (Quake3Arena, 640x480 "normal" settings, timedemo of demo001).
This article is a good comparison (using Q3, with and without tweaking) showing similar results. The update at the end, using the newer win2k drivers gave even larger improvements for win2k.
--
qube http://www.quake3arena.co.uk
Umm, until this article, I thought Linux was "lean" and optimized more than Windows. Well these tests prove that wrong. You might say "Its a damn game, who cares". Well games are a VERY important factor as to why Windows is so successful. Stuff runs faster on "bloated" Windows. All my question is "Why?". As much hype as Linux is getting about its speed and optimization, why do these figures show otherwise?
I remember you can enable Q3 to be multithreaded in Win32, which is all you need on NT and Win2K to take advantage of SMP. I know a little about threads on Win32 to understand it's not difficult to use them. How easy is it to use threads on Linux?
Driver writers need to understand threads too. Come to think of it I don't even know if there's such a thing as a multithreaded driver.
Use Evolution instead of Outlook? Bewa
With that attitude, you better be posting that comment using a 300baud modem, on an original IBM PC.
The only reason I ever have to reboot my machine is when Q3F takes a crap. I've got a 3dfx Voodoo3 and when a game crashes it COMPLETELY HOSES all video. The only way to get it back is to Ctrl+Alt+Backspace from X and blindly running /sbin/reboot. I've played with the *textmode commands and try as they might they are unable to recover the display... they do make it better, but I wouldn't say its anywhere near usable.
Other than the display being foobared, the sytem is fine. I just wanna know if this is normal behavior under Linux, and if its fixed in XFree4, or with other card/driver combos.
I'm amazed at how quickly people jump to the defense of Linux by saying it has more overhead, and doesn't let the game have total system control. I thought Windows was huge and bloated! I thought it could never be as efficient as Linux, in terms of CPU time given to an application -vs- OS overhead. I read that distributed.net clients ran faster in Linux than Windows! This reminds me of the Mindcraft scenario, where noone could live up to Linux having a flaw. (In that case, the IP stack)
That having been said, the lesson we need to fess up to is that if Linux is to compete on the desktop, development in these areas needs to progress. Microsoft is behind in servers, and reliability. Linux is behind on the desktop. Both are working toward improving in the strengths of the other.
Business can prioritize, attack, and resolve. Open source must do the same. If we continue to cop-out the problem will not be solved, and open source will get a scar for being unwilling(not unable) to see the big picture.
Your observation was completely true in the days of DOS and Win95, but not anymore. A game running on NT is just as isolated from the hardware as a game running on Linux (and Win98 is only slightly looser). e.g. a Win32 application must use the Win32 API to play sounds, same as Linux apps must use the
3D graphics is exceptional, since direct interaction with the hardware is required for acceptable speed. Both Win32 and the above X drivers provide just such a direct-rendering capability.
(essentially the X drivers let OpenGL functions act like system calls, rather than going through the X IPC mechanism like 2D drawing commands. On Windows *all* drawing is done this way)
Why does not someone make an OS that does not give a crap about security, or multi-user stuff. I am the only one that uses my computer, so why should I be brought down by crap that I dont need. Someone needs to make an OS and API that is dedicated to FAST graphics/sound/Network performance. And no MAC OS does not fit the bill. I think this "new" OS would be very successful.
yet another source of FUD for M$. I can already see the web page they're gonna put up next to their BS "Linux Myths" page.....
mas cerveza, por favor politically incorrect stu
Reading articles that have floated through Slashdot recently, it is apparent that X-Windows has some limitations to it's abilities. Do these impact on 3D performance? And if they do is X-Windows always going to limit 3D performance?
tangent - art and creation are a higher purpose
postmoderncore - art and creation are a higher purpose
Under Windows you're drivers can talk near directly to the hardware, and their are less layers of protection slowing things down. Under Linux, their are more layers of protection between the hardware and the drivers, not to mention things have to talk more directly to a windowing system like X.
This means that in general linux games, etc.. will be more stable than their Windows equivalents, and if they crash, again in general, your system should be able to survive. (Even if you have to telnet in and reboot.)
Here's where DRI and XFree 4.0 come in. With DRI a driver can talk much more directly to the hardware, and generally speed things up, and provide more features.
So in Windows you get a slight speed increase in drivers at the sacrifice of stability. Of course, anything dealing with hardware/drivers can cause complete system lockups, its just less likely in Windows than in Linux.
A better comparison would be either Linux to NT 4.0, or Linux with XFree 4.0 to Windows 2000.
As it stands, you can get a nice idea of the slowdown by the GeForce drivers, which are ownly slightly faster in Windows than in linux. Of other note, is that the drivers in Linux are not optimized for games, as the article brought out several times. It would be VERY interesting to see some Linux vs. NT/2000 benchmarks in workstation operations(cad/design/etc...)
------ 24.5% slashdot pure
That said, having a networked client/server GUI beats the shit out of a single-user, single desktop GUI anytime.
The purpose of DRI was to allow card drivers to bypass the client/server architecture and go straight to hardware (for fullscreen only, IIRC). Linux still has some other issues, but i'm pretty sure they'll be resolved soon. And someone slap nvidia and tell them to use DRI, so they don't have compatibility problems...
Looks like it's hard to argue with the fact that in Linux, if you don't_want_to_do_it_our_way, you can't do it at all right now.
Damn the once a week (once an hour, don't you guys ever configure this stuff??) BSOD's and full steam ahead!
It's not so much Linux sucks for games. It's more that very few games have been released. But seriously, I stopped playing games on my computer long ago. I've always prefered game consols over computers. The only PC games I buy anymore are little desktop divesions, and LucasArts style adventure games (Grim Fandango was the last computer game I bought) and it's a real shame that it is a dying genre.
I realize that computers do have a lot more power, and have better graphics, but my Playstation has never crashed, is user friendly, and doesn't require an upgrade of my graphics card/processor every 6 months in order to play the latest games. People who write games for consoles seem to always find a new way to exploit the little processor power they have rather than require you have the latest hardware in order to play their games. I remember old SNES games that were truely amazing considering how limited the SNES was. New PS games are pushing the outer limits of its capabilities. And the nextgen consoles are looking amazing. A 300 investment in a console lasts for 5 years before its sucessor is even released, and maybe 7 years before the console is no longer developed for. I can spend $300 on a video card that is obsolete before I open the box.
My point is, a computer is a waste of money if all you are going to use it for is a gaming console. If you are going to run Windows just so you can shoot aliens, and 5 BSODs a day is worth it, you are insane. Use a gaming console for games, goddamnit! And use a computer for work.
I've heard the opposite. I'm not inclined to believe either view without an attempt at proof. Are there publicly available benchmarks for this sort of thing?
"Linux would have to decide whether it wants to be on desktop or on the Server side."
:o)
A good gaming OS does not a necessarily have to be a good Desktop OS - the two are entirely different things.
As far as a good desktop OS, I find myself much happier with Linux/X/KDE than with Windows. It's easier (and more consistant) to use, and I find myself much more productive than under Windows. (It's more functional, and more stable.)
A gaming OS needs to reduce the number of obstacles between the hardware and software - this is (more or less) something Windows has done - however it's done so at the cost of stability
(functionality is a different matter - it's less functional than KDE, but I believe that the functionality that is there is the primary cause of its' instability.)
A year ago, I believed that Windows still had a place in computing, in the role of the desktop; that Linux wasn't going to evolve into a useful desktop OS.
This has changed - I still believe that Windows has a place, but now as a game OS, as Linux has proven (to me) that it can be more "user-friendly".
I'm not so convinced that this will remain to be true - next year may see Linux on top of Windows for game performance. (But I'm not holding my breath
Someone give me an example of a situation where well written code on Linux will run significantly slower than on Windows.
Just because it CAN be done, doesn't mean it should!
Wow, an epiffiany! Thanks to your post, I realize I'm one of the thousands of loosers (millions, actualy) out there who wastes time playing video games! Movies, books, video games, music, everything is just such a waste of time And remember, idle hands are the devils hands. And things that are fun should never be done. What was I thinking. All this time I've wasted playing quake3 I could have been staring at a wall or something! or working feverishly on a lame, content free, website!
Thanks for showing me the way albamuth!
ReadThe ReflectionEngine, a cyberpunk style n
How many think the following experiment would be interesting:
Set up a fake "benchmark report" on a reputable website (get the owners in on the gag) that purports to prove Windows is marginally better than Linux at doing something.
Next month, set up a very similar benchmark on the same site that proves Linux is marginally better than Windows at the exact same thing.
Then, compare the slashdot comments. My hypothesis is that we'd flush a lot of handwaving and armchair OS design... How many people giving authoritative views here even have a flying **** about the ways 3d games *really* work? Heh. Not me, that's for sure!
-konstant
Yes! We are all individuals! I'm not!
-konstant
Yes! We are all individuals! I'm not!
Windows is still somewhat based on DOS, which never was too big on multitasking. Games completely take over your system,
This is so completely wrong, how the hell did it get modded up? Windows isn't based on dos any more then Linux is because of Dosemu and lodlin.com. The only reason the system boots in dos mode (in win9x) is for compatibility sake. In fact, this no longer happens in WinME.
And as far as taking over your system, it simply doesn't happen. The game can take over your screen with directX, and use a Realtime priority thread if it wants to hog the CPU, but it probably doesn't. This is no different then what happens in Linux. Win32 programs don't use DOS at all, that's why they run on NT and 2000, witch don't use DOS code at all.
Is it so hard for people to know what they're talking about before doing so?
ReadThe ReflectionEngine, a cyberpunk style n
Q3 under 98 = 68 fps
Q3 under 2000 = 62 fps
This is about 9% slower.
I wish the Linux games guys had tested against Windows 2000 as well.
Hboy
test.
I have a Celery 400 OCed to 500, Voodoo3 2000 AGP.
I get 1 fps higher in linux than I do in windows. Running at normal settings, at 640x480, I get 62 fps in win98, and 63 fps in linux.
Sticking feathers up your butt does not make you a chicken - Tyler Durden
Erhmm yes in fact windows is heavily based on DOS but that is not relevant...
If you knew what you were talking about you would know that in a system designed to support multiple users with differing priviledges performance penalties are incurred because of context switches.
You might also know that win 9* is not really a protected mode operating system and you might have a clue as to what that means at the hardware level...
Oh and win32 programs ALWAYS run on NT, any game can run on NT just fine yessiree...
Putz
Unfortunately this study didn't really reflect the real and significant performance differences between OS 3D support.
An ostensibly fill limited benchmark will not reveal any disparity beyond state issues in the hardware. The geometry performance is not going to be significantly different in a software implementation of T&L provided there is at least a moderate effort by the vendor to support instruction sets like SSE or 3DNow on Linux.
What will be really telling in future is the difference between dispatch mechanisms & kernel level support for graphics when we get some heavy duty T&L support and a benchmark which exercises it. This will become significant for many games and is already significant for 'serious' 3D applications.
You are simply not going to get anything near parity unless we can persuade Cox & Torvalds that 3D is important enough to distribute the kinds of kernel hacks needed for efficient 3D support. Right now they are intransigent and have even lambasted 3D experts like Jon Leech in public over kernel support issues. In addition we'll need to remain flexible about driver implementation frameworks we want to support. Something like the DRI is OK now but when you need to send 15 million triangles a second to the graphics card it's not going to cut it.
Unfortunately Linux currently does not give anything near parity with hardware T&L. Even with an excellent driver implementation and the best will in the world it cannot and it may not for some time to come.
Most games will NOT run under...
{am I about to make a fool of myself? did microsoft finally realize that some people want directx in nt?}
[checking MS site...]
Microsoft DirectX®, the programming environment that's an integral part of Windows 98, was not supported in Windows NT. By incorporating DirectX, Windows 2000 outguns its predecessor and eliminates the need for dual booting. Windows 2000 is therefore able to support considerably more games than Windows NT, without sacrificing any of the latter's signature stability.
Well I'll be damned! I actually learned about an honest-to-god feature of Windows because of Slashdot!
This has been a message from someone who works with VMS.
--
5-10% speed improvement? Wha? I have talked to at least twenty people who have done this and have not seen that result. All had a drop in performance. I'd like to know where you got your numbers. Gamecenter shows 98 beating 2000 by 13fps and 10fps on identical machines. That roughly matches my own tests. The only game I've heard of even posting an increase was UT, and it was by 3fps - negligable.
--
I think there is a world market for maybe five personal web logs.
blocking sites? Nope we can't have any subversive activity.
I'm currently logged in as my redundant backup account as my primary failed over.
As for Windows, I haven't lost any data that I care about for quite some time... part of this may be the Pavlovian Alt-F-S I smack every time I stop to think, but my BSOD's are nearly zero, and when it does crash it's while doing something that isn't available on Linux anyway! And my fridge does stop working and spoils all my food once in a while. It sucks, and it's called a power outage, but we're all used to it in a region where the lines ice over every winter.
Present performance is all that matters. "Tweaking" shouldn't cause a 400% speed increase. Linux is a great OS but it am not a good game OS. =)
Yet...
--
--
The gravitational constant of protein has been changed[...] Also, rabbit carcasses no longer weigh as much a
GNU/Linux. The Freshmaker.
This shouldn't be surprising. Windows 98 simply does not have the amount of overhead that Linux or any multiuser OS with a good security model has. Things like that just take more CPU cycles, period. Windows will always have a bit of a performance edge for anything, but at the same time it will be a lot easier to crash it, write to memory that you're not supposed to, etc. Hardcore gamers will almost always stick with Win 98/ME. For the rest of us, it seems like Linux really is coming along. Again, it's just a measure of driver support, this test. Obviously Nvidia has more of it right now. Given two mature drivers for Linux and Windows, though, and Windows ought to win.
--
I think there is a world market for maybe five personal web logs.
But gaining a lot of stability.
If you give people this choice, do you think
people would pick the less stable version? I
don't think there would be a wide audience
for an unstable gaming platform.
What good is more speed if you waste time
rebooting your system because it has crashed?
Here's my .02 on linux driver support. None of the linux drivers seem to take advantage of SMP, as they do in Quake3 under windows NT. The difference on my dual 550 system is noticeable. That's not reflected on any of those tests-they were on a uniproc system. While the newest round of drivers from Nvidia is definitely usable, they're missing a nice perk(as are the 3dfx drivers, for the voodoo2 at least). It almost makes me want to boot back into NT to play quake. Unlike some of the differences in those tests, the difference from 50-70 fps is a pretty big one. I just hope that Nvidia deals with some of those issues. This lack shows how unfortunate Nvidia's choice was to keep the drivers closed.
I agree... especially with all those custom/niche/nonmainstream apps that only exist on Windows. My own personal stab on this is that I was trained on a program called SolidWorks for our CAD, and after going to try AutoCAD, I saw my boss' point that it was an antiquainted waste of time to learn. Unfortunatley, I'll probably have to learn it some day, because it is the popular choice, and I like to work and eat.
Well, you're rather lucky. I've used Office more than I wish I had and it's lost a few of my docs now and then, even on NT. Not a lot, but enough to really annoy me. With the chancyness Office and Windows combined, it's not surprising that people save their docs every minute out of fear that one of the two will have some strange error.
Not Tested?
Sorry, couldn't resist. Especially after my NT machine just BSODed for the second time this week.
I don't know if this functionality has bee extended to include mounting physical disks as folders.
Yes MS finally saw the light, W2K will let you mount physical disks or partitions as folders. I triple boot on my computer (DOS, W2K and Linux) and I access the DOS partition from W2K as a folder. Just go to Disk Management in the Computer Management Console to mount other partitions and drives.
We've been getting increasingly insane TPC-C 3.x numbers for the last year - that benchmark is not a realistic test of current-generation hardware and databases. Even the TPC committee knows this - if you look around on their site, you'll find whitepapers for an upcoming (heavily modified) version of TPC-C which will not have all the work optimized away and running in memory. They don't quite come out and say "TPC-C is currently worthless" - the vendors paying their bills wouldn't care for that ! - but it's there if you read between the lines a little.
Expect vendors to keep quoting figures from the old version after the new comes out. Insist on knowing the TPC-C version number when somebody throws one of these at you!
It *was* a really good test in 1996 or so, it would be handy if they fix it right.
Useablity is certainly a factor at the moment under Linux, but this is changing with desktop environments such as Gnome, which put user considerations at the top of the list. All OSs have some distance to go with ease of use and can all learn from each other, but making an easy to use interface is an art that is harder to perfect in certain cases, than creating a good kernel, where there are hard and fast rules.
It is the demand of today that rule decide the direction of tomorrow!
Jumpstart the tartan drive.
Quake 3 doesn't support SMP under Linux. Enabling it does nothing.
Considering how long WIN has been running and tweaking how both 3D hardware and software work with the OS in relation to games, I'd say Linux is doing great.
---
How long have you been listening to the world's famous?
'Bout six weeks.
Six weeks!
I just leave myself logged in as root on console #6. I wrote a script that contains this:
/root/etc/textmode.dat /usr/local/glide/bin/test3Dfx
restoretextmode -r
consolechars -fgr8x16
xinit -e
and if q3 bites it I just hit Ctrl+Alt+F6 and run the script and everything's all happy again.
Thanks again for the info
Windows isn't based on dos any more then Linux is because of Dosemu and lodlin.com.
Wrong. Win95/98 still have good old MS-DOS underneath the hood. In other words, you cannot run 95/98 without DOS (just because you don't see it on bootup doesn't mean it's not there). This is probably true in WinME, though much of the DOS command.com functionality has apparently been removed. True, NT is not DOS based, but can run DOS applications in a special subsystem. I'm not sure what to make of your comment about Win32.
One of the reasons 95/98 is faster than NT/Linux for some applications is the fact that DOS is so light. In fact, many have argued that DOS is not really an OS but a library of hardware routines.
My hope for the next NT is that MS finally gets rid of drive letters. This holdover really, really tells of NT's deep roots in DOS.
You can have it both ways. Since the GUI, X, is not tied into the kernel, it's possible to have a system with or without it. Want a stable server? No X. Like to live dangerously on the desktop? Throw it on! Seroiusly, when the GUI stuff isn't mixed into the kernel (which is starting to change somewhat with accellerated video drivers, such as NVidia's - but even that's a module), Linux users can have their cake and eat it too.
The thing ignored here: perhaps Quake3Arena is better optimized in Windows than it is in LINUX.
--
er
i don't have any files by those names
Yes, it would be feasible with a dual-Athlon or 512MB+ of RAM, but it wouldn't be stable.
--
Let's not all suck at the same time please
Let's not all suck at the same time please
no linux won't crash with non-beta drivers. The creative drivers were buggy closed source drivers. Creative has released the source to these and now my sblive works perfectly. I've had non-beta 3c9000b drivers from dell crash a windows nt machine and put it in a non recoverable state. The fact that you have to reinstall the service pack to nt every time you install something shows just how rediculous nt is.
It has been statistically shown that helmets increase the risk of head injury.
...The very reason I'm still on Windows rather than Linux. Linux has an uphill battle in these arenas right now mostly because MS has almost everyone in the world developing/optimizing/etc. for their OS because it's the quickest way to hit the largest market. However, this shows that there are passable Linux drivers out there, and that the multimedia infrastructure is steadily increasing... Now, it just remains for one of you Linux-lovin' wizards to create the must-have game for Linux ONLY and start pulling those fans over. BTW, if you use GPL'ed libraries (this would be the equivalent of the MS Foundation Classes/ActiveX whatever), I'm assuming you have to open the source, right? If so, it may be necessary to kill off this clause and allow the multimedia developers/game writers to keep the IP they have worked to build...
Sorry to go offtopic here, but I would much appreciate it if you could answer a question for me. The display brightness slider in Q3A seems to work differently in NT, I find the optimum brightness is at around 40% of the way (all the way and it gets pretty dark) but its still not as bright as it should be. This is with TNT, TNT2 and GeForce, so it might be an nvidia thing, or it might be an NT thing, but if you have a "fix" for this problem I'd appreciate it.
There is Win95/98 and there is NT/2000. They differ a lot.
As for Linux being such a superior OS -- that must be compared on an issue-to-issue basis. It's true for some, not true for others (usability).
A little bit less generalization, please.
If all's good, there won't be any Windows twenty years from now and nothing left to argue. And yes, I want everyone to use Linux because then my family will use it and I will be able keep everyone from deleting important system files.
It is true that if you keep your console for say four years, the PC hardware will have leapt on by leaps and bounds in that time. This for me is an argument for buying a console. If I buy a game in four years time for my console it will be a good, fast solid game. The graphics will probably be better for it too as the software tehnology will have improved. If you buy a game for your PC in four years time it probably won't even run and if it does it will be really slow and almost unplayable because it will be written for the latest greatest hardware which you haven't got yet.
So if you use your PC for games, expect to be dissapointed soon unless you pay for regular hardware upgrades/new PC's. I on the other hand can play nice fast games until such time as the lates greatest console drops in price to practically nothing and I can be bothered to buy one.
Regards
Linux doesn't need to decide. We need some more optimizations and compile-time options, but we need different distros focused at different solutions. We could have a Debian based distro with a kernel compiled for maximum speed with some stability checks left out in order to make it faster, with XFree configured for maximum graphics performance as well. A set up like this would work for some one who's primary purpose is gaming, and general things.
A second distro could concurrently exist with the kernel tweaked for maximum io for servers. It wouldn't need anything more than the standard install, it would just be slightly different.
These two distros could just be Debian variants on one CD so a person could pick the one they wanted at install time.
Nascantur in Admiratione. (Let them be born in Wonder)
What does that have to do with anything? Dell laptops have been running NT forever. And NT bluescreens quite a bit. Win2k is better than NT as far as features go because they inserted third-party add-ons and code from win98. It bluescreens less, but it can still die just like anything. What part of his post do you think is a lie?
CP
Why do you think DRI bypasses X? Answer: because of X's limitations; because all X calls necessarily involve IPC; because X is inherently slow (by comparison, mind you -- I don't mean it's horribly slow).
The "X bashers" have a point. And a lot of people who do understand X quite well (Zawinski comes to mind) have said the same things.
"Supposedly, there is still an occasional crash"
If the system freezes while playing Q3 and you have a GeForce, it might be the motherboard - apparently some motherboards (specifically cheaper ones) do this. Its not the driver and its not nVidia's fault either. We have that problem where I work (we do 3D graphics apps.)
We suspect that the cheaper motherboards cut some corners and can't handle the power requirements of the GeForce (16W) .. the GeForce2 uses less power, 9W, so it is quite possible the problem won't exist on the GeForce2.
I don't know if this is the prob you are referring to, but it sounded like it might be.
I purchased my Voodoo 3 because of the new DRI stuff in XFree86 4. I've tried the card under W2K and was amazed with its quality and smoothness even at high resolutions.
Under Linux, the story is different. While bearable, the card has visible chunking even with undemanding quake levels, and slows down a lot with advanced features like reflective surfaces. It is understandable, given the drivers have to go through the old glide API, but still, I want my speed!
I can't stand to use Windows, so I can do nothing but wait for better software under Linux. Meanwhile, my gaming friends mock Linux as a gaming platform, and rightfully so.
Somebody really didn't do their homework on 3Dfx drivers here.. The author seems to shrug off the XFree86 4.0 drivers for 3Dfx as being unstable or otherwise unready for public consumption, and instead chose to use the older, indirect XFree86 3.3.6 driver. If this person had built up enough courage to play with CVS, he/she might have discovered that the DRI are in fact closer to release quality than the NVidia drivers, and were even back when NVidia did their release a month ago. The bottom line is, both the NVidia and 3Dfx drivers for XFree86 4.0 are screaming fast and not yet release quality. Why this reviewer chose to use the beta driver put out by NVidia, but not the 3Dfx driver remains a mystery. It really paints an incorrect picture of the state of things. IMHO it's too bad.
Hear hear. My 98 box at home crashes (always Explorer =) maybe twice a day doing intensive work (coding and gaming). My NT box at work (work only, frown) has crashed once in three months. Why an OS this reliable gets such bad press for "crashing every hour" is beyond me.
Yes, I'm carefully sidestepping any mention of security here =).
--
--
The gravitational constant of protein has been changed[...] Also, rabbit carcasses no longer weigh as much a
>Linux would have to decide whether it wants to be on desktop or on the Server side.
:-)
...the amount of desktop apps it has
...and also the amount of games that has been released for it
:-)
You have Linux (the kernel) and Linux Distribution (most everything else) confused. The linux kernel simply gives software easy access to the hardware (among other tasks). Currently distributions like RedHat and Mandrake focus on the desktop/ease of use market (IMHO), and distributions like Slackware focus on the server market (again, IMHO).
Of course, there isn't any reason why someone can't invent "desktop Linux". I suppose that _was_ what LinuxOne had in mind...
>Windows is a clear winner on the Desktop for its Ease of use
Your opinion, not mine. Show a little proof from a reputable source (no, this doesn't include any ZD publications), and this point will make sense.
>Windows is a clear winner on...
Again, your opinion, not mine. Personally, I have found Linux can do more for me desktopwise than Windows -- and at a price I can afford. What can you do in Windows that cannot be done (perhaps in a different way) in Linux?
>Windows...
This I agree with. There are many more games for windows than there are for Linux. Of course, I think that console systems are best suited to games. Nowadays you save money by buying a console over an uberfast computer, and end up with a games machine that is faster and better suited to the task. So, if you want to buy $1000 worth of Linux computer and a $300 game console, you now have the same gaming capabilities (and game availability) of the $2000 windows game computer.
>Linux is powerful for its robustness on the Server side.
Thanks...
>I dont mind a desktop that crashes twice a day
How many people would say that about any other home electronics/appliances. "Hey hon, the fridge crashed again." "Just push the reboot button". Personally, any crashes are unacceptable to me. They lead to data loss, and time wasted. Two things that royally p*ss me off, and most certainly make the boss mad (fortunately, not at me).
>But I dont want a stable OS to run on my desktop with no games to be played.
Buy a console and a BT848 TV capture card. A lot of people won't even realize it isn't being played on the computer and the games aren't any less fun.
If you could be told what you can see or read, then it follows that you could be told what to say or think - BoC
Don't forget the fact that X also goes through a network layer, so the program and the display can be on entirely different computers (even different OSes). So, you could RUN Quake on a computer with no video card and make it DISPLAY on a computer with a GeForce. However, I doubt that even Gigabit ethernet is as fast as the loopback interface, so performace would be "less than optimal", to say the least.
For a long time, it was my impression that 3dfx had one big selling point behind the Voodoo3: Linux support. I heard it was good; really fast and ran Quake3 (the true test). Matrox had the same deal (but they had features on their card, like 32-bit color). Now, thanks to the fact that NVidia held up their statement that they would release good, new drivers, gaming (or anything in GL) in Linux is a FAR more viable option. NVidia has put out wonderful cards, AND DRIVERS, in the past, continues do do so, and now, they have eclipsed Voodoo and Matrox.
The GeForce ran at the highest percentages of its windows performace, and had FPS's like: 59.8 fps @ 1024x768x32 (HQ).
In short, great job, NVidia, and keep up the good work.
ALSO: Someone needs to let [The] Carmack get his hands on ALL of the code for these drivers. He, or MANY other people in the Open Source community good do these drivers a GREAT benefit. I await the day when someone cares MORE about Linux drivers than Windows drivers. I hope NVidia will be the first. I would like to see the percentages on that site be above 100%... currently, they are getting effectively that, with NVidia getting almost exactly the same score in Linux as in Windows (they could easily exceed it, seeing they are using the SAME code base).
NVidia: Great job, keep it up... and release the source!
part of this may be the Pavlovian Alt-F-S I smack every time I stop
to think</i><p>
You shouldn't have to live in such fear of data loss, should you?<p>
And about the power outage example - what's your point? An outside force (say, HW failure) will bring down any PC, regardless of OS.
Linux isn't a gammer platform.. Not news to me...
[This is "News for nerds, Stuff that matters" it's just not news to me..]
Windows has much of what games need. Plus it has the support from 3D card makers.
Windows is an afterthought.. and much of it's problems stem from that...
However for the PC graphics is an after thought... 3D graphics even more so...
and it's worse on Linux....
Windows knows how to handle all of this...
It's a driver patch system. So patching it for a new technology isn't a problem...
Linux is more of a hacker system. Upgrade, modify, hack, improve, anything..
and it's based on Unix...
Now mind you Unix dose not mean Command line... there are GUI Unix systems... NeXT and Explorer come to mine (Exporer NuBus later Mac NuBus... and now NeXT is MacOsX)
But Linux isn't one of those systems.. With Linux graphics is very much an after thought...
and Unlike Dos.. Linux dosn't get out of the way...
So whats the point of this artical?
To scare away Gammers from using Linux?
Hard core gammers arn't likely to use Linux to start with... No games.. no support and it's not well suted to the job.
Why the demand for games to be ported to Linux then?
Simple.. more and more games are multiplayer. That means game servers... Linux might not make a good game client... but Windows absolutly stinks as a game server...
You want a Linux admin to run a game server when the client is Windows only? Let me laff in your face...
Linux admin do buy the best hardware etc etc etc just like Windows game fanatics.
They'll buy the best games becouse they get bord. They may not play as fanaticly as Windows gammers. But then if they don't play... they won't run the servers.. if they don't... who will....
Thats why game develupers should make games for Linux. Not becouse Linux is a better game platform. But becouse it's a better game server...
I don't actually exist.
When this happends on Q3 I Ctrl alt - and then a Ctrl Alt +
The video (wigged out) Linux is ok... the drivers just never reset the card....
I think this isn't purely a Linux issue... My card is 3DFX Voodoo 3 for the PCI...
But mine is a normal PCI slot not an extended meta slot... and the Linux driver may be assuming full slot and operating accordingly...
Anyway the video can be fixed... just drop the resolution down and then back up.. it resets the card and unscrambles the screen...
I don't actually exist.
Thanks much for the info, i'll give it a shot when I get home from work.
BTW, The only time Q3F crashes is because I have a horrible (56k) net connection and when things get too laggy Q3 has some trouble.
I'm quite aware that you couldn't do hardware accelerated rendering over the network, but a software rendered game of Quake II should still be possible, right?
I tested QuakeII years ago with a Voodoo I, it ran consistently faster under Linux than it did under Windows 95, by only a few FPS.
I have seen benchmarks of Matrox G400's with the GLX driver running 30% faster than the Windows 9x drivers. That was when it was just a labour of love by users and John Carmack.
Some idiots will see these benchmarks and forever think Linux is slow because the drivers being compared are incomparable.
War crimes, torture, lies, illegal spying... Would someone give Bush a blowjob, already, so he can be impeached?
Preach it bruvvah! I can't tell you how many times I've gone on service calls to re-map a drive letter (no, really!). And when I ask the user what the server and share name were, "Um... O:?" Argh! My current office can all repeat my mantra, "Drive letters are evil!"
Little Debian: America's #1 Snack Distro!
Are there reasons why one would expect slightly better 3D performance under Windows than Linux, given drivers that are about equally optimized on each system? I'm interested to know if we should expect to match or even beat Windows 3D performance, or if 95% is about the best that can be done. If 95% is the best we can do, that's quite alright so long as we have good reasons (e.g. solid design, stability). I've always believed it's good to sacrifice a bit of performance to keep things manageable for software people, and stable for end-users.
Why does linux have to 'decide'? Linux just does it's own thing. The fact that one camp wants it on the desktop, and another on the server doesn't mean it can't do both. Sheesh.
linux wasn't invented to take on windows. Linux was invented Just Because.
There is a type of FUD called "exaggerating differences", where a small % advantage for one platform or another is turned into a big deal. Let's not engage in FUD here.
-E
Send mail here if you want to reach me.
There excist multithreaded graphics drivers for most of the highend 3d cards.
http://www.3dlabs.com bought Oxygen not long ago that were one of the
pioneers in multithreaded graphics drivers. These graphics card had up 8 GPU
( long before nvidida ) which work against all the
CPU's in your main system
"One World, One Web, One Program" - Microsoft Promotional Ad
I know not what course others may take; but as for me, give me liberty or give me death!
I think that the issue here may go beyond simply driver optimizations (except in the case of 3dfx). Windows is still somewhat based on DOS, which never was too big on multitasking. Games completely take over your system, when they run. Linux, being a true multitasking system, does not allow the game to have a completely free reign over system resources, which of course makes it more stable, but also imposes a mild performance hit. Windows, on the other hand, lets the game do whatever it wants with resources, leading of course to inherent instabilities we are all familiar with.
Thank you Bill Gates, for giving me an extra 5 fps, packaged with the extra 5 BSODs per day.
WARNING: there is a trojan on your
when the GUI stuff isn't mixed into the kernel ...Linux users can have their cake and eat it too.
At the cost of some performance.
Trees can't go dancing
So do them a big favor
Pretend dancing stinks!
Roughly speaking, anything better than human persistence of vision is going to look nice and smooth (17-25fps, as a guide).
However, the major confounding factor is that the FPS isn't constant. My machine here will do a comfortable 40-50 fps in quake when there's not much going on, but it will still drop to 10-12 on a busy level with a rocket fight going on.
And when a rocket fight is going on is the last time you want the frame rate to drop!
Ideally, you want a figure which keeps the fps above 30 *all the time*. Typical estimates are that you want a bit over 60fps in a timedemo to assure this.
Jules
-- Any sufficiently advanced technology is indistinguishable from a perl script.
The conclusion of this article was that performance under Linux is "comparable" - not good enough. When people are prepared to overclock systems or spend hundreds on a video card alone to get the most out of their gaming, 88-99% as fast as windows just isn't an option. Linux has to get BETTER than windows or forget it - the gamer market needs a compelling reason to switch, and for those guys the only compelling reason is MORE FPS. In fact Linux has a disadvantage to overcome - most games are not even available. Can't wait for Terminus!
"Does Windows performance really blow away Linux performance?" So far.
For all those win2k bigots out there who claims 'win2k is so much faster then win98' please go do some research at http://www.tomshardware.com. You will find that Win2k lags behind win98 by about 10%. Thank you, drive through.
-- You can be a geeklord too
It will be secure until the latest release of redhat shows up and then a dozen remote root exploits. To tell the truth for a desktop I don't really have a use for the multi user stuff. Maybe ppp for a dialup, but certainly not a ftp/www server. I don't need user security if I'm the only one who uses it.
Only the State obtains its revenue by coercion. - Murray Rothbard
It's not suprising that Windows outperforms Linux in the area of graphics acceleration. With all the $$ that Microsoft has it will be the best at whatever it wants to be. Recall what happened to Netscape because Microsoft felt that it needed to enter the brower market. If windows went open source tomorrow Linux would be dead the day after. The fact that Linux is open source and free is what is keeping it alive.
obviously you've never had the displeasure of using the Active Desktop. the bastards at Microsoft who forced that on us should be shot in their fucking heads.
-- You see, there would be these conclusions that you could jump to
why 33 fps isn't enough? even on the voodoo3 at 1024x768, linux still pulls 33fps. that's better than tv or film. The last game i played with any level of enthusiasm was doom II, and that ran at 320x240, probably 25 fps. i never had any complaints with that.
What's wrong with 1024x768 at 33fps? Is it even distinguishable from 60fps?
confused,
mark
In the article, they complained about the nVidia drivers being unstable, and they suggested that this bug would be fixed if the drivers were open. But then they said that they were not using the latest drivers due to "issues" with Q3A.
I am using the latest drivers. There are no "issues". They work perfectly. What's more, the instability they mentioned WAS FIXED (mostly) in the newer drivers. Supposedly, there is still an occasional crash, but I have tried to make it happen with no success.
------
Thats a problem with your software, buddy...not the OS
Um, because memory leaking is when the operating system fails to reclaim memory that has been marked as reusable. Thus, stopping netscape will not solve the problem
I find that hard to believe - unless netscape is allocating shared memory.
This sentence has exactly zero relevance. The corporate purchasers that purchase the vast majority of NT licenses simply don't care about technical issues such as stability. And it's not like there are many choices out there...Microsoft has a monopoly on desktop operating systems for PCs.
It was relevant with regards to the millions of disatisfied customers statement. If you say sell 100 billion copies of NT (yeah i know that's grossly inflated) and you have 1 million people who are disatisfied, it's excellent. And what's more if it's software it's even more excellent, since software engineering is still very bad.
As Linux grows because of its capabilities in other areas and its openness, it will gain market share, and the disparity between the two will decrease. Just give it time
---
This sig has been temporarily disconnected or is no longer in service
SE stands for Special Edition, fool
There is nothing preventing companies from writing commercial software for Linux, and keeping the source closed. There are, however, excellent multimedia libraries that are open source and would make the porting or writing job easier.
Proof: Look at Loki's popular games, released under closed-source. At the same time, however, Loki is giving back by making their library (SDL) LGPL'ed.
The more interesting question, IMHO, is the reason behind the good performace of the Matrox G400 and Nvidia, compared to 3dfx. The Nvidia card's performance is due to vendor tweaking, hence the close performance measurements. The Matrox G400, I think, is due to the wide popularity of the card, detailed vendor specs, and - dar I say - the wide use by the XFree86 team. 3dfx had a greater discrepancy becuase it is an older card and developers have lost interest, perhaps?
But of course, just how much can you trust the benchmarks? They ran it on one game, using a particular configuration, for a specific kernel. How can one generalize the results? I just have to wonder ...
A general pattern in those tests is that Linux is slower than Windows. We all know that Windows 95 (98 or Millenium) has little protection: most applications can tweak global system properties, whereas Linux puts severe restrictions on what applications can do (a bit less so when run as root).
I would like to know whether the performance is hit because of some of the protection stuff, for instance for transfers to video RAM (even with mmap, it's not necessarily easy on Linux because the system has to work a lot getting DMA to work around virtual memory). If there are knowledgeable people around here...
(Yes, that's a good question: how does Linux handle DMA transfers from userland to PCI busmaster devices? Temporary buffers?)
This is what about 95% of the business world would consider real work, and, hell, I could do that on an old PowerMac for all I care!
You mean like 3d animation? CAD stuff? Most variety for that is still on Windows... Including the non-mainstream stuff I consider best.
Well, I guess Linux serves and crunches data the best of all the OS's (it better damn well, since that's what Unix was written for)...
The point of the article is trying to help Linux go from a robust to a robustly *useable* OS where you could do anything your nasty little heart pleases without having to write it youself.