EduardoS (no.delete@this.spam.com) on June 29, 2013 9:31 pm wrote:
> anon (anon.delete@this.anon.com) on June 29, 2013 9:15 pm wrote:
> > Then why has the market (almost every general purpose CPU from high end smartphones
> > to high end servers) overwhelmingly spent effort on FP performance?
>
> Because it is important to win benchmarks?

Is that a question, or an answer?

It is important to win benchmarks that are vaguely relevant. It is important to run on real code that matters. How important is it to win irrelevant benchmarks.

> Because it is easier/cheaper than improving any other
> part of the chip?

Is that a question?

What do you mean, easier/cheaper? If it "improves" the chip for things that don't matter, then it is not actually an improvement, is it?

You could easily and cheaply "improve" a chip for an artificial test simply by adding a vast amount of slow L4 cache to it, for example. Does not require any logic changes.

So your question-answer doesn't really answer much.

> Ask chip designers, looking from distance designing a FPU that doesn't sucks
> very badly doesn't look as hard as designing a scheduller that performs exceptioanlly well.

Your logic simply doesn't hold up. If they just wanted to get flops, the CPU would look like a GPU.

>
> > *That* was my question.
>
> So you formed you question really bad.
>

And your question-answer things are formed atrociously.