Amd fx8120 got higher rating that the stronger fx8150???

[dingo44]

Has anyone recently been on cpu benchmark?
if so could someone help explain this why the amd fx8120 is ranked 6th with 9996 points but the strongest amd fx8150 has been rated 19th with 8635 points?

Source: http://www.cpubenchmark.net/high_end_cpus.html

I tried to upload a picture couldnt figure out how to, noob moment :lol:

 

[jimmysmitty]

I don't see it.

 

[dingo44]

Dont you?? hey your right, thats odd cause it was like that not that long ago, fortunately I took a snapshot for everyone to look at

 

[sarinaide]

Oh good another synthetic benchmark, as if the 8120 is better than the 2600K or is it just the "more cores is better bias".

 

[iam2thecrowe]

what a pointless benchmark. everyone knows that the i7 2600k is faster overall than the amd 8 core. read reviews, dont judge a cpu by that useless passmark utility.

 

[dingo44]

No i wouldnt have a clue, because what I am trying to figure out is how the 8150 is bellow the 8120 when the 8150 has a higher clock rate and same amount of cores, and where the 8120 went from this chart in the last couple of hours.

thanks i have read plenty of reviews but i am trying to figure out what is up with cpu benchmark website, sure it might be as accurate as WEI but an amd fx8120 doesnt just appear and beat its higher clocked amd fx 8150 by 15 places??? get what im saying

a bit fishy if you ask me :lol:

 

[statikregimen]

I was looking at their charts yesterday, and noticed this. Now the 8120 has completely disappeared. I can tell you first-hand, however, that the results were from overclocked CPUs and that's probably why they were removed. My fx-8120 @ 4.5ghz scored 10,249 overall in passmark. I don't get why they didn't simply move the results to the OC chart, but rather they appear to be completely gone....

EDIT: btw, passmark isn't useless. Any benchmark that can produce repeatable results is a good measuring stick. Their online charts, however, should be taken with a grain of salt, as they are averages of all submitted scores. The benchmakr tool itself is by far not useless...

 

[geekapproved]

It's because the architecture is a bit ahead of it's time softwarewise, and not recognized by most apps, games and benchmarks. Even Windows 7. That is the reason it performs so poorly in the things we use most. This processor will perform much better in the server environment, which is what it was really designed for, heavy multithreading enviroments.

The problem is even windows 7 doesn't know where to place the threads optimally because it doesn't recognize the architecture. Windows 8 should address this.

Anand:
"Windows 7 isn't really all that optimized for Bulldozer. Given AMD's unique multi-core module architecture, the OS scheduler needs to know when to place threads on a single module (with shared caches) vs. on separate modules with dedicated caches. Windows 7's scheduler isn't aware of Bulldozer's architecture and as a result sort of places threads wherever it sees fit, regardless of optimal placement. Windows 8 is expected to correct this, however given the short lead time on Bulldozer reviews we weren't able to do much experimenting with Windows 8 performance on the platform. There's also the fact that Windows 8 isn't expected out until the end of next year, at which point we'll likely see an upgraded successor to Bulldozer."

"Currently Windows 7 is unaware of the shared nature contained within the FX-8150 processor. As a result, there are opportunities for resource sharing or activate higher turbo core frequencies are missed"


I'd be interested in seeing how it performs with OS that does recognize the architecture and can optimize it.

 

[statikregimen]

It's because the architecture is not recognized by most apps, games and os's, including Windows 7. That is the reason it performs so poorly in the things we use most. This processor will perform much better in the server environment, which is what it was really designed for, heavy multithreading enviroments.

The problem is even windows 7 doesn't know where to place the threads optimally because it doesn't recognize the architecture. Windows 8 should address this.

Anand:
Windows 7 isn't really all that optimized for Bulldozer. Given AMD's unique multi-core module architecture, the OS scheduler needs to know when to place threads on a single module (with shared caches) vs. on separate modules with dedicated caches. Windows 7's scheduler isn't aware of Bulldozer's architecture and as a result sort of places threads wherever it sees fit, regardless of optimal placement. Windows 8 is expected to correct this, however given the short lead time on Bulldozer reviews we weren't able to do much experimenting with Windows 8 performance on the platform. There's also the fact that Windows 8 isn't expected out until the end of next year, at which point we'll likely see an upgraded successor to Bulldozer.

I'd be interested in seeing how it performs with a OS scheduler that does recognize the architecture and knows where to place the threads.

While I agree with this fully, the performance gain is only going to be realized in lightly threaded workloads and the improvment will be <20% in most cases (really, more around 5-10% as shown by tests people have done where they force threads to various cores via affinity). Where I think we'll most likely see more marked improvments, is through n-threaded software and the new instruction sets. Sadly, however, that means Bulldozer is well ahead of its time, and it wont be for several more generations of CPUs before software takes advantage of these things, as a rule, rather than exception.

However, as has been stated many times, AMD has been able to push innovation in software rather effectively, in the past; e.g. x86-64 architecture. But using that as an example, we see that the first AMD64 CPUs came out in 2003, and we're only just getting to where operating systems and applications are widely available and used in the last couple of years. If that trend happens again, Bulldozer will be all but irrelevant by the time it is performing at its peak, and CPUs of that generation will be 10x faster from both manufacturers (and who knows: by then, totally new companies maybe competing, and we'll be using fiberoptic gate technology running at >10ghz on 10000 cores ).

 

[dingo44]

yeah its a mystery the fx8120 completely disappears from cpu benchmark, which by the way that is a pretty nice overclock. I like using passmark because it is simply a good chart to compare all the cpu performances against each other.

it will be interesting to see when the fx8120 reappears on passmark and where it will place itself on the chart.

Amd have a way with future technology, just like there new apu's

 

[gamerk316]

It's because the architecture is a bit ahead of it's time softwarewise, and not recognized by most apps, games and benchmarks. Even Windows 7. That is the reason it performs so poorly in the things we use most. This processor will perform much better in the server environment, which is what it was really designed for, heavy multithreading enviroments.

The problem is even windows 7 doesn't know where to place the threads optimally because it doesn't recognize the architecture. Windows 8 should address this.

Anand:
"Windows 7 isn't really all that optimized for Bulldozer. Given AMD's unique multi-core module architecture, the OS scheduler needs to know when to place threads on a single module (with shared caches) vs. on separate modules with dedicated caches. Windows 7's scheduler isn't aware of Bulldozer's architecture and as a result sort of places threads wherever it sees fit, regardless of optimal placement. Windows 8 is expected to correct this, however given the short lead time on Bulldozer reviews we weren't able to do much experimenting with Windows 8 performance on the platform. There's also the fact that Windows 8 isn't expected out until the end of next year, at which point we'll likely see an upgraded successor to Bulldozer."

"Currently Windows 7 is unaware of the shared nature contained within the FX-8150 processor. As a result, there are opportunities for resource sharing or activate higher turbo core frequencies are missed"


I'd be interested in seeing how it performs with OS that does recognize the architecture and can optimize it.

Thats a cop out. While it is true that using the second SMT core in a BD module, you take a hit in performance, but again, theres no reason why AMD couldn't simply call it a HT core via CPUID, which would have solved the entire issue. Blame the marketers for that one...

And no, BD is hardly revolutionary; 2-way SMT has been around for over a decade...

 

[dingo44]

Best answer selected by dingo44.

 

[Maziar]

This topic has been closed by Maziar