CPU for gaming - state of the union 2014

[TheRealSintel]

Last year anandtech did a feature on "best fit for gaming" cpus
http://www.anandtech.com/show/7189/choosing-a-gaming-cpu-september-2013/10

The conclusion there was that even an AMD A8 is mostly sufficient for single-GPU gaming (CIV V being the notable exception), though it was at -quite GPU limited- 1440p max settings.

Now someone I know is on a budget and even though I recommended the i5, he's wondering if it truly is worth the investment. This is for single-GPU gaming on a 1080p60hz screen with a R9 270X - mostly interested in racing/fifa/fps.

We're 2014 now and even though I remember reading about CPU-hungry games on-and-off, funnily enough I can't find any evidence that the AMD lacks power in games (except those that need huge GPU power as well). He's quoting these passmark scores but we all know they hold little value http://www.cpubenchmark.net/cpu.php?cpu=AMD+FX-6300+Six-Core

Is it truly the case that the old AMD is still sufficient for driving most games and an upgrade to i3 or i5 would be negligible? Are there any example games you can name that support your thesis (whatever it is)?

I know the Intel i3 and i5 CPUs will clobber it in all other categories, and that the platform is much more modern, but that's not the question.

 

[Erenhardt]

fx6350 can handle 2X770 SLI.

2014 AT holiday guide:
http://www.anandtech.com/show/8750/holiday-guides-2014-cpus

 

[TheRealSintel]

Thanks, I forgot to check the holidays guide, I thought the article from last year was the most recent.

As for an equal price comparison, found that the i3 4160 is exactly the same cost as the FX-6300 and the following review (the 4340 has more 1MB more L3 cache though)
http://www.hardcoreware.net/intel-core-i3-4340-review/

So at least for that set of games Intel has a direct competitor doing better.

 

[Craig234]

Why are the i7-4790 and i7-4790k the same?

 

[ShintaiDK]

Why are the i7-4790 and i7-4790k the same?

Its so GPU limited that the chart is worthless. Prescripted benchmark makes it even worse.

 

[Craig234]

Its so GPU limited that the chart is worthless. Prescripted benchmark makes it even worse.

But it seems strange the CPU has enough role that the numbers change for other CPU's, but not those two.

 

[Erenhardt]

Its so GPU limited that the chart is worthless. Prescripted benchmark makes it even worse.

Srsly... So we should run dual R9 295X2 with settings to low and resolution reduced to 320x240...

This is SLI benchmark, with FPS around 100.

Does BF4 even have canned benchmark?

It clearly shows that 100$ FX6350 provides 90% of performance you get from much much more expensive CPUs in High end GPU configuration

 

[BSim500]

It clearly shows that 100$ FX6350 provides 90% of performance you get from much much more expensive CPUs in High end GPU configuration

In some games yes. In many others, no.

 

[el etro]

FXs processors really handle well all games you throw on it. My recommendation to FX processors start in FX6300, and up to the cheapest core i5 available.

 

[ShintaiDK]

Srsly... So we should run dual R9 295X2 with settings to low and resolution reduced to 320x240...

This is SLI benchmark, with FPS around 100.

Does BF4 even have canned benchmark?

It clearly shows that 100$ FX6350 provides 90% of performance you get from much much more expensive CPUs in High end GPU configuration

Do you remember the 800Mhz Haswell, prescripted benchmark vs ingame?

Arcording to you, 800Mhz Haswells should be plenty.

Oh in case you forgot, Tomb Raider. 800Mhz Haswell 74FPS in benchmark, 22FPS ingame with 7950CF.

 

[frozentundra123456]

But it seems strange the CPU has enough role that the numbers change for other CPU's, but not those two.

Because with the weaker cpus, you are at least partially limited by the cpu. With those two fastest cpu's the limitation shifts totally to the gpu and cpu speed no longer matters.