Recent content by darkblu

  1. D

    Wii U hardware discussion and investigation *rename

    I assume you're suggesting that multiple threads on the SIMD unit could emulate a single Flipper pipeline, as AMD's VLIW architectures have exactly 5 (or 4) ALUs per clock per thread on the SIMD.
  2. D

    Wii U hardware discussion and investigation *rename

    That's part of my concerns re such benches as well - CPU caches are strictly domestic animals compared to the GPU caches in the wild.
  3. D

    Wii U hardware discussion and investigation *rename

    Yes, but we don't know how the multiple banks in there factor in - are fb accesses interleaved across all banks, or are they interleaved across a few, and if the latter, what do the 'idle' banks do - are they open for access for other purposes? Thank you. I'm really curious to see Sandra's...
  4. D

    Wii U hardware discussion and investigation *rename

    That much is clear - the device should have its 'proper' LDS. Question is, how close the eDRAM block would need to be to have a chance of being accessed by the SIMDs, Evergreen style. Yes, according to the official floorplan, the LDS block sits right on the left of the SIMD0 block, and not so...
  5. D

    Wii U hardware discussion and investigation *rename

    Which block on the die shot do we assume to be the LDS?
  6. D

    Wii U hardware discussion and investigation *rename

    No, you're not. But 3dilettante was exclusive talking of texture and vertex fetches (i.e. GATHER and VFETCH ops) and missed one third kind of data access - LDS. Such access is done from within ALU clauses, without the interference of the scheduler. Or put in other words, VLIW SIMDs *can* make...
  7. D

    Wii U hardware discussion and investigation *rename

    I sense a misunderstanding in the making here. So let's backtrack a bit. You: Nintendo could have used a Bobcat. Me: Why? Outside of 'Just because I think so.' You: Asking why is unfair. Nintendo choice was BC driven. Well, guess what - a dual-core 1.6GHz Bobcat could have given Nintendo...
  8. D

    Wii U hardware discussion and investigation *rename

    Perhaps this one could do?
  9. D

    Wii U hardware discussion and investigation *rename

    I'm yet to see a coherent argument why the above would have been a better choice than the present Espresso setup. People just keep suggesting alternatives to what nintendo did for no other reason than believing they could do better than nintendo.
  10. D

    Wii U hardware discussion and investigation *rename

    I have access to a 7447A, which should be as close to the Atom on paper as it gets - 7447's SIMD cluster is entirely in-order, including permutes, with a sole FMADD unit (actually, a sole FP SIMD unit).
  11. D

    Wii U hardware discussion and investigation *rename

    In fp SIMD? Well, we'd have to decide what the 'real' SIMD added to the G3 would be. Are you ok with AltiVec?
  12. D

    Wii U hardware discussion and investigation *rename

    Compared to some recent x86 netbook cpus. True. Just imagine what a sad bunch those recent netbook x86 cpus would've been if those ancient G3s had the slightest resemblance of a SIMD unit. *shudders at the thought*
  13. D

    Wii U hardware discussion and investigation *rename

    Esrever, is there a particular reason why you keep calling those cedar SP's 'bobcat'?
  14. D

    Wii U hardware discussion and investigation *rename

    Sure. I didn't see any extrapolations of performance differentials in that post, though.
  15. D

    Wii U hardware discussion and investigation *rename

    No, I'm just saying the comparison is flawed. As to what and why nintendo did - the jury is still out on those.
Back
Top