AMD FX-8150 Review: From Bulldozer To Zambezi To FX

[youssef 2010]

[citation][nom]techpops[/nom] Intel has the best answer for you right now on price / performance / upgrade path.[/citation]

I agree with everything you said except the upgrade path part. With AMD you'll have a better upgrade path than any Intel architecture you buy. When buying Intel, know that the platform will be a dead end. You'll need a new motherboard to support new Intel CPUs, while that's not the case with AMD

 

[youssef 2010]

[citation][nom]amk-aka-phantom[/nom]I love their graphics cards... somehow AMD is great on that market, no matter how evil Intel and nVidia conspire against them Why is that?[/citation]

Because of ATI's talented engineers

 

[amk-aka-Phantom]

I agree with everything you said except the upgrade path part. With AMD you'll have a better upgrade path than any Intel architecture you buy. When buying Intel, know that the platform will be a dead end. You'll need a new motherboard to support new Intel CPUs, while that's not the case with AMD

Core 2 Quad is still rocking. X58 is still rocking. Sandy Bridge will be rocking for quite some time, and if these rumors of Ivy Bridge working on P67/Z68 are true... even if not, SB is great. So, dead-end, but a great one.

 

[ChromeTusk]

Finally read the entire article and I am a little disappointed in Bulldozer. I was hoping for a bit more competition with Intel to help bring prices down for everyone. Looks like my next build will be Intel based

 

[hsir135]

I can hardly wait to put Bulldozer on my Vista machine!!
Intel can now change the Ivy Bridge name to Steamroller.

 

[shadyinc]

Guys, For heaven's sake, PLEASE..!!!
This launch, has been delayed by 2 tears..!
That is hell lot of a time..and that is the time in which Intel has made all the "advancments"

So lets look at it this way:technically, this is a 2 year old CPU, available for sale now..(SAD!!!)
And for a 2 year old part, the performance is commendable..!
It still almost betters or matches 2600k in threaded applications.

No wonder AMD is gonna try and compensate for the delay by launching a new iteration of BullDozer Arch. every year till 2014..!!!

 

[shinkueagle]

I USED to be an AMD Fanboy..... Protecting the name as much as I can....But now.... I have FORGETTEN what "AMD" stands for... Hmmm....Always Meaningless and Dysfunctional?

 

[shinkueagle]

[citation][nom]shadyinc[/nom]Guys, For heaven's sake, PLEASE..!!!This launch, has been delayed by 2 tears..!That is hell lot of a time..and that is the time in which Intel has made all the "advancments"So lets look at it this way:technically, this is a 2 year old CPU, available for sale now..(SAD!!!)And for a 2 year old part, the performance is commendable..!It still almost betters or matches 2600k in threaded applications.No wonder AMD is gonna try and compensate for the delay by launching a new iteration of BullDozer Arch. every year till 2014..!!![/citation]

Next thing you know the CONSTRUCTICONS will be the next best thing in processor technology....

 

[hsir135]

I USED to be an AMD Fanboy..... Protecting the name as much as I can....But now.... I have FORGETTEN what "AMD" stands for... Hmmm....Always Meaningless and Dysfunctional?

Another Marketing Disaster??
Another Missed Deadline??

 

[mrmaia]

Sucks on everything, including overclocking... and marketing. What a shame.

 

[larkspur]

[citation][nom]shadyinc[/nom]No wonder AMD is gonna try and compensate for the delay by launching a new iteration of BullDozer Arch. every year till 2014..!!![/citation]

Yes and a 10-15% increase per year through mainly clock increases (which we can do by OCing) is not going to keep them competitive in the mainstream market segment (see the Intel P4). A 3 year old Intel chip (clocked @2.66 with modest turbo) is beating them in gaming benchmarks. At this point all of BD's "chips" are riding on the future POSSIBILITY of software optimizations. Their current value is only present in highly threaded apps and even then they are alarmingly inconsistent. BD does appear to have a great future in servers, but I'm talking mainstream desktops and unless there is a bug in there somewhere, Zambezi falls way short.

 

[dragonsqrrl]

[citation][nom]hawkeye70[/nom]you obviously can't read. I said it would be better then the X4 series cpu and no where did I say it would achieve similar performance gains... just that it would be better then the existing x4 platform which most people are going to buy as they don't need 8 cores anyway... understand?[/citation]
x4 is a platform now? That's news to me...

Honestly dude? Okay, so that's what you meant when you said, "Funny I don't remember seeing extremely high multi-threaded scores on the phenom II x4 cores. Surely you don't expect the new 4100 series fx processor to do worse in the heavy multi-threaded apps when the 8150 does so much better?"

...and, "I would bet you it will be slightly better than a 980 except in lightly threaded apps and will blow it away in most of the multi-threaded apps at stock"

Sorry, I must have just misinterpreted what you were saying, thanks for the clarification... lol, I guess sometimes we just can't admit we were wrong. Have you had look at the FX-4100 benchmarks I linked in my previous comment yet? Just in case you missed it:

http://www.guru3d.com/article/amd-fx-8150--8120-6100-and-4100-performance-review/4

 

[Why_Me]

The nice thing about the 2500K is that with a sub $30 USD cpu cooler, a person can hit an easy 4.4Ghz while using little power and producing very little heat. Double the $30 USD and your looking at or near 5.0Ghz.

No need for an expensive cooling set up.

 

[amk-aka-Phantom]

The nice thing about the 2500K is that with a sub $30 USD cpu cooler, a person can hit an easy 4.4Ghz while using little power and producing very little heat. Double the $30 USD and your looking at or near 5.0Ghz.

No need for an expensive cooling set up.

I agree, but how is this relevant? Are you sure that this response is not for SB-E liquid cooler thread? Or did I miss something?

 

[techpops]

[citation][nom]youssef 2010[/nom]I agree with everything you said except the upgrade path part. With AMD you'll have a better upgrade path than any Intel architecture you buy. When buying Intel, know that the platform will be a dead end. You'll need a new motherboard to support new Intel CPUs, while that's not the case with AMD[/citation]

The thing is Bulldozer didn't need a new chipset, they could have allowed it to be dropped straight into any Am2+ or AM3 motherboard. I can work that out as they're doing exactly that with the C32 and G34 server boards. But no, you need a super expensive 990FX motherboard to power a Bulldozer CPU that gives you no extra speed over the old 1100T CPU's. You see the problem here? It's an unnecessary leap to a new chipset that seems to be more about trying to get a tiny bit more performance out of benchmarks so Bulldozer doesn't look so bad. So right now you do have no upgrade path buying into anything pre Bulldozer. Buy Bulldozer now and pay through the nose for the motherboard and sure you have an upgrade path ahead but to what? 15% speed bump per year AMD is predicting. So that top end FX-8150 will be beaten by I presume the FX-8170 and will feel exactly the same to us in 12 months time where Intel will have surely dropped a little in price by then if not having given us faster CPU's for the money and again Bulldozer will look like something ancient. Put it this way, in two years time according to AMD's predictions, a 1100T 6 core chip could be overclocked to be as fast as anything Bulldozer has. That's not an upgrade path I want to be part of. Intel do switch up motherboards far too often but at least what's out now has some legs and any new board you buy into now has a future over the next few years at least.

 

[gxpbecker]

I really want to see how the 4170 does at gaming. "quad" core and 4.2gig (OCing ability may not be that great though). I am dissapointed, but we many have a light at the end of the tunnel (gaming at least).

 

[Marcus52]

The good news for me is, I won't be tempted to build a new rig before I planned.

Joining the crowd, AMD, as someone who longs for you to take the spotlight again. Get a clue - we don't care what your advertisements say, we care about actual performance.

Somebody. Anybody. Give Intel a serious run for their money again. (I like Intel, but believe real competition is good for us all, including Intel. I favored AMD at one time, now - not so much.)

On the flip side, I don't really understand why the modern CPU is mostly left idle by so many apps. Can you say multi-core, software engineers? In that scene, Bulldozer here looks better, and I'm not so sure I fault AMD for software developers not getting their thinking out of yesterday.

 

[sonofliberty08]

if the true quad core FX-4 will beat the 4 thread 2 core only i3, and are cheaper than it or equal, then they will win on the mainstream market...... or maybe AMD should count 1 BD module as 1 core with 2 thread, and fight hand to hand with the 2 core 4 thread i3

 

[pepsibottle1]

It's a shame it's a real disapointment because the news was looking really promising. As an AMD fan, I was eager to see if it could go head to head against Intel's Sandy Bridge CPU's but the old Phenom II's are just as fast. Real shame because I was hoping that this could turn the tide for the boys in green. Now Intel can once again get away with their ridiculous pricing and can ease up development knowing that they are once again far and beyone better than AMD at the given moment.

Still, I remain true to my brand and will continue to be a devout AMD customer. I've been building systems since the days of the K-6 and have not once had a AMD chip DOA or fail on me, while I cannot begin to count the problems I've had with Intel. And for the price, you can't beat AMD either. Hopefully they can hit the lab and make some improvements to up the ante a little bit

 

[amk-aka-Phantom]

And for the price, you can't beat AMD either.

$220 i5-2500K > $240 FX-8150; 'nuff said.

 

[peroludiarom]

[citation][nom]IgnoranceIsBliss[/nom]@amk-aka-phantomdude, you tripping a little there. I'll admit that BD gets beat hands down in most the benches, but then in some strange benches where it gives the i2500 and even the i2600 a good spanking, a poor performing CPU should not be able to do that, it should fail all round and for sure it should not be able to spank the i2600, the damn thing is all over the place, in some places it trails the i2500 in others it somewhere between i2500 and i2600 and other times it tops them other times it even loses out to an older phenomim pretty sure Mr Angelini did crack a laughter, cause thats one hell of a schizo chipand the intel compiler thing is no conspiracy, it's fact, the FTC slapped them over the hand and told them to fix it, which to this day they have failed to do so, what if i told you that if you faked the chip's ID it would give your chip a boost in performance irrespective of architecture, doesn't matter how hard AMD works hard to get their chips to work with the software if they are getting artificially crippled then they wont ever reach full potential (VIA got a 47% boost to their chip's performance just by pretending to be an intel chip), again you fail to understand what the compiler issue is, it's not a compatibility issue, it's intel purposely crippling performance on a non intel chip even though the chip maybe capable of morelet me put it into layman terms, it's like your Car refusing to go faster then 30mph cause you wont put Shell gas in the tank therefore the other brands of gas must be inferior right[/citation]

I can add one more thing. Now because Intel didn't completely remove cripling code from compiler, US Federal Court has started file agaist them! And the man who found this "Issue", in his blog, posted that in June 2011 the code is STILL THERE. So the point is that, I WANT to see BENCH with OpenSource compiller. All others are sh*t to me. Now the compiller chooses slowest path, and so the benches show little performance.

 

[Why_Me]

I can add one more thing. Now because Intel didn't completely remove cripling code from compiler, US Federal Court has started file agaist them! And the man who found this "Issue", in his blog, posted that in June 2011 the code is STILL THERE. So the point is that, I WANT to see BENCH with OpenSource compiller. All others are sh*t to me. Now the compiller chooses slowest path, and so the benches show little performance.

This still doesn't explain why AMD's are heat producing power hogs.

 

[masterasia]

Was really hoping this would make me switch back to AMD. Guess I'll just have to stick with my i7 920 for another 3 more years until something worthwhile comes out. I got my i7 920 3 years ago and Bulldozer can't even top it....come on AMD step it up.

 

[Guest]

Zambezi already needs a 50euro/dollar price slice to be competitive.

 

[cangelini]

[citation][nom]amk-aka-phantom[/nom]You've got to be kidding! Wait ANOTHER year? What, 3 years was not enough?So, now AMD fanboys are finding all kinds of excuses for Bulldozer underperforming... Intel compiler conspiracy, huh? Well, BD was supposed to be a gaming beast, and i5-2500K beat it in all games, ESPECIALLY CPU-heavy (WoW). Say what you will, but Bulldozer failed miserably for its intended audience, I hope it will be worth it for a worstation; I'll wait for some more benchmarks - maybe it'll be good for something else than gaming... can't imagine what, though, since it lost to 2500/2600K in nearly all tests.Before you whine about how all these programs are Intel-biased, let me remind you that Mr. Angelini here tested Bulldozer in many, many popular applications and benchmarks... he was probably shocked and was desperately trying to find at least SOMETHING where BD would excel, but failed. I don't blame him, hell no, great review - I'd be shocked, too! I imagine how he was going through the benchmarks, each time the same thing... Core i5/i7 on top... and Phenom X6 beating BD must've caused him to either crack with laughter or triple-check all components. Mr. Angelini, am I right? TL;DR: you might say that the benchmarks and games used here are Intel-biased, but that was the most popular and widespread software. If AMD can't make their CPU work with them...Next AMD fan excuse prediction: Tom's got a faulty chip for testing, aka Bulldozer w/o a shovel.Oh, and before you flame me and call me an AMD hater, I love their graphics cards... somehow AMD is great on that market, no matter how evil Intel and nVidia conspire against them Why is that?[/citation]

You're correct in that I double- and triple-checked the results. I even sent my entire list of results to AMD a week prior to the review going live to let them look at the performance I was seeing. None of what I turned up was unexpected.

I still maintain that something is wrong with F1 2011 on the AMD platform. But I've been telling AMD for more than a year now (since F1 2010) that there's something going on there. At this point, shoot, maybe it's just *that* processor-bound?