Comments by "IIIRattleHeadIII" (@badass6300) on "why can’t computers have thousands of cores?" video.

  1. Also, a big factor is that many programs have linear logic. Amdahl's law shows how well a task scales with multiple cores depending on how parallel it is. For 50% parallel tasks above 4 cores is pointless. For 75% parallel above 16 cores is pointless. You just don't gain performance and that is baked in the logic of the task. Many cores are great when doing multiple of the same task without caring which task is completed first.
    635
  2.  @mewsermeow8683  if the gpu has the instructions for it, but 99% of the time yes.
    26
  3. ​ @hjups  True, but GPU architecture is getting close to CPU architecture with the passing generations. AMD GPUs since RDNA1 have hardware schedulers and might get OoO Execution in the future. Then again with chiplets they might get a whole CPU to themselves for certain tasks. Or vice-versa, integrated GPUs might get good, or both.
    9
  4.  @PyromancerRift  Zen didn't shake anything. It just forced intel to release the proper amount of cores. We are talking about concepts here, the concept of linear, dependent logic. These concepts don't change, just like the 2 + 2 is always 4.
    7
  5.  @utubekullanicisi  true, that's what I said in the end of my comment.
    2
  6.  @brianwesley28  Rendering is the best case scenario and based on the architecture it's 98-99.9% parallel.
    2
  7.  @Joker22593  Thanks!
    1
  8.  @alexolivers9476  Yeah, AMD wanted all the floating point operations to be done on the iGPU, that's why they developed the hUMA controller and Bulldozer/Piledriver/Steamroller CPUs/APUs have weak FPUs... guess what happened... Developers didn't develop for it and they still kept using the CPU and its FPU for floating point tasks. Now we have AVX 512... horrible, power hungry, space taking instruction...
    1
  9.  @alexolivers9476  yeah, we have GPUs for that.
    1