AMD FX-8150 Bulldozer On Ubuntu Linux Benchmarks

[koorosh]

and I really don't understand why people would want something this new on the consumer side of a Linux OS...

It doesn't matter it's consumer or enterprise it should have linux support soon enough. Things has changed and you don't need to wait years for linux to support latest hardware. It may be become custom to use linux on your old hardware but it's not necessary, linux runs great on new hardware too.

 

[nt300]

Sounds like vague BS...Still doesn't help the poor people that paid for these turds thinking they would be a good chip but got fooled by the name....and on FM2....OK so in other words AM3+ is dead already...F*ck off AMD...at least the ATI division is still worth something.....the B4 stepping wont ever happen....They will likely abandon this and move on...

and I really don't understand why people would want something this new on the consumer side of a Linux OS...

R U kidding? The Bulldozer is going to sell out everytime it released. AMD won't be able to meet demand which sucks, it will just take me longer to get one. We do know a B3 revision is coming, we also know AMD know Bulldozer needs to be faster. If they can tweak the design enough it may perform much better than what we have today.

 

[Deleted member 67555]

It doesn't matter it's consumer or enterprise it should have linux support soon enough. Things has changed and you don't need to wait years for linux to support latest hardware. It may be become custom to use linux on your old hardware but it's not necessary, linux runs great on new hardware too.

I understand that but Linux is better imo at using way less resources than a Windows OS..So I don't see the need for Top of the line or newer hardware...oh wait this chip sucks ass...nevermind

R U kidding? The Bulldozer is going to sell out everytime it released. AMD won't be able to meet demand which sucks, it will just take me longer to get one. We do know a B3 revision is coming, we also know AMD know Bulldozer needs to be faster. If they can tweak the design enough it may perform much better than what we have today.

You're right because most people don't pay attention to sites like this or are loyal Fanboi's that can't except it's a turd...
As for the revision..Read between the lines..They are gonna try Higher Clocks and a more efficient IMC...that's not gonna help that much because as it's pointed out again and again it's design flaw in the pipeline...same as Netburst....I'm guessing they will take the same route Intel did with Piledriver

 

[Super XP]

I understand that but Linux is better imo at using way less resources than a Windows OS..So I don't see the need for Top of the line or newer hardware...oh wait this chip sucks ass...nevermind



You're right because most people don't pay attention to sites like this or are loyal Fanboi's that can't except it's a turd...
As for the revision..Read between the lines..They are gonna try Higher Clocks and a more efficient IMC...that's not gonna help that much because as it's pointed out again and again it's design flaw in the pipeline...same as Netburst....I'm guessing they will take the same route Intel did with Piledriver

O.K. I will post this again. Not sure if true, but it makes sense with the revisions based on process steppings. And once again can these minor improvements be enough to "at the very least" blow away any Phemon II CPU out of the water? In my estimation we should be able to see a 10% to 15% bump in performance with this B3 stepping clock for clock vs. the B2 stepping Bulldozer. That IMO is AMD's goal to at least ensure it's faster than the older based gen i.e.: Phenom II's. Acually it now makes a lot of sense and the reason why you cannot find FX 8100's anywhere, because AMD scraped the B2 stepping in favour for the B3 coming either late Dec 2011 or sometime in Jan 2012.

So what about this so called B3 stepping. This was posted on semiaccurate by somebody that calls themself ATInsider. If what he says is true, how much performance can a B3 give Bulldozer before Piledriver gets released with another added 10% increase?

AMD FX – Series B3 revision is more than just a basic stepping:
I have direct knowledge of a possible B3 revision for the AMD FX line of CPUs. I cannot disclose performance projections at this time, but be assured AMDs processor division is working vigorously on a (B3) stepping revision with minor architectural tweaks. The base architecture will not be changed at this time.
Within the B3 stepping revision, expect minor tweaks to the following:
1) L1, L2 and L3 latencies
2) Cache Thrashing Issues
3) Modified Algorithms for Branch Prediction
4) Healthy Bump in Processor Frequency
5) Slight Frequency increase via NB Controller
6) “Total Intelligent Control” For example programs and applications should look at the module design approach and the ability for the processor to intelligently turn off and/or turn on specific cores that it believes is hindering performance for maximum performance. (May be for Socket FM2, not sure at this time).
7) Power will be improved but not my much. We will have to wait for Socket FM2 or a future B4 revision for the AM3+ platform for better power efficiency especially when Over-clocked.
ATInsider

http://semiaccurate.com/2011/10/17/bulldozer-doesnt-have-just-a-single-problem/comment-page-1/#comment-11399

 

[erocker]

What does this have to do with the current FX chip and Linux? Nothing. Please stay on topic.

 

[Super XP]

What does this have to do with the current FX chip and Linux? Nothing. Please stay on topic.

O.K. but I was answering a question. Anyhow my point was the B3 revision may bring performance increases to have them perform better in benchmarks.

 

[BenchZowner]

allow me to add a few linux benchmarks at 3.6G ( turbo off ) against stock i5 & i7.

.:: BenchZone ::. FX-8150 Preview 1: Linux Performance

 

[unsmart]

Just me or do the colors on those grafts burn your eyes too
anyway I think Intel and AMD have a deal. AMD makes low end cpus and Intel doesn't enter the GFX market.

 

[BenchZowner]

Just me or do the colors on those grafts burn your eyes too
anyway I think Intel and AMD have a deal. AMD makes low end cpus and Intel doesn't enter the GFX market.

Burn my eyes no, but I'm a bit hardcore, even 12 hrs of continuous battlefield 3 gameplay can't burn my eyes.

I can agree that the colors are a bit funky, not my fault, Swiff Charts Pro decides the colors everytime ( you can change them, but I didn't really thought about the colors it picked this time )

 

[unsmart]

It,s prob my Asus monitor it has the no touch buttons and I tend to have piles on my desk that set them off and f the settings up.
On topic though I think we will see some improvement in linux/win7 with core optimization but in the end it is what it is. A Ford is not a Porsche no matter how much you tweak it.

 

[Super XP]

Bulldozer does O.K. in those benchmarks. But it doesn't say anything about disabling the onboard graphics on the i7 and i5 CPU's.

 

[BenchZowner]

The integrated GPU was disabled.
Enabled or disabled it wouldn't make any difference at all anyway.

 

[nt300]

I did somemore digging about todays Benchmarks. check this out

Vapor
Posted on: 10/19/2011 02:17 AM
--------------------------------------------------------------------------------
The real reason that Bulldozer did not stack up in the benchmarks is the compiler used for for each of the benchmarks. All of these closed-source benchmarks are compiled on the standard Intel compiler with the Intel libraries. It is not optimized to support any instructions beyond SSE3 for any processor other than Intel chips. SSE4.1, SSE4.2, AVX, and FMA4 significantly increase the floating point performance of AMD processors, but are not used by code compiled on an Intel compiler. Intel was sued for this and settled rather than fix it.

If you look at the integer performance of the benchmarks, AMD almost always out-performs the intel chips and shows a 15-30% increase in performance over the Phenom II x6 processors. If the compiler used was completely optimized for both Intel and AMD, floating point performance would also show similar gains.

Lastly, under full load where all of the threads are being used, the Intel chip is not physically capable of beating the AMD chip. 4 cores that complete one instruction each per cycle cannot physically beat 8 cores completing 1 instruction each per cycle, when threads are continually running.

Lets compare My favorite to Intels best as it is easier to see the elephant in the room.

AMD fx-4170
4.2ghz 1866mhz RAM 12mb Cache total. 1mb L2 per core

Intel 2600K
3.6ghz 1333mhz RAM 9mb Cache total. 1mb L2 for all cores

Intel Fanboys will say it supports 1600RAM but it will run at 1333 unless you overclock it and AMD can overclock RAM to 2500 easy so lets compare fairly. Also intels Hyperthread causes stuttering. It requires ineffiency in order to gain a boost, by having gaps in code it threads other processes in those gaps. Being ineffective makes hyperhthread run well, but since it only has 1mb L2 cache having to store any extra data to hyper-thread creates stuttering
I would also like everyone to understand many of these benchmarks online are biased. Some were re-re-posted before final silicone was ever in existence. Some have obvious falsehoods like showing hyperthread on a 2500's CPU-Z readout. They usually fix them when you comment on this too.

 

[erocker]

I did somemore digging about todays Benchmarks. check this out

LOL! Yeah, I have both a 8150 and a 2500K. I don't find myself using the 8150 too much. Why? The 2500K is faster in everyday use and gaming all while using way less power. For servers, BD is a good buy due to it's price... as long as you don't care about the electricity bill.

You know, I see your posts all over the internet.. on other forums. I'm really starting to believe you actually work for AMD and you're just spreading your BS to save face. Am I wrong? If so, what is your motiviation?

it will just take me longer to get one.

If you really want one, PM me. I do ship to AMD headquarters.

 

[cadaveca]

I did somemore digging about todays Benchmarks. check this out

Still just BS. Show me proof, show me that every user can buy a chip, plug it in, and get different results, and I might start listening. Conjecture on compiler optimizations doesn't help the end user, and until it does, that whole thing is just plain old BS.


Thanks anyway.


And yes, I have a very large BIAS...TO THE TRUTH.

I've noticed that Turbo isn't working properly on many boards, with cores sitting @ 1400 MHz(idle), while two clock up, even though all should be clocking up to 3300MHz with my chip. This same issue could be happening in every review...do I care?

Nope!

What matters is not what is possible, but what the end user will experience. Change that experience, and then speak up, as until then, it just seems like pulling at straws.

 

[seronx]

For servers, BD is a good buy due to it's price... as long as you don't care about the electricity bill.

I would wait for you to see Interlagos/Valencia reviews before you infer Zambezi is clocked the same as Interlagos and Valencia

 

[erocker]

I would wait for you to see Interlagos/Valencia reviews before you infer Zambezi is clocked the same as Interlagos and Valencia

Um. Okay? You missed my point as I'm not really talking about Interlagos/Valencia not to mention it's not the topic of this thread.

 

[seronx]

Um. Okay? You missed my point as I'm not really talking about Interlagos/Valencia.

You are inferring that Bulldozer on the desktops is representative of Bulldozer on the servers

For servers, BD is a good buy due to it's price... as long as you don't care about the electricity bill.

 

[erocker]

You are inferring that Bulldozer on the desktops is representative of Bulldozer on the servers

No, I'm not. I'm inferring that a Bulldozer desktop chip can be used as a servicable server chip. Either way, it doesn't matter. I own the chip and I deem it a pretty good failure all around. Cheers to waiting for a new stepping or whatever. Now please, lets see some charts and graphs that I don't care about.

 

[Damn_Smooth]

No, I'm not. I'm inferring that a Bulldozer desktop chip can be used as a servicable server chip. Either way, it doesn't matter. I own the chip and I deem it a pretty good failure all around. Cheers to waiting for a new stepping or whatever. Now please, lets see some charts and graphs that I don't care about.

Congrats on your 27,000th post.

 

[seronx]

I'm inferring that a Bulldozer desktop chip can be used as a servicable server chip.

A workstation chip you mean then like the LGA 1366/LGA 2011 platform

 

[erocker]

A workstation chip you mean then like the LGA 1366/LGA 2011 platform

Yes, though not as efficient.. I think. I don't have much experience with LGA 1366/2011. Brass tacks, BD needs better single threaded performance and lower leakage. Fix that and it's a big win for everyone.

 

[seronx]

Yes, though not as efficient.. I think. I don't have much experience with LGA 1366/2011. Brass tacks, BD needs better single threaded performance and lower leakage. Fix that and it's a big win for everyone.

Improved single threaded performance is coming 2013 with Steamroller
Leakage improvements is coming 2012 with Piledriver/2014 with Excavator

^-- there is more stuff involved but that is the general idea what you get from the SOG

 

[Damn_Smooth]

Improved single threaded performance is coming 2013 with Steamroller
Leakage improvements is coming 2012 with Piledriver/2014 with Excavator

^-- there is more stuff involved but that is the general idea what you get from the SOG

They need improved single threaded performance yesterday, not 2 years from now.

 

[erocker]

Improved single threaded performance is coming 2013 with Steamroller
Leakage improvements is coming 2012 with Piledriver/2014 with Excavator

^-- there is more stuff involved but that is the general idea what you get from the SOG

This I know.

They need improved single threaded performance yesterday, not 2 years from now.

This I agree with.