Slashdot Mirror


Nouveau NVIDIA Driver To Enter Linux 2.6.33 Kernel

An anonymous reader writes "Not only is DRBD to be included in the Linux 2.6.33 kernel, but so is the Nouveau driver. The Nouveau driver is the free software driver that was created by clean-room reverse engineering NVIDIA's binary Linux driver. It has been in development for several years with 2D, 3D, and video support. The DRM component is set to enter the Linux 2.6.33 kernel as a staging driver. This is coming as a surprise move after yesterday Linus began ranting over Red Hat not upstreaming Nouveau and then Red Hat attributing this delay to microcode issues. The microcode issue is temporarily worked around by removing it from the driver itself and using the kernel's firmware loader to insert this potentially copyrighted work instead."

289 comments

  1. I'm not an Avid Linux User... by Monkeedude1212 · · Score: 0

    But does this mean I could run a Video game?

    Does this mean the "But does it run on Linux" Jokes will come to an end?

    1. Re:I'm not an Avid Linux User... by Game_Ender · · Score: 5, Informative

      No it means that linux will ship with an open source alternative to the closed source Nvidia drivers.

    2. Re:I'm not an Avid Linux User... by Anonymous Coward · · Score: 1, Insightful

      And it will suck major ass in comparison to nvidia's.

    3. Re:I'm not an Avid Linux User... by Anonymous Coward · · Score: 2, Interesting

      For 3D. For 2D it's already better. Good 2D is underappreciated, but matters most for a lot of stuff that people casually use computers for.

      Obviously, gamers care about 3D, but good 2D matters also more than you might know for gamers into 80s/90s emulation - it's quite disappointing that even today, emulators sometimes fail to reliably vsync, really doesn't recreate the classic experience of amiga or snes gaming if frame rates aren't a rock solid tear-free 60Hz (or 50Hz depending on territory).

    4. Re:I'm not an Avid Linux User... by pantherace · · Score: 1

      It means that there will be an open source Nvidia driver that isn't (frankly) crap.

      As it is, you can run a number of games on Linux, there are some native ones (sadly not that many). However, many games will run on Wine. Wine has gotten pretty good.

      I was recently comparing 32-bit vs 64-bit, and using blender as an example on both Kubuntu 9.10 and Windows 7. So I decided to try the Windows version on Wine/Kubuntu. The win64 version would not run, but the win32 version did, and was faster under wine than running on Windows 7. (Not by a lot, about 2%, but it is still impressive that Wine does some win32 faster than Windows 7. Also, amusingly some of the runs of Wine/Blender are faster than Native Linux/Blender ~1%)

      (2.49b->2.50-alpha-1 is about 60% of the render time, 32->64 bit is about 60-70% on Kubuntu 9.10 and 78-85% on Windows 7, total from 2.49b 32->64bit is about 37% on Linux and about 49% for Windows, comparing 2.50-alpha-1 64-bit (native) Kubuntu takes about 75% of the time Windows 7 does.)

    5. Re:I'm not an Avid Linux User... by Anonymous Coward · · Score: 0

      Linux already has a closed source driver from Nvidia that works pretty damn well. The problem has never been in the drivers, it's in the games being compiled for Windows and Windows-specific libraries.

      That's why cool people compile Linux versions of their games. World of Goo and the Penny Arcade game both have Linux versions IIRC.

    6. Re:I'm not an Avid Linux User... by JeffSpudrinski · · Score: 1, Funny

      Now if we could just get a stable nVidia driver in Windows, we'll be set.

      -JJS

    7. Re:I'm not an Avid Linux User... by LOLLinux · · Score: 3, Insightful

      The people who are using nvidia's driver obviously care about 3d performance otherwise they'd already be using the open source driver with 2d support. Also, I doubt nouveau has the hardware accelerated playback of mpeg-2, vc-1 and h.264 like the closed drivers.

    8. Re:I'm not an Avid Linux User... by LOLLinux · · Score: 2, Insightful

      or 2D it's already better.

      Do you have any benchmarks that aren't 2+ years old and were comparing nouveau to the ancient NV drivers?

    9. Re:I'm not an Avid Linux User... by LOLLinux · · Score: 0

      Hmm, looks like I was right. You probably googled and saw those ancient comparisons from 2007. Here is a modern day comparison showing that the binary driver rapes nouveau in 2d performance.

      In terms of performance, however, it’s clear that nouveau has a ways to go, at least on my hardware. I used glxgears (yes, it’s not a good tool for benchmarking overall video performance, but it’s a useful basis for standardized comparison of FPS rates under different video drivers) to measure video frames per second under the nouveau, nv and nvidia (closed-source) drivers. The tests were run with desktop effects turned off and the system idling. The average FPS rates were as follows:
      nouveau: 355.3
      nv: 475.8
      nvidia (version 96, from Ubuntu repositories): 1993

    10. Re:I'm not an Avid Linux User... by V!NCENT · · Score: 1

      *sigh*

      The DRM module we're talking about is the Direct Rendering Manager. That means bypassing userspace layers and directly talking to the kernel which then talks to the graphics card.
      You could run hardware accelerated graphics on Linux for ages and ages and ages ago. What stone have you been living under?

      I was playing Quake 3 since the Pentium3 days, with hardware OpenGL acceleration on Linux.

      Maybe you should watch a video in which somebody plays Crysis on Linux? -> http://www.youtube.com/watch?v=fZ147bcoLi0

      Fscktard...

      --
      Here be signatures
    11. Re:I'm not an Avid Linux User... by Anonymous Coward · · Score: 1, Insightful

      What the hell does FPS (in all cases well above typical 50-100Hz display rates anyway) of a 3D program (glxgears) have to do with jitter and tear free 2D? Not a whole lot, that's what.

    12. Re:I'm not an Avid Linux User... by Anonymous Coward · · Score: 1, Insightful

      Being open source does not make it automatically not crap. I suspect it will always be far more unstable for 3D applications. As it is now it can't even handle Blender without crashing, and that's an immediate-mode app that isn't exactly using esoteric extensions.

    13. Re:I'm not an Avid Linux User... by ozmanjusri · · Score: 1

      AS a regular Blender user, and one who's been impressed by the stability, I'm astonished you've been able to crash it. Do you have any logs/scenes you can provide to show how you managed that?

      --
      "I've got more toys than Teruhisa Kitahara."
    14. Re:I'm not an Avid Linux User... by John+Whitley · · Score: 1

      The people who are using nvidia's driver obviously care about 3d performance otherwise they'd already be using the open source driver with 2d support.

      Not so much. I have an Ubuntu desktop box with an NVidia card (work system; I didn't spec it), and the open source driver seemed to have a number of issues when running dual head. I flipped to the binary driver, it just worked, and haven't looked back. It sounds like the Nouveau driver will bring some substantial forward progress to open source NVidia drivers. ( No thanks to NVidia themselves. :-P )

    15. Re:I'm not an Avid Linux User... by RMingin · · Score: 2, Insightful

      Ok, I may be dating myself here, but way back in 2000 we were telling people in the Nvidia support channel that GLXGEARS IS NOT A BENCHMARK. Nvidia spent a LOT of time over the years putting absurd amounts of time into over-optimizing the performance of that small snippet of code, and a few others, simply because they noticed that certain clueless noobs were using it as a performance metric.

      The ONLY purpose of glxgears as it was designed is to indicate if software or hardware acceleration of 3D was happening. It's now useless for that even, since CPUs have become fast enough to crank out absurd FPS numbers as well.

      --
      The preceding comment is my own, and in no way construes an opinon of the Emperor of Mankind.
    16. Re:I'm not an Avid Linux User... by Anonymous Coward · · Score: 0

      Now if we could just get a stable ATI driver in Windows, we'll be set.

      FTFY.

      To be fair, I hear it's changed, but driver support is why I ditched ATI for nVidia so very long ago, and haven't had a driver issue since.

    17. Re:I'm not an Avid Linux User... by Anonymous Coward · · Score: 0

      Yes, good find. However, one thing you forgot to mention is that glxgears doesn't fucking test 2D performance.

    18. Re:I'm not an Avid Linux User... by h4rr4r · · Score: 1

      No that is what wine has meant for ages. Heck, crossover(the pay for version of wine) added l4d2 support the night the game came out.

    19. Re:I'm not an Avid Linux User... by Anonymous Coward · · Score: 0

      Here is a modern day comparison showing that the binary driver rapes nouveau in 2d performance.

      I used glxgears (yes, it's not a good tool for benchmarking overall video performance, but it's a useful basis for standardized comparison of FPS rates under different video drivers) to measure video frames per second under the nouveau, nv and nvidia (closed-source) drivers.

      (emphasis mine) Isn't glxgears a 3d "benchmark" and not a test of 2d performance?

      http://wiki.cchtml.com/index.php/Glxgears_is_not_a_Benchmark

    20. Re:I'm not an Avid Linux User... by Anonymous Coward · · Score: 0

      ... or 3D performance. It used to be at least suitable as a test to see if 3D accel was basically working, but even modern CPU rendering gives glxgears FPS like accelerators of a few years ago.

      People were saying "glxgears is not a benchmark" back when I had a voodoo 3, never mind a hd4870

    21. Re:I'm not an Avid Linux User... by hairyfeet · · Score: 1

      I switched from Nvidia to ATI after the bad solder mess, and frankly I don't know what everyone's complaints about ATI drivers in Windows are. I've had them run much MORE stable than the Nvidia drivers did for me, and that is even after adding an ATI USB TV Tuner to the mix. Frankly it shocked the hell out of me that TV Tuner drivers, which are the most flaky things I've ever had to deal with, not only didn't crash themselves but didn't mess up anything with my ATI 4650.

      So what exactly is supposed to be wrong with the ATI drivers? Because I have 3 ATI cards now, from an old X1950 to my 4650, on everything from an old single core XP box to a new quad running Windows 7 HP x64, and even after running games and other GPU intensive apps for days without a reboot I haven't seen so much as a single hiccup. Surely I couldn't have gotten THAT lucky that three different boxes with different OSes and ATI hardware all "just work", could I?

      --
      ACs don't waste your time replying, your posts are never seen by me.
    22. Re:I'm not an Avid Linux User... by JeffSpudrinski · · Score: 1

      Actually it has changed quite a bit for the better in the past few years.
      I just couldn't resist the obvious "cheap shot". I manage a small / mid size network with about 100 users. I swear by nVidia cards. We use serveral different 3D design packages that push the hardware to the limit (and beyond sometimes). There truly are less real "crashes" now. There are still a number of glitches, but those can just as easily be attributed to the application itself as to the video driver.
      -JJS

    23. Re:I'm not an Avid Linux User... by bencoder · · Score: 1

      He didn't say he crashed blender. His graphics card/driver crashed when running blender. Very different things.

    24. Re:I'm not an Avid Linux User... by rikkitikki · · Score: 1

      As far as casual 2D desktop usage is concerned, ATI is probably fine. The problems, for me, come when needing hardware OpenGL drivers. ATI has been working on it and seems to possibly be improving, but in years past, ATI's support for OpenGL has been a crap shoot at best. First, was whether OpenGL even worked on whatever ATI card you put in. Next, was what bugs to be expecting. Basic things like gl_FragCoord (in fragment shaders) were completely broken...for a long time. And there were several other bugs a developer had to deal with. As an OpenGL developer, you pretty much had the following expectations:

      Nvidia driver: OpenGL worked as spec'd. Very few if any bugs.
      ATI driver: OpenGL was a crap shoot. Sometimes worked well enough. Sometimes had a ton of bugs that needed working around.
      Intel driver: Revert to _software_ OpenGL rendering because Intel's drivers were so completely useless.

      If a new extension comes out, or a new version of OpenGL comes out:

      Nvidia: expect a new driver within weeks to support it.
      ATI: expect support somewhere 6 months to a year later.
      Intel: OpenGL? what's that?

      ATI seems to be working to improve their situation, but they're still not quite to the level of OpenGL support that NVidia has provided.

    25. Re:I'm not an Avid Linux User... by Anonymous Coward · · Score: 0

      But GLXGears is an OpenGL (ie 3D) application?!

    26. Re:I'm not an Avid Linux User... by Anonymous Coward · · Score: 1, Interesting

      Nvidia's drivers arn't even that good at general 3D. Actually, they suck for anything other than popular FPS games.

      About vsync: Nvidia cards are capable of outputing PAL and NTSC RGB signals on the VGA ports, which gives you an excellent picture quality on old CRT TVs. This is useful for emulators and watching SDTV the way it's supposed to look. But unfortunately the implementation of interlacing and vsync is wrong and broken in Nvidia's driver, which means video fields get played in an effectively random order - and there is nothing you can do about it. The nv driver doesn't support interlace at all and the developers refuse to implement it. Nouveau supports interlace in these modes, and it would actually be possible to fix the vsync bug here.

    27. Re:I'm not an Avid Linux User... by hairyfeet · · Score: 2, Interesting

      But I thought AMD/ATI opened up their specs? I know that here on /. we hear all the time "if they would only open up their specs we would take care of the rest". So shouldn't the ATI drivers for Linux be much better now? I know the Nvidia driver isn't "free as in freedom" and if AMD/ATI have fully opened up their specs shouldn't ATI be having the better experience in Linux now?

      --
      ACs don't waste your time replying, your posts are never seen by me.
    28. Re:I'm not an Avid Linux User... by pnewhook · · Score: 1

      OpenGL support under ATI is pitiful. If you use OpenGL you are almost forced to use NVidia

      --
      Tesla was a genius. Edison however was a overrated hack who liked to torture puppies.
    29. Re:I'm not an Avid Linux User... by Anonymous Coward · · Score: 0

      That's not even close to true. NVIDIA is the first to tell you that glxgears is not a benchmark.

    30. Re:I'm not an Avid Linux User... by mikechant · · Score: 1

      But I thought AMD/ATI opened up their specs? I know that here on /. we hear all the time "if they would only open up their specs we would take care of the rest". So shouldn't the ATI drivers for Linux be much better now?

      The full specs were provided relatively recently and this is a highly complex project which will probably take another year or two to produce reasonably mature drivers.

    31. Re:I'm not an Avid Linux User... by DJCater · · Score: 1

      What is a good benchmark?

      --
      Sig Appended to the end of comments you post. 120 chars.
    32. Re:I'm not an Avid Linux User... by hairyfeet · · Score: 1

      That brings up a good point: If the hardware is complex enough does releasing full specs help at all? Because it seems to me if it is gonna take them a seriously long time to get them up to snuff, and the speed at which new GPUs come out, it would look like in the end all they would be supporting is old crap nobody can buy at retail anymore anyway.

      And that is another thing that strikes me as odd and backwards about the Linux way of "release and we'll get it into the mainline and support it in the kernel" way of doing things...if I am a consumer hardware manufacturer the Linux way actually helps me not one damned bit. What I mean by that is the average shelf life of consumer grade hardware is 6 months to a year, but it would take so long for the change in the kernel to "trickle down" into the distros that by the time they have the driver for my device foo I'm not even selling it anymore. So what would be the point of me even releasing the code?

      Which is why I think that ultimately to get traction Linux needs to shake the kernel devs until they agree on a stable driver ABI. Because if I release a device with Windows drivers on the CD not only can I sell the device to Windows users RIGHT NOW, but I can also sell that device for years if I so choose, because it is such a long time between Windows Driver Model changes. I mean .VXD lasted for a decade, The Win2K/XP driver model closer to a decade and a half by the time XP leaves support, and Windows 7 is a hit and from the looks of it they'll be sticking with the Vista/Win7 driver model, probably for another decade. That means just FOUR drivers and I can support every Windows user from 2001-2014, no need to rewrite, no need to keep developers writing drivers for old devices, etc.

      So while the Linux way may benefit those like RMS that love to and actually have the skills to hack, for me the device manufacturer or Joe and Jane consumer the current model doesn't really work. You can't get the driver trickled down into the distros in time for it to help the manufacturer before the device is already considered old. Joe and Jane have to play paperweight roulette because they don't have drivers on the CD nor a penguin on the box because the distros are in a constant state of change and any binary driver I release today might not work when they try it in the next Distro release. And raw code don't help Joe and Jane who certainly don't have the skills to compile and tweak source code. So surely something will have to change, or else Linux just won't grow. Because if I can't even shop for new devices for your OS without playing paperweight roulette or doing studying, how likely is it I'll buy your OS? And if releasing specs doesn't get me more customers NOW, as a manufacturer why would I bother?

      --
      ACs don't waste your time replying, your posts are never seen by me.
    33. Re:I'm not an Avid Linux User... by RMingin · · Score: 1

      They do say that. Since you insist on using glxgears as a performance metric, they continue to invest manpower into inflating that score.

      You're not disagreeing with me.

      --
      The preceding comment is my own, and in no way construes an opinon of the Emperor of Mankind.
    34. Re:I'm not an Avid Linux User... by RMingin · · Score: 1

      It's different for everyone but generally the best benchmarks are the ones that relate to work you want done. For 'pro' users, that might be SpecViewPerf. For gamers, it's almost certainly a game. Very few people are looking to interact with glxgears on an ongoing basis, and it's not related to any real workloads I can picture.

      Since you're not running glxgears as your primary workload, don't use glxgears as a benchmark.

      --
      The preceding comment is my own, and in no way construes an opinon of the Emperor of Mankind.
    35. Re:I'm not an Avid Linux User... by 3.1415926535 · · Score: 1

      The nv driver doesn't support interlace at all and the developers refuse to implement it.

      Interlace support for G8x and up:
      http://cgit.freedesktop.org/xorg/driver/xf86-video-nv/tree/src/g80_display.c#n434
      Interlace support for pre-G80:
      http://cgit.freedesktop.org/xorg/driver/xf86-video-nv/tree/src/nv_dac.c#n169

    36. Re:I'm not an Avid Linux User... by TheTyrannyOfForcedRe · · Score: 1

      I've been using the open source ATI drivers on an RV635 for many months. They're awesome. Zero stability issues. They're 2d only right now but 3d support will be released around March.

      --
      "Liechtenstein is the world's largest producer of sausage casings, potassium storage units, and false teeth."
    37. Re:I'm not an Avid Linux User... by Anonymous Coward · · Score: 0

      One presumes that, after the drivers support 2D/3D in some graphics drivers, that things only need to be tweeked rather then re-written for new hardware, and new features added rather then re-writing a driver around them.

      I expect it to take something like 1/2 years to get the base driver down, which fully supports some less-then-recent graphics cards. Then, newer cards will be much easier to add support for.

    38. Re:I'm not an Avid Linux User... by badkarmadayaccount · · Score: 1

      Start work on specs before you hit the market? Seems reasonable.

      --
      I know tobacco is bad for you, so I smoke weed with crack.
  2. Why not just use the Windows driver model.. by Anonymous Coward · · Score: 0, Flamebait

    .. and directly load those superior and polished drivers?

    1. Re:Why not just use the Windows driver model.. by Wonko+the+Sane · · Score: 4, Insightful

      Because Nouveau works on a more architectures than Windows has ever been ported to.

    2. Re:Why not just use the Windows driver model.. by iammani · · Score: 1

      Can be done, at the cost of performance.

    3. Re:Why not just use the Windows driver model.. by Anonymous Coward · · Score: 0

      Patch please

    4. Re:Why not just use the Windows driver model.. by rumith · · Score: 4, Insightful

      Now if NVidia cards would work on those architectures, too :-)

    5. Re:Why not just use the Windows driver model.. by Wonko+the+Sane · · Score: 1

      Not all architectures have PCIe, but some people still have PCI video cards that they'd like to use.

    6. Re:Why not just use the Windows driver model.. by Anonymous Coward · · Score: 3, Insightful

      You clearly know nothing about writing drivers, let alone video drivers, for Windows.

      Windows video drivers do not "generally include kernel components". That's complete bullshit. The driver itself can be considered a "kernel component". But otherwise, any Windows graphics driver just implements a certain well-defined interface, and only calls certain well-defined kernel functions.

      There is nothing technical stopping the Linux kernel, the FreeBSD kernel, the Solaris kernel, the Mac OS X kernel, and whatever other popular x86 or x64 operating system you use from implementing similar interfaces. Many of the functions would just need to be stubs that do nothing.

      And "virtualization" does not mean what you think it means. There is absolutely no need to modify the machine code of the graphics drivers, let alone fake the hardware underneath it using software.

      Any restrictions would likely be imposed by licensing and ideology, rather than any technical obstacles.

      Please refrain from spouting out your pure bullshit in the future. Or at least try to write a Windows graphics driver before you pretend to know anything about them.

    7. Re:Why not just use the Windows driver model.. by Anonymous Coward · · Score: 0

      Bravo, sir. Bravo.

    8. Re:Why not just use the Windows driver model.. by TheNetAvenger · · Score: 1

      Because Nouveau works on a more architectures than Windows has ever been ported to.

      Technically not true, especially in the context of any platform that has NVidia hardware.

    9. Re:Why not just use the Windows driver model.. by fat_mike · · Score: 1

      If you seriously don't have the money to even buy an AGP Nvidia card then you've got bigger problems.

    10. Re:Why not just use the Windows driver model.. by Wonko+the+Sane · · Score: 1

      It's not just a matter of money. I might want to use a reasonably modern graphics card on an older workstation that doesn't have AGP. There's plenty of hardware out there that works just fine even though you can't install Windows 7 on it.

    11. Re:Why not just use the Windows driver model.. by Wonko+the+Sane · · Score: 1

      Close enough. I could install a PCI version of an Nvidia card on an Alpha workstation. While it's true that there was a version of Windows ported to Alpha architecture that version of Windows isn't sold anymore and Nvidia doesn't make drivers for it.

    12. Re:Why not just use the Windows driver model.. by Anonymous Coward · · Score: 0

      Or as good.

    13. Re:Why not just use the Windows driver model.. by socrplayr813 · · Score: 1

      Quite right. I'm currently tossing around the idea of building a small server/MythTV box out of barebones Atom or Nano system. Very few or none of these have decent video integrated. The ones that do have acceptable video don't have the outputs I'm looking for. They also don't have AGP or PCIe. Solid 2d support out of the box for a PCI Nvidia card would a definite bonus for this build.

      Similarly, I have an ancient Dell that is currently being used as a headless server. If/when I get the new one running, I'll likely want to put a graphics card in that for use as a MythTV frontend. That will have to be a PCI card as well.

      --
      The confidence of ignorance will always overcome the indecision of knowledge.
    14. Re:Why not just use the Windows driver model.. by sznupi · · Score: 1

      Where have you been? PCI video cards were being released long after AGP died. And were quite expensive then (and just before, when coexisting with AGP)

      --
      One that hath name thou can not otter
    15. Re:Why not just use the Windows driver model.. by Anonymous Coward · · Score: 0

      My PowerMac G5 disagrees with you.

    16. Re:Why not just use the Windows driver model.. by omnichad · · Score: 1

      And that gets the driver working in maybe x86 and x86_64. Linux supports a whole lot of architectures that wouldn't benefit at all from this.

    17. Re:Why not just use the Windows driver model.. by Anonymous Coward · · Score: 0

      thank god x86 and x64 are the only relevant ones for 3d gaming market (and a lot of other things)

    18. Re:Why not just use the Windows driver model.. by Corporate+Troll · · Score: 1
      I have a Point of view Atom 330 motherboard with 4Gig RAM (back then they claimed 4Gig supported since it's a 64-bit CPU, but nooooooo!). The graphic chip is a NVidia 9400M as I've been told (and so this is ontopic!) It has been heralded as the saviour for the dog-slow Atom Chip. Well, it is not... The thing is dog-slow running Ubuntu 9.10 with the NVidia proprietary drivers. I didn't do anything scientific, but the i386 Ubuntu version seems to run faster than the x86_64 code.

      I have it now replaced with a AMD Athlon 2600+ with 1GB RAM..... Way faster. The money I put in that Atom was a pure waste, if a 6++ year old computer beats it.

    19. Re:Why not just use the Windows driver model.. by socrplayr813 · · Score: 1

      I appreciate the input, but I still think it's the way I want to go.

      The Atom is a low-power solution for embedded machines and things that don't need speed. A fileserver/MythTV box doesn't need a lot of processor power, since almost all of the work will be done by the tuner card and video card (except for commercial flagging, which doesn't need to be fast). Since I'll be leaving it on all the time, I want to save some electricity.

      --
      The confidence of ignorance will always overcome the indecision of knowledge.
  3. How does it compare with the other NVidia drivers? by Lord+Lode · · Score: 4, Interesting

    I'm a Linux user using the official binary NVidia drivers, they work good - very good even, many modern Windows games work in Wine without any performance loss.

    How do the Nouveau Nvidia drivers compare to the official ones? Do they have the same performance, no little annoying bugs or differences, etc...?

  4. What card to buy today? by Just+Some+Guy · · Score: 3, Interesting

    My Dell at work has an ATI RV635 card. You know: the one that might, someday, support 3D but hasn't yet in the couple of years it's been out? I switched from Ubuntu Karmic to Fedora Core 12 a couple of weeks ago to see if the experimental drivers worked, but ended up with a non-working X.

    If I want to buy a card that has working accelerated 3D today - not next week, not "maybe if I download a hack from North Korea that might work or might catch fire" - so I can do basic stuff like get smooth compositing in KDE, what should I get? Again, this is going into my computer at work, so $500 gaming cards are right out. I'm positive I can get the hardware guy to order a reasonably priced card for me (and another for himself) if it'll work on Linux, though.

    BTW, let me preemptively say that I'm not gonna Google it. There are 5,000,000 outdated and spurious reports. I'd much rather discuss it with a group of peers than try to decode what some kid in Sri Lanka came up with.

    --
    Dewey, what part of this looks like authorities should be involved?
    1. Re:What card to buy today? by John+Hasler · · Score: 1

      > I'd much rather discuss it with a group of peers than try to decode what
      > some kid in Sri Lanka came up with.

      Then why are you asking on Slashdot?

      --
      Warning: this article may contain humor, sarcasm, parody, and perhaps even irony. Read at your own risk.
    2. Re:What card to buy today? by Sir_Lewk · · Score: 4, Insightful

      If you need/want to use a Free driver, get an older ATI card. I have a card in the R500 series and the Free 'radeon' driver works wonderfully for what I ask from it (urban terror and mplayer). Anything up to the R500's have good support atm, the R600/700 support is getting there...

      If you don't care about that, get an Nvidia card and use the non-Free driver. This option will also get you the best preformance.

      --
      "linux is just DOS with a UNIX like syntax" -- Galactic Dominator (944134)
    3. Re:What card to buy today? by Anonymous Coward · · Score: 0

      Pretty much anything from Nvidia. You can go to their web page and check for Linux drivers for chipset of your interest. ATi + Linux usually means problems.

    4. Re:What card to buy today? by gazbo · · Score: 3, Informative
      My work laptop has a GeForce 9600M GS (according to lspci) and once I installed the binary driver with a simple `yum install kmod-nvidia` it just worked. Dual screens with different resolutions set up fine with the nvidia utility (don't use the standard Linux display stuff) and performance on compositing is great. Only difference is I'm using Gnome not KDE.

      And I know fuck all about Linux, so it must work easily. I read nvidia cards worked well, and it certainly seemed to go smoother than the Radeon in my old laptop.

    5. Re:What card to buy today? by Just+Some+Guy · · Score: 2, Insightful

      "Peer" doesn't mean "infallible expert", or at least not among my peer group.

      --
      Dewey, what part of this looks like authorities should be involved?
    6. Re:What card to buy today? by Anonymous Coward · · Score: 0

      How sure are you that some kid in Sri Lanka wont reply to your message now?

      Go with nvidia.

    7. Re:What card to buy today? by diegocg · · Score: 2, Informative

      A motherboard with an integrated intel graphic card. They are not as fast as ATI/Nvidia, but they work great for things like desktop compositting, and the driver is the most complete and stable driver available in the FOSS world.

    8. Re:What card to buy today? by socrplayr813 · · Score: 2, Insightful

      If you're against closed drivers, all I will say is good luck.

      If you're okay with using proprietary drivers, any Nvidia card should work fine. If you don't need fancy games or similar, the run-of-the-mill $50 cards will be plenty.

      I know anecdotes are not evidence, but I haven't had any issues in the last 2 years or so getting Nvidia cards to work on my personal computers (three separate machines). My one ATI machine though, still barely manages 2d and crashes if I install the proprietary driver. I've heard and read many similar stories.

      --
      The confidence of ignorance will always overcome the indecision of knowledge.
    9. Re:What card to buy today? by Anonymous Coward · · Score: 0

      I would (and in fact, I did with my Thinkpad x200 and my recent workstation) buy an intel graphics card (or rather a mainboard with the integrated graphics chip, such as the GM45).

      They do have free drivers which work just fine, they don't use much power, they support KMS and they have 3D acceleration (not the best for gaming, but for your "smooth compositing in KDE" it should be enough).

    10. Re:What card to buy today? by Anonymous Coward · · Score: 0

      Using a 9500GT from NVidia which was plug&play, including Compiz support at full screen resolution on two screens. Apt-get install and you're off toggling Compiz settings.

    11. Re:What card to buy today? by LWATCDR · · Score: 1

      I have an nVidia card and use the nVidia closed drivers and have no real issues.
      If you are getting a new PC and FOSS purity is important to you then get Intel graphics and you will be good to go.
      I have heard good things about ATI but I have no real experience with them myself..

      --
      See my blog http://ilovecookes.blogspot.com/ for light hearted technical information.
    12. Re:What card to buy today? by Anonymous Coward · · Score: 0

      It's a shame that Intel don't make add-on cards, their integrated graphics, while not being killer performers, are very solid for 3d(compiz easily and openarena is playable at low res) in my netbook.

      I'd suggest just grabbing yourself an nvidia, something cheap like the 6200 would do for basic, sensible, work stuff. They're good cards and the binary drivers are pretty solid.

    13. Re:What card to buy today? by Anonymous Coward · · Score: 0

      I have been using almost 100% nVidia for the last 10 years just because they have always supported Linux the best. Any graphics card will an nVidia chipset will do although I prefer EVGA these days.

      I have tried ATI a few times over the years and it sucked every single time. Especially their OpenGL support which to this day continues to suck-ass.

    14. Re:What card to buy today? by endoftheroadmatt · · Score: 2, Informative

      If you're looking for a specific card, this is in my office Dell Optiplex workstation, I think it's an EVGA, it was ~$50 at the time.

      01:00.0 VGA compatible controller: nVidia Corporation GeForce 8600 GT (rev a1)

      Debian Lenny with nvidia debs from non-free, dual 19 inch DVI monitors

    15. Re:What card to buy today? by Just+Some+Guy · · Score: 1

      A motherboard with an integrated intel graphic card.

      And my IT department won't mind if I just pop in a new motherboard? :-)

      In truth, no one would care (as long as it worked afterward), but I'd much rather swap in a replacement graphics card.

      --
      Dewey, what part of this looks like authorities should be involved?
    16. Re:What card to buy today? by h4rr4r · · Score: 1

      An intel integrated should do the job fine. I have all the wizbang bullshit effects on my dell mini 9 and that has an intel GMA950. I turn them right off an use metacity though.

    17. Re:What card to buy today? by Anonymous Coward · · Score: 0

      Anything nVidia. ATI is dead to me. I have three heads running on KDE4/Fedora 12 at work (2 on GeForce 6200LE and 1 on GeForce 6150SE) and an older GeForce 7600GT in my desktop at home. All of them work great. This open nv driver seems like a waste of time to me, because it doesn't have support from nVidia and therefore is likely to never work 100% properly on 100% of the boards.

    18. Re:What card to buy today? by GameboyRMH · · Score: 1

      Seconded. In my experience onboard Intel cards have been the least troublesome by a long shot. Usually they work fine with no tweaking, 3D acceleration and all.

      --
      "When information is power, privacy is freedom" - Jah-Wren Ryel
    19. Re:What card to buy today? by bfields · · Score: 2, Informative

      For composited desktop with all the wobbly windows and such, tuxracer, watching DVD's, etc., the integrated intel chips are more than adequate and have great open-source support (everything except GMA 500 they use in netbooks).

      I'm told they aren't so great for the latest games.

    20. Re:What card to buy today? by Anonymous Coward · · Score: 0

      From personal experience:

      nVidia: fast, reliable and well supported but closed.
      Intel: slower but reliable and open, therefore even better supported. But beware that not all Intel chipsets have open source support, you better check first.

      Heard thru the grapevine:
      ATI, fast, open but buggy. Ask around to know how's their drivers status today. I can't tell about those cards since the only one I have is the built in one in my old PPC Mac Mini (Debian runs it fine though).

      To summarize, today if you need fast 3D the magic word is nVidia, while for anything else about every built in video chipset will do the work.

      My personal choice on two non gaming machines (one Linux, one Windows) was the nVidia 7300 (the LE model if I recall). Fast enough for high res video or non demanding stuff (to give a figure, Half Life 2 runs well at mid settings under WINE, don't even think about Crysis or similar games) but it's totally fanless, therefore 100% silent and ideal as an office/development machine.

    21. Re:What card to buy today? by boa · · Score: 2, Informative

      ATI Radeon 4890 with binary ATI driverse works well for me under Ubuntu 9.10.

    22. Re:What card to buy today? by SarahAnnAlien · · Score: 2, Interesting
      As one of the other posters said, buy an older R500 based ATI card. I'm trying this out myself; I ordered a card this week and (hopefully) will find out how well it works with the Radeon driver. But everything I've read suggests it will work pretty well.

      Read more here: http://en.wikipedia.org/wiki/R520#Variants

      I also ordered a (cheap) R600 (or R700? don't remember...) based card to experiment with or put on the shelf for a few months; people seem to be working quite hard on ironing the bugs out, and those will probably be well supported in the future.

      I ordered my R580+ based card from a place called compuvest. It was my first order with them so I can't say whether or not they're worth recommending. They seem to have a lot of generation-ago equipment. Another place to try would be geeks.com, which is where I got the R600 based card.

      I'm not sure how well this will work out, but that's what I'm trying...

    23. Re:What card to buy today? by mikechant · · Score: 1

      Nvidia 8300GS - cheap low end card, works perfectly with the binary driver (suspend/resume/compiz/fullscreen video/3d games).

    24. Re:What card to buy today? by True+Grit · · Score: 1

      If you're against closed drivers, all I will say is good luck.

      If the GP is against closed drivers, then they should do as the other poster suggested and get a slightly older R500 based AMD-ATI card, use the fglrx driver for now, then switch to AMD's open radeon driver when it goes mainstream in a year or so. Then they'll have all the hardware-accelerated goodness, without any of the binary-blob badness.

      I know anecdotes are not evidence

      So why bother?

  5. Re:How does it compare with the other NVidia drive by LOLLinux · · Score: 2, Informative

    How do the Nouveau Nvidia drivers compare to the official ones?

    Slower on every machine I've tested.

  6. Re:How does it compare with the other NVidia drive by Meshach · · Score: 1

    I'm curious about this too. I've used the official NVidia driver in Linux for many years and never had a problem. What was the compelling reason to reverse engineer?

    --
    "Maybe this world is another planet's hell"
    Aldous Huxley
  7. Re:How does it compare with the other NVidia drive by Idiot+with+a+gun · · Score: 5, Insightful

    I'll agree with you, they work good, when they work. The problem with the official drivers is that they're a binary blob, thus most distributions (none I've ever seen) ship with them enabled. This is an issue if the default nv driver crashes your machine. Because of this, I'm going with ATI next time, I've heard they're way more Linux friendly now.

  8. Re:How does it compare with the other NVidia drive by Sir_Lewk · · Score: 2, Interesting

    As I understand it (I have an ATI card, not an Nvidia), Nouveau currently has 2d hardware support, and 3d support is in progress. Don't expect it to replace the proprietary driver for anything requiring preformance anytime soon, but this is good news for people who dislike the proprietary drivers, and for distros that cannot/willnot ship with them by default.

    --
    "linux is just DOS with a UNIX like syntax" -- Galactic Dominator (944134)
  9. Re:How does it compare with the other NVidia drive by Monkeedude1212 · · Score: 0, Troll

    I don't know about you, but I've never paid for a driver in my life.

  10. Re:How does it compare with the other NVidia drive by binarylarry · · Score: 4, Informative

    The official closed source driver creates a proprietary dependency on an otherwise open OS kernel.

    This irks some free software hippies and it also makes using Nvidia hardware on unsupported hardware platforms more difficult.

    --
    Mod me down, my New Earth Global Warmingist friends!
  11. This is great - sort of by ClosedSource · · Score: 3, Interesting

    I've often wondered why more reverse engineering isn't done to create Linux drivers rather than just complaining about the manufacturer of the hardware. The only unfortunate thing about this project is that Linux drivers already exist (according to other posts here).

    Wouldn't it be better to reverse-engineer hardware to create Linux drivers that don't exist?

    1. Re:This is great - sort of by Corporate+Troll · · Score: 2, Informative

      Wouldn't it be better to reverse-engineer hardware to create Linux drivers that don't exist?

      That would be way more time-intensive and way harder....

    2. Re:This is great - sort of by Sir_Lewk · · Score: 4, Insightful

      Video drivers are generally considered the thing most lacking in linux. Last I heard/tried, USB wifi cards are a nightmare, but besides those, most high-profile hardware is pretty well supported. You'll always find the odd usb controlled nerf gun turret or whatnot that lacks a driver, but that's not really an issue for most people.

      Furthermore, it is in error to think that people reverse engineering video cards would otherwise spend their time reverse engineering other hardware. These people do not necessarily specialize in other sorts of hardware. In linux, more people working on A does not really mean less people work on B. It's not like there is a manager at the top assigning and moving people around from task to task.

      --
      "linux is just DOS with a UNIX like syntax" -- Galactic Dominator (944134)
    3. Re:This is great - sort of by david.given · · Score: 2, Informative

      I've often wondered why more reverse engineering isn't done to create Linux drivers rather than just complaining about the manufacturer of the hardware.

      Because these days it's really, really hard.

      A modern graphics card, for example, is actually a complete computer. It's got RAM, a processor, a bunch of peripherals, a complete miniature operating system... and you don't even know what type of processor it is. A lot of peripherals work like this; a wireless card is typically an ARM processor with some RAM attached on one end to the radio and on the other to an I/O controller that talks to the computer.

      So in order to reverse engineer a graphics card you both need to decompile your way through multiple megabytes of binary blob driver that runs on the PC, but also decompile your way through multiple megabytes of binary blob operating system image that runs on the card... and you don't even know what the CPU architecture is!

      It's actually more productive to hassle the manufacturers into releasing documentation. Which in some ways is a pity; it would be really cool to be able to run your own bare-metal OS on the GPU.

    4. Re:This is great - sort of by skine · · Score: 1

      Indeed.

      Also, given that >95% of the Linux community contributes no code (or so I've heard), it's not surprising that most people rely on others to get their hardware to work (and work well).

    5. Re:This is great - sort of by sproingie · · Score: 1

      Video cards run nothing like an OS on the hardware, and you won't be getting one running yourself. Most GPUs don't have branch instructions (newer ones kind of do, but only in the "skip forward" sense). It's kinda hard to run an OS without those.

      Now it is fair to say that the _driver_ is a sort of microkernel OS all its own (and also has an optimizing compiler at that), but it does still use your CPU to execute it.

    6. Re:This is great - sort of by myrdos2 · · Score: 2, Insightful

      Because reverse-engineering is very hard and complaining is very easy.

    7. Re:This is great - sort of by RAMMS+EIN · · Score: 3, Interesting

      ``A modern graphics card, for example, is actually a complete computer. It's got RAM, a processor, a bunch of peripherals, a complete miniature operating system... and you don't even know what type of processor it is. A lot of peripherals work like this; a wireless card is typically an ARM processor with some RAM attached on one end to the radio and on the other to an I/O controller that talks to the computer.''

      The big difference here is that we usually think of as a computer typically has enough specifications published that you can program them, and there are really just a couple of flavors. By comparison, even though each wireless network card does pretty much the same things as the next one, they are usually programmed completely differently and we're not told how. We have standards and heaps of documentation when it comes to CPUs, but when it comes to graphics accelerators or wireless network cards, it's a mess of undocumented, proprietary, incompatible interfaces.

      --
      Please correct me if I got my facts wrong.
    8. Re:This is great - sort of by Culture20 · · Score: 1

      You'll always find the odd usb controlled nerf gun turret or whatnot that lacks a driver, but that's not really an issue for most people.

      http://www.qbik.ch/usb/devices/showdev.php?id=4022
      http://www.thok.org/intranet/python/usb/README.html

    9. Re:This is great - sort of by gr8_phk · · Score: 2, Interesting

      The reverse engineering of the nVidia driver took advantage of the close nVidia driver. Because the OS is open, they were able to put in hooks to intercept all data moving between the computer and the card. They can even monitor communication during specific rendering events. While it's an enormous amount of effort for a video card, it pales in comparison to trying to figure out how even a simple device works with no documentation and no existing way to manipulate it. In other words, the existence of the closed driver is what made the effort possible in the first place.

    10. Re:This is great - sort of by Anonymous Coward · · Score: 0

      You'll always find the odd usb controlled nerf gun turret or whatnot that lacks a driver, but that's not really an issue for most people.

      Actually, there's software for that now.

      http://blog.taragana.com/index.php/archive/how-to-control-usb-missile-launcher-on-linux/

    11. Re:This is great - sort of by david.given · · Score: 1

      Video cards run nothing like an OS on the hardware, and you won't be getting one running yourself. Most GPUs don't have branch instructions (newer ones kind of do, but only in the "skip forward" sense). It's kinda hard to run an OS without those.

      How are Cg subroutines and jumps compiled, then?

    12. Re:This is great - sort of by kostmo · · Score: 1

      That's old. Try sudo apt-get install pyrocket -- works for nearly all of the turrets out today.

    13. Re:This is great - sort of by ClosedSource · · Score: 1

      "Because these days it's really, really hard."

      It's always been hard. Every company other than Atari that created video games for the 2600 had to do this (or pay somebody else to do it) because the specs were a trade secret.

    14. Re:This is great - sort of by sproingie · · Score: 1

      The compiler does a lot of unrolling. Newer GPUs do have actual branch instructions, but they have enough locality restrictions that you'd have a hell of a time making them fully general. Restrictions on what you can write back are also not particularly fun.

      It's still turing-complete all right, but the memory fetch latencies and other restrictions would make it absolutely hellish to run an OS on a GPU, which is why they don't run one.

    15. Re:This is great - sort of by badkarmadayaccount · · Score: 1

      Dear IBM^H^H^HSanta Clause, I wish the PC world had channel I/O, so they could quit wasting silicon.

      --
      I know tobacco is bad for you, so I smoke weed with crack.
  12. Re:How does it compare with the other NVidia drive by LOLLinux · · Score: 0, Flamebait

    Because of this, I'm going with ATI next time, I've heard they're way more Linux friendly now.

    Have fun enjoying your extremely crappy drivers. Both the open source and the proprietary ATI drivers suck.

  13. Re:How does it compare with the other NVidia drive by LOLLinux · · Score: 1

    If all that mattered was that it was free you might as well stick to the generic 2d drivers. Most people want performant 3d drivers.

  14. Re:How does it compare with the other NVidia drive by Jon_S · · Score: 0, Troll

    Sigh...

    Gratis vs. Libre

    Look it up.

  15. Just for those who wonder... by Corporate+Troll · · Score: 5, Informative

    DRM in this context means Direct Rendering Manager and not Digital Rights Management

  16. Re:How does it compare with the other NVidia drive by Anonymous Coward · · Score: 0

    Hahahahahahahahahahahah!

    Oh god that's funny

    My computers have ATI ... their drivers make me cry. Once I had the binary driver working. Then I upgraded. It stopped working. (For i = 1:20, repeat the last two steps). Then they said my year old IGP card was too old, and stop supporting it with new driver releases.

    Open source stuff works OK, but not as good for hi-def video.

  17. Re:How does it compare with the other NVidia drive by LOLLinux · · Score: 0, Troll

    Which is a distinction most people won't care about when their "gratis" driver can't properly play their 3d games.

  18. Re:How does it compare with the other NVidia drive by Anonymous Coward · · Score: 0

    The official ones are free, too.

  19. Re:How does it compare with the other NVidia drive by Anonymous Coward · · Score: 0

    Free as in freedom jackass.

  20. Re:How does it compare with the other NVidia drive by Lord+Lode · · Score: 2, Insightful

    Hardware video card support is pretty darn important these days, especially with more and more calculations (even not graphics related) being possible on the GPU and non-game applications using 3D acceleration to render 2D things faster, so I really, really, hope that Linux (and the free software in general) will have a good solution to run stuff on any GPU as good as it can run stuff on a CPU right now, because otherwise it'll lag behind and prevent applications that use that instead of the classical CPU + software rendered 2D graphics combination.

  21. Re:How does it compare with the other NVidia drive by icebraining · · Score: 3, Insightful

    Most people don't care about /. either, and here we are.

  22. monolithic kernel by Icegryphon · · Score: 1, Insightful

    monolithic kernel is monolithic!
    Soon 2.6 will have support for the kitchen sink!

    1. Re:monolithic kernel by Anonymous Coward · · Score: 2, Insightful

      Yes indeed. Years of dissing Windows for integrating the graphics drivers into the kernel, and what does Linux do ...

    2. Re:monolithic kernel by josh61980 · · Score: 1

      This needs a motivational poster.

    3. Re:monolithic kernel by Anonymous Coward · · Score: 0

      Yeah, how they dare, adding functionality to the kernel. Those bastards.

      Can't we have a barely functioning lean system, like in the Good Old Times (TM)?

    4. Re:monolithic kernel by Just+Some+Guy · · Score: 1

      And this causes what problems in practice?

      --
      Dewey, what part of this looks like authorities should be involved?
    5. Re:monolithic kernel by Anonymous Coward · · Score: 0

      Soon? I thought support for kitchen sinks has been in the kernel since 2.2.

    6. Re:monolithic kernel by heson · · Score: 1

      Enjoy your Hurd. An I'll enjoy the sink.

    7. Re:monolithic kernel by Anonymous Coward · · Score: 0

      monolithic kernel is monolithic!

      Soon 2.6 will have support for the kitchen sink!

      [*] monolithic kernel
      [ ] kitchen sink
      [M] ???
      [*] PROFIT!

    8. Re:monolithic kernel by Anonymous Coward · · Score: 0

      Are you really complaining about device support?

    9. Re:monolithic kernel by badkarmadayaccount · · Score: 1

      Context switches. Cooperative multitasking and compiler based protection FTW! And screw x86!

      --
      I know tobacco is bad for you, so I smoke weed with crack.
  23. Debugging advantage of a Free driver by tepples · · Score: 3, Informative

    You've probably paid for it with operating system crashes. Your time spent waiting for a reboot, re-creating lost work, and troubleshooting the failure is probably worth money. If a driver is Free (in the GNU sense), developers of the kernel and the X server can trace into it to see what's going wrong. Interactions with black boxes are much harder to debug.

    1. Re:Debugging advantage of a Free driver by jsnipy · · Score: 1

      Well to person who wants the driver to plays games and is not willing to or able to trace an issue and contribute a fix -- what the difference?

      --
      -- if you mod me down, I will become more powerful than you can possibly imagine
    2. Re:Debugging advantage of a Free driver by LOLLinux · · Score: 1

      Yeah because no open source drivers ever crash or anything. Oh wait...

    3. Re:Debugging advantage of a Free driver by jimicus · · Score: 1

      Well to person who wants the driver to plays games and is not willing to or able to trace an issue and contribute a fix -- what the difference?

      With any luck, in time it will become more stable and more reliable - you won't upgrade your Linux distribution to the latest and greatest only to find that your 3d graphics has magically stopped working. Not because you traced issues, but because other people did.

    4. Re:Debugging advantage of a Free driver by Tekfactory · · Score: 1

      I haven't traced or debugged code in a few years, not since my last C++ class.

      I've never contributed a fix to OSS.

      That wouldn't in this case stop me from sending a copy of the crash debug report to the developer. Maybe opening a bug report, which is dead easy.

      Before open source drivers I would have no choice but to use binary blobsk, and wouldn't even be able to send in the crash reports.

    5. Re:Debugging advantage of a Free driver by tepples · · Score: 2, Informative

      Well to person who wants the driver to plays games and is not willing to or able to trace an issue and contribute a fix -- what the difference?

      If the driver is Free, the user can pay someone else who is a kernel or X developer to fix it. That's the point of buying Linux support from a company like Canonical. But with a non-free driver, NVIDIA and ATI can just say "sucks to be you; we'd be glad for you to take your business elsewhere."

    6. Re:Debugging advantage of a Free driver by Al+Dimond · · Score: 1

      The difference is not much on Windows. ATI and Nvidia care enough to trace issues and fix them on Windows. They don't care enough to do it on Linux. They don't care enough to implement many X features in standard ways either.

      The difference is that with a Libre driver when there's an annoying bug it will get fixed. With the gratis Linux drivers there are plenty of annoying bugs that will probably never get fixed. In another post on this story I mentioned ATI's fast-user-switching bug. It causes X to go down completely when you use it. A lot of KDE and Gnome devs probably want to fix this bug, and are perfectly capable. I want to fix this bug, and I'm probably capable. But none of us can.

      The difference is that you'd get better quality in many respects with a Libre driver.

    7. Re:Debugging advantage of a Free driver by tepples · · Score: 1
      Let me put it this way:
      • All sufficiently complex software has defects.
      • In free software, anyone can fix the defects.
      • In non-free software, only the original publisher can fix the defects, or nobody can fix them at all if the original publisher has a business reason not to.
    8. Re:Debugging advantage of a Free driver by Anonymous Coward · · Score: 0

      In free software, anyone can fix the defects.

      See, thats the fallacy. We are talking about drivers for a modern graphics card. Please show me a patch by someone who isn't paid very good by a large company to do that.

      The truth is no one will ever fix the video driver for himself. This isn't something you can accomplish in your free time.

  24. Reverse engineering by OzPeter · · Score: 3, Insightful

    Reverse engineering a complete video driver is an impressive feat. However it is a reactive process and not a proactive process. Presumably when NVidia changes their driver architecture (to suit future hardware) won't this all have to be done over from scratch?

    --
    I am Slashdot. Are you Slashdot as well?
    1. Re:Reverse engineering by Anonymous Coward · · Score: 0

      so what you are saying is that new things are new?

    2. Re:Reverse engineering by Anonymous Coward · · Score: 0

      Presumably when NVidia changes their driver architecture (to suit future hardware) won't this all have to be done over from scratch?

      New hardware will need updates to the driver (hopefully not "from scratch") but the open source driver is completely independent of nVidia's own driver's architecture.

    3. Re:Reverse engineering by gr8_phk · · Score: 3, Informative

      A number of reverse engineering tools were developed for the Nouveau effort. Some of that can be used for similar efforts with other hardware. Most of that can be used should a new graphics driver architecture come out of nVidia. I have often wonder how long it will take if the Nouveau status matrix gets a new column, for that new generation of card to get support. It's been like 3 years for all the existing ones.

    4. Re:Reverse engineering by StormReaver · · Score: 1

      ...won't this all have to be done over from scratch?

      New card models will require more reverse engineering to get access to the new features. All features in those models that are fully backwards compatible will still work. Once a particular model is reverse engineered, it doesn't matter what NVIDIA does with future drivers since NVIDIA's drivers are no longer needed.

    5. Re:Reverse engineering by recharged95 · · Score: 1

      Agreed, will all the brainpower, testing and effort put into this, they might as well had prototyped a non-nvidia hardware design, opensourced the design to someone like intel to create a mass producible ref design and then licensed it to Nvidia. Then we'd have stuff like GMA's architecture (and plays very easy with FOSS code) with Nvidia power.... and everyone wins.

      The OP's approach is really a nose in the air, we're smarter than everyone else and going it alone approach--if Nvidia takes note to compete, it will bite this effort in the butt down the road.

    6. Re:Reverse engineering by Lemming+Mark · · Score: 2, Insightful

      I'd have said the hardest part is reverse engineering how to talk to the hardware. If new hardware comes out they'll have to reverse engineer the new features / changes in that but they shouldn't have to start again from scratch.

    7. Re:Reverse engineering by martin-boundary · · Score: 3, Insightful
      You've actually got it backwards, this is indeed proactive. With an open source driver, whenever the Linux kernel (or Xorg etc) changes, the driver can be modified to still work. With a closed source driver, if change happens, your driver is effectively bricked. The problem is that Nvidia isn't here for the longhaul, at least nowhere near as long as open source.

      There are many reasons why Nvidia might stop updating a driver for a particular card: they might go out of business, they might refuse to support obsolete cards that they no longer sell, etc. So over time, even if you still own their hardware, it will stop working at full potential (or at all) with the latest Linux. You'd need to download a newer driver, and if that's no option, you'd need to buy youself a new card, or you'd need to refuse to upgrade your Linux etc.

      If you expect your hardware to always work with Linux, then with closed source you need Nvidia to react to all changes that may happen to keep your card working, whereas with an open source driver there's potentially millions of people who can be persuaded to keep your card working.

  25. Devil's advocate by tepples · · Score: 1


    And the vast majority of popular video games aren't Libre, so why should the driver be?
    </devilsadvocate>

    1. Re:Devil's advocate by Anonymous Coward · · Score: 0

      <devilsadvocate>
      And the vast majority of popular video games aren't Libre, so why should the driver be?
      </devilsadvocate>

      It's not a video game driver, it's a graphics driver. There's plenty of graphical Libre software.

    2. Re:Devil's advocate by Al+Dimond · · Score: 4, Insightful

      Because it's in the kernel of their operating system. Because the fact that the driver is not Libre prevents other desktop-related stuff from working because the one vendor doesn't care and nobody else can fix it. Here's an example:

      Using the gratis ATI driver, running two X servers on the card crashes the driver and leaves X and the card in an unusable state (you have to ssh into the box to reboot it cleanly). This has apparently been a bug in the ATI driver for ages. And because multiple X servers are used to implement "fast user switching", ATI's crap driver blocks fast user switching.

      This sort of bug would be fixed in a libre driver. It's 100% reproducible, incredibly annoying, and affects a feature in desktop environments with millions of users and thousands of developers. If I had the source code to ATI's driver I could probably fix this bug. But ATI doesn't care.

      It's impossible for the Linux kernel team and X.org to design interfaces and a good model for how kernel drivers should interact with userspace X drivers to provide rendering in a way that fits in with X's model when the two biggest GPU makers will just ignore it, write their own kernel modules and their own interfaces. With a Libre driver new X.org standards and interfaces would be adopted much quicker and the drivers would fit into the system better. Nvidia and ATI care about this for Windows (to some degree) and so their drivers fit well there. On Linux they don't. But lots of other people do care, and non-Libre drivers prevent them from doing anything about it.

    3. Re:Devil's advocate by Tekfactory · · Score: 1

      Because the Card that the Driver is um... Driving, wasn't Free as in Beer.

    4. Re:Devil's advocate by tepples · · Score: 1

      It's not a video game driver, it's a graphics driver.

      And for applications of graphics other than 3D video games and 3D modeling, the 2D-accelerated driver works fine.

      There's plenty of graphical Libre software.

      But do they use 3D graphics with lots of triangles and complex shaders? Or do they use the big flat planes of pixels that even an Intel GMA can accelerate?

    5. Re:Devil's advocate by tepples · · Score: 1

      Because it's in the kernel of their operating system.

      Code lower than the kernel is non-free: the CPU's microcode. But you have a good point about ATI refusing to fix the problem that prevents fast user switching.

    6. Re:Devil's advocate by Anonymous Coward · · Score: 0

      So the FREE (as in herpes) ATI driver doesn't have that problem? Oh wait, there is no FREE (as in do exactly what stallman says) ATI driver. But if there was, somebody would fix all the bugs. Thanks for clarify.

    7. Re:Devil's advocate by Anonymous Coward · · Score: 0

      ?
      (lameness filter: activate!)

    8. Re:Devil's advocate by Al+Dimond · · Score: 1

      I hate to feed the trolls, but last I checked you could run the VESA driver for pretty much any card. It offers a couple basic 2-D acceleration primitives that are in the VESA standard, which is usually enough for me.

      Some people are working on a Libre ATI driver. It takes a lot of work, even given specs. It doesn't support my ATI board yet, so I'm not using it yet (I'm using my onboard Intel until then). When the Libre ATI driver is ready I'll put my ATI card back in and give it a whirl. It almost certainly won't have this stupid bug in it. And even if perf is a little worse than the gratis driver, it's worth it for that reason.

    9. Re:Devil's advocate by True+Grit · · Score: 1

      Some people are working on a Libre ATI driver.

      FYI: Some of those people are actually AMD employees.

      Your earlier rant about NV & ATI was correct, but ATI no longer exists, they're a sub-division of AMD now, and AMD at least seems to care about openness and supporting non-Windows OSes more than NV or ATI ever did. Not saying AMD is perfect, just a hell of an upgrade from what we had before.

    10. Re:Devil's advocate by Al+Dimond · · Score: 1

      I am not convinced AMD/ATI cares that much. Their driver situation is *still* bad enough that I'm using an integrated Intel chip that can barely run Google Earth until the Libre driver supports my ATI card. It doesn't say much for their regard for Unix that they haven't fixed a long-standing bug, 100% reproducible, affecting a standard Unix feature accessible on modern desktop environments in two clicks. It doesn't say much for their regard for openness that they haven't even acknowledged that it exists.

      To be fair, the reason I got an ATI card in the first place is that they released their specs; I figured there would be passable Libre drivers eventually. I didn't anticipate how awful the Gratis ones would be.

    11. Re:Devil's advocate by True+Grit · · Score: 1

      I am not convinced AMD/ATI cares that much.

      That's fair enough.

      Give them a little more time to get the open driver they're working on now ready for the mainstream.

      I'm one of the an early adopters who is using that new driver stack (right from their source repos on kernel.org and x.org). Overall, its already better than either the old radeon driver or fglrx, except in terms of raw OpenGL performance (and that will come once the new infrastructure is in place), so I've seen the evidence, but anyone who hasn't tried the in-development stuff can understandably think that very little is happening.

      About a year ago, no ATI driver supported my IGP (I had to use the vesa driver!), now I've got hardware accelerated 2D & 3D as well as other things (like a high-res accelerated console mode available at boot), so I've seen the rapid improvements, but you'll need to wait for a distro using 2.6.33 (or .34) and the next major release of Xorg (or possibly the one after that) to see all of what I'm seeing (some of this, e.g, KMS, is already in 2.6.32 & will be seen in the next Xorg).

  26. Re:How does it compare with the other NVidia drive by socrplayr813 · · Score: 4, Interesting

    I've heard some absolutely nightmarish stories about getting ATI cards to work properly in Linux and they haven't gotten much better. In the most recent releases, they may have even gotten worse.

    They might be more Linux-friendly now than they were in the past, but that doesn't make them good. They're certainly nowhere near as Linux-friendly as Nvidia.

    --
    The confidence of ignorance will always overcome the indecision of knowledge.
  27. Re:Just for those who wonder... by cptdondo · · Score: 1

    And I really, really wish they'd change that. It's really confusing. Especially since a big chunk of D[igital] R[ights] M[anagement] seems to be preventing the dreaded video analog hole.

  28. What about BSD? by rhavenn · · Score: 2, Interesting

    So, currently there is an issue with xorg 7.5 being imported into FreeBSD due very Linux specific driver "hacks", specifically in the latest Intel drivers and the ATI radeon drivers. Is this the same issue? Will this Nouveau driver work on anything else or is "open source" becoming synonymous with "if it runs on Linux, that's good enough". Linux has achieved great strides, but far too many "open source" developers target Linux only and have blinders on to any other open source OS or UNIX'esque OS where this stuff should really be able to run.

    1. Re:What about BSD? by Anonymous Coward · · Score: 0

      Well, it is open source. If *BSD needs to make changes to make drivers work, they can do it. But you hardly can expect Linux developers to code for *BSD now can you? Remember, X.org parts of these drivers are MIT lisenced, so go hack on these drivers to make them to work on your chosen platform. I really do not know about the kernel side of these new kernel mode setting drivers and their licensing though.

    2. Re:What about BSD? by mvdwege · · Score: 1

      Since when is it the responsibility of Linux driver writers to port their drivers to a different kernel?

      The source is out there. If the BSD folks want to have open-source drivers, let them port it. They are the ones who have the need and the knowledge of their kernel architecture after all.

      Mart

      --
      "I know I will be modded down for this": where's the option '-1, Asking for it'?
  29. Cost of Drivers is a Tax Now? by i_ate_god · · Score: 1

    If Linux has free alternatives to nvidia drivers and I don't use nvidia drivers, then I should get a discount on my next purchase of an nvidia card since part of that cost goes into development of the drivers.

    --
    I'm god, but it's a bit of a drag really...
    1. Re:Cost of Drivers is a Tax Now? by Anonymous Coward · · Score: 0

      Yes you should. Good luck with that.

    2. Re:Cost of Drivers is a Tax Now? by Sir_Lewk · · Score: 1

      If I buy a car, then paint it myself, do I get a discount on the car since some of the cost payed for painting it originally?

      I mean, give me a break, I'm probably considered huge FOSS appologist and proprietary software hater, but even I think that is stupid.

      --
      "linux is just DOS with a UNIX like syntax" -- Galactic Dominator (944134)
    3. Re:Cost of Drivers is a Tax Now? by Anonymous Coward · · Score: 0

      Slashdot needs a +1, Car Analogy

    4. Re:Cost of Drivers is a Tax Now? by omnichad · · Score: 1

      If I buy a toaster and throw away the box, shouldn't I get reimbursed for the box? You can only buy what's on the market. I totally agree with you.

  30. Re:How does it compare with the other NVidia drive by zoward · · Score: 1

    nVidia can arbitrarily stop supporting old graphics cards at any time. ATI did this with my R600-based laptop chipset; the newest ATI Catalyst linux drivers now longer support my two-year-old laptop. Since linux has a smaller user base, it's a "safe" place to cut costs by not having to feature-test against older hardware with every proprietary driver release. Having an open source driver would prevent you from suddenly becoming unable to use your hardware on newer linux releases.

    --
    "Can't you see that everyone is buying station wagons?"
  31. Good Point! by Cassini2 · · Score: 1

    Good Point! I was a little confused by the reference too.

  32. Re:Is linus being an arsehole here? by mellon · · Score: 2, Insightful

    He's forcing the point. If you are the one upon whom the point is being forced, I guess you could see it in the way you've described, but it's just a tactic for making the right thing happen. To the extent that you can say anything is his job, this is his job. Linux wouldn't be where it is today, for better or for worse, without Linus being the benevolent tyrant.

  33. Re:How does it compare with the other NVidia drive by diegocg · · Score: 1

    Just check the feature matrix

    3D features are not really supported, but except that, most of the basic things seem to be supported - KMS, KMS-based FB, suspend/resume, dual head/randr, 2D, video...

  34. But tyrant still. by Anonymous Coward · · Score: 0, Insightful

    But a tyrant still.

    And not always benevolent (cf BK vs Tridge).

  35. Re:How does it compare with the other NVidia drive by Sir_Lewk · · Score: 1

    Gallium3d is the planned future of graphics drivers in linux. OpenCL is what is going to be used for general purpose computing on GPUs.

    This area really isn't my forte, but you can find a bit more information at: http://zrusin.blogspot.com/2009/02/opencl.html

    --
    "linux is just DOS with a UNIX like syntax" -- Galactic Dominator (944134)
  36. Re:Just for those who wonder... by Again · · Score: 2, Informative

    DRM in this context means Direct Rendering Manager and not Digital Rights Management

    Thanks. I was reading through the comments looking for the usual DRM rants.

  37. 3d support is in Gallium3D by Anonymous Coward · · Score: 1, Informative

    And here is the page that describes current 3d support in Gallium3D: http://www.x.org/wiki/GalliumStatus

    Notice the sea of grey under the nVidia chipsets. So if you want games, keep using the binary for now.

  38. Re:How does it compare with the other NVidia drive by Anonymous Coward · · Score: 0

    Closed components in the ecosystem slow down development and make everything more difficult for us free software developers.

  39. Re:How does it compare with the other NVidia drive by CannonballHead · · Score: 1

    I have had nothing but trouble with my ATI linux drivers. Granted, it was an older and a mobile card, but it was a pain to get 3D working at all. Actually, not that old - X1400 Mobile, I believe.

  40. Re:How does it compare with the other NVidia drive by pantherace · · Score: 2, Insightful

    None ship them enabled because nvidia doesn't let them by default.* I think at least one distro has distributed them (Mandrake) possibly in one of their pay products. Most have an option to download them after install. (Kubuntu, Gentoo being the last two I checked, though you could argue that's still in the install for gentoo.)

    Frankly, I think you'll be disappointed in the support ATI on Linux has.

    *I just looked, and they now allow it, provided nothing is modified. They didn't last time I looked.

  41. Re:How does it compare with the other NVidia drive by Tetsujin · · Score: 1

    Free as in freedom jackass.

    Freedom isn't free!

    --
    Bow-ties are cool.
  42. Re:How does it compare with the other NVidia drive by tepples · · Score: 1

    Open source stuff works OK, but not as good for hi-def video.

    Is the hi-def video open content?

  43. Re:How does it compare with the other NVidia drive by paziek · · Score: 1

    For a few months now I had issues with both HD 3850 at home, and HD 4870 at work. It keeps probing my monitor for available modes, and that makes a lot of stuff slower/non-workin. Wine usage if out of question. At home I could still watch videos, its not possible at work tho. But at work I had more updated version of Linux, so perhaps thats why. Right now I'm using Windows as host for Virtualbox where I run my Ubuntu - it works better that way. Mind, that this error is also true for Gentoo (or it was a few months ago, when I ditched Linux at home). I think that ubuntu 7.10 works good with Radeons, so u might try that. Just don't update it (not just drivers, X as well).
    Anyway, Linux friendliness could be true, as I hear they are trying to open up documentation for their drivers so that community can make open-source ones better. But it doesn't mean, that right now Radeon drivers are better. At least not the one provided by AMD.

  44. Re:How does it compare with the other NVidia drive by Tetsujin · · Score: 1

    I'll agree with you, they work good, when they work. The problem with the official drivers is that they're a binary blob, thus most distributions (none I've ever seen) ship with them enabled. This is an issue if the default nv driver crashes your machine. Because of this, I'm going with ATI next time, I've heard they're way more Linux friendly now.

    For what it's worth: I decided to go the ATI route this time around. I mostly use it for running Blender. I've been pretty happy with it overall - but I wouldn't say I've found the drivers to be particularly more or less troublesome than the NVidia ones.

    --
    Bow-ties are cool.
  45. Re:Just for those who wonder... by tepples · · Score: 1

    Nothing can stop the analog hole for noninteractive video. It is always possible to camcord the screen, and an MPAA representative actually recommended it. The only surefire way to plug the analog hole is to make video games instead of movies.

  46. Re:How does it compare with the other NVidia drive by FauxPasIII · · Score: 1

    I've heard lots of horror stories about the ATI/AMD graphics drivers for Linux (and there are plenty of them in responses to your post), but I'll contribute that I just got a laptop with a FireGL M7740, got "fglrx" out of apt, and everything worked immediately and with good performance. It's a big binary blob, but once you've resigned yourself to that, I don't find it to be any less performant or reliable than the big nVidia binary blob.

    --
    25% Funny, 25% Insightful, 25% Informative, 25% Troll
  47. Re:How does it compare with the other NVidia drive by LOLLinux · · Score: 5, Insightful

    Do you think any end user cares? The nvidia binary driver provides hardware accelerated playback of all high-def formats. The open source one doesn't. That's all that matters.

  48. Re:How does it compare with the other NVidia drive by Minwee · · Score: 5, Insightful

    It also causes my Inspiron 8200 to crash hard when I try to use ACPI functions. Nvidia has expressed no interest in fixing this bug and that pushes it from "mildly unacceptable to free software hippies and people with obscure unsupported hardware" to "completely useless crap masquerading as software".

    I'm not bitter about it but it's a good example of a problem which could easily be fixed in open source software, but can't even be touched in something as closed as the nvidia video driver.

  49. Re:How does it compare with the other NVidia drive by vadim_t · · Score: 3, Informative

    This irks some free software hippies and it also makes using Nvidia hardware on unsupported hardware platforms more difficult.

    It also irks people who noticed that a huge amount of devices didn't get 64 bit Windows drivers, because it was a lot more profitable to get people to buy new scanners, printers and webcams. Precisely thanks to this I now have a perfectly good color laser printer and scanner that my brother can't use anymore.

    Experience shows that if you trust the manufacturer will release updated drivers when they become needed, you're going to get screwed sooner or later. His new scanner (also made by Canon, guess he doesn't learn) looks nearly identical, and has pretty much the same specs. The only difference is that the light has been replaced with LEDs, but really he didn't gain anything from the new model.

  50. Re:How does it compare with the other NVidia drive by Anonymous Coward · · Score: 0

    Ah, ATI. Lovely... just lovely memories of ATI under Linux. Like how it would somehow stutter the visuals on Puzzle Pirates last time I tried them. Let me repeat that: It would stutter the visuals on a dead-simple 2D sprite-based Java game. Not some bullet hell, eighty bazillion sprites on the screen, special effects whoring, framebuffer coder's wet dream 2D game. Puzzle. Pirates. One install of an older NVidia card fixed that right up.

    Yes, I know, durr hurr javas slow so thats y it so slow im so funy durrrrrr hurrrrrr. Cute. Go fuck yourself. No video card, nor their drivers, open source or not, should be the cause of even the slightest bit of slowdown when trying to render Puzzle Pirates, of all games.

    I've not touched an ATI card since then under Linux, and I've not had a single video-based problem.

  51. Re:How does it compare with the other NVidia drive by castironpigeon · · Score: 1

    Linux developers can arbitrarily stop supporting whatever they damn well don't feel like supporting any longer so they can go program extra functions into their USB foam dart cannon. Just sayin'.

    --
    mmmm...forbidden donut
  52. Re:Just for those who wonder... by Anonymous Coward · · Score: 0

    Crap... I was just going to sent them DMCA toke-down notification. Oh well, next time.

    nVidia IP Officer

  53. Linus in a snit by Anonymous Coward · · Score: 2, Insightful

    Reading Linus' remarks, I can't help but think what a childish, whinging prat he is. He makes Theo look calm, cool, and collected.

    1. Re:Linus in a snit by True+Grit · · Score: 1

      Reading Linus' remarks

      Linus isn't a diplomat, he's a kernel hacker, and he's not the only kernel hacker who is not especially diplomatic. That's because being a diplomat is not his job.

      a childish, whinging prat

      If that was really what he is, then no one would pay attention to his rants. Instead, just days later, his rant accomplished real change, and for the better. NV hardware users (who want/like Nouveau) should be glad Linus is someone people listen too, no matter how undiplomatic he may be.

      Besides, having read through that entire LKML thread (as of yesterday), I don't actually see Linus behaving particularly badly there (for him). He's said some especially stupid things in the past, but here, not so much, and I think he actually had a fair point about the upstreaming issue (which is probably why this change happened so quickly).

      Me? I'd much rather have that 'whinging prat' in charge of the Linux kernel than a real diplomat.

  54. They're not by VMaN · · Score: 1

    Believe me, I've had nothing but headaches with my Radeon 3650 mobility, and being a laptop, I'll have to live with it.

    But for the foreseeable future, any ATI chip in a laptop is a dealbreaker

    1. Re:They're not by S.O.B. · · Score: 1

      I also have the 3650 in my laptop and like you I've had no end of troubles. I was even forced to run the onboard Intel chip for a while until ATI got around to fixing their driver. I stopped buying ATI after they stopped supporting OS/2 (yeah, I know). They have a history of providing little or no quality support for alternative operating systems. Oh, they'll support it for a while and only as much as they feel they have to until they get you hooked on their hardware but after they get your money support is just a loss to them. They don't consider that the money they put into support will get them the next hardware purchase.

      For me, Nvidia is the ONLY video card vendor I will consider (I believe I'm on my 7th or 8th Nvidia card) and in the future I won't make the same mistake of buying a laptop that has any of that ATI crap in it.

      I'll take a vendor that gives me a binary-only driver that just works over a vendor that gives me an open source driver that wastes hours of my time troubleshooting.

      --
      Some of what I say is fact, some is conjecture, the rest I'm just blowing out my ass...you guess.
  55. You mean ForkBSD by Anonymous Coward · · Score: 0

    Why don't they just fork it like they do EVERYTHING ELSE so that they can be ideologically licensoligically "pure"?

  56. I lol'd by GameboyRMH · · Score: 1

    Linux already has a closed source driver from Nvidia that works pretty damn well.

    A driver that breaks suspend and hibernate, needs power management tweaks to allow the PC to boot without freezing up and makes compiz hang on a regular basis "works pretty damn well?" Maybe you're talking about the Nvidia control panel for Linux? It's very handy for helping you fiddle with settings in a futile attempt to get the awful thing to work.

    I'm looking forward to this new driver but I already learned my lesson about Nvidia and Linux.

    --
    "When information is power, privacy is freedom" - Jah-Wren Ryel
    1. Re:I lol'd by atrus · · Score: 2, Informative

      My Quadro based laptop with the NVidia driver has no issues with suspending and hibernating, and runs generally flawlessly.

    2. Re:I lol'd by wsanders · · Score: 1

      This dates me too, but the binary Nvidia driver for my ancient GeForce MX 400 has been working flawlessly since, oh, about the time Nividia EOLed the GeForce 2 hardware.

      --
      Give a man a fish and you have fed him for today. Teach a man to fish, and he'll say "WHERE'S MY FISH, YOU IDIOT?"
    3. Re:I lol'd by phantomcircuit · · Score: 1

      Count yourself lucky my Quadros 140M won't.

    4. Re:I lol'd by mikechant · · Score: 1

      A driver that breaks suspend and hibernate, needs power management tweaks to allow the PC to boot without freezing up and makes compiz hang on a regular basis "works pretty damn well?"

      All this works fine for me. The question is whether your experience or mine is typical. Judging from the Ubuntu forums and other sources such as /., I get the impression suspend/resume/compiz work OK for most people.
      Of course, a good quality FOSS driver is the best option and I'll certainly try it out once I get onto the kernel version referred to in TFA.

  57. Re:How does it compare with the other NVidia drive by Anonymous Coward · · Score: 0

    SO how does Nouveau work with hibernate/sleep?

  58. Re:How does it compare with the other NVidia drive by binarylarry · · Score: 1

    I'm willing to bet it's much easier for you to fix Nouveau's implementation than to get access to Nvidia's code and fix the bug in the proprietary driver.

    Which is really the point of not completely relying on proprietary code in the kernel.

    --
    Mod me down, my New Earth Global Warmingist friends!
  59. Re:How does it compare with the other NVidia drive by gmack · · Score: 4, Insightful

    I have had more luck now that the Open Source ATI driver added 3D accel support for my card. The official ATI drivers suck badly with barley working 32 bit drivers and mostly useless 64 bit support. The open source drivers actually make me like using my Dell Vostro again and it's actually to a point where I would rather use ATI than NVIDIA.

  60. Re:How does it compare with the other NVidia drive by Anonymous Coward · · Score: 0

    That was true 3-6 months ago. The latest open source drivers work perfectly, and the fglrx linux drivers worked great on my 3xxx series card, but not my 4770. Furthermore X1400 is either R4xx based or R5xx based, *NOT* R600 based, and thus supported by the open source drivers (since the r400/500 were basically just evolutions of the R300 chips, and documentation for them was release over a year ago.)

  61. PS: your answer is "yes" by Anonymous Coward · · Score: 0

    PS: your answer is "yes". But you think for a good reason.

    1. Re:PS: your answer is "yes" by mellon · · Score: 1

      No, being an "arsehole" is when you do something annoying without a good reason. Having a good reason for doing something doesn't stop it from being annoying, but the worst "arseholes" are the ones who never do anything that might annoy anybody, and in the process let whatever project they're running go straight to hell.

  62. Re:How does it compare with the other NVidia drive by koogydelbbog · · Score: 1

    i tried Fedora 12 and the last thing i see is the nouveau messages and then the screen goes black. the laptop continues to boot, i just can't see what it's doing. if i hit ctrl-alt-delete it'll reboot and the wifi light comes on so i think it's all ok apart from the screen. (12 beta was fine as it didn't use nouveau but i did one upgrade too many. have tried the F12 live cd and it's the same)

    have tried nouveau.modeset=1 whilst booting (a list of other options would be nice, but i couldn't find one). have tried disabling the module. i've deleted all the nouveau drivers from the root partition in an attempt to force it to use something else but i guess it's using the versions in the initrd file(?). have tried booting single user. nothing.

    other people have raised this as a bug but all the developers do is ask them to post further details. but they can't because they can't see what it's doing. is kinda frustrating.

    (this with a newish 130M btw)

  63. Re:How does it compare with the other NVidia drive by Anonymous Coward · · Score: 0

    For me small things like security and stability.
    The blob has a history of security problems and stability problems, and on my machine nouveau is just so much more stable then the nvidia binary driver.
    Then features like KMS, having the possibility to rebuild for new API/ABIs yourself, and so on is just nice bonuses.

    The only thing nouveau is lacking currently is stable 3d, but it has become better, and with kernel inclusion I have a feeling nouveau will get more exposed and more fixed, which leaves the current devs free for mesa-tasks.
    But speedwise in 2d it is nearly comparable to nvidia-binary on my card.
    It is just a point of - like with radeon and intel - have the latest pieces on the right places (xorg-server-1.6 has known speed problems with nouveau and so on).

  64. Re:How does it compare with the other NVidia drive by sarhjinian · · Score: 1

    Doesn't matter. The problem is that, with many combinations of card and driver, there's no accelerated video support, and that video tears like a sonofabitch even in non-HD forms.

    --
    --srj/mmv
  65. Re:How does it compare with the other NVidia drive by GameboyRMH · · Score: 1

    I'm a Linux user using the official binary NVidia drivers, they work bad - very bad even, they handle suspend/resume and hibernation like a falling egg handles the ground and no combination of settings fixes the horrendous diagonal tearing during video playback.

    How do the Nouveau Nvidia drivers compare to the official ones? Do they actually work, no glitching/freezing issues or other similarities to the official driver, etc...?

    --
    "When information is power, privacy is freedom" - Jah-Wren Ryel
  66. Re:How does it compare with the other NVidia drive by Anonymous Coward · · Score: 0

    You're also forgetting issues that come up on supported hardware platforms, but aren't marked as a priority of NVIDIA's in-house developers to fix. NVIDIA also drops support for older cards, so a newer driver may not support your older card. If the last driver version to support your older card has a bug in it? Tough luck!

    Having access to the source and the freedom to modify it allows people that are motivated enough to fix these issues.

  67. Re:How does it compare with the other NVidia drive by GameboyRMH · · Score: 1

    I'm thinking of going with Intel on my Linux machines from now on. the only problem is that the TV-out on laptops is a crapshoot (at least with some of the older 9xx adapters, in my experience). Apart from the laptop TV-out issues I've only had one Intel adapter not work perfectly right off the bat - on a server running CentOS.

    --
    "When information is power, privacy is freedom" - Jah-Wren Ryel
  68. Re:How does it compare with the other NVidia drive by Anonymous Coward · · Score: 0

    This irks some free software hippies and it also makes using Nvidia hardware on unsupported hardware platforms more difficult.

    That's a little ingenious. Video drivers are complex and can bring a machine down. People complain to the kernel developers about it, but there's no way they can look for bugs in a closed source package from the likes of NVidia.

    Should you need to update the kernel for a simple security fix or new feature, you are at the mercy of NVidia releasing another binary release to match the kernel. They drag their feet doing this, so if you want your core code up to date, you have to dump NVidia's driver or wait several weeks. Not everyone wants to do that.

    Finally, both NV and ATI cheat in their drivers to gain better performance test numbers. Having an open source driver is really going to put pressure on them, as the project allows others to wade in and offer improvements.

  69. Re:How does it compare with the other NVidia drive by Hurricane78 · · Score: 0, Troll

    Don’t do it. Whatever you heard. I have personal experience with the “drivers” right now. And compared to them, the nVidia ones are a wonderful dreamland of milk and honey. (Talking about the proprietary fglrx drivers.)

    The Linux “team” of ATi is a one-man-show, and focuses only on workstations. Everything else is simply ignored.
    Sometimes out of luck, they drop some crumbs of features that are useful. But they only got compositing some weeks ago.
    Video playback still is horrible, with only mpeg2-style codecs accelerated. And so badly rendered, that you basically have to turn it off.
    Like applying a lens effect on the histogram. glowing white, deep black, nearly nothing in-between.
    You can expect weird crashes, that do not happen when I then boot with the nVidia onboard card and try again.
    They only just updated the driver to support the new kernel interfaces, after the old ones they used were so outdated, that they were completely removed from the kernel!
    There are other weird graphics errors with areas not being redrawn.
    And if X crashes, you can expect to press the cold reset button, because you can’t get to any console anymore.
    Also don’t dare to switch to a console and back too quickly, or do it at all, without atieventsd running. Because then you also have to reset.

    And that is for the 4850! I don’t think that with the 5xxx series it will be better than back when the 4xxx was new. (When neither compositing, nor Xinerama or video playback worked at all. Let alone all of them together.)

    ATi have great hardware, but really really bad drivers. For Windows too (ask any game or demo developer). Just there it’s less visible because everybody knows how to route around.

    If you’re not planning on using a card that’s at least one or two generations back, with the open radeon (or radeonhd) drivers, stay faar away from ATi.

    --
    Any sufficiently advanced intelligence is indistinguishable from stupidity.
  70. Re:How does it compare with the other NVidia drive by Sir_Lewk · · Score: 1

    I agree with most of this, but "But they only got compositing some weeks ago." is flat out false.

    I remember playing with Beryl on my ATI card using fglrx back in 2007. Makes me question some of your other more specific claims. (I can't speak for much currently, I've been happily using the free radeon driver)

    --
    "linux is just DOS with a UNIX like syntax" -- Galactic Dominator (944134)
  71. Never A Crash by TheNinjaroach · · Score: 1

    I use NVidia's closed drivers for Linux on my 8800GT and I have never, ever experienced a crash or lockup. Even when playing games like Team Fortress 2 with a maxed out resolution for hours at a time, never a crash. On the other hand, my completely open Intel drivers on this here work PC has caused three hard locks just this week while running the glxgears screen saver.

    --
    I went to eat some animal crackers and the box said, "Do not eat if seal is broken." I opened the box and sure enough..
    1. Re:Never A Crash by True+Grit · · Score: 1

      I use NVidia's closed drivers for Linux on my 8800GT and I have never, ever experienced a crash or lockup.

      Some people are just damn lucky...

      Otherwise your point is?

    2. Re:Never A Crash by Anonymous Coward · · Score: 0

      Some people reply to Slashdot messages.
       
      Otherwise, your point is?

    3. Re:Never A Crash by True+Grit · · Score: 1

      Otherwise, your point is?

      Personal anecdotes can not *prove* a general statement, they can only be used (sometimes) to disprove a general statement.

      I've used NV cards in the past (2 of them before switching to a mobo with an AMD-ATI IGP), however, my personal anecdotes do not match the GP's post (hence my 'lucky' reference).

      The fact that things worked great for the GP does not change the fact that they did not work great for others - thus his personal anecdote proves nothing.

      But thanks for playing anyway, Mr. AC.

  72. Re:How does it compare with the other NVidia drive by risinganger · · Score: 1

    It costs folks like you and me

  73. Re:How does it compare with the other NVidia drive by TheKidWho · · Score: 1

    So because the ACPI is a little buggy the software is complete crap? Never mind every other feature that the software has.

    I'm sure the competitions graphics cards are even better.

  74. Re:How does it compare with the other NVidia drive by ChrisMP1 · · Score: 1

    Well yes, but the code is still out there, and can still be compiled with newer kernels, unlike the binaries, which tend to only work with certain versions. Just sayin'.

    --
    <sig>&nbsp;</sig>
  75. I can finally install Ubuntu on my computer by Culture20 · · Score: 1

    ...using the GUI that is. Ubuntu's Live CDs have had problems with the nv driver on the Live CD producing green vertical lines with certain video cards for a long while. Also, the VESA driver is broken and reloads xorg every half-second (the nv driver is better because at least a virt term works).

  76. Try it yourself by gr8_phk · · Score: 1

    How do the Nouveau Nvidia drivers compare to the official ones? Do they actually work, no glitching/freezing issues or other similarities to the official driver, etc...?

    Download a Fedora 12 live CD. Make a bootable USB flash drive from it and just try it. Bugs vary from GPU to GPU, so this is a great way to try Nouveau with whatever you've got. Of course this is 2D only, but you're interested in video playback, so that's fine.

  77. Re:How does it compare with the other NVidia drive by Minwee · · Score: 4, Insightful

    So because the ACPI is a little buggy the software is complete crap? Never mind every other feature that the software has.

    Here is a lovely pastry. It was made with the finest butter, the flour was hand ground by monks, and it is served with cream and tiny bits of shaved chocolate.

    Oh, and it is also covered with sprinkles of bacillus anthracis which will cause you to die in agony after you eat it. But just look at all the other wonderful features it has!

    Don't you want to eat it? Sure the antrax does pose a teeny tiny little problem, but maybe you could just eat a little bit of it.

    (Or do we need a car analogy to explain the problem here?)

  78. Re:How does it compare with the other NVidia drive by bjourne · · Score: 1

    They are seen as more Linuxfriendly because they have released the specifications for their 3d chipsets which Nvidia hasn't. Unfortunately, and contrary to what slashdot-commenters believed, that hasn't lead to someone creating high quality free drivers for ATI cards.

  79. Re:How does it compare with the other NVidia drive by FudRucker · · Score: 1

    3D on a computer monitor is a delusion, you have height and width, that third dimension is the delusion that is implemented by software to give the appearance of depth. it is not like you can reach in to your screen like it really had depth, (nothing wrong with good hardware acceleration with decent screen resolution and color definition (at least 16, but 24 is better)

    --
    Politics is Treachery, Religion is Brainwashing
  80. Re:How does it compare with the other NVidia drive by quickOnTheUptake · · Score: 1

    Either I missed the force of your post or you mixed up gratis and libre.
    I assume you are meaning to say that people don't care whether it's open source if it doesn't work as well as the free-but-closed-source alternative.

    --
    Mod points: Guaranteed to remove your sense of humor.
    Side effects may include gullibility and temporary retardation
  81. Mandriva also ships with Nouveau by Anonymous Coward · · Score: 0

    Mandriva also ships with Nouveau, but that don't seems to be mentioned. Also other distros ship it?

  82. Re:How does it compare with the other NVidia drive by Sir_Lewk · · Score: 2, Informative

    Uhh..... No shit?

    2d acceleration refers to things like textured video, for playing movies or whatnot. 3d acceleration refers to things like the rendering of 3d primitives on screen, stuff like Quake. Different sorts of math are used for each.

    I certainly hope you are trolling because if not, this is a new level of ignorance that I was not aware existed.

    --
    "linux is just DOS with a UNIX like syntax" -- Galactic Dominator (944134)
  83. Re:How does it compare with the other NVidia drive by Anonymous Coward · · Score: 0

    I don't hibernate my nvidia machine, but sleep works with my Quadro NVS 140M just fine. Better than nvidia in fact (which like to go back to sleep after waking up). Some minor corruption before the screen is unlocked, but is doesn't last. It's worked since August-ish with Fedora 11 and continues into Fedora 12.

  84. Re:How does it compare with the other NVidia drive by Anonymous Coward · · Score: 0

    Yes, but if the devs do this, *someone* can do the support in the future. If nvidia drops your chipset, you're SOL.

  85. Re:How does it compare with the other NVidia drive by goltzc · · Score: 1

    I guess if we need a car analogy we need a car analogy.

    As much as I enjoyed the dessert analogy I think you went a little to the extreme.

    It's more like if you had a car but the radio doesn't work. You can fully enjoy the car but a feature of the car that is completely disconnected from the core purpose of the vehicle doesn't work.

    --
    Our bugs are smarter than your test scripts.
  86. Re:How does it compare with the other NVidia drive by TheKidWho · · Score: 1

    Your analogy fails because the ACPI bug doesn't cause the entire system to become unstable and unusable... Unlike your killer pastry.

    It would be more analogous to driving around in a BMW that works perfectly, except for an air vent in the rear passenger side of the vehicle which seems to get no air flow.

  87. Difficult? by Anonymous Coward · · Score: 0

    It makes using older nvidia hardware with recent software *impossible*. Nvidia cuts support for old stuff to force you to upgrade even if you're not an uber-gamer and are happy with your current performance.

    Enjoy the hardware upgrade hamster-wheel.

  88. Re:How does it compare with the other NVidia drive by bfields · · Score: 1

    "Because of this, I'm going with ATI next time, I've heard they're way more Linux friendly now."

    I haven't been following this so closely, but my impression was that ATI is doing a better job of working with the community at this point--giving adequate access to specs and so on, as opposed to just dropping a binary driver in our laps--which means in months and years to come ATI is likely to work better. But for now it may still have catching up to do.

    From my limited experience in recent years, Intel hardware is the only stuff with first-class open-source support. (But of course the best drivers in the world don't help if their hardware isn't good enough to run your game. I wouldn't know--for compiz, watching DVD's, and the occasional game of Tuxracer, the Intel stuff is fine.)

  89. Re:How does it compare with the other NVidia drive by Anonymous Coward · · Score: 0

    Windows driver signing can be to blame, too. There's a lot of money lost in that pit that gets sent to Microsoft when new drivers are made.

  90. Re:How does it compare with the other NVidia drive by Anonymous Coward · · Score: 0

    NVIDIA has already effectively done this with all the pre-GeForce 5xxx cards. You can still get their last drivers, but they won't work on any remotely modern distribution.

  91. ENOUGH about "copyrighted" work! by Anonymous Coward · · Score: 0

    It's ALL "copyrighted" at least in the US. If I write some screed it is "copyrighted." LINUX IS PROTECTED BY COPYRIGHT. Anything GPL IS PROTECTED BY COPYRIGHT. STOP confusing "copyright" with "proprietary" or "non GPL" it does nothing but cloud the issue further in the mind of an ignorant public that believes you are only protected by copyright if you're a giant corporation or a slave to one.

  92. Good news. Next stop: The Future by Lemming+Mark · · Score: 2, Insightful

    This is good stuff, I think. They're not going to shove the whole Nouveau device driver into the kernel, it'll follow the modern X.org / Linux model of having kernel modesetting and a DRM driver in the kernel and a whole load of other stuff living in userspace.

    Kernel modesetting (KMS) means that one entity, the kernel, always controls what graphics mode the video card is in. That's useful because pre-KMS, X.org might have changed the mode *without the kernel knowing*. That's one reason Linux can't easily have a Blue Screen Of Death - the kernel doesn't know what it can send to the graphics card to display it. BSOD isn't a feature you want to *see* but if you have a kernel panic, it'd be a lot more useful to actually see it, rather than it being hidden by your frozen X server! I'm not aware of graphical kernel panics currently being supported but at least it could be done in principle now, AIUI. KMS also reduces unnecessary modeswitching "to make sure" that you otherwise get, so switching between console and X should be quicker, as should switching between X sessions (fast user switching). KMS is also what's used by the new bootsplashes, like RedHat's Plymouth (which other distros, e.g. Mandriva) are also moving towards. DRM, in this context, is the Direct Rendering Manager and is how GL apps get direct rendering access to the graphics card, in a controlled way. I don't know so much about that though ;-)

    The Nvidia open source driver "nv" doesn't support KMS or any 3D. The Nvidia proprietary driver doesn't support KMS but does support 3D (with good performance). Many distros have tended to use nv by default, some do ship nvidia though. Either way, you don't get the nice boot splash and neater terminal switching that a KMS driver would get you. The Nvidia proprietary driver is good performance-wise but it also tends to lag the open source drivers in terms of features a little; I think Nouveau (at one point? may not still be true) was aiming to support Xrandr features that Nvidia's did not. I've also heard that Nvidia's driver has issues with suspend.

    Because of all this, expanding Nouveau support is a good thing. Nouveau are also in the process of reverse engineering for 3D support but they have some way to go. However, I've had the impression that it's getting towards being better than the 2D-only nv driver. So at *least* it will mean that when installing on your system you can expect a decent boot experience and correctly-working basic 2D graphics, with suspend/resume behaving sensible, etc. So it could be making life better for users *soon*. But as 3D support improves, things should get better still.

    The Linux kernel devs generally take a stance these days that all kernel code ought to be merged into Linus's tree as soon as possible. It's really rather impressive to see this process working and the kernel devs (mostly) really following through on this.

  93. Real time safety and Nvidia. by Anonymous Coward · · Score: 0

    This is perhaps not the most common use of Linux, but some of us use it for recording.

    The official NVidia drivers are not real time safe, which can cause problems for people using Linux for low latency audio work.
    There is no way to fix this without having the source code.

    If you use an RT patched kernel, the official Nvidia drivers will sometimes not even install.

    The open source 2d Xorg drivers are OK most of the time, but some people use PD/Gem or Blender+Ardour etc and need 3d *and* good audio performance.

  94. Welcome new comer! by msimm · · Score: 3, Insightful

    I love posts like this because it demonstrates a dramatic (and frequent) misunderstanding. The idea seems to be (and I don't mean to fault anyone for this) that there's a great big general pool of Linux/driver developers that get together and decide what to do. We regularly see suggestions (like the one above) directed at how to make better use of this imaginary pool.

    But the truth is much of what gets done in terms of development is done by people like yourself, with interests of their own and probably more frequently then you imagine, on their own time. So while the project might not make sense to every possible user, particularly in the terms of some great imaginary directed labor pool, like many open source projects it's intended to scratch the developers own particular itch. And I don't know about you, but when I sit down to program in my free time I like to do something that I'm personally (preferably even passionately) interested in.

    --
    Quack, quack.
    1. Re:Welcome new comer! by ClosedSource · · Score: 1

      Well, sometimes Linux supporters claim that the work is mostly done by paid professionals, not by volunteers. Sometimes the claims go the other way.

      It seems that the assertion given on any occasion is the one that puts Linux in the best light given the current discussion.

    2. Re:Welcome new comer! by msimm · · Score: 1

      You know, like most things in life, depending on the context (or project) either can be true.

      But either way I'm sure if you take the time to make suggestions like this to closed source projects you'll see the only consistent difference is with open source projects if you feel strongly enough about [insert thing] you can dig in and start to do it yourself.

      --
      Quack, quack.
  95. Re:How does it compare with the other NVidia drive by Anonymous Coward · · Score: 0

    But is it noticeably slower, or just a benchmark thing?

    Or is this one of those videophile things like gold plated HDMI connectors and frame rates faster than the monitor refresh?

  96. Re:How does it compare with the other NVidia drive by omnichad · · Score: 1

    I'll be the 30th person to chime in and say that I just threw out another ATI card on a Linux Media Center and bought an NVIDIA card. You don't want to go anywhere near ATI. Especially if you're hooking the card up to a TV that gives bad EDID data.

  97. Re:How does it compare with the other NVidia drive by Anonymous Coward · · Score: 0

    It's only videophiles who care about the exact amount of pixels on the screen.
    Most normal people can't tell the difference between uncompressed video and lossy anyway.

    Of course, there will always be some who like 'hi-def' and gold plated cables for their game console!

    http://www.pspworld.com/sony-psp/accessories/accessory-overkill-goldplated-video-cable-for-psp-010808.php

  98. Re:How does it compare with the other NVidia drive by icebraining · · Score: 1

    Some Linux developers are also users of said cards, so they have a real need to fulfil.

    The Mach64 drivers received updates until 2007 (when the dev's laptop failed), although it only supported 10 year old cards, like the ATI RAGEs.

  99. Re:How does it compare with the other NVidia drive by Anonymous Coward · · Score: 0

    I am willing to bet that both is equally impossible to almost everyone.

  100. Re:How does it compare with the other NVidia drive by cerberusss · · Score: 1

    Experience shows that if you trust the manufacturer will release updated drivers when they become needed, you're going to get screwed sooner or later

    I'm not sure that what you are asking, is reasonable. You bought a scanner that worked with the PC you bought. Then you upgraded your PC or bought a new one. And now you complain that the vendor dos not help you for free with the upgrade.

    Let me put this in a car analogy. You have a car and you bought a baby seat with it. Then you buy a new car. Now the baby seat does not fit. The vendor could easily supply you with a cheap, plastic piece that makes it fit safely and snugly. You want it for free, while the vendor has a newer model available that fits.

    Just because it is a printer driver, which is software, does not mean that it is easy or free to make. Or that it was included in the printer price.

    --
    8 of 13 people found this answer helpful. Did you?
  101. VDPAU support? by batkiwi · · Score: 1

    Does this have VDPAU support ready? Planned? On ANY horizon?

    If so, hooray!

    If not, worthless for everyone who chose nvidia over ATI for this killer feature.

    1. Re:VDPAU support? by clarkn0va · · Score: 1

      Agreed. If not for VDPAU, which I enjoy at least weekly (sometimes more), I could be happily running just about any video card and free driver under the sun. VDPAU is a game changer, however, and I shall continue to use my 8600GT and its closed binary driver until some viable alternative makes me reconsider.

      --
      I am literally 3000 tokens away from the chaotic crossbow --Stephen
  102. Re:How does it compare with the other NVidia drive by Anonymous Coward · · Score: 0

    What? I don't understand.

    Car analogy now!

  103. Re:How does it compare with the other NVidia drive by smussman · · Score: 1

    The official ATI drivers suck badly with barley working 32 bit drivers and mostly useless 64 bit support.

    I hear the NVIDIA drivers are working much more ricely. Some people have even said they're amaizing

  104. Re:How does it compare with the other NVidia drive by vadim_t · · Score: 1

    I'm not sure that what you are asking, is reasonable.

    Thing is, you've not seen the device I'm talking about. It's not scanner "2.0", which is much improved over the previous version. It's not even that old, it was bought maybe a year before the switch to the 64 bit version.

    No, the new model is pretty much version 1.1. It looks identical. It has identical specs, as far as I can tell. The changes made to it are effectively cosmetic. It probably even talks pretty much the same protocol as the previous one. Linux certainly has no problems with talking to both of them, it's only Windows what has an issue.

    You bought a scanner that worked with the PC you bought. Then you upgraded your PC or bought a new one. And now you complain that the vendor dos not help you for free with the upgrade.

    See, that's not the way I see it. The way I see it, is that I didn't buy a scanner for Win XP, I bought a scanner, period. It shouldn't ever stop working for any reasons besides physically breaking, becoming so technically obsolete that it makes no sense to keep using it, or becoming impossible to connect (like if USB some day disappears).

    Linux fulfills this idea of mine. I can still use a Creative Webcam 5 on it, which is positively ancient by modern standards (it's a USB 1 webcam). There are no XP 64 drivers of course, but 64 bit Linux works perfectly fine with it.

    I think long term, so I consistently choose standards and openness whenever possible, because it's not in my interest to replace things that work perfectly fine just because the manufacturer would prefer to have a bigger number in the bank account.

    Just because it is a printer driver, which is software, does not mean that it is easy or free to make. Or that it was included in the printer price.

    So give me specs and the source, I'll fix it, and even submit a patch to the upstream.

  105. Re:How does it compare with the other NVidia drive by gmack · · Score: 1

    They work better than ATI's fglrx but they are far from amazing. NVIDIA has working 64 bit support and they only seem to lag the kernel by a little bit.

    Need to boot an old kernel? You will need to reinstall your NVIDIA drivers. Install a new kernel? Reboot then install the new drivers instead of compiling everything in advance. It also won't install while X is running and if the Direct Rendering module isn't loaded X won't start either.

  106. Thank you by polle404 · · Score: 1

    Thank you.

    not much more to say, really.

    --

    ~men are from earth. women are from earth. deal with it.~
  107. Re:How does it compare with the other NVidia drive by Anonymous Coward · · Score: 0

    You heard wrong or are just outright lying. I choose the latter because you reek of fanboi.

    ATI and Intel both release documents for their hardware and actively work on X.org themselves to ensure that their hardware works. The default ati and intel drivers in Mesa are maintained by their respective companies now. Every release gets magnitudes more performance with their cards as X.org becomes more capable of taking full advantage of them. Meanwhile, nVidia becomes even more of a headache for everyone because other people have to do the work to maintain backwards compatibility with their binary blob.

    nVidia simply refuses all efforts to be a good neighbor, or even act competitive in this case. But because many of the Linux world still live in 2005, when ATI (pre-AMD buyout) refused to give Linux any attention, still cling to your nVidia blob driver regardless, they have no incentive. Plus, if you haven't noticed, Linux isn't exactly the gaming market nVidia is after anyway.

    OPs opinions are not the facts, they are just FUD. nVidia is the very distant last in terms of Linux support. Even VIA does more to support Chrome. You can find actual benchmarks around http://www.phoronix.com/

    Disclaimer: I rub elbows with some people from X.org.

  108. Several companies competing for support dollars by tepples · · Score: 1

    Please show me a patch by someone who isn't paid very good by a large company to do that.

    The advantage of free software is that you can pay any sufficiently large company, like Canonical, to do the work. Non-free software limits you to a single publisher that can refuse outright to do the work.

  109. Re:How does it compare with the other NVidia drive by andreyvul · · Score: 1

    I actually notice the difference between under 9000 and over 9000 FPS, you insensitive clod!

    --
    proud caffeine whore
  110. Re:How does it compare with the other NVidia drive by Draek · · Score: 1

    Hence the "unsupported hardware platforms" bit, I believe. Whenever we Free Software supporters use the phrase, for some reason most people automatically think "NetBSD on a toaster" instead of "64-bit CPUs". Hopefully as 64-bit CPUs increase in use and more people get screwed by lazy manufacturers, they'll start to take notice.

    --
    No problem is insoluble in all conceivable circumstances.
  111. Re:How does it compare with the other NVidia drive by Draek · · Score: 1

    ATI's closed-source driver is Hell on Earth. The one that shipped with Ubuntu 9.04 caused the first crash I had ever seen of a Linux machine not caused by hardware problems *five* minutes after first use, and even the latest ones still have problems playing full-screen video.

    The open-source drivers on the other hand are reliable, solid, and even support 3D acceleration in many (though not all) chips. The only problem is that it's development is pretty fast so if you don't run a rolling release distro ala Arch or Gentoo, you end up reading about lots of cool features you can't try 'til your distro bothers with making a new release.

    --
    No problem is insoluble in all conceivable circumstances.
  112. Re:How does it compare with the other NVidia drive by wall0159 · · Score: 1

    I would say that the Nouveau project is evidence that there are some users who would trade some convenience for freedom.

  113. Re:How does it compare with the other NVidia drive by Anonymous Coward · · Score: 0

    I have an ATI card with Ubuntu 9.10. It is a nightmare. It is not stable with the open driver and it simply does not work with ATI binary driver.
    I used to have a laptop with NVidia, it was easier and more stable.

  114. Re:How does it compare with the other NVidia drive by Pecisk · · Score: 1

    Actually AMD/ATI has changed their attitude in last two years and have pushed lot of stuff which is obsoleted from binary driver into open source 'radeon' one, giving docs to developers along the way. It might be slow and some cards in the middle are left out in the cold (when distros and Xorg fail to include open sourced support in time), but it's improving. For two years I had laptop which has ATI Radeon Mobility X1600 (I think), which support in binary drivers starting with Ubuntu Gutsy was nightmare. And vola, in next release support was already in free 'radeon' driver! I was quite surprised.

    In result, I have a little bit more faith in AMD than with Nvidia, which in fact have no plans at all to ever opensource or provide docs for any cards, even stuff they don't even sell anymore. Therefore nouveau effort is that important. When Nvidia will be gone or they will drop support for old cards, there will be nowhere else to go but with nouveau.

    --
    user@ubuntubox:~$ stfu This server is going down for shutdown NOW!
  115. Re:How does it compare with the other NVidia drive by Eunuchswear · · Score: 1

    The "official" Nvidia driver doesn't work in IBM 370 based systems.

    Not all the world is x86 you know.

    Also there's this little idea of "Freedom" (or "Openness" if you're an ESR fan).

    --
    Watch this Heartland Institute video
  116. Re:How does it compare with the other NVidia drive by mikechant · · Score: 1

    Linux developers can arbitrarily stop supporting whatever they damn well don't feel like supporting any longer so they can go program extra functions into their USB foam dart cannon. Just sayin'.

    True...but lots of old drivers probably just need recompiling to keep working fine, with no real 'support' required. That's probably why my CanoScan N650U still works with the latest Linux kernel despite not being supported by any Windows version after XP.

  117. Re:How does it compare with the other NVidia drive by Anonymous Coward · · Score: 0

    Do you think any end user cares?

    I care, alot. Lemote Yeeloong + coreboot + GNU GRUB + gNewSense.

  118. Re:How does it compare with the other NVidia drive by thegrassyknowl · · Score: 1

    Experience shows that if you trust the manufacturer will release updated drivers when they become needed, you're going to get screwed sooner or later

    Or if you want to run on unsupported platforms (like FreeBSD 8.0-AMD64) you're S.O.L. NVidia seem to have zero interest in fixing the deficiency of their closed source driver there, and the FreeBSD people seem to have little interest in implementing a couple of Linux-specific features just so the driver will boot.

    A good free version can be ported to wherever it's needed and might mean I can finally use FreeBSD-64 on my desktop like I'd like to.

    --
    I drink to make other people interesting!
  119. Re:How does it compare with the other NVidia drive by Hurricane78 · · Score: 1

    Sorry, but you are flat out false! :) Unintentionally, but still.
    I know what you meant. But you misunderstood me.

    The 4800 series did not even exist in 2007 back then. And that is what I was talking about.

    Because compositing did not work before version 9.8 for those cards. That version is only some weeks old.

    --
    Any sufficiently advanced intelligence is indistinguishable from stupidity.
  120. Re:How does it compare with the other NVidia drive by Hurricane78 · · Score: 1

    Hey moderator! Have you even had contact with the driver developers? Have you debugged the drivers? Have you sat for a whole week straight, trying to get pre-9.8 drivers to work with a 4800? I have!

    Come here to my place! I’ll show you that every single of my statements is a fact! Don’t hide in your basement! Get here! And then you can think about moderating topics you don’t know shit about!

    So don’t tell me I’m freakin’ trolling you trollerator!

    --
    Any sufficiently advanced intelligence is indistinguishable from stupidity.
  121. Got burned by adopting early by drinkypoo · · Score: 1

    I got an nVidia GT 240 1GB and it's not supported by even the BETA Linux driver, it works but I'm getting 3d trashing and GLX failures that I wasn't getting with my 9600 GT. In the 180 driver, it wasn't even correctly identified... it seemed slow, but wasn't so crashy. Very odd.

    Anyway, nVidia isn't always the answer. Either way, you have to buy older hardware, not the latest stuff, if you want it to work properly on Linux.

    --
    "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
  122. Re:How does it compare with the other NVidia drive by mvdwege · · Score: 1

    In case you hadn't noticed, ACPI is a fairly crucial part of the system on a laptop. Unless you enjoy getting your legs fried and no battery lifetime at all. And yes, nVidia's power-saving functions are severely broken. Last time I tried the official driver on a laptop, I found out I had to run a uniprocessor kernel to get any support for ACPI at all. If you think giving up half your CPU is a good trade, I have only one description for you: fanboi.

    Mart

    --
    "I know I will be modded down for this": where's the option '-1, Asking for it'?
  123. Re:How does it compare with the other NVidia drive by mvdwege · · Score: 1

    [ATI releasing specs] hasn't lead to someone creating high quality free drivers for ATI cards.

    Are you kidding? The radeonhd driver is just fine. It doesn't support all chipsets 100% just yet, but the ones that are supported work great. Unless of course you define 'high quality' as 'highest framerates possible and damn the stability of the system' like nVidia fanbois tend to do.

    Mart

    --
    "I know I will be modded down for this": where's the option '-1, Asking for it'?
  124. Re:How does it compare with the other NVidia drive by True+Grit · · Score: 1

    Actually AMD/ATI has changed their attitude in last two years

    ATI 'changed their attitude' the day after AMD bought them. :)

    AMD has always been a more open/FOSS friendly company than either NV or ATI.

    The buyout happened more than 2 years ago, its just taken some time for AMD to sort through everything and get up to speed with their support of non-Windows on the graphics side of things, since after all, they basically had to start from scratch (ATI was just as uninterested in non-Windows arches as NV still is).

  125. Re:How does it compare with the other NVidia drive by True+Grit · · Score: 1

    Hey moderator! ... every single of my statements is a fact

    This one isn't:

    The Linux “team” of ATi is a one-man-show, and focuses only on workstations. Everything else is simply ignored.

    You missed the part where AMD has devs working on the open driver as well.

    As for the rest of your rant, everyone knows the fglrx driver sucks on Linux. What has changed is that ATI is now a subdivision of AMD, and the future of ATI graphics on Linux will be the open driver that is being (rapidly) developed as we speak, and in the end it'll be much better than anything NV is willing to provide.

    Anyone brave enough to use that in-development version of AMD's open driver already knows what the future is going to look like. Give AMD another year or two to stabilize the open driver and bring it up to speed on chip support, performance tweaking, and handling corner-cases, and once that work makes it into the mainstream Linux releases, they will end up changing the world of Linux graphics support (as we've known it) forever.

    After all, their driver will be entirely found within either the kernel (KMS) or Xorg itself (DRM/Mesa), so you'll no longer need a separate binary blob/package just to get hardware-accelerated 2D & 3D (anyone with AMD-ATI hardware will thus get all this goodness right out of the box, as soon as they install Linux). And thats just for starters...

  126. Re:How does it compare with the other NVidia drive by True+Grit · · Score: 1

    They're certainly nowhere near as Linux-friendly as Nvidia.

    NV is Linux-friendly? Does that mean MS is merely 'Linux-neutral'? :)

    No offense, but I think your definition of what a friend is... needs a little work.

    As for AMD-ATI, give them another year or so, and check their open driver again. I'm sure that you'll be pleasantly surprised then (I'm using that in-development open driver now).

  127. Re:How does it compare with the other NVidia drive by True+Grit · · Score: 1

    I certainly hope you are trolling because if not, this is a new level of ignorance that I was not aware existed.

    Oh I don't know, I usually find ignorance and trolling actually go hand-in-hand...

    :)

  128. Re:How does it compare with the other NVidia drive by Sir_Lewk · · Score: 1

    The good trolls are often pretty funny, or at least clever in construction in my experiance. Generally I consider unintelligent trolls to be more flamebait.

    --
    "linux is just DOS with a UNIX like syntax" -- Galactic Dominator (944134)
  129. Re:How does it compare with the other NVidia drive by Minwee · · Score: 1

    It's more like if you had a car but the radio doesn't work. You can fully enjoy the car but a feature of the car that is completely disconnected from the core purpose of the vehicle doesn't work.

    But ACPI isn't completely disconnected from the core purpose of a notebook computer. The effect is more like having the engine fall out onto the street every time you put the car into reverse than not being able to listen to "Queen's Greatest Hits" while driving.

  130. Re:How does it compare with the other NVidia drive by True+Grit · · Score: 1

    The good trolls are often pretty funny, or at least clever in construction in my experiance. Generally I consider unintelligent trolls to be more flamebait.

    I agree, the clever ones are actually very entertaining. Unfortunately, for every intelligent troll, there seems to be at least 50 moronic ones, hence my post.

    As for the 'troll' & 'flamebait' monikers, alas there is no common standard definition for them. Besides, from what I've seen here, intelligent trolls are rarely moderated down anyway (I guess everyone is so tired of the dumb ones that we 'reward' the smart ones), so 'troll' & 'flamebait' are largely being used interchangeably (even though they shouldn't).

  131. Re:How does it compare with the other NVidia drive by Anonymous Coward · · Score: 0

    I Think the correct word would be more open, nvidia supports Linux reasonably well considering its marketshare, AMD/ATI just choose a more effective solution, by releasing specs they don't have to bother with supporting every OS on the planet, While it might take a bit more time for good drivers for all cards to appear for less used platforms atleast everyone has been given the tools needed to make it happen, Companies like Microsoft and Apple can even write their own drivers for old AMD/ATI hardware if it becomes necessary in the future.

    So AMD/ATI isn't more Linux friendly, they're just more friendly towards everyone.

    I still use nvidia though since they had the best proprietary Linux drivers when i last upgraded and the ATI opensource drivers weren't good enough for me.

  132. Re:How does it compare with the other NVidia drive by bjourne · · Score: 1

    They don't provide hardware accelerated 3D graphics last time I checked, which is a fairly major flaw if you ask me.

  133. Re:How does it compare with the other NVidia drive by mvdwege · · Score: 1

    Erm. They do. Just not on all chipsets yet, as I already said. When was the last time you checked? Last century?

    Mart

    --
    "I know I will be modded down for this": where's the option '-1, Asking for it'?
  134. Re:How does it compare with the other NVidia drive by True+Grit · · Score: 1

    AMD/ATI just choose a more effective solution, by releasing specs they don't have to bother with supporting every OS on the planet, While it might take a bit more time for good drivers for all cards to appear

    Except releasing specs isn't the only thing they're doing. They've also got at least 3 of their own devs (AMD employees) working on the new driver stack. Once that new stack hits the mainstream, people's perceptions of AMD support for Linux are going to change dramatically (I'm using that new stack now).

  135. Re:How does it compare with the other NVidia drive by Ginger+Unicorn · · Score: 1

    so a more appropriate car analogy would be having a car with the drive belts missing. You can drive it around for a short period of time, but after that the engine gets very hot and the battery goes flat.

    --
    (1.21 gigawatts) / (88 miles per hour) = 30 757 874 newtons