Page 2 of 4 FirstFirst 1234 LastLast
Results 11 to 20 of 38

Thread: Reasons Mesa 9.1 Is Still Disappointing For End-Users

  1. #11
    Join Date
    Feb 2013
    Posts
    112

    Default

    How important is OpenCL support? I've only been following fairly technical Linux websites for a few months and I've only seen LLVM being mentioned or focused from that general category so apart from forgetting about OpenCL I had the impression it either wasn't needed or some other method for the same result was applied.

  2. #12
    Join Date
    Oct 2008
    Posts
    104

    Default

    Quote Originally Posted by DanL View Post
    Personally, the most disappointing thing is radeon VDPAU acceleration, because it's being held up by legal reasons. The fruit's in the tree, but a team of AMD lawyers are guarding it waiting to put their wingtips up anyone's rear who gets near it. Not cool.
    Blame Hollywood and their stupid "Everything must be copy protected even though no sane pirate would bother trying to read content out of the video card buffers therefore documentation for the video decoding hardware would be bad" BS.

  3. #13
    Join Date
    May 2007
    Location
    Nurnberg.
    Posts
    323

    Default

    Quote Originally Posted by DanL View Post
    Unrealistic goal, so of course anyone expecting this is going to be disappointed. I didn't find a million dollars on the sidewalk today either...
    WRONG!

    As i have proven with Q3A, there is no real reason for this to be so. The lima driver will have performance which matches the binary driver, the only thing which can slow us down is the design of Mesa itself, and that can be kicked and kicked until it behaves.

    That some hw designs are depending on software optimization that is the problem of the hw design though.

  4. #14
    Join Date
    Oct 2008
    Posts
    3,130

    Default

    Quote Originally Posted by libv View Post
    That some hw designs are depending on software optimization that is the problem of the hw design though.
    There's a pretty big difference between hardware that costs $5 to produce, and hardware that costs $500 to produce.

  5. #15
    Join Date
    May 2007
    Location
    Nurnberg.
    Posts
    323

    Default

    Quote Originally Posted by smitty3268 View Post
    There's a pretty big difference between hardware that costs $5 to produce, and hardware that costs $500 to produce.
    And that difference is:
    * selling millions of units as opposed to thousands.
    * battery life is massively important so none of the available resources should ever be needlessly wasted.

    This is exactly what made the mali-400 such a contender, and is exactly why i have such an easy time making my free software driver perform equally good or usually even better than the ARM developed binary driver.

    Honestly though, those 495 dollars are not going towards hardware or software design. It is going towards low quantity. And therefor also to halfarsed hw design. If you really believe differently, then you do need some help. Mobile chips simply cannot afford the luxuries of the infinite powerdraw of desktop high-end GPUs and have no option but to shore up either design or to really shore up software (like pvr). And the overhead of maintenance and support of PVR is really killing imaginations business atm.

    Face it. The desktop is dying. The desktop style GPUs are also dying. We can no longer afford to waste electrons, and this means that we better have solid hw. Nvidia and ATI desktop style GPUs, where power draw does not matter, are an evolutionary dead end.

  6. #16
    Join Date
    Nov 2007
    Posts
    1,353

    Default

    I couldnt possibly disagree more There will always be a market for highend GPU's... The only real question is how big that market will be.....

    It's -EXACTLY- that same attitude that fucked AMD in the CPU market... It was believed that competing with Intel at the very high end wasnt a very good idea because the market was shrinking... But in the end since they failed to develop a high end design, they didnt have anything to scale down and offer to mid and low end customers.... Now they are being forced to scale up.... And that is infinitely harder than scaling down. I sure as hell hope that they don't make the same mistake in the GPU market.
    Last edited by duby229; 02-21-2013 at 10:43 PM.

  7. #17
    Join Date
    Oct 2008
    Posts
    3,130

    Default

    Quote Originally Posted by libv View Post
    ...
    /end rant.

    Yeah, we get it, you don't like ati or nvidia.

    Anyway, I'm not going to argue with you. I'll just point out that both companies have sunk a lot more money into optimizing their hardware and drivers for performance than the ARM manufacturers have. ARM manufacturers have optimized for cost and power use, which obviously makes sense for their market. But it means that trying to draw conclusions about drivers for one and assuming they are equally true for the other is a risky proposition.
    Last edited by smitty3268; 02-21-2013 at 11:44 PM.

  8. #18
    Join Date
    Jan 2009
    Posts
    1,402

    Default

    Quote Originally Posted by DanL View Post
    Mesa may be behind Khronos, but they will always be behind the spec.
    I'm not sure about that. Intel seems like they've put a great deal behind getting mesa up to ogles3 compliance. It would seem like you would only need another 4-6 full time devs working on mesa to keep abreast of the standard.
    Keep in mind, they only issue updates maybe once per year, and add to that I'd imagine at least some of the mesa developers were kept in the loop as to what the update would consist of (but maybe they had to sign an NDA until the official release).

  9. #19
    Join Date
    Oct 2008
    Posts
    3,130

    Default

    Quote Originally Posted by liam View Post
    I'm not sure about that. Intel seems like they've put a great deal behind getting mesa up to ogles3 compliance. It would seem like you would only need another 4-6 full time devs working on mesa to keep abreast of the standard.
    Keep in mind, they only issue updates maybe once per year, and add to that I'd imagine at least some of the mesa developers were kept in the loop as to what the update would consist of (but maybe they had to sign an NDA until the official release).
    They actually have hardware that supports ogles3, though. That's true up to the current GL spec right now with Ivy Bridge, but typically speaking Intel has not been extremely fast about even having hardware that can run brand new DX versions, let alone having the software drivers ready for it.

    So I think it's probably likely that we will continue to see lags when trying to add features that require new hardware. However, new GL versions that can be implemented on existing hardware (like minor bumps from 4.2 -> 4.3) could be done reasonably quickly.

  10. #20
    Join Date
    Jan 2009
    Posts
    1,402

    Default

    Quote Originally Posted by smitty3268 View Post
    They actually have hardware that supports ogles3, though. That's true up to the current GL spec right now with Ivy Bridge, but typically speaking Intel has not been extremely fast about even having hardware that can run brand new DX versions, let alone having the software drivers ready for it.

    So I think it's probably likely that we will continue to see lags when trying to add features that require new hardware. However, new GL versions that can be implemented on existing hardware (like minor bumps from 4.2 -> 4.3) could be done reasonably quickly.
    I was only taking about recent Intel behavior. They've clearly decided to take Linux graphics a bit more seriously (maybe due to Googles efforts with android and chrome is). Additionally they are taking graphics more seriously by offering pretty decent performance at low power budgets.
    Having said that I agree that lag is still likely even if only because I don't see 4-6 devs being hired to work on mesa exclusively anytime soon.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •