Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 



Forgot your password?
typodupeerror
×
Graphics Software Linux

AMD Releases Open-Source R600/700 3D Code 307

Michael writes "AMD has just released code that will allow for open-source 3D acceleration on their ATI R600 and R700 graphics cards, including all of their newest Radeon HD 4xxx products. This code consists of a demo program that feeds the commands to the hardware, updates to their RadeonHD driver, and a Direct Rendering Manager update. With this code comes working 2D EXA acceleration support for these newer ATI graphics processors as well as basic X-Video support. AMD will be releasing sanitized documentation for these new ATI GPUs in the coming weeks. Phoronix has an article detailing what's all encompassed by today's code drop as well as the activities that led to this open-source code coming about for release."
This discussion has been archived. No new comments can be posted.

AMD Releases Open-Source R600/700 3D Code

Comments Filter:
  • Re:Hallejulla! (Score:2, Informative)

    by AndrewBuck ( 1120597 ) on Tuesday December 30, 2008 @12:35AM (#26265247)

    If I recall correctly this isn't the first code ATI has released and hopefully it won't be the last. I think we are beginning to see companies starting to realize that although there may not be a huge number of linux users, we sure do buy a lot of computer hardware.

    -Buck

  • Re:this is either (Score:5, Informative)

    by QuantumG ( 50515 ) * <qg@biodome.org> on Tuesday December 30, 2008 @12:38AM (#26265267) Homepage Journal

    Or just good economical sense.

    "Hey Bob, these kids on the Internet want to write Linux drivers for our cards."
    "Oh really? Have we had any customer requests for Linux drivers lately?"
    "Yeah, a couple."
    "Send 'em that dev code we did last week, see what they come up with."
    "Ok."

    Shocking!

  • Re:Hallejulla! (Score:2, Informative)

    by In hydraulis ( 1318473 ) on Tuesday December 30, 2008 @12:42AM (#26265287)
    I have an R350 chipset, you insensitive clod!

    Seriously though, I'm not seeing much progress with respect to older processors. FTFA,

    Two weeks after the initial R500 3D documentation release, AMD had released an R300 3D register guide. This programming guide concerning their older graphics hardware was previously only available through Non-Disclosure Agreements to select developers.

    Well, so far my experience with the open source R350 drivers is lukewarm. They do work to an extent, in that they can run Tux Racer and its forks, but FlightGear remains beyond their capabilities.

  • by Timothy Brownawell ( 627747 ) <tbrownaw@prjek.net> on Tuesday December 30, 2008 @12:49AM (#26265323) Homepage Journal

    I realize the above is a troll but what is he referring to with the licensing fee? I've seen this in a few stories and have always wondered what it was.

    A while back, SCO tried to claim that they owned Linux, and that anyone using it had to buy licenses at $699 each (I think this may have been related to their lawsuit against IBM, before Novell stepped in). A couple of companies actually paid up, and were duly ridiculed here.

  • by LoRdTAW ( 99712 ) on Tuesday December 30, 2008 @01:00AM (#26265375)

    I am looking forward to see what this means for Linux, OpenCL and other GP-GPU goodies. With OpenCL working along side OpenGL, a tightly integrated kernel ATI driver that handles the GP-GPU/OpenCL stuff we will really see some interesting stuff come our way. To my understanding OpenCL allows someone who is writing an algorithm to implement it in OpenCL and let OpenCL take care of diving up the work load between GPU's and CPU cores. Damn I am really excited to see the OSS community tie all this stuff together and release the computing power of the GPU to more general yet compute intense applications.

    A system with a quad core CPU and four ATI cards would be a force to be reckoned with! Fast trans-coding/cracking of Blu-ray, rapid key sniffing for air crack, even networked applications could be sped up like IPsec and SSH. We could have fast rendering in blender and ray tracing can be done with high precision as well as speed (maybe even real time!). Gimp plug-ins can be given a boost in speed and video editing a breeze. Even a laptop with a slower dual core could benefit from its on board GPU's number crunching power. Useful for cracking WEP/WPA keys.

    And AMD/ATI arent the only ones getting on board the OpenCL bandwagon, Apple developed it, and Intel along with Nvidia are also going to support it. So OpenCL will allow us to run our apps on the hardware of our choice.

                     

  • Re:this is either (Score:5, Informative)

    by Tubal-Cain ( 1289912 ) * on Tuesday December 30, 2008 @01:06AM (#26265401) Journal
    I think they know how to write off such a contribution as charity.
  • Re:Hallejulla! (Score:3, Informative)

    by afidel ( 530433 ) on Tuesday December 30, 2008 @01:43AM (#26265557)
    LOL, I haven't had a BSOD from nvidia drivers since the early days of XP and at least their driver's don't require the freaking bloated .NET CRL to even install! In fact I can't remember a time in the last 4 NVidia cards I've owned (going back to a Ti-4200) that I had an issue related to the driver.
  • FAQ (Score:5, Informative)

    by MostAwesomeDude ( 980382 ) on Tuesday December 30, 2008 @01:50AM (#26265585) Homepage

    Based on what's been on IRC in the past few hours.

    Q: Wait, what?

    A: Code for radeonhd and the kernel providing acceleration for Radeon HD 2400 and newer. Kernel parts are already pretty much integrated; radeonhd is integrated as well, although stuff still needs to be copied to radeon.

    Q: So what does this mean for the user?

    A: EXA means faster GUI responsiveness. Xv means fast video. Kernel DRM is the basis for all acceleration unification (OpenGL, etc.)

    Q: Speaking of OpenGL...

    A: Lawl, no. Not for another few months. Most of the code we're gonna write will target Gallium, so--

    Q: Gallium?

    A: Gallium is the next generation of GPU acceleration. Once we get drivers ready, it'll be awesome. Linky to TG: http://www.tungstengraphics.com/wiki/index.php/Gallium3D [tungstengraphics.com]

    Q: So this is just docs and some basic code?

    A: Nope, no docs. AMD couldn't agree on docs before their vacation time, so I guess we'll see those in a month or so. On the other hand, we've got enough here to do a lot of stuff. It'd be nice if we had more devs, though. :3

    Q: So why is there only code for radeonhd? Will radeon support this too? Why two separate drivers?

    A: The reason for two separate drivers is a very long and largely silly story. I don't feel like repeating it, and I probably couldn't tell it fairly anyway.

    I'll get radeonhd code ported over to radeon once my vacation's over, assuming nobody does it sooner. I can't do the HDMI audio setup without testing hardware, though; does anybody want to donate an HDMI audio-enabled monitor? :3

    ~ C.

  • Re:Hallejulla! (Score:3, Informative)

    by GigaplexNZ ( 1233886 ) on Tuesday December 30, 2008 @01:59AM (#26265619)
    I've been getting a few BSODs lately regarding the Nvidia driver. Just because you haven't seen any on your particular machine doesn't mean the code is perfect. Also, you don't need .NET to install the ATI drivers. You only need it for the catalyst control panel.
  • by Anonymous Coward on Tuesday December 30, 2008 @02:41AM (#26265765)

    Radeon HD 4670 is ~$80 and will play most games, period.

  • Re:Dammit (Score:5, Informative)

    by innocent_white_lamb ( 151825 ) on Tuesday December 30, 2008 @02:46AM (#26265787)

    Eh? Intel has had fully open-source drivers available for quite some time now. ATI is currently playing catch-up in that regard. (And Nvidia isn't playing at all.)

  • Re:Wow (Score:2, Informative)

    by innocent_white_lamb ( 151825 ) on Tuesday December 30, 2008 @02:49AM (#26265799)

    whatever happened to noveau anyway?
     
    It's still being worked on, apparently. http://nouveau.freedesktop.org/wiki/ [freedesktop.org] The last update was on November 16, so it's not being worked on really fast....

  • by dfn_deux ( 535506 ) <datsun510&gmail,com> on Tuesday December 30, 2008 @04:54AM (#26266207) Homepage

    The parent has pegged a round hole with a square question. Hardware support in Linux works well if you build your own machines, or happen to get one with supported hardware. How do you find a system that is fully supported and for which distributions?

    Anything with an Intel Centrino logo /should/ have a full array of linux supported hardware. The intel centrino chip "package" includes wifi, video, cpu, acpi, sata, and sound all with known working mainline kernel supported hardware. Not that I work for or endorse their products necessarily, they just happen to be the only vendor who has bothered with providing the code, documentation and (in the case of their wifi chipset) firmware for all the same hardware that they include in their logo certification program. Probably not the top of the line hardware, especially the video, but it's hard to argue with a product that fits so neatly into the HCL for any recent linux distro.

  • Re:FAQ (Score:4, Informative)

    by Kjella ( 173770 ) on Tuesday December 30, 2008 @06:49AM (#26266577) Homepage

    Since you're a Mesa/X.org dev, maybe you can answer this.
    Why has there apparently been no interest in implementing XvMC?

    XvMC only supports MPEG2 acceleration, it was designed around that waaaaay back and would have to undergo major changes to support anything else. Since pretty much every computer can do MPEG2 with both hands tied behind their back, even HD MPEG2 as in ATSC, HDV and a few HDDVD/Blu-Rays, there's very little interest. What everyone wants is H.264 / VC-1 acceleration so you can play back modern media like Blu-Ray, AVCHD and almost everything off the net.

    So far Intel has talked a little about extending XvMC, someone made VA API that lacks implementation, AMD has been mumbling about XvBA which is a copy of the DirectX video acceleration support for their closed driver and no agreement on whether it'll be supported in open source at all, some effort to implement it as generic GPGPU algorithms, but in the end the only way you'll have full working video acceleration under Linux today is having a nVidia card and the proprietary drivers which support VDPAU. They've gone from zero to hero on this in no time, but it's now in varying degrees implemented on mplayer, ffmpeg, MythTV, xine and VLC. Shortly all of these will support it in an official release, the 180.xx driver introducing this still doesn't have a stable release either.

  • Re:Hallejulla! (Score:5, Informative)

    by RMingin ( 985478 ) on Tuesday December 30, 2008 @10:11AM (#26267533) Homepage

    Problem is that the drivers you're referencing and the Carmack's comments on them date from around 2000. Lots has changed in the meantime. FWIW, the Carmack was referring to Rage128 era hardware/software, which was one unusable ball of software workarounds for hardware bugs and hardware workarounds of legacy software bugs. ATI threw it all away and started fresh roughly around the time they ditched the Rage architecture and had released drivers on the newer codebase when they released the second-gen Radeons. The hardware wasn't fully new-gen and pretty until roughly Radeon 9700.

    They repeated the process on a smaller scale again roughly the time the X1K cards were released (software restart) and around the time the HD2K cards came out (completely new hardware generation).

    This is all just a lot more info than you needed, but the simple answer is 'Yes, everything has changed since the paleolithic quote's time. Twice'.

    The current quality of ATI/AMD's Windows drivers is debatable, but I'd be entirely comfortable saying 'they are very comparable in quality to Nvidia's current drivers'.

  • Re:FAQ (Score:3, Informative)

    by MostAwesomeDude ( 980382 ) on Tuesday December 30, 2008 @10:20AM (#26267587) Homepage

    Since you're a Mesa/X.org dev, maybe you can answer this.
    Why has there apparently been no interest in implementing XvMC?

    XvMC requires very specific hardware support (which AMD hasn't been able to get legal clearance for) or complex shaders. The latter is already in the Gallium tree, but we don't have a working driver that can run it yet.

    Also, each of the big three has gone ahead and crafted their own goddamn standard. Intel's VAAPI, AMD's XvBA, and nVidia's VDPAU. Eventually, at least one of those will probably be added to Gallium. (Probably Intel's pick, since they put so much money into this.)

    ~ C.

  • Re:Hallejulla! (Score:3, Informative)

    by kv9 ( 697238 ) on Tuesday December 30, 2008 @12:27PM (#26268613) Homepage

    There is no reason we- the OSS community, can't have the best drivers, like we have the best web browsers.

    Opera isn't open source.

  • Re:Hallejulla! (Score:4, Informative)

    by WiiVault ( 1039946 ) on Tuesday December 30, 2008 @06:31PM (#26273033)
    Who said Opera? I was refering to Gecko and KHTML/ Webkit.
  • Re:FAQ (Score:3, Informative)

    by Kjella ( 173770 ) on Tuesday December 30, 2008 @09:15PM (#26274537) Homepage

    CoreAVC can let you playback bluray-quality h.264 on an Atom 330 without too much stress (I assume a 230 and the Nano can handle it just fine too).

    1. No, it doesn't.
    2. No, it doesn't.

    I had to check it out since you claimed it... the 330 can barely decode a 1080p RIP with about 20-25% the bitrate of a Blu-Ray. Furthermore, CoreAVC is well threaded so with about 50% of the power the 230 wouldn't even be able to decode the rip without stuttering. In short, there's still a very good market for hardware decoders for many years to come.

    What happens when they're "old shit" like MPEG2?

    Well, there's been no significant new codec for the last five years and as it's now heavily entrenched in Blu-Ray and many HD broadcasts that will take a very, very long time to replace. Probably much longer than DVD, maybe if you're talking in a 20+ years timeframe something new and better will appear but for now it seems fairly close to optimal, close enough that a revolutionary improvement is highly unlikely.

Remember, UNIX spelled backwards is XINU. -- Mt.

Working...