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

#1
reverze
ill be getting a bulldozer next month :) gonna save money on the CPU and not to mention the mainboard also for top performance ( more than i even need )
Posted on Reply
#2
erocker
by: JrRacinFan
If they want to boast numbers, show actual screenshots. Just a few slides does not have me beleiving this.
I don't even care about screenshots. Show BD vs. 980x vs. 2500/2600K in gaming. Hmmm, what will come out on top? I already know.

Judging by the few gaming benchmarks (especially F1 2010) it looks like BD's IMC is just a bit behind Intel's older gen. i7.

It's going to be a great chip for the price.

by: newtekie1
That isn't what AMD says. From their own marketting slide: "4 extra cores"...
They can speak all the marketing jargon they want. Performance wise the chip either is bad at multi threading "real" cores or, it's really good "hyper threading".
Posted on Reply
#3
cdawall
where the hell are my stars
by: heky
If YOU cant notice a difference, that doesnt mean its not there. Not by a long shot.
Never said there was no difference. Said to my eyes I could not tell. Was playing @1080p on a projector so that may have hidden some.
Posted on Reply
#4
NC37
Like you'd expect AMD internal benching to be completely accurate. Stop getting your undies sweaty over this. You all wanted benches and now your complaining cause they aren't independent. But AMD did give you benches finally.

Performance is on par with what I was expecting. Something that can put them in the range of i7, even if low end, and something better than i5. Truthfully that is all that matters is that they at least have something they can be competitive with.

In the end, FX will probably be akin to the Radeon 3870. It couldn't beat NV but it got in at a price point which saved ATI till 4k series. As I said before, I'd be more interested in Piledriver than BD. BD to me seemed like a small step up from Phenoms while PD was looking to change more things.

Let the stopgap wars begin!
Posted on Reply
#5
Benetanegia
by: cdawall
How about the fact that nothing except for core 2 has ever been "50%" faster than the last gen and even then it wasn't "50%" faster than the first set of core architecture chips.
And that has nothing to do with what I said. SB-E will have 50% more cores and will be 50% faster than SB in multi-threaded apps. Fact.

BD seems to be on par with SB. Fact(?) according to AMD's internal benchmarks.

A+B= SB-E will be 50% faster than BD at least in those apps where BD is faster, due to having more cores.

SB-E is NOT the next gen. It's the high-end part of the current gen, while SB was the mid-range. It should have been obvious considering the prices for SB.
I have been an AMD fan for a while now as everyone on here knows however. I still argue off of what i own in games i noticed no difference between a >4ghz clocked xeon 4c/8t with faster ram and my 1090T @4ghz. I have not personally played the new SB chips nor do i care too there is not enough performance gain. Your Q6600 having a crap FSB setup probably accounts for 90% of your speed increase.
PII was never much faster than Intel's Conroe. If a huge difference exists between Conroe and SB, it also exists between SB and PII. Many TPUers went to SB from PII's and have accounted for this massive difference. I guess it's good for you that you don't?
AMD can if this launch is done correctly. All they have to do is get a hold of more than HP/Compaq and Acer. You would be surprised the number of inexpensive machines sold vs expensive ones at BBY. If I had to choose between an Acer quad core AMD and Dell quad core Intel both with IGP's I am sorry but the AMD/Ati package makes more sense to me even if I loose 10-20% cpu performance I gain all of that back in GPU performance. Think that doesn't sell? your wrong.
Oh sure IF they can snatch those vendors, but that's a big if because Intel can always undercut AMD's price where required.

And BD does not have IGP so I don't know wtf are you saying there. SB on the other hand does have IGP, which is why it will most probably win all those vendors. Ivy will have an even better IGP.

Llanos and Piledrivers sure, but that's a story for another day. We are discussing Bulldozer.
Posted on Reply
#6
Dent1
by: Shihabyooo
Applause for what ? A self-claimed increase in performance against dated hardware ?
by: Shihabyooo

Or maybe it's outstanding value compared to a niche product aimed for a very limited section of the market ??
So which hardware should they have review it against :)


So the i5 2500k and i7 2600k which came out this year are dated?

by: Shihabyooo



You should know better than to believe companies self-made benchmarks. Just wait for a unbiased source to do a review on them. If these numbers turn out to be accurate (against rivalling products from the same class/price range) then I'll clap my hands till they bleed.
Obviously the benchmarks are somewhat bias they are from AMD. But we can only go by the information we have at the moment whether true or not - and the information which we have right now is the AMD Bulldozer is the superior CPU. Until unbiased AMD reviews become available AMD deserve the benefit of the doubt and some applause. Let them have their victory until a unaffiliated website gets a chance to review it.
Posted on Reply
#7
Benetanegia
by: Dent1
So which hardware should they have review it against :)
Sandy Bridge. But in every benchmark, not only in those where it could win, while using an old and expensive CPU for gaming benchmarks and then say something as stupid as "$800 doesn't buy much in terms of performance improvement." God, stfu AMD and use a 2500k, then try to make the same claim with a straight face.
Posted on Reply
#8
Dent1
by: Benetanegia
Sandy Bridge. But in every benchmark, not only in those where it could win, while using an old and expensive CPU for gaming benchmarks and then say something as stupid as "$800 doesn't buy much in terms of performance improvement." God, stfu AMD and use a 2500k, then try to make the same claim with a straight face.
Yes it would be nice to see Sandybridge in every benchmark and AMD's quotes were in bad taste.

However, a few days ago the same people in this forum were saying Bulldozer will be slower than Sandybridge. Heck you guys were saying Bulldozer wouldnt touch any of the i7/i5 family including the Gulftown. The fact Bulldozer has been seen atleast on a FEW occassions to beat out Sandybridge (even if was cherrypicked results) demonstrated that AMD still proved you guys wrong.

Edit:

by: Shihabyooo



The Xeon 5150 and the i7 980x are..
Maybe so, but the 2500k and 2600k not old


by: Shihabyooo


I would've given them that, if they hadn't called the 6990 "the world's fastest graphics card" :shadedshu .
That has nothing to do with Sandybridge vs Bulldozer. Stay on topic.
Posted on Reply
#9
Shihabyooo
by: Dent1
So which hardware should they have review it against :)
by: Benetanegia
Sandy Bridge. But in every benchmark, not only in those where it could win, while using an old and expensive CPU for gaming benchmarks and then say something as stupid as "$800 doesn't buy much in terms of performance improvement." God, stfu AMD and use a 2500k, then try to make the same claim with a straight face.
^ this

by: Dent1

So the i5 2500k and i7 2600k which came out this year are dated?
The Xeon 5150 and the i7 980x are.

by: Dent1

Obviously the benchmarks are somewhat bias they are from AMD. But we can only go by the information we have at the moment whether true or not - and the information which we have right now is the AMD Bulldozer is the superior CPU. Until unbiased AMD reviews become available AMD deserve the benefit of the doubt and some applause. Let them have their victory until a unaffiliated website gets a chance to review it.
I would've given them that, if they hadn't called the 6990 "the world's fastest graphics card" :shadedshu .
Posted on Reply
#10
NdMk2o1o
by: Benetanegia
stfu AMD and use a 2500k
They did and a 2600k or did you just decide to omit that because you felt like it?

Jesus, I can't wait until BD is released so all the fanboyism (from both sides) will stop, and as has been the case with AMD for the last 5 years even if they don't take the performance crown from the prices they are going to go for should be about right for the performance they give.
Posted on Reply
#11
erocker
by: Dent1
Yes it would be nice to see Sandybridge in every benchmark and AMD's quotes were in bad taste.

However, a few days ago the same people in this forum were saying Bulldozer will be slower than Sandybridge. Heck you guys were saying Bulldozer wouldnt touch any of the i7/i5 family including the Gulftown. The fact Bulldozer has been seen atleast on a FEW occassions to beat out Sandybridge (even if was cherrypicked results) demonstrated that AMD still proved you guys wrong.
With things like gaming and other things where more memory bandwith is utilized, AMD FX series is slower. How much will remain to be seen. A comparison of a SB rig vs. a BD rig running CrossFire or SLi will clearly show this. That being said, BD is a winner at its price point.
Posted on Reply
#12
RuskiSnajper
by: mtosev
internal amd tests aren't relevant.somebody else who isn't associated with the company should review the cpu:)
no brainer ;)
Posted on Reply
#13
TheLaughingMan
by: erocker
With things like gaming and other things where more memory bandwith is utilized, AMD FX series is slower. How much will remain to be seen. A comparison of a SB rig vs. a BD rig running CrossFire or SLi will clearly show this. That being said, BD is a winner at its price point.
Even I will have to watch for TPU to see BD in Crossfire.
Posted on Reply
#14
MxPhenom 216
Corsair Fanboy
8 cores needed to beat 4 cores? hahahahahahahaha:roll:
Posted on Reply
#15
de.das.dude
Pro Indian Modder
by: nvidiaintelftw
8 cores needed to beat 4 cores? hahahahahahahaha:roll:
the haha is 4 cores cost more than 3times of 8 cores. shows how you are getting ripped off.
Posted on Reply
#16
Dent1
by: nvidiaintelftw
8 cores and 8 threads needed to beat 6 cores and 12 threads hahahahahahahaha:roll:
I fixed it :)
Posted on Reply
#17
erocker
by: TheLaughingMan
Even I will have to watch for TPU to see BD in Crossfire.
Hopefully it will be done, but if not I will post results.
Posted on Reply
#18
cdawall
where the hell are my stars
by: Benetanegia
And that has nothing to do with what I said. SB-E will have 50% more cores and will be 50% faster than SB in multi-threaded apps. Fact.
NO, NOT all multithreaded apps are the same, SB will not be faster in the vast majority which cap out at 4 cores. It will not be faster in any of those same games, etc. You are also assuming multithreading is linear. It is not take cinebench a "normal HT-less" quad will often score 3.79X when multithreaded vs a single core. thats not 100% scaling just incase your math needs help.
Posted on Reply
#19
MxPhenom 216
Corsair Fanboy
by: de.das.dude
the haha is 4 cores cost more than 3times of 8 cores. shows how you are getting ripped off.
uhhh if my math is correct $219 is not 3x more then $245
Posted on Reply
#20
cdawall
where the hell are my stars
by: nvidiaintelftw
uhhh if my math is correct $219 is not 3x more then $245
if my math is correct they were trading blows. wait lets use your logic intel is probably still faster at superpi no matter what. better chip cause of it right?
Posted on Reply
#21
MxPhenom 216
Corsair Fanboy
by: cdawall
if my math is correct they were trading blows.
yeah but he was like 4 cores that a priced 3x higher then 8 cores shows how much you got ripped off.
Posted on Reply
#22
Benetanegia
by: NdMk2o1o
They did and a 2600k or did you just decide to omit that because you felt like it?

Jesus, I can't wait until BD is released so all the fanboyism (from both sides) will stop, and as has been the case with AMD for the last 5 years even if they don't take the performance crown from the prices they are going to go for should be about right for the performance they give.
Oh I must be blind, or maybe you can't read my post. This is not meant to be offensive, it's just that either of us is reading something wrong.

Show me the benchmark where BD is put against SB (2500k or 2600k I don't care) in gaming benchmarks. I can't find that chart. They deliberately used the slower and much more expensive 980X for gaming and the fewer core SB for multi-threaded apps. Also whenever price or perf/price is mentioned, the 980X is again used, instead of the faster and far far cheaper SB.
Posted on Reply
#23
Dent1
Benetanegia. But that only holds true for gaming.

Putting gaming aside until we have better gaming benchmarks available. We all know the 980X is faster than the Sandybridge overall (due to additional cores/threads/cache).

It is clear the Bulldozer is as faster than the 980X, thus the Sandybridge overall. Again putting gaming aside.

Considering that gaming is just 1 cateogry, it's fair to say the Bulldozer is faster than the Sandybridge overall and is the more well rounded choice according to the information we have right now.
Posted on Reply
#24
Benetanegia
by: cdawall
NO, NOT all multithreaded apps are the same, SB will not be faster in the vast majority which cap out at 4 cores. It will not be faster in any of those same games, etc. You are also assuming multithreading is linear. It is not take cinebench a "normal HT-less" quad will often score 3.79X when multithreaded vs a single core. thats not 100% scaling just incase your math needs help.
You just proved my point thanks. In any and every benchmark where Bulldozer is faster because it's 8 cores supose an advantage over less cores, SB-E will necessarily be (almost) 50% faster than SB. ANY code capable of extracting the most out of an 8 core CPU will ALWAYS extract the same, most probably MORE, performance out of 6 cores. If in those same benchmarks, the 8 core BD is just as fast as the 4 core SB, it will necessarily be almost 50% slower than the 6 core SB.

by: Dent1
Benetanegia. But that only holds true for gaming.

Putting gaming aside until we have better gaming benchmarks available. We all know the 980X is faster than the Sandybridge overall (due to additional cores/threads/cache).

It is clear the Bulldozer is as faster than the 980X, thus the Sandybridge overall. Again putting gaming aside.

Considering that gaming is just 1 cateogry, it's fair to say the Bulldozer is faster than the Sandybridge overall and is the more well rounded choice according to the information we have right now.
That's irrelevant. AMD clearly used SB and the 980X in the way they did (that is, where they are respectively weaker), because that's the only way to show BD in a "good light". Otherwise they would have been consistent on using either the 2500k, 2600k or 980X for every benchmark.

- If they had used 2500k the price/perf argument would be invalidated, especially in gaming, where the 2500k would be faster AND cheaper, which is why they didn't do that direct comparison.

- If they used 2600k it would have been a close fight in price/perf in almost all categories, according to AMD's own benchmarks. But the only thing Intel would need to do is lower 2600k price. Matching is bad when you opponent can price you out f the market if so he wishes.

- If they had used the 980X in multi-threaded benchmarks a 2 year old architecture would have consistenly beaten BD, which would not be good marketing.
Posted on Reply
#25
the54thvoid
In the anandtech article on SB, the 980 loses 7 out of ten tests to the core i7 2500 and i7 2600 (in gaming). To be honest guys, it's a marketing gimmick. If it shows the BD barely beating the 980, then it clearly by logic is a long way off beating SB.

That being said, the 980 bested SB on two runs of Civ V (heavily core dependent) and DA:O.

There is no point comparing BD with the 980 as it is an obsolete chip for most people. The 980 (like SB-E) is a part designed for heavy multi threaded use and in particular workstations. If BD beats the SB architecture at heavy multi-tasking then it's good for AMD. But it's a bit poo for the desktop gamers (most of whom are arguing here). If a cheaper core i7 2500 can beat an i7 980 gaming (which has been shown by AMD's own slide to beat BD in most gaming cases shown) then BD coming in at more than a 2500 is bad for AMD.

http://www.anandtech.com/show/4083/the-sandy-bridge-review-intel-core-i7-2600k-i5-2500k-core-i3-2100-tested/20
Posted on Reply
Add your own comment