Proprietary SW interfaces and hardware vendors

Article: AMD's Analyst Update
By: Anon (no.delete@this.email.com), February 11, 2012 4:12 am
Room: Moderated Discussions
Anon (no@email.com) on 2/11/12 wrote:
---------------------------
>David Kanter (dkanter@realworldtech.com) on 2/10/12 wrote:
>---------------------------
>>Anon (no@email.com) on 2/10/12 wrote:
>>---------------------------
>>>David Kanter (dkanter@realworldtech.com) on 2/10/12 wrote:
>>>---------------------------
>>>>Anon (no@thanks.com) on 2/9/12 wrote:
>>>>---------------------------
>>>>>David Kanter (dkanter@realworldtech.com) on 2/9/12 wrote:
>>>>>---------------------------
>>>>>>The bottom line is that I don't think Nvidia is in a position to push a proprietary
>>>>>>standard and get it adopted by the broader industry. HPC folks are probably fine,
>>>>>>but I have a very hard time seeing mainstream software developers voluntarily choosing CUDA over the alternatives.
>>>>>>
>>>>>>Now you can argue that perhaps they will choose CUDA + something else, but I'm skeptical.
>>>>>
>>>>>Don't know about AMP and OpenACC, but when discussing potential implementations
>>>>>of an app that would benefit from massive parallelism, the developers with experience
>>>>>that I talked to unanimously said to use CUDA unless you absolutely had to run on
>>>>>non-NVidia HW. They went so far as to suggest developing/debugging in CUDA and then
>>>>>porting to OpenCL if necessary, just because the developer >experience is so much better.
>>>>
>>>>Yes, and that's what I meant. It will change over time.
>>>>
>>>>Also, while developers might like CUDA (for good reasons), it doesn't matter if
>>>>it cuts your potential revenue in 1/3rd.
>>>>
>>>>>Also, there does seem to be a limited opening of CUDA going on: http://pressroom.nvidia.com/easyir/customrel.do?easyirid=A0D622CE9F579F09&releasejsp=release_157&prid=831864
>>>>>
>>>>>Not sure how seriously to take it. Longer-term, I agree >that CUDA needs to become an open standard to survive.
>>>>
>>>>That announcement was meaningless marketing. Control of the CUDA standard is what
>>>>matters, not that someone else can make a compiler.
>>>
>>>On earth why?
>>
>>Because Nvidia could always choose to redefine CUDA to favor their next-gen GPU
>>and disadvantage Intel, AMD and anyone else. And everyone knows that's exactly
>>what Nvidia would do if someone else started using CUDA.
>
>And break all the pre-existing CUDA Code? you are really stretching here, you know.
>If other implementations were existant, functional, and used, NVidia would only damage themselves by doing this.
>Of course until there are such other implementations, it is different.
>
>>
>>You saw this with DX as well, both AMD and Nvidia probably wrote code that was
>>optimal for their GPUs, but ran poorly on the competition. That was partially due
>>to the very different underlying architectures, a factor which probably hit AMD's
>VLIW4/5 harder than Nvidia's designs.
>
>I am unable to decode what you are claiming here, it makes no sense to me at all,
>especially considering what you write next..
>
>>
>>>That seems like a completely closed minded view.
>>>how 'open' is DirectX, for example?
>>
>>You are missing the point. DirectX is closed but controlled by an agnostic company.
>>MS really just wants DX to work on all their partner's hardware. They gain little
>>benefit from favoring one partner over another, although they might have a lead partner in a given generation.
>
>In which case you need to look at the history of DirectX a little more closely,
>at nearly every generation there has been strong evidence of MS favouring one particular
>'camp' (although I hate the term myself..) - MS have well and truely shown that
>they are happy to use DX as a tool to push things the way they want.
>
>>On the other hand, having the GPU programming interface controlled by a single
>>vendor that makes GPUs is a recipe for disaster. There's a reason why Glide was
>>never adopted by anyone else...and those reasons are just as pertinent today to
>>GPGPU as they were 15 years ago to graphics.
>
>Because Glide was both rubbish and extremely hardware specific? two things that CUDA is not?
>
>Personally I think it doesnt matter, I notice you carefully avoid the content I
>wrote about dual codepaths, and the fact that CUDA easily outperforms OpenCL due
>to basic limitations of OpenCL..

Oops, my apologies, I seem to have forgotten that that was in a different post! serves me right for rushing through at crazy hours! I still stand by the fact that dual codepaths matter in consumer applications though (and we already have them, in abundance.. to support different generations of AMD/Intel CPUs..)

>as long as that stays true, both will continue
>to be used/supported, and hardware targets using the OpenCL path will tend to suffer a performance disadvantage.
>Of course OpenCL could also be fixed, but there is little sign of that at present, lets hope...
>
>There is an interesting parallel in Cg/GLSL/HLSL.
>It may surprise you to know that Cg is still heavily used, even though it started
>life as an NVidia solution. It is still THE standard for DCC, and HLSL ended up mirroring it quite closely.
>GLSL, the 'open' standard, even in OpenGL, has much less traction.
>
>I am sure there are a lot of people who would love to see Intel and AMD have CUDA
>Support, and NVidia could well be one of them..
>
>
>
< Previous Post in ThreadNext Post in Thread >
TopicPosted ByDate
Fused Multiply-Add and HSARichard H.2012/02/07 07:29 PM
  Fused Multiply-Add and HSADavid Kanter2012/02/07 11:54 PM
    "CUDA is on its way out"pk2012/02/09 05:37 AM
      "CUDA is on its way out"David Kanter2012/02/09 11:49 AM
        "CUDA is on its way out"Bryan Catanzaro2012/02/09 12:44 PM
          "CUDA is on its way out"David Kanter2012/02/09 05:48 PM
            "CUDA is on its way out"Anon2012/02/09 07:11 PM
              "CUDA is on its way out"David Kanter2012/02/10 12:00 AM
                "CUDA is on its way out"Michael S2012/02/10 04:54 AM
                "CUDA is on its way out"Anon2012/02/10 06:42 PM
                  Proprietary SW interfaces and hardware vendorsDavid Kanter2012/02/10 09:07 PM
                    Proprietary SW interfaces and hardware vendorsAnon2012/02/11 04:08 AM
                      Proprietary SW interfaces and hardware vendorsAnon2012/02/11 04:12 AM
                      Proprietary SW interfaces and hardware vendorsDel2012/02/11 08:39 AM
                  "CUDA is on its way out"Del2012/02/11 08:33 AM
                "NVIDIA is on its way out"Brendan2012/02/10 10:42 PM
                  "NVIDIA is on its way out"anonymous2012/02/11 11:37 AM
              "CUDA is on its way out"Del2012/02/11 08:22 AM
                "CUDA is on its way out"Anon2012/02/11 04:11 PM
                  ispc on GPUsMatt Pharr2012/02/12 10:02 AM
                    ispc on GPUsAnon2012/02/13 03:34 PM
          "CUDA is on its way out"Erwin Coumans2012/02/09 09:39 PM
            "CUDA is on its way out"Brett2012/02/10 12:54 PM
              "CUDA is on its way out"David Kanter2012/02/10 04:00 PM
              "CUDA is on its way out"Anon2012/02/10 06:45 PM
                "CUDA is on its way out"Brett2012/02/10 08:20 PM
                  "CUDA is on its way out"David Kanter2012/02/10 08:58 PM
                    "CUDA is on its way out"Brett2012/02/11 07:15 AM
                      Macroscalarnone2012/02/11 08:40 AM
        "CUDA is on its way out"Anon2012/02/10 06:39 PM
          "CUDA is on its way out"David Kanter2012/02/10 09:19 PM
            "CUDA is on its way out"Anon2012/02/11 04:33 AM
          "CUDA is on its way out"NN2012/02/14 04:24 PM
            "CUDA is on its way out"Anon2012/02/15 04:16 PM
        "CUDA is on its way out"bakaneko2012/02/11 05:31 AM
      "CUDA is on its way out"Aaron Spink2012/02/10 12:29 AM
  Fused Multiply-Add and HSAEduardoS2012/02/08 03:11 PM
    Fused Multiply-Add and HSAFoo_2012/02/09 06:24 AM
Reply to this Topic
Name:
Email:
Topic:
Body: No Text
How do you spell green?