• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

HD 7970: Bulldozer vs. Sandy Bridge vs. Nehalem

Yep, rather interesting the only time BD pulls ahead(by a VERY small margin) is at super high Res. I am glad I went with my 2600K I mean look at the pic fullinfusion posted. I mean c'mon 1.54V to hit 5Ghz?? That is just pure silliness. It is already a power soak at stock as it is...

Takes a brave man to pick up a BD and have aspirations of big overclocks. I really hope AMD fix this with Piledriver, I mean I said it when BD came out they should have just scrapped it and worked on Piledriver as that seems to be this CPU done right(if we are to believe the info we have)
Granted I haven't been with AMD since my Opty 170 rig and it was great to the point that it still felt "snappier" than the C2D E6600 I replaced it with. After seeing this I KNOW I chose wisely with SB.
 
Awesome review and enlightening.

Makes me glad I didn't waste my money upgrading to a bd. With 3.7 and 4.1 turbo I have the same clocks and it still works well for what I need.
 
Very interesting comparison. Thanks for the review.

Makes me even happier about just buying a 2500K. :D
 
Ya know, I'll just say. It is good that BD is at least delivering playable fps in those benches. Even if it doesn't win. I'll give it that much.

Glad to see Hard Reset made the benching cut. Bout time. Great game and indie bunch. Knew from the demo that it would be a title I'd want to see benched.

What I'd really like to see is a comprehensive upgrade guide. Take boards and setups from the last 5 years. Then run them on a stock set of benchmarks and list which should be upgraded.

I know others do per year benching but each year they change the benches. Really if I want to know how a new system is going to run on say, L4D or whatever else...I'll want to see it. But you see it one year, then the next, gone. Yeah you know it will be better, but its still nice to see it. Metro has been one of the only titles I've been able to count on being in the lists. Glad it still is.
 
wow bull...nice thanks for review but sandy is better ..ie..ie:roll:
 
question for ya w1zz: could you compile the results differently, so we have a DX9/11 comparison between the CPU's?


basically its to see if DX11's multithreading changes the results, since we have so many DX9/poorly threaded games which could skew the overall result towards the intel CPU's, with their better performance per thread, vs AMD's more threads.


edit: to make more sense, one of your 'total performance averaged out' charts, but one for DX9, one for DX11 titles.
 
Last edited:
Very nice review, thanks.
Bulldozer does alright in higher resolutions. Not bad at all, hopefully AMD can enhanced, fix and tweak Piledriver so it can better complete with Intel.
For me going with a FX-8120 was a big performance improvement over my late Phenom II, and it did not cost me an arm and a leg for the best Socket AM3+ mobo and Ram.
 
:shadedshu

I have a FX for gaming and its the best gaming machine I have ever owned. :) :rockout:

Does that make his statement invalid? No. My rig is the fastest I've ever used, does that make it the fastest rig period?
 
why is the ram underclocked ?
 
The results would be a minimal difference. It sounds like a waste of time.

With the great memory bandwidth of BD, I think you'd be surprised.
 
With the great memory bandwidth of BD, I think you'd be surprised.

Nowhere near as surprised as I was at the difference between SB and BD.
 
Great review! thank you for taking time out of your personal holiday's to make my day. :)

Lol i'm still going to wait till either Ivy Bridge or later to upgrade my Q6600
 
I'm especially impressed with the results of the starcraft 2 tests. CPU must really be a limiting factor with that game.
 
core 2 cpu comparasions

How about comparing it with a core 2 quad cpu (e.g, Q6600 or QX9650)? Do I really need to upgrade the CPU if I play at 1920x1080 all settings max?
 
Thank you for doing this review. It's nice to have reaffirmed with data what I suspected (and hoped really) that I had no need to replace my i7 920. I knew, but looking at actual numbers was reassuring.
 
It's reviews like this that come out of nowhere that really impress me about TPU. Great work, W1zzard! :respect:
 
2500K, Chuck norris use this CPU.
thanks for the megabench w1zz, i know how long and frustrating are all those tests

Chuck who ? I think you need to read up on this guy he kicked his ass more than ounce :) Long live 'They Call Me Bruce'!!! Bruce would = the 2500k performance and stamina.:toast::rockout::respect:

Great fun you had Wizz over the Christmas holiday.We gonna have a Crossfire one too?:respect:master reviewer :respect:
 
Last edited:
I was expecting BD to be shit, but it performed better then I thought, the cpu is not as bad as everyone makes it out to be ,candy bridge is still obviously the better choice ^_^
 
If You think about it that is still less voltage than what a Athlon XP required even overclocked.

Yep, rather interesting the only time BD pulls ahead(by a VERY small margin) is at super high Res. I am glad I went with my 2600K I mean look at the pic fullinfusion posted. I mean c'mon 1.54V to hit 5Ghz?? That is just pure silliness. It is already a power soak at stock as it is...

Takes a brave man to pick up a BD and have aspirations of big overclocks. I really hope AMD fix this with Piledriver, I mean I said it when BD came out they should have just scrapped it and worked on Piledriver as that seems to be this CPU done right(if we are to believe the info we have)
Granted I haven't been with AMD since my Opty 170 rig and it was great to the point that it still felt "snappier" than the C2D E6600 I replaced it with. After seeing this I KNOW I chose wisely with SB.
 
I was expecting BD to be shit, but it performed better then I thought, the cpu is not as bad as everyone makes it out to be ,candy bridge is still obviously the better choice ^_^

I wonder how many people can honestly see that few percent difference in the chips...
 
My BD has been doing me very well and I have no regrets from it at all. I'd be interested in seeing a review when the Windows 7 BD FIX patch comes out and I like Mussle's Idea of an average split between DX9 games and dX11 games due to DX11's multithreading optimization.
 
If You think about it that is still less voltage than what a Athlon XP required even overclocked.

Well that's just it isn't? Generations later and you'd think lower wattage would come with progress.
 
Naah. It's slower than SB but holds up good against .. that older Intel chip. But yeah, if you're a hardcore gamer intel is the way to go.
You're forgetting that the 2500K is cheaper. Seriously, I got mine and a Z68 board for $300 flat, brand new.
 
Well that's just it isn't? Generations later and you'd think lower wattage would come with progress.

so u want wine with that cheese lol
 
question for ya w1zz: could you compile the results differently, so we have a DX9/11 comparison between the CPU's?

all benchmarks are dx11. except for hard reset and star craft II
 
Back
Top