Announcement

Collapse
No announcement yet.

Intel's Poulsbo Driver A Bloody Mess?

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

  • #21
    Thanks for bringing this up.

    To complete this mess, Intel does not have only one badly(non?) functioning, closed-source GMA500 driver... No! They have two closed-sourced, bad functioning GMA500 drivers:

    Intel Embedded Graphics Driver (IEGD)

    Comment


    • #22
      It wouldnt surprise me if there are some customers (either those building PowerVR IP into their SoCs/chipsets/etc or those building actual devices featuring chips containing PowerVR IP) that would rather NOT see open source drivers for the PowerVR stuff in their chips/devices.

      Comment


      • #23
        Originally posted by Ant P. View Post
        Ow... that just completely crushed my enthusiasm for getting a Pandora. My eee may be technically inferior, but at least I know it's supported...
        well the pandora it is supported
        by the imtec drives
        i don't nou way intel did not use the imtec Linux drives
        they are bloop drives still they work far better than the intel tungsten drivers
        p.s pandora people have the drives and they did show a 3d demo

        Comment


        • #24
          Originally posted by argor View Post
          well the pandora it is supported
          by the imtec drives
          i don't nou way intel did not use the imtec Linux drives
          they are bloop drives still they work far better than the intel tungsten drivers
          p.s pandora people have the drives and they did show a 3d demo
          I'm sorry, but if you're referring to the Byterapers' Treed3D 2.0 demo, it's using the crippled*, closed source OpenGL ES 2.0 driver. Intel commissioned the development of their own driver because there was actually no (full) OpenGL implementation available for the SGX series.

          * The PowerVR SGX specifications exceed the OpenGL 3.0 (not ES!) requirements.

          Comment


          • #25
            Originally posted by drag View Post
            Note that the PowerVR SGX stuff is used in much more then just the Dell Mini-12.

            It's used in other architectures. For example the OMAP3 being used in the OpenPandora Handheld and the developer oriented BeagleBoard has the PowerVR SGX chipset.
            Thankfully (it's a mixed bag, though...), the OMAP3 crowd has drivers written by Imagination Technologies and the GPL compatible kernel wedges showed up last month. The Pandora kernel team's working on verification of things right now on that front. We have working 3D that works against modern kernels and distributions...for now...

            So it's especially bad because the proprietary x86 driver won't work in x86-64, nor will it work in the ARM platform. People shipping PowerVR-based graphics in ARM will need their own, seperate, proprietary Linux driver.
            Actually... They will only need ONE driver. The one TI's providing to people right at the moment- there's a GPLed kernel wedge and then two proprietary ES 1.1 and 2.0 API libraries you link against. This is for ANYTHING using the OMAP3 right at the moment, including Beagleboard, Pandora, and Nokia's N9XX series devices.

            It's still a problem, but it's not as bad as you're making out to be. I'd dearly love beyond words to see ImgTec turn around and provide info on how to push the limits on their chip. But, that's not happening and as long as they're doing the "right thing" a' la NVidia, for now, that'll work.

            It's a clusterf*k alright. With a open source driver you'd be able to update your code as well as created a unified driver that will be easily made to work across lots of different devices irregardless of the actual architecture.
            You're preaching to the choir. VIA's apparently figured this one out. Intel's done it too (They can't do it with the GMA500 as it's NOT theirs to do it with, but with the other GMA's and Larabee, they are...). Same with AMD- and it's the one with the most potential and promise on the desktop. It's just Imagination, NVidia, and S3 that're being a pain right at the moment.

            Comment


            • #26
              Originally posted by HyperDrive View Post
              I'm sorry, but if you're referring to the Byterapers' Treed3D 2.0 demo, it's using the crippled*, closed source OpenGL ES 2.0 driver.
              what is so crippled about the driver they were were the beta OpenGL ES
              drives yes they are blop drives whith GPL compatible kernel wedges
              just like Nvidia
              Intel commissioned the development of their own driver because there was actually no (full) OpenGL implementation available for the SGX series.
              * The PowerVR SGX specifications exceed the OpenGL 3.0 (not ES!) requirements.
              did powervr not provided intel whith complete OpenGL reference drivers
              it is up to intel whatever to use them or not
              here is also interesting read http://forum.beyond3d.com/showthread.php?t=50430

              Comment


              • #27
                Originally posted by argor View Post
                what is so crippled about the driver they were were the beta OpenGL ES
                drives yes they are blop drives whith GPL compatible kernel wedges
                just like Nvidia
                Ideology aside (since I personally lean towards "the Stallman way" instead of "the Torvalds way" of doing things), OpenGL ES 2.0 is a strict subset of OpenGL 2.0. According to the specifications (i. e., unless Imagination Technologies is lying), the PowerVR SGX is fully OpenGL 3.0 capable. That's deliberate crippling in my book.

                did powervr not provided intel whith complete OpenGL reference drivers
                it is up to intel whatever to use them or not
                here is also interesting read http://forum.beyond3d.com/showthread.php?t=50430
                No, they didn't, according to the 8th post in the thread you mention.
                Last edited by HyperDrive; 05 February 2009, 09:18 AM.

                Comment


                • #28
                  Originally posted by HyperDrive View Post
                  Ideology aside (since I personally lean towards "the Stallman way" instead of "the Torvalds way" of doing things), OpenGL ES 2.0 is a strict subset of OpenGL 2.0. According to the specifications (i. e., unless Imagination Technologies is lying), the PowerVR SGX is fully OpenGL 3.0 capable. That's deliberate crippling in my book.
                  no it is up to TI what they support or not they did desited to only provided support for opengl es 1.1 and 2.0
                  just like Intel could provided support for opengl es
                  it just depend on what marked you are in what is useful or not
                  No, they didn't, according to the 8th post on the thread you mention.
                  that post dose not say anything
                  the licenser they desited what they want just like ti wanted only OpenGL ES/VG drivers for OMAP3
                  Last edited by argor; 05 February 2009, 01:45 PM.

                  Comment


                  • #29
                    Does anyone have access to this page:
                    https://blueprints.edge.launchpad.ne...ackaging/+edit ? My launchpad account sure does not

                    Comment


                    • #30
                      Originally posted by HyperDrive View Post
                      One thing is for sure, my acquisition of an MSI Wind U115 netbook depends on the availability of a free and open source GMA500 Xorg driver.
                      Just for the record, since I just don't have time/patience/whatever to wait for the end of this "soap opera", I voted with my euros and got myself an Eee PC 901 yesterday. Iff the PowerVR is reverse-engineered, maybe I'll get myself a BeagleBoard to hack around.

                      Comment

                      Working...
                      X