Return of FX : GA-990FXA-UD7 - OC'd AMD FX 8150 / 6990 Performance Comparison Review

[grimpr]

This is more flattering of the Crown.

 

[polyzp]

LOL at the wishful thinking that AMD will be able to magically get 15-20% higher IPC from a slightly tweaked architecture on the same process node.

Its called science :whistle

Intel seems to be happy with a mere 3% increase in IPC, while Trinity IPC is looking great in Engineering Sample benchmarks. We will see May 15th!

 

[LOL_Wut_Axel]

Its called science :whistle

Intel seems to be happy with a mere 3% increase in IPC, while Trinity IPC is looking great in Engineering Sample benchmarks. We will see May 15th!

You realize those are more than likely fake, right?

And Intel is more than happy with a 5% IPC increase because they're already kicking AMD left and right. Even if AMD make a big leap forwards (which, BTW, they won't) Intel would still be ahead by a mile.

Sorry, but AMD CPUs are slow as hell compared to Intel's and a small architectural revision won't change that.

 

[minitron]

I don't think I've seen so many doctored results and made up math on the interwebs since last week!

Obviously we should trust this guy instead of the well-documented and thorough reviews by legitimate websites!

 

[polyzp]

I don't think I've seen so many doctored results and made up math on the interwebs since last week!

Obviously we should trust this guy instead of the well-documented and thorough reviews by legitimate websites!

Unless you have an FX 8150 that overclocks to 4.9 Ghz and a 6990 that overclocks to 990/1500 you are not qualified to judge whether my results are "made up". Any one with an FX 8150, or with common sense (36.1% clock scaled up from reviews at stock on CPU) can confirm my results within error.

 

[MustangSVT]

in b4 lock!

 

[minitron]

Unless you have an FX 8150 that overclocks to 4.9 Ghz and a 6990 that overclocks to 990/1500 you are not qualified to judge whether my results are "made up". Any one with an FX 8150, or with common sense (36.1% clock scaled up from reviews at stock on CPU) can confirm my results within error.

Obviously you can't comprehend the concept that performance does not directly scale with clock speed.

Anyone with common sense can tell your results are BS.

 

[Maximilian]

Unless you have an FX 8150 that overclocks to 4.9 Ghz and a 6990 that overclocks to 990/1500 you are not qualified to judge whether my results are "made up". Any one with an FX 8150, or with common sense (36.1% clock scaled up from reviews at stock on CPU) can confirm my results within error.

Translation:

I am right, everyone else is wrong, nobody else can prove me wrong because nobody else was stupid enough to pair such a high power GPU with such a lousy CPU.

 

[Rvenger]

Translation:

I am right, everyone else is wrong, nobody else can prove me wrong because nobody else was stupid enough to pair such a high power GPU with such a lousy CPU.

:ninja:

Who would even run an FX 24/7 at 4.9ghz? I probably would have to distribute the power between 2 circuit breakers so my electrical socket doesn't burn.

My FX6100 setup sucks down more juice than my whole IVB setup.

 

[Black96ws6]

As they should, since the FX-8150 is a cheaper less expensive part compared to those other CPU. Thuban die size is larger than 8150. Intel CPU are on a smaller more advanced process.

Fact remains, it goes from being 48% behind at single threaded to DEAD even in the multithreaded test, compared to 2500k. If you are going to call that "bad scaling", than what is wrong with Intel's scaling such that they lose so much ground when going to the multicore test?

How is an 8 core CPU running even with a 4 core CPU a good thing?

And if CPU A (Bulldozer) gets destroyed by CPU B (2500k) in single threaded, and breaks even in multi-threaded....and CPU A uses more power and can't push SLI\Crossfire setups as hard.....why would you ever...ever... buy CPU A?

 

[SickBeast]

IMO this thread is potential evidence that AMD is paying people to post this nonsense on the forums. The OP has posted this same thread across just about every tech site out there.

 

[Maximilian]

IMO this thread is potential evidence that AMD is paying people to post this nonsense on the forums. The OP has posted this same thread across just about every tech site out there.

Yup, there was two threads before this one, one got locked, the other was forgotten about.

I was chuffed with my 2600k and its overclock/performance, i didnt go to every tech site and start a thread about how my 2600k roxors ur soxors. People who do that shit have an agenda and people who have an agenda need to gtfo.

 

[chimaxi83]

Yea, this dude keeps coming back for more, despite having threads locked and basically being flamed out of a lot of his threads on other sites. Glutton for punishment?

I remember picking apart all his little graphs and other BS because he thinks comparing his setup to others at random clock speeds, RAM speeds, different video cards, drivers, basically a bunch of random shit... proves a point. Meh.

Bulldozer is ok at some stuff, sure. If you want to zip the hell out of some files, all day every day, go BD. Want to play some Fritz chess? Go BD. Otherwise, you're better off sticking with Intel for the foreseeable future.

 

[soccerballtux]

have you done any 4-thread testing on Bulldozer, so we can see "without module contention, bulldozer is this much faster per clock than Ph2"?

 

[Abwx]

Obviously you can't comprehend the concept that performance does not directly scale with clock speed.

Anyone with common sense can tell your results are BS.

At least for the CPU part , perfs scaling with frequency is almost perfect
with a BD , so who s spreading BS ??...:biggrin:

 

[AtenRa]

have you done any 4-thread testing on Bulldozer, so we can see "without module contention, bulldozer is this much faster per clock than Ph2"?

No it's not

http://atenra.*********/2012/02/01/amd’s-bulldozer-cmt-scaling/

 

[Abwx]

have you done any 4-thread testing on Bulldozer, so we can see "without module contention, bulldozer is this much faster per clock than Ph2"?

Dont know if it match a phenom , but using four modules
for four threads improve performances by as much as 25%
compared with 2Modules/4 threads.

http://www.hardware.fr/articles/842-9/efficacite-cmt.html

 

[minitron]

At least for the CPU part , perfs scaling with frequency is almost perfect
with a BD , so who s spreading BS ??...:biggrin:

We're talking about GPU performance and CPU scaling. Hence the title/topic of the thread.

 

[polyzp]

have you done any 4-thread testing on Bulldozer, so we can see "without module contention, bulldozer is this much faster per clock than Ph2"?

I will be in the third part of my review. I laugh at any one who thinks my results are BS because you simply have no concrete evidence to make such a idiotic claim. My results are mostly CPU based so do some basic first grade math with OC scaling by % from stock results and voila my results are within error. For you to claim my results are fabricated, then any single result ever posted by any website "COULD" be fabricated. For a benchmark review you simply take them at their word, there is no point in me making up these numbers. It sounds like you secretly think the FX 8150 is "too terrible" to have the results i am showing, with absolutely no specific evidence to validate this. Keep thinking I work for AMD you guys make me laugh.

I have linked every screenshot for my results, stop trolling.

 

[polyzp]

I don't think I've seen so many doctored results and made up math on the interwebs since last week!

Obviously we should trust this guy instead of the well-documented and thorough reviews by legitimate websites!

Show me a side by side comparison proving my results are false. I would LOVE to see you try.