Mark Roulo (nothanks.delete@this.xxx.com) on May 14, 2013 5:37 pm wrote:
> RichardC (tich.delete@this.pobox.com) on May 14, 2013 12:09 pm wrote:
> > Ricardo B (ricardo.b.delete@this.xxxxx.xx) on May 14, 2013 11:18 am wrote:
> > > Actually, this question should be different.
> > > If you're going to use FPS as the benchmark for games, does high end CPU
> > > performance even matter or should you just go and get a faster GPU?
> >
> > The benchmarks I saw showed that 4C/4T vs 4C/8T at 3.8GHz was
> > pretty equal; but 4C/4T 4.0GHz was significantly faster than 4C/4T 3.8GHz.
> > That's quite strong evidence that cpu performance does make a difference,
> > but that hyperthreading is not particularly helpful for current game engines.
>
> Can you provide a link? Naively, I would expect a ~5% increase in clock speed to result in a ~5% increase
> in measured performance (frames per second, whatever). Maybe a bit less. And I wouldn't describe a 5%
> increase in performance as "significantly faster." It would be useful to see the actual numbers.

Here's one: http://www.overclock.net/t/671977/hyperthreading-in-games

I wouldn't expect 5% core clock increase to give a 5% overall speedup.
Too much stuff depends on DRAM. And of course for gaming the GPU is very
much involved as well.

However, by "significant increase" I didn't mean to imply as much as 5%.
Just that HT vs non-HT was a nothingburger, and 3.8GHz vs 4.0GHz was a
detectable and consistent speedup ("significant" more in the statistical
sense of being outside the range of random variation).