AMD Bulldozer / FX-8150 Desktop Performance Review @ [H]ardOCP

I wonder what (if any) AMD's reaction will be. Now that Tom's Hardware, here, and many other sites have all released their reviews that are far from glowing. ALL the reviews I have seen said the same thing..disappointing with AMD, poor single thread performance etc etc.

I would love to see how AMD spins this disaster. I think I will keep my 1090T running @ 4Ghz for the time being since there is ZERO advantage to upgrading to a FX chip any time soon.
 
will still buy the Fusion/Llano products for family/friends but BD should have been released a couple years ago when they had all the hype, 2 years ago this would have been a good cpu


but when it comes down to it, it seems more often then not software isnt taking advantage of all that hardware can offer anyways.
 
very true, most software doesn't come close to harnessing the full potential of much of the available hardware (GPU's aside).
 
That's because most people are busy playing shooter console ports. Shooters have never demanded much CPU power.

Try some strategy titles which actually need processing power and the differences in old and new CPUs( and Bulldozer and SB) start to really show.
 
eastmen said:
Thuban is only 904M tranistors. Just doubling that would give you 1.808B tranistors almost 200m tranistors less than bulldozer.


So they oculd have made a 6 core cpu with a smaller die size than bulldozer on 32nm .

Deneb is 758m tranistors , so for 1.516B tranistors they could have built a true 8 core cpu.

Anyway you look at it i think it would have been a win over bulldozer
Click to expand...

That's what I've been thinking. A die shrunk Thuban would have been better.
 
guess I will keep my x4 @ 3.9ghz for a few more months.

still likely going to go bulldozer, don't want to buy a new mobo too. Those power consumption numbers seems really off, I hope it's a matter of the process needing to mature, but I thought llano was on this same process.
 
Thanks for the thorough reviews. I've used a mixture of AMD and Intel CPUs in the past and Bulldozer is more of what I've come to expect from AMD; I think I'm going full-blown Intel fanboy after this. heh :)
 
you can buy a 32nm thuban on F1 socket I think, maybe it's a lesser core. would be interesting to see how those overclock.

sorry double post.
 
I'm personally interested in replace the 965 125W with the 95W 4100 for a virtualization server that I already have. It would be neat to have my virtual firewall be able to do hardware AES.
 
a price reduction is needed for 8150 to be 200 dollars and 8130 to be 150 dollars and drop the x6 to 120 and x4 to less than 100 to compete
 
On all accounts the perfornance here really is atrocious. The review really is too kind. The single-threaded performance is horrible, IPC, and power-draw are unreal. The added cores weren't even able to make up for the obvious lack of overall performance.

So obvious, in fact, that it begs to wonder just how something like this was allowed to happen? Was it a failure on Glofo's part? They've been having issues with producing quality 32nm llano chips at high enough quantity. Has this passed on to Bulldozer? Is it not a silicon issue but an architectural one? When you look at the architecture, it's quite clear they've hedged their bets on increasing the frequency and lengthening the pipeline. Why? If i'm not mistaken, wouldn't that essentially lead to the powerdraw issues we've seen here? It didn't even increase their overall per-core performance.

I mean... what the hell? It seems like a monumental failure for both AMD and GloFo at this point. I really wanted them to do well. I was even expecting them to do well. As a longtime AMD fan since the T-bird days, I can comfortably say that my next build will have an Intel in it. I feel like i'm being forced to give money to the bad guy here, but that's how disappointing this really is.

To recap:

45nm>32nm and we've seen an increase in power draw, lower per-core performance, lower IPC. That's ~2billion more transistors and they've only managed going backwards in all the places you'd expect an increase in performance. Netburst is about the only word I can come up with.
 
Last edited:
I remember a while a go when there was some other talk about Bullsozer I said that I suspect it'll be slower even than the i7-900s. It would appear it wasn't wrong to suggest that although it would of been nice to have some real competition, even though Intel seems quite happy to fight with itself in the ring and keep producing better processors.
 
Guess my next build will be Intel. Again. :( Unless I can find an 8150 and board for $250 total, and even then I feel like it would just be a sidegrade from my Q9550...

Thanks for the comprehensive review, Kyle. :)
 
stealthy123 said:
guess I will keep my x4 @ 3.9ghz for a few more months.

still likely going to go bulldozer, don't want to buy a new mobo too. Those power consumption numbers seems really off, I hope it's a matter of the process needing to mature, but I thought llano was on this same process.
Click to expand...

I mean I don't wanna say the fanboy word but do you really believe that everyone's review numbers of the power consumption are off??
 
As with shitty console ported pc games, I will not buy this cpu just to support the company and because competition is good. If they want my money they need to show me the benefits. Not my job to make their executives even richer.
 
shoota said:
I mean I don't wanna say the fanboy word but do you really believe that everyone's review numbers of the power consumption are off??
Click to expand...

I don't mean the reviews are off, but that there is problems with current leakage and other crap with the current spin of the bulldozer silicon.

I was just suggesting it could be a simple CPU revision and the next release will be more in line with not such nutty power draw.
 
stealthy123 said:
I don't mean the reviews are off, but that there is problems with current leakage and other crap with the current spin of the bulldozer silicon.

I was just suggesting it could be a simple CPU revision and the next release will be more in line with not such nutty power draw.
Click to expand...

Ah ok good clarification. I hope you're right but there doesn't seem to be much hope here. As it stands right now there is only a small segment of people out there that this would be an advisable upgrade.
 
I just ordered one.

I see a lot wrong with this review. No mention of the stepping that I can see, and it seems like the main complaints were with the 8120 not performing as high as you wanted it to?

What is everyone's obsession with purchasing a step lower and expecting it to be just as good as the one that costs a few bucks more? The processors may be physically the same, with different tolerances indicating what becomes 8120 vs 8150, just like with the 6970 and 6950 (which is also AMD, see the trend?).

I'm building my next rig to play BF3 first and foremost, and it seems to perform (even in this review) more than adequately for my liking. Hate all you want, this thing at *least* competes with Intel SB CPU's, despite the flaws pointed out in the review.
 
AlexStenka said:
The thing you guys are all forgetting these babies are 8 core CPUs, when software start to take advantage of all 8 cores where does that leave intel?
Click to expand...

[facepalm]

savantu said:
When software takes advantage of all 8 cores, Intel will be happy since it means all of SB's 8 threads will be put to use. Intel is already there.
Click to expand...

-Dragon- said:
Did you even READ the articles and LOOK at the benchmarks? They tested lots of software that takes advantage of all 8 cores and even then the best AMD's so called "real" cores could muster was keeping up with intel's 4.
Click to expand...

^^^ These. ^^^

bcorcoran said:
What is everyone's obsession with purchasing a step lower and expecting it to be just as good as the one that costs [strike=]a few bucks more[/s] less?
Click to expand...

FTFY. Maybe you're not reading it correctly, but an 8 core $205 FX-8120 basically got it's ass handed to it by a 4 core 2500K which technically is cheaper ($179 @ B&M Microcenter).
The only one in this case that costs more is the BD. Why would you pay more for lesser performance?

bcorcoran said:
Hate all you want, this thing at *least* competes with Intel SB CPU's, despite the flaws pointed out in the review.
Click to expand...

It doesn't "compete"; it loses to SB at the same price point. That's not competing, that's losing.
 
Last edited:
God damn it. What a disappointment. Fuck AMD, who the hell puts out a new product that is worse than the previous one? I guess my workhorse Q6600 will serve me a little longer while I ponder whether I need the i7 or can make do with the i5.
 
bcorcoran said:
I just ordered one.

I see a lot wrong with this review. No mention of the stepping that I can see, and it seems like the main complaints were with the 8120 not performing as high as you wanted it to?

What is everyone's obsession with purchasing a step lower and expecting it to be just as good as the one that costs a few bucks more? The processors may be physically the same, with different tolerances indicating what becomes 8120 vs 8150, just like with the 6970 and 6950 (which is also AMD, see the trend?).

I'm building my next rig to play BF3 first and foremost, and it seems to perform (even in this review) more than adequately for my liking. Hate all you want, this thing at *least* competes with Intel SB CPU's, despite the flaws pointed out in the review.
Click to expand...


thank you for supporting the competition with your wallet

then you will see "a lot" wrong with every review ive seen thus far.

the price when you factor in the mobo and higher end PSU and additional power consumption does not make this any more attractive, slice it anyways you want, in the end its not even a best bang for buck scenario. AMD had that in the fusion line with Llano, but they missed the mark on BD, BD needs to drop in price alot to get into informed peoples shopping carts

sure it competes with the cheaper, mid range SB...for more money.
 
Yeah, can't see any practical reason someone would get a BD. It loses out in pretty much everything, no matter which way you see it. Price, single-threaded performance, most multi-threaded performance, has no chance in strategy titles( Starcraft2, X3, etc.)...The only thing it manages to do okay in is console ported fps games which require little from the CPU.
 
Great review and thank you for adding in the 4Ghz clocks.

However one thing to point out in this quote

"The fact that your 4.6GHz overclocked Bulldozer system will be sitting there idling at 120 watts is not a bad thing at all.

Seeing the older overclocked i7-920 surely puts it all in to perspective though.

Probably what is most impressive is the i7-990X. The die size process between that CPU and the previous 920 generation is easily seen, and keep in mind that the 990X is clocked at 4.54GHz."

Those numbers must be way off on that 920 chip for you to be pulling that wattage at idle means you don't have speedstep enabled and what voltage are you using on the old c0 chip??

My numbers from my APC UPS which isn't 100% accurate as I have my modem,router and a few other items plugged in show 150 watts idle if I uplugged the rest of the crap it would probably be closer to 120watts idle.

I think it would be worth while to mention in the review you don't have any of the power saving featured enabled on the 920 setup.

Unless of course i'm wrong.


As for AMD chip nothing to see here move along.

a 6 core gulftown chip will be my next upgrade!
 
It's really disappointing that AMD has decided to release a processor UNWORTHY of the FX line.

I used to own a FX-60 chip, and that thing was the bee's knees.

This shit is fucking unworthy of the FX name. AMD should have known better than to release a chip 2 years behind the times instead of doing something actually worth a shit.

Hey, if you wanna know what I'm talking about, in 2006 when the FX-60 was released, AMD was doing so good, it's stock price was at $40ish, now? $4 . . . . Tells you all you need to know about how far behind AMD is. You can look it up yourself.

This chip is about 2 years late and $36 dollars per share short.
 
I just "upgraded" last night(as a stop gap measure, and before I read this review) to a lowly Phenom 2 545(20 buck on fleabay), and am pushing 3.7GHZ without even touching my voltages. I was saving my "real money" to buy one of these new Bulldozers. Guess I wont be now. This is the same disappointed feeling I had with the original Phenom, but they ended up releasing some fairly decent procs with the Phenom II line. Is there ANY bright spots to this architecture as far as future steppings and revisions(like with the jump from Phenom 1 to Phenom II), or is this just a giant turd without hope?
 
Fail said:
It's really disappointing that AMD has decided to release a processor UNWORTHY of the FX line.
Click to expand...
The last time a CPU was "worthy" of the FX line was before the Core 2 Duo came out in 2006.
 
I'm glad I bought a 990x when I did and not wait for bulldozer. Nice review! Still though the price of bulldozer is a nice compared to the Intel extreme chips.
 
Mr. Bluntman said:
Thanks for the comprehensive review, Kyle. :)
Click to expand...

Thanks for the kind words from all you guys. I tried my best to make sure that we gave a fair and objective look at what the Bulldozer was all about and I think we did a good job.

Trust me, from a business perspective, it would have been a lot better for our bottom line to have a true competing technology emerge. It would have been great for a lot of stories and advertising revenue. But, that just did not happen. Our first priority is making sure you guys get our solid and honest analysis and opinion from our staff. And I hope you think that is exactly what we delivered.
 
zero2dash said:
FTFY. Maybe you're not reading it correctly, but an 8 core $205 FX-8120 basically got it's ass handed to it by a 4 core 2500K which technically is cheaper ($179 @ B&M Microcenter).
Click to expand...

I was referencing 8120 vs 8150, but your point stands. My gripe with complaining about the 8120's performance is that there IS an 8150 sitting right there which is rated for higher clocks. I don't see any logic in why the reviewer chose to focus on the 8120 and used that to come to his conclusion, other than that it was cheaper and *could be* (likely that it is) from the same die.

AMD doesn't just toss a coin to decide which chips get the 8120 or 8150 badge. If they did, they'd be in for some real bullshit involving many many lawyers. Without seeing any numbers I can all but guarantee you that the 8150 will overclock better than the 8120. Just like the Opteron 165 in my sig overclocked way better than the consumer-level equivalent at the time. It's not a direct relationship in that sense, but it's similar.

Fact of the matter is, and I know this isn't in the spirit of overclockers in general, but if you want the best performance and your goal in a review is to review the PROCESSOR and not the overclocking performance, then at LEAST focus on each version individually rather than focus mostly on the #2. There's no merit to saying "well, enthusiasts will probably buy this chip, so I'll focus on that". That's just doing a disservice to your readers.
 
umcpgrad said:
people like to lie to themselves let them
Click to expand...

hey we need people to support them for the next cpu :) ..just hope it isn't 4 years and 2 years too late again.. Intelreally does need some enthusiast level competition


thankfully the Llano and Radeon line seems do do well.
 
So my question is, weren't these initially targeted to the server market first? Those power consumption numbers are horrendous, I suspect that will really hurt their server market...
 
I'm glad I didn't wait for BD and bought a standard AM3 board for cheap with a Phenom X4... these are going to get here expensive because they're new, and they're not worth it.

The game is just the same. Sad.
 
Well looks like I'll stick with my phenom x4 9850 @ 3.01 until the piledriver comes out. Oh well. I only use this system for hardcore gaming and encoding.
 
bcorcoran said:
I was referencing 8120 vs 8150, but your point stands. My gripe with complaining about the 8120's performance is that there IS an 8150 sitting right there which is rated for higher clocks. I don't see any logic in why the reviewer chose to focus on the 8120 and used that to come to his conclusion, other than that it was cheaper and *could be* (likely that it is) from the same die.

AMD doesn't just toss a coin to decide which chips get the 8120 or 8150 badge. If they did, they'd be in for some real bullshit involving many many lawyers. Without seeing any numbers I can all but guarantee you that the 8150 will overclock better than the 8120. Just like the Opteron 165 in my sig overclocked way better than the consumer-level equivalent at the time. It's not a direct relationship in that sense, but it's similar.

Fact of the matter is, and I know this isn't in the spirit of overclockers in general, but if you want the best performance and your goal in a review is to review the PROCESSOR and not the overclocking performance, then at LEAST focus on each version individually rather than focus mostly on the #2. There's no merit to saying "well, enthusiasts will probably buy this chip, so I'll focus on that". That's just doing a disservice to your readers.
Click to expand...

Enthusiasts buy chips not because they overclock well, but because they perform well and, as a bonus, overclock well. I'm fairly certain AMD wasn't marketing these chips to the 2 or 3 dudes in their garage with some spare LN and 1500W PSUs laying around. Overclocking performance is only explored and compared after the processor is benchmarked at stock. And comparing the 8150 and 8120 neglects that both processors are poor performers and poor price : performance bargains at both stock and overclocked speeds. The [H] review caters to the more tech savvy, and us folk generally overclock the chips we get. So reading a review that centers on the chips OC'd performance isn't an unfair representation of the chip's performance. As was stated below, anandtech did a great review of bulldozer as well and walked away with the same conclusions.

Kyle, would you guys care to go into why these things performed the way they did? I'm absolutely floored by the benchmarks I've been reading. Was it a silicon issue? Poor architecture? Pipes too long? It looks to me to be a case of the netbursts but with more cores sprinkled in.
 
Last edited:
Although I don't think it hurts conclusions, I gotta question the systems setup, more so the 4ghz "stock clocks" but also the overclocking results.

why no real stock clocks? why settle for 4ghz when none of the cpus tested run at that speed at stock? Its like mild overclocking vs real overclocking tests.

And then the OC results, I think they mean very little as it doesn't represent in any way the real world overclocking potential of either AMD nor intel cpus.

4.8 is hardly the top speed for a 2500k/2600k, and we don't know the real potential of AMD cpus yet.

I think the Gaming performance review did a much better job, that's the one that I'll take into account when deciding what will be my next upgrade.
 
You must log in or register to reply here.
Back
Top