AMD FX-8350 powering GTX 780 SLI vs GTX 980 SLI at 4K

[monstercameron]

AMD FX-8350 powering GTX 780 SLI vs GTX 980 SLI at 4K

Read more at http://www.tweaktown.com/tweakipedi...g-gtx-780-sli-vs-gtx-980-sli-at-4k/index.html

interesting scaling or weird errata?

 

[mistersprinkles]

Seems bogus. 4930K would give you a 16x/16X link at PCIE 3.0. AMD rig would give you PCIE 2.0...

 

[desprado]

I7 4940k is not good for gaming.If they used 4770k or 4790k than u did not even need to posted this topic.

 

[SlowSpyder]

Seems bogus. 4930K would give you a 16x/16X link at PCIE 3.0. AMD rig would give you PCIE 2.0...

I don't think PCIE 2.0 would be a significant bottleneck yet.

 

[Enigmoid]

Looks like errata.

This is extreme.

Scores are very low.

Should be over 10k.

 

[monstercameron]

Looks like errata.

-snip-

This is extreme.
-snip-

Scores are very low.

Should be over 10k.

these are 4k benchies, so that may be the differential.

 

[Shivansps]

This does not look right at all.

How could a FX8350 have a inferior total score but superior graphics score on the 980 SLI? makes no sence, thats directly impling that the 8350 is slower but somehow the gpus performs better with it.

And the numbers are very low, they should be over 10K for a 980 SLI.

 

[SlowSpyder]

these are 4k benchies, so that may be the differential.

Was going to ask the same thing about that, if the 10k+ score was 4k resolution. Here is a 5.2GHz 4770K with GTX980 SLI @ 1440P. He scores 10,211 in Firestrike Extreme at that resolution. So I think the score would be lower, as the Tweaktown review shows, at 4k.

http://forums.evga.com/GTX-980-SLI-Benchmarks-m2219434.aspx

 

[SlowSpyder]

This does not look right at all.

IMG

How could a FX8350 have a inferior total score but superior graphics score on the 980 SLI? makes no sence, thats directly impling that the 8350 is slower but somehow the gpus performs better with it.

And the numbers are very low, they should be over 10K for a 980 SLI.

Because the physics and other CPU scores could have been a bit lower on the FX. But maybe Nvidia's driver interacts better with the FX, SLI could work differently. Maybe the graphs are mislabeled. I really don't know. But, I have an FX and know it is an excellent CPU, I'm not shocked to see it grab a win here and there.

 

[Shivansps]

it does not work that way, there is too much of a gap, look at the result on the 780SLI a full 1200 points dif on graphics!!!!! INSANE, even for a sli.

 

[davie jambo]

Somebody post that AMD picture

First person to mention skyrim or starcraft 2 benchmarks gets a gold star

 

[USER8000]

People are forgetting the more important thing though - at least with the games tested,either setup is as playable or as unplayable at each other,and that moving to a faster card seems to provide more gains that a faster CPU,at least for the games tested.

 

[davie jambo]

I just read that article , does seem like AMD are the kings of 4k

That was a $300 cpu and motherboard bundle up against $1000 intel and it won every test. Impressed

 

[SlowSpyder]

it does not work that way, there is too much of a gap, look at the result on the 780SLI a full 1200 points dif on graphics!!!!! INSANE, even for a sli.

I'm not sure if it'll help clear anything up since my rig is so different than those tested in this review (single 7970, 19x12 resolution). But, I should at least be able to compare my score with other 7970's + Intel CPU's to see what things look like in comparison.

 

[BSim500]

interesting scaling or weird errata?

Very much the latter. In fact, I've been left wondering about the 'quality' of TweakTown's 'benchmarks' even since a stock Pentium G3258 beats an i7-4790K OC'd to 4.7GHz in 4K gaming... (And the same G3258 is also somehow "faster" at 3.2GHz than 4.5GHz):-
http://www.tweaktown.com/articles/6...niversary-cpu-gaming-performance/index10.html

They seriously need a new test rig / methodology.

 

[Burpo]

This sure is pretty..

 

[SlowSpyder]

Very much the latter. In fact, I've been left wondering about the 'quality' of TweakTown's 'benchmarks' even since a stock Pentium G3258 beats an i7-4790K OC'd to 4.7GHz in 4K gaming... (And the same G3258 is also somehow "faster" at 3.2GHz than 4.5GHz):-
http://www.tweaktown.com/articles/6...niversary-cpu-gaming-performance/index10.html

They seriously need a new test rig / methodology.

Meh, that's within the margin of error, obviously a GPU limited benchmark (talking about Heaven, the others are more or less the same story or the 4770k is faster). It is less than 1% faster, nothing there that would make me question it really.

 

[MiddleOfTheRoad]

I just read that article , does seem like AMD are the kings of 4k

That was a $300 cpu and motherboard bundle up against $1000 intel and it won every test. Impressed

Agreed -- At 4K resolutions, the GPU is the critical piece of the equation. AMD CPU's post nearly identical performance #'s to Intel CPU's for 4K gamers...... It's all about the GPU once resolutions get that high.

 

[Enigmoid]

Agreed -- At 4K resolutions, the GPU is the critical piece of the equation. AMD CPU's post nearly identical performance #'s to Intel CPU's for 4K gamers...... It's all about the GPU once resolutions get that high.

20% higher on firestrike extreme graphics is something wrong. Seriously, you should not see differences that big even if you were comparing to the G3258.

 

[frozentundra123456]

At best it is an apples to oranges comparison, the 4930 k without overclocking is certainly not Intel's best gaming processor.

 

[SlowSpyder]

At best it is an apples to oranges comparison, the 4930 k without overclocking is certainly not Intel's best gaming processor.

But I don't think they overclocked either CPU. Unless I missed it, that's apples to apples to me. (Entirely possible I may have missed it if they did, I've had a fair amount of Miller High Life at a friend's tonight... haha)

 

[frozentundra123456]

But I don't think they overclocked either CPU. Unless I missed it, that's apples to apples to me. (Entirely possible I may have missed it if they did, I've had a fair amount of Miller High Life at a friend's tonight... haha)

I think you are right, but what I am saying, is if you aren't overclocking, the 4790k is the appropriate chip to test. It is also closer in price to the 8350.

 

[SlowSpyder]

Got ya. But in fairness the 8350 isn't AMD's best gaming CPU either.

 

[MiddleOfTheRoad]

20% higher on firestrike extreme graphics is something wrong. Seriously, you should not see differences that big even if you were comparing to the G3258.

My guess would be a video card driver issue -- but the 4930k seemed to be held back across the board with the 780 SLI setup, not just on that one benchmark.

 

[frozentundra123456]

Got ya. But in fairness the 8350 isn't AMD's best gaming CPU either.

OK, I was trying not to stir the pot too much, but now I will just be blunt. Testing a generation old more expensive Intel CPU instead of a newer, cheaper, and faster one just seems like stacking the deck in favor of amd especially when price is considered, which inevitably it will be. I am not saying a 4790k would have been faster, but it would have been a more logical comparison.