Saturday, September 24th 2011

AMD FX 8150 Looks Core i7-980X and Core i7 2600K in the Eye: AMD Benchmarks

The bets are off, it looks like Intel is in for a price-performance shock with AMD's Bulldozer, after all. In the press deck of AMD FX Processor series leaked by DonanimHaber ahead of its launch, AMD claims huge performance leads over Intel. To sum it up, AMD claims that its AMD FX 8150 processor is looking Intel's Core i7-980X in the eye in game tests, even edging past it in some DirectX 11 titles.

It is performing on par with the Core i7-2600K in several popular CPU benchmarks such as WinRAR 4, X.264 pass 2, Handbrake, 7Zip, POV Ray 3.7, ABBYY OCR, wPrime 32M, and Bibble 5.0. AMD FX 8150 is claimed to be genuinely benefiting from the FMA4 instruction set that Sandy Bridge lacks, in the OCL Performance Mandelbrot test, the FX 8150 outperforms the i7-2600K by as much as 70%. Lastly, the pricing of the FX 8150 is confirmed to be around the $250 mark. Given this, and the fact that the Core i7-2600K is priced about $70 higher, Intel is in for a price-performance shock.
Source: DonanimHaber
Add your own comment

854 Comments on AMD FX 8150 Looks Core i7-980X and Core i7 2600K in the Eye: AMD Benchmarks

#526
bear jesus
Musselsthe point was that heaps of new models come out with zero performance improvements. they might be cheaper (5870 vs 6870) and rarely (FX series) far worse.


a new model does not always mean higher speeds.
Well yes i agree on that but i was thinking more along the lines of the improvement coming from more cores and that there not being a decrease per core, my main thought is about the supposed benchmarks is decreases in performance as in it being slower so not even the same speed clock for clock even though there is supposed to be an increase of the IPC.

I admit i was heavily suggesting an increase per core or an increase overall due to it being the next generation but would you not agree it would be unlikely for it to be a reduction over the previous generation?

Even when there is no improvement between generation due to whatever reason is it not very rare that there is a decrease in performance?
Posted on Reply
#527
eidairaman1
The Exiled Airman
Musselsthe point was that heaps of new models come out with zero performance improvements. they might be cheaper (5870 vs 6870) and rarely (FX series) far worse.


a new model does not always mean higher speeds.
ya it can mean reduction in power and production costs but then also overclocking headroom to boot
Posted on Reply
#528
Inceptor
Musselsthe point was that heaps of new models come out with zero performance improvements. they might be cheaper (5870 vs 6870) and rarely (FX series) far worse.


a new model does not always mean higher speeds.
5870 vs 6870 is not a valid comparison.
Their numbering scheme may be similar, but they are not equivalent performace tiers.
5970 vs 6990 is valid, and there is a performance increase in the 6990.
5870 vs 6970 is valid, and there is a performance increase in the 6970.
5850 vs 6870 is valid, and there is a performance increase in the 6870, albeit a slight increase.
5770 vs 6850 is valid, and there is a performance increase in the 6850.

It's amazing that even over a year since release and explanation, the 5870 vs 6870 and 5850 vs 6850 mistake is still made, constantly. What flavour of gpu core that is in each card is irrelevant, what matters is the performance tier in which the card is meant to compete.
Posted on Reply
#529
ensabrenoir
.....this story is already writen in 2 week ....hopefully some 3rd party reviews that will show tons of Future possibilities and marginal present gains. World wide web war breaks out between sides. Sandy bridge e drops....expensive powerful. Insert arguement about cost / performance increase etc. Onto piledriver vs ivybridge
Posted on Reply
#530
NAVI_Z
havin much fun reading everyone's posts!:laugh:

some folks here say to wait and see what the real world numbers will look like before buying.

very wise imho. its still fun to read the opinions over speculation ;)

will have fingers crossed until after Oct 12th.... :D:toast:
Posted on Reply
#531
TRWOV
I think that the french magazine is very much right since AMD themselves said that BD would be targeted at Core i5-2xxx performance.
"The AMD FX-8120 is probably the model of the new series "Bulldozer" that offers the best price/performance ratio. It's able to compete with the i5-2500K in most computing applications even if it lags behind in video games. But don't forget that overclocking capabilities are available with no additional cost. Faced with the old Phenom X4 980, this is a very good alternative."

"Offered at a slightly lower price than the Intel Core i7-2600K, the FX-8150 is currently the most powerful model of the new "Bulldozer" architecture from AMD. Unfortunately, it fails in comparison, at best it can match its direct competitor in some media processing applications but is always behind in games."
library.madeinpresse.fr/samples/MPqY2Vg2v45Z-f p.9 "Nos choix" panel.


Nobody thought that BD could "beat" SB, not even AMD (if they did they would have said so) BUT with the announced prices and the fully unlocked multiplier I think that BD would be a very good choice for gaming on a budget. People with larger budgets will go for the highest performance regardless of cost but that isn't AMD's turf right now.

Clock for clock BD will fall behind SB, of course, but that only matters in technical discussion. For the end user it's about the price/performance and I hope that AMD can deliver. :respect:



By the way, does someone know how to convert from AMD's TDP to Intel's TDP? :confused: I think that AMD should market their consumer CPUs with the same measure. People are going to see the 95w v 130w and probably choose SB because it seems to be "greener". :nutkick:
Posted on Reply
#532
YautjaLord
2ensabrenoir:

Pile Driver have to be monstrous car to crash Ivy's Bridge; i'd like to see the Ped Basher vs Army Hangar. :laugh: jk Just having laugh @ the AMD's & Intel's CPUs naming scheme; though AMD reminds me why i love Carmageddon & this firm's CPUs. They had SledgeHammer, don't they? :) How's that ambulance from Carmageddon Splat Pack called? Fits in it's nature to AMD names. :toast:
Posted on Reply
#533
TRWOV
YautjaLord2ensabrenoir:

Pile Driver have to be monstrous car to crash Ivy's Bridge; i'd like to see the Ped Basher vs Army Hangar. :laugh: jk Just having laugh @ the AMD's & Intel's CPUs naming scheme; though AMD reminds me why i love Carmageddon & this firm's CPUs. They had SledgeHammer, don't they? :) How's that ambulance from Carmageddon Splat Pack called? Fits in it's nature to AMD names. :toast:
You know, If AMD had chosen to name BD CPUs as "Phenom III" I think that there wouldn't be such an uproar over performance numbers. By using the FX moniker people's expectations just went apeshit. :roll:

Calm down people.:slap: It's just two more weeks.
Posted on Reply
#534
Damn_Smooth
TRWOVYou know, If AMD had chosen to name BD CPUs as "Phenom III" I think that there wouldn't be such an uproar over performance numbers. By using the FX moniker people's expectations just went apeshit. :roll:

Calm down people.:slap: It's just two more weeks.
Finally, some words of wisdom.
Posted on Reply
#536
Wile E
Power User
MelvisYea ok i can understand that, more cores to get the 50%. But that doesn't mean it will perform 50% better.

It will be an increase that's for sure, but it wont be 50% no way, regardless if it gets 50% more cores.

Ok to me if its going to be a 50% increase i would expect the new SB-E to be 50% faster then there current 980/990X CPU's ( both 6 cores of course) , if not then its not going to be anything to write home about in my eyes. As SB vs the old 1366 975 isnt much different in performance at all if any. Just more affordable.
SB is faster clock for clock/core for core than 1366.

Besides, I never decided the context of the 50% comment. That was established earlier by someone else as being 50% faster than current SB and BD. And while 50% more cores doesn't give exactly a 50% increase, it's damn close in multithreaded apps. 50% was being used as a ballpark figure. And in that context, it's pretty much correct.

Now, that says nothing of pricing. The 50% increase over BD/SB will cost an arm and a leg.
mastrdrver50% faster (probably) in well threaded programs. Otherwise it will be just as fast as SB regardless of how much PR Intel throws at it.
Or able to run 50% more tasks without a performance penalty. Still 50% faster. You just have to use it to it's potential. That's like saying a Ferrari isn't faster than a 370z. Just because you don't use the available performance on your drive to work, doesn't mean it isn't there when you do want or need it.
Posted on Reply
#537
YautjaLord
TRWOVYou know, If AMD had chosen to name BD CPUs as "Phenom III" I think that there wouldn't be such an uproar over performance numbers. By using the FX moniker people's expectations just went apeshit. :roll:

Calm down people.:slap: It's just two more weeks.
lol'd

True, more or less what i say: while i participate here, why not have a good laugh. BTW: the ambulance from Carma Splat Pack called BloodMobile; will AMD use that name for some future CPU? :roll: jk

Seeya all in 10 days (actually few hours less lol) from now. :toast:
Posted on Reply
#539
erocker
*
Damn_SmoothSigh, when will people learn to quit posting bullshit from OBR? The troll has already failed.
On other forums people get infracted for posting his stuff. No joke.
Posted on Reply
#540
mastrdrver
Wile ESB is faster clock for clock/core for core than 1366.

Besides, I never decided the context of the 50% comment. That was established earlier by someone else as being 50% faster than current SB and BD. And while 50% more cores doesn't give exactly a 50% increase, it's damn close in multithreaded apps. 50% was being used as a ballpark figure. And in that context, it's pretty much correct.

Now, that says nothing of pricing. The 50% increase over BD/SB will cost an arm and a leg.



Or able to run 50% more tasks without a performance penalty. Still 50% faster. You just have to use it to it's potential. That's like saying a Ferrari isn't faster than a 370z. Just because you don't use the available performance on your drive to work, doesn't mean it isn't there when you do want or need it.
Maybe fast clock for clock but only core per core because the uncore is clocked higher. 90% of desktop work that is going to be done is not going to see this magical 50% fast part either. A SSD (from what I hear) would be worth it more then a SB-E system.

And where is this mythical performance left untapped you talk of? It is in well threaded programs. Of which very few exist that will see work on a desktop PC. Server is a different story and I would agree with the 50% faster on them.

You also drive a Ferrari instead of a 370Z because you have the money and want to show it off. Not necessarily because it's faster, that's a secondary concern. Ferrari's also are higher maintenance then a 370Z. What are you trying to say about SB-E? :p (j/k)
Posted on Reply
#541
Wile E
Power User
None of your concerns change the fact that it is 50% faster, give or take. Whether you use it all the time or not is completely irrelevant. Why SB is faster than the 1366 i7's is also completely irrelevant, all that's relevant is that it is faster clock for clock.

And all you need to do to see the 50% increase is encode some videos. Not an uncommon task at all.
Posted on Reply
#542
Covert_Death
Super XPThese numbers look like they came out of an aris. More nonesense without any backing...
here's what i don't get...

they OC'd an i7 more then the 8150, i know it's "clock to clock" but that is utterly pointless really, if the 8150 has as much overhead to OC as we are seeing how about we do either "stock to stock" or "max to max" because really those are the two things that matter, nobody is going to clock an 8150 to 4.2 just because its a good number, they will either leave it stock or crank it up high enough to feel good about themselves. now sure SOME people might land at 4.2 but its retarded to think that its a good comparison to OC them to the same clock, when that comparison doesn't mean anything...

if the i7 is faster at the SAME clock, but the 8150 can OC much more then the i7 then what good is that test? its pointless
Posted on Reply
#543
Damn_Smooth
erockerOn other forums people get infracted for posting his stuff. No joke.
I can see why. The guy admitted himself that he can't be trusted.

I wouldn't blame you if you wanted to start doing that yourself. :D
Posted on Reply
#544
mastrdrver
Wile ENone of your concerns change the fact that it is 50% faster, give or take. Whether you use it all the time or not is completely irrelevant. Why SB is faster than the 1366 i7's is also completely irrelevant, all that's relevant is that it is faster clock for clock.

And all you need to do to see the 50% increase is encode some videos. Not an uncommon task at all.
It's relevant if the performance does not justify the extra price.

It was like that with Lynnfield vs Bloomfield. The only real reason to get Bloomfield was for triple SLI/Crossfire and 6 core CPU support. You could easily (about 2 yrs ago) build a Lynnfield system for a lot cheaper then a Bloomfield one.
Posted on Reply
#545
Wile E
Power User
mastrdrverIt's relevant if the performance does not justify the extra price.

It was like that with Lynnfield vs Bloomfield. The only real reason to get Bloomfield was for triple SLI/Crossfire and 6 core CPU support. You could easily (about 2 yrs ago) build a Lynnfield system for a lot cheaper then a Bloomfield one.
Who decides what justifies the extra price? I found it justifiable enough to buy a 980X.

We weren't talking price/performance here. We were talking raw numbers. 50% faster is 50% faster. Price doesn't change that.
Posted on Reply
#546
Mussels
Freshwater Moderator
Inceptor5870 vs 6870 is not a valid comparison.
Their numbering scheme may be similar, but they are not equivalent performace tiers.
5970 vs 6990 is valid, and there is a performance increase in the 6990.
5870 vs 6970 is valid, and there is a performance increase in the 6970.
5850 vs 6870 is valid, and there is a performance increase in the 6870, albeit a slight increase.
5770 vs 6850 is valid, and there is a performance increase in the 6850.

It's amazing that even over a year since release and explanation, the 5870 vs 6870 and 5850 vs 6850 mistake is still made, constantly. What flavour of gpu core that is in each card is irrelevant, what matters is the performance tier in which the card is meant to compete.
you really are missing the point. as far as everyone was concerned on launch, thats exactly what it meant. and thats why everyone thought a slower, cheaper product was released as a succesor. this is a direct reply to the initial comment someone made that slower hardware is never released to replace something - in this case, it was. they redid the naming scheme, but but direct numerical successor was slower when they did.
Posted on Reply
#547
Covert_Death
yes but a new tearing system means you work from the top down.... you compare 1st to 1st 2nd to 2nd .... not 2nd to 1st just because " the numbers have 3 digits in common"
Posted on Reply
#548
Super XP
Hey everybody 9 days left for Bulldozer's launch. I am hoping there will be no more delays :twitch: :D
Posted on Reply
#549
Inceptor
Musselsyou really are missing the point.
mea culpa, mea maxima culpa.:eek:
Posted on Reply
#550
sk
official Test

hello techp forums and peeps.
Heres an official bench, posted this weekend.

8150 is about 17 proc's down @ 8,600+
www.cpubenchmark.net/high_end_cpus.html

so im guessing the 8170 released down the road will be up to the Intel EX SB's.
not bad not bad,

this will be my 1st "personal" AMD build in 10yrs, but i've built alot for others...
i cant wait
Posted on Reply
Add your own comment
Nov 23rd, 2024 02:51 EST change timezone

New Forum Posts

Popular Reviews

Controversial News Posts