Announcement

Collapse
No announcement yet.

People incorrectly assume that AMD drivers suck. They don't.

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • #41
    Originally posted by RussianNeuroMancer View Post
    I doesn't check this one, but you may try to google for "Pole to Pole mkv".
    Provide a direct link, please. I'm in no mood to trawl through cyberlocker indexers before getting to a valid link. Can even be a torrent if need be, though you should probably give that one in a PM.

    Originally posted by RussianNeuroMancer View Post
    Because different people may notice tearing, or may doesn't notice it. If someone doesn't see tearing (and need special samples that expose issue) that doesn't mean there is no tearing.
    That's actually a valid argument. Why do you always have to go through several posts of trolling before you make a valid point? Though there's still the part where you were wrong about the Nvidia devs.

    Comment


    • #42
      Originally posted by RussianNeuroMancer View Post
      Try to search for latest news about upgrade 10.7 to 10.8. A lot of software doesn't work anymore, even some OS functions stop working. Oh, by the way: simple page scrolling in web-browser on latest MacBook with Retina give FPS below 30. I doesn't think this is what a Desktop should be.
      I would, but all of my issues have already been resolved by Applecare. You see, Mac users don't have to "search for the latest news" when we have an issue. We simply call Applecare. It's a completely different ecosystem and philosophy.

      Fans running at 100% all the time? Call Applecare.
      Superdrive making a motorboat sound? Call Applecare.
      Batman AA crashes? Call Applecare.
      Keyframe animation borked in Finalcut? Call Applecare
      Font have a kerning issue? Call Applecare

      I find the majority of Mac-whine-blogs out there are customers that have recently migrated from the PC world, and do not yet have an understanding of how the Apple support model works. This is not to say that their issues don't exist or are invalid, but that that are choosing to whine rather than accept the assistance that they paid for.

      F

      Comment


      • #43
        Originally posted by Gusar View Post
        I'm in no mood
        Direct link special for you.
        Originally posted by Gusar View Post
        Why do you always have to go through several posts of trolling
        Oh my, then "I'm in no mood" is trolling too.
        Originally posted by Gusar View Post
        before you make a valid point?
        Because this valid point is obvious. Honestly, I can't image someone can not understand such simple things.

        Originally posted by russofris View Post
        I would, but all of my issues have already been resolved by Applecare. You see, Mac users don't have to "search for the latest news" when we have an issue. We simply call Applecare. It's a completely different ecosystem and philosophy.
        Okay, let's take a look at this thread: https://discussions.apple.com/thread/3203487
        First post:
        I have problem with wake up from sleep.
        when i just use my mac, it works normaly
        but after wake from sleep, graphic performance
        down to about half.

        for example i got 70 fps around. in WOW.
        after wake, it drops to 20 fps around.
        Cinebench 11.5 shows 16fps after wake.
        but 30 fps from first reboot condition.

        Mac is 2008 2.8 octacore with Radeon 5770
        lion looks heavy than SL. more heavy feeling to
        click something. it reacting bit slowly than SL
        and this made feel heavy. please fix this problems.
        Latest post:
        A year later, 10.7.4, and this is still an issue. Seriously Apple?
        No, Apple support is doesn't work as you think.
        Last edited by RussianNeuroMancer; 07-29-2012, 12:21 PM. Reason: Add answer to russofris message

        Comment


        • #44
          Originally posted by Rallos Zek View Post
          From a post taken of Slashdot...



          http://linux.slashdot.org/comments.p...7&cid=40776403

          I seen posts and blogs about this issue before, stuff written for Nvidia hardware sucks on AMD/Intel hardware cause Nvidia cares not about specs and correctness, just speed and hacks in their OpenGL stack.


          Thoughts?
          There are reasons Nvidia's "The way it is meant to be played" campaign is oft referred to as "The Way it was meant to be code sabotaged"

          Put bluntly, Nvidia makes semi-decent hardware. The performance per watt ratio is generally awful, and the feature per die space ratio also leaves something to be desired when compared to competitor products from XGI, ATi, ATi/AMD, Via/S3, SiS, Matrox, PowerVR, Trident, Intel, and even going back to 3DFX.

          Pub bluntly, Nvidia makes awful drivers and software development kits.

          Nvidia loves to leverage proprietary OpenGL calls, the most famous possibly being UltraShadow as used in the IDTech3 engine. Nvidia's love of non-standard graphics calls were one of the reasons City of Heroes never ran right on AMD/ATi graphics cards. People on the CoH forums loved to complain that the problem was the ATi/AMD drivers... but ATi/AMD was not the culprit. Nvidia's deliberate code level sabotage was the culprit.

          Nvidia loves proprietary / non-free development. We've already heard about the business contracts lost because Nvidia has no open-source strategy. Yes, let us be clear here, Nvidia-GLX is not an open-source strategy. Nvidia has already been called out by the /Linux kernel developers and the Linux Foundation... and managed to get itself branded as the worst company the Linux Foundation had ever worked with; including a very explicit "Fuck You" from Linus T. himself.

          * * *

          What Nvidia does have... is marketshare: http://store.steampowered.com/hwsurvey/

          As of June 2102 Nvidia based cards make up over 47% of the graphics cards used by those participating in the Steam Survey.

          Nvidia tends to wield this marketshare like a club, using it to force developers to use Nvidia SDK's and Nvidia products when developing commercial games. The result is the code-level sabotage in recent games like Batman: Arkham Asylum and Arkham City.

          The result is a Microsoft-like cycle. People buy Nvidia cards because Nvidia has marketshare and that marketshare results in software that is written to perform better on Nvidia hardware: Which fuels the marketshare.

          By the same token. People who use Nvidia cards on /Linux do not do so because they support /Linux development or open-source ideals... and if you say do ... you do not. End of Story. Can it Deary.

          To be blunt, if you support /Linux development or open-source ideals, you would not be using Nvidia cards at all. You would be using either Intel or AMD cards on x86-platforms.

          Developers who implement software applications that make calls to Nvidia API's generally do so because, again, Marketshare.

          * * *

          Admittedly both Microsoft and Khronos have taken steps to reign in Nvidia's shenanigans. DirectX 11, OpenGL 3.x, and OpenGL 4.x have stricter definitions for supported calls, as well as supporting more functions in the core specification. Standards like OpenGL ES 2.0 and ES 3.0 truncate the list of API supported calls even further, giving less room for proprietary problems.

          Admittedly Nvidia has also been quicker to adopt new protocols or functions compared to Intel or AMD. VDPAU is a very good example of this. Nvidia had their API out and ready for usage... and leveraged marketshare like a club. So, instead of having a universal standard for Video Acceleration API's... we have a proprietary standard being resold as an open-standard.

          Admittedly AMD and Intel have been slow to react to other advances in the open-source market. Case in point would be Kwin+OpenGL on AMD hardware: http://blog.martin-graesslin.com/blo...gacy-hardware/

          * * *

          However, these last two situations are slowly being addressed. AMD and Intel have an open-source strategy and are carrying their strategies out by fixing problems, getting appropriate specifications released under Open-Licenses, and generally being good Open-Source citizens.

          Comment


          • #45
            Originally posted by RussianNeuroMancer View Post
            Har Har.

            Now you've proven 100% that you're not here to have an actual conversation, but only to troll. What's the matter? Are you afraid that if you provide the link, the video will decode perfectly on my machine?

            Comment


            • #46
              Originally posted by Gusar View Post
              Are you afraid that if you provide the link
              You are afraid to click first link in Google? What's the matter?
              Last edited by RussianNeuroMancer; 07-29-2012, 03:05 PM.

              Comment


              • #47
                This actually fixed in Catalyst 12.6.

                Comment


                • #48
                  Originally posted by RussianNeuroMancer View Post
                  You are afraid to click first link in Google? What's the matter?
                  Yep, you *are* afraid that the video will decode fine on my machine.

                  Oh, and just BTW, the first link on google is this: http://www.filestube.com/p/planet+ea...80p+sample+mkv <- like I said, I'm not gonna trawl through cyberlocker indexers, just because *you* are afraid that I'll *again* prove you wrong.

                  Comment


                  • #49
                    You funny guy indeed. Good luck in your fighting with fear of links and trollphobia.
                    Last edited by RussianNeuroMancer; 07-29-2012, 03:31 PM. Reason: Typo fix

                    Comment


                    • #50
                      Originally posted by nej_simon View Post
                      That could be true. But if I have to choose between a driver that follows specs but doesn't work properly and a driver that uses various hacks but does work and is stable I'd definitly choose the later. Who cares if the driver is "correct" if it crashes the computer?
                      "Doesn't work properly" is a pretty vague statement though, don't you think? Also that's partly what the quoted text in the first post refers to: that people get the impression the drivers don't work properly, although it's due to sloppy programming in the target application, because nvidia's drivers (to quote another slashdot comment I read a few months ago) "let you get away with murder".
                      Why should you care about correct drivers? Specifications are there for a reason: to allow any company to bring their goods to the table and let customers sort things out. Clean competition is a Good Thing (tm).

                      Comment

                      Working...
                      X