ATi Support on infinityOS

[darkphoenix22]

My suggestion is if infinityOS users with ATi cards want 3D gaming and 3D graphics, my suggestion is to use Windows, as the fglrx drivers in their current state can not even provide full functionallity in those areas.

The quaility of the fglrx drivers pales in comparision to the Nvidia and Intel, and ATi should be ashamed. They are equals to the drivers for theMatrox cards, a comapny that went out of favour 10 years ago.

For now for users with ATi cards who want 3D gaming, I highly advise that they use WIndows. Due to a lack of effort on the part of ATi, sadly they can not currently take full advantage of their hardware on Linux.

[bridgman]

Good to hear the open source drivers are working for you; thanks for giving them a try.

I still encourage ATi to adopt the Radeon driver codebase as the basis of their Linux drivers, but I recognize that you have corporate interests in mind. I can only appeal and ask that you keep the needs of your users in mind, as it is us who will be buying your products and recommending them to others. It is never a good idea to put the needs of your corporation ahead of the needs of your consumers.

I don't understand your last comment - the reason we offer proprietary drivers is not because of "the needs of our corporation".

Some of our *customers* need features and performance which are not feasible to deliver in a Linux-specific code base (because the Linux market share is too small to support the work required) but which can be supported by sharing code with other OSes.

Even if we shut down fglrx work tomorrow and moved all of the Linux-specific developers across to the open source driver stack that would not be sufficient to meet the needs of our 3D workstation customers.

[darkphoenix22]

I don't understand your last comment - the reason we offer proprietary drivers is not because of "the needs of our corporation".

How is not opening up your drivers due to "code-sharing" not due to a corporate interest?

The Radeon guys are doing great for mostly being on their own (admittingly with a bit of help from you guys). Why would the fglrx drivers have to be depricated while you are working on their replacement? Both could easily exist side-by-side for the short term.

PS I was looking over your documentation and I have a quick question. Has ATi released any documentation at all for the the r700 and Evergreen based cards (ie. the HD 4000 and HD 5000 series)?

[bridgman]

If we opened up the proprietary driver code and were unable to supply a robust DRM solution for other OSes we would lose the ability to sell into perhaps 95% of the market and would effectively be out of business. I guess you could call that a corporate interest but I regard that as more of a hard constraint.

Regarding fglrx vs open drivers, you're not listening to what I'm saying. EVEN IF WE MOVED ALL OF THE FGLRX DEVELOPERS TO WORK ON THE OPEN SOURCE DRIVERS (whether in the short term or long term, with a sharp cutover or after a parallel effort) THAT WOULD NOT BE ENOUGH TO MAKE THE OPEN SOURCE DRIVERS COMPETITIVE WITH FGLRX FOR SOME OF OUR CUSTOMERS. The details of how we move the developers is not the issue - Linux doesn't have the market share to support anything like the development effort needed for competitive 3D support - that's why anyone active in the workstation market is sharing code across OSes. Why is it a Good Thing when NVidia does it but an evil greedy corporate thing when we do it ?

Regarding docs, the 6xx and 7xx programming model is close enough that we use the same docs for both. The 3D guide covers the differences between the generations. For Evergreen, the shader core has been documented (that's the longest lead part of the support effort) and we're working on docs for the rest of the pipe now.

http://www.x.org/wiki/DataSheets

Alex would probably not be happy with your comment about "with a bit of help from you guys" since he wrote nearly all of the 2D and video code you have been praising. You might want to take a hard look at the commit logs over the last couple of years before making comments like that.

[darkphoenix22]

Alex would probably not be happy with your comment about "with a bit of help from you guys" since he wrote nearly all of the 2D and video code you have been praising. You might want to take a hard look at the commit logs over the last couple of years before making comments like that.

And I thank him deeply for doing so as I am currently using his work in my media center. If it was not for his hard work and effort, I would still be stuck using my noisy Nvidia 8600 card.

[Armin]

The Radeon guys are doing great for mostly being on their own (admittingly with a bit of help from you guys).

I'm seriously wondering how someone who claims to develop/ maintain a linux distribution can not know that the radeon driver is mainly developed by AMD/ATI (and RedHat). With a little research (git repos) u would know that this driver is highly backed up by AMD/ATI (not meant to undermine the work of the RedHat guys or independent developers like Marek or Corbin).

I followed the whole thread and it seams like you learnt due to the anwsers here that AMD/ATI is developing/ supporting the open source driver.

I also don't understand why AMD/ATI should stop fglrx! They should increase their effort (especially the 2D stuff), because that's the driver people like me (Mechanical engineers) use for there daily work. People that need decend 3D performance rely on fglrx (even though they also curse about the 2D performance :-))!

Cheers
Armin

[darkphoenix22]

Why is it a Good Thing when NVidia does it but an evil greedy corporate thing when we do it ?.

Because the Nvidia drivers actually WORK.

[darkphoenix22]

I'm seriously wondering how someone who claims to develop/ maintain a linux distribution can not know that the radeon driver is mainly developed by AMD/ATI (and RedHat). With a little research (git repos) u would know that this driver is highly backed up by AMD/ATI (not meant to undermine the work of the RedHat guys or independent developers like Marek or Corbin).

I apologize for greatly understating the work ATi has done on the open source Radeon drivers. However, they still can and should be doing much more.

[bridgman]

Sorry, I don't get the connection between driver maturity and good/evil.

NVidia moved to code-shared proprietary drivers for Linux over 10 years ago (not sure of the exact date), while we supported open source driver development or Linux-specific drivers until ~2006, with the first code-shared driver (at least the 3D part) appearing in late 2007. Since then there has been pretty significant process (does anyone disagree ?) and we expect that to continue.

Armin, there is a new 2D acceleration implementation in the works; you've probably seen it mentioned here a couple of times. In the meantime, Felix wrote a "backclear" patch for the X server which replaces the old "no-backfill" patch that distros were previously shipping but which eliminates most of the side-effects that resulted in the no-backfill patch being removed and certain 2D operations becoming very slow on XAA drivers.

[bridgman]

Stinkin' edit limit. No more trips for Michael until that gets fixed

Armin, if you aren't using the backclear patch you should really give it a try. There are pre-patched X servers available for most distros, although you have to hunt around to find them sometimes.