you are viewing a single comment's thread
view the rest of the comments
[–] 54 points 4 months ago* (last edited 4 months ago) (2 children)

TL;DR:

Mr Krabs saying "Money"

But really, I don't buy that Nvidia can't regularly release improved GPUs for consumers. Consumer hardware is reaching a peak but there's still a lot more that can be done in terms of efficiency. Just look at how great modern flagship phone chips are. The M5 Max macbook pro for example is sooooooooo much more efficient than a typical gaming laptops, and it's in a slim chassis.

Nvidia just stopped caring because they make too much money selling to data centers now, and they are releasing regular improvements for their high end server hardware. Consumer GPUs are just pocket money to them now, and there's practically no competition so why would they bother releasing something new?

  • source
  • hideshow 2 child comments
  • [–] 5 points 4 months ago*

    The power efficiency difference between x86 and ARM boils down to legacy baggage in the instruction set. ARM doesn't suffer from that baggage, and that's why Apple was able to make their chips so efficient.

    GPUs are a different story though, and I don't even know how you would compare the here. Maybe the voltage difference could explain some of that power efficiency gap.

  • source
  • parent
  • load more comments (1 reply)