xfire msi 290 oc with 8350

[Martini1991]

Ideally just needs to bench some games against LTMatt, none of this Firestrike rubbish

You have Avenged and Ras both stating bottlenecks in games with a single R9 290X and clocked FX83's, so yeah...

 

[Avenged7Fold]

I am going to get called an intel preacher now?

Best change my sig/projectlog or i am not going to have any credibility as one.

 

[LtMatt]

I'll provide some comparisons in game benchmarks, if needed.

 

[Kaapstad]

So if I ran this bench mark my graphics score would less than someone with a 4930k

The graphics score is independent. Any of the intel i5s or i7s running at the same clocks will allow the cards to score roughly the same Graphics score. If you do score less on graphics score with a 8350 it means there is a bottleneck. The one thing you must not compare is the overall score as this will have Physics added in which gives the i7s an advantage.

 

[Choimaloy75]

So if I bench my cards will my graphics score not be higher than someone with a 4930k as kaapstad has.

 

[Martini1991]

Valley is an awful bench to use as it is rubbish on any decent GPU setup regardless of whether it's on an intel or AMD CPU, this is why I did not use it. Intel will get an advantage on it though.

Doesn't that in its nature make it closer to how games would be though? Not that I think synthetics are good for bottleneck testing.

Certainly not Firestrike.

 

[Martini1991]

So if I bench my cards will my graphics score not be higher than someone with a 4930k as kaapstad has.

You're wording it a certain way to muddy the point.

 

[Kaapstad]

I'll provide some comparisons in game benchmarks, if needed.

I have got to go to work now but if you and anyone else run the sleeping dogs bench (using the settings in that thread) with the 290P/Xs at stock that will also be a good comparison. I will run it tomorrow when I get in.

 

[Avenged7Fold]

So if I bench my cards will my graphics score show a massive bottleneck?

Yes it will.

 

[Kaapstad]

Doesn't that in its nature make it closer to how games would be though? Not that I think synthetics are good for bottleneck testing.

Certainly not Firestrike.

Firestrike is closer to a game than Valley is, another good bench that is very game like is Heaven 4.

 

[Martini1991]

Firestrike is closer to a game than Valley is, another good bench that is very game like is Heaven 4.

I truly disagree on both points

If you want to test bottlenecks for games, you test games, it's that simple.

We've seen Haswell i3's barely bottleneck a titan in Heaven, how is that anywhere near a game ?

 

[Kaapstad]

So if I bench my cards will my graphics score not be higher than someone with a 4930k as kaapstad has.

the 4930k is not the best CPU for gaming.

 

[Choimaloy75]

I've benched mine with Firestrike amd my graphics score at stock 21337.

 

[Choimaloy75]

This is more than kaapstads with the 4930k

 

[Tonester0011]

I've benched mine with Firestrike amd my graphics score at stock 21337.

Then you didn't bench using stock clocks

 

[Martini1991]

Oh, now you've done it.

No offence Kaap, but I think you're doing far more harm than good in the discussion of bottlenecks

Running GPU intense synthetics for CPU bottlenecks isn't indicative of bottlenecks for gaming..

 

[Tonester0011]

I truly disagree on both points

If you want to test bottlenecks for games, you test games, it's that simple.

We've seen Haswell i3's barely bottleneck a titan in Heaven, how is that anywhere near a game ?

This ^
I've used a celeron on Heaven and had like 30-50 fewer points compared to a clocked 2600k (with a 670)

 

[Choimaloy75]

Yes I did 1000 core 1250 memory.

 

[Martini1991]

FX8350 better than an 4930K, that's me switching to AMD.
Derp.

 

[Choimaloy75]

Your point