Monday, December 19th 2011

AMD FX 8150 with Microsoft KB2592546 Put Through 'Before and After' Patch Tests

To the surprise of many, last week, Microsoft rolled out a patch (KB2592546) for Windows that it claimed would improve performance of systems running AMD processors based on the "Bulldozer" architecture. The patch works by making the OS aware of the way Bulldozer cores are structured, so it could effectively make use of the parallelism at its disposal. Sadly, a couple of days later, it pulled that patch. Meanwhile, SweClockers got enough time to do a "before and after" performance test of the AMD FX-8150 processor, using this patch.

The results of SweClockers' tests are tabled below. "tidigare" is before, "nytt" is after, and "skillnad" is change. The reviewer put the chip through a wide range of tests, including synthetic CPU-intensive tests (both single and multi-threaded), and real-world gaming performance tests. The results are less than impressive. Perhaps, that's why the patch was redacted.

Source: SweClockers
Add your own comment

96 Comments on AMD FX 8150 with Microsoft KB2592546 Put Through 'Before and After' Patch Tests

#1
trickson
OH, I have such a headache
by: DrPepper
Can't really draw a conclusion based on a patch that was leaked then pulled. I'll wait till they release the finished product before making judgements, however I doubt there is a magic software fix for this.
I think it would be far better for AMD to put out products that work with out all the hot fix fuss . Just kind of makes one mad at them for putting out crap . We waited for years for BD and got this crap ? Come on !
Posted on Reply
#2
eidairaman1
The ones who are thread bashing sound like broken records, we all know the FX series aren't the best; they certainly aren't crap though. Stop beating a dead horse and help people who actually need their machines fixed. Or are y'all not intelligent enough to act like men and not a bunch of boys?
Posted on Reply
#3
trickson
OH, I have such a headache
by: eidairaman1
The ones who are thread bashing sound like broken records, we all know the FX series aren't the best; they certainly aren't crap though. Stop beating a dead horse and help people who actually need their machines fixed. Or are y'all not intelligent enough to act like men and not a bunch of boys?
Wouldn't it have been better for AMD to put out a chip that worked in the first place ? :wtf:
But I guess you are right . I do think that AMD will get it all sorted out soon and every one with a BD will be happy . SOON ! Just have to wait till they get it together is all . Pulling a bad patch that lowered performance was a good thing IMHO .
Posted on Reply
#4
eidairaman1
In the past both companies have had MS release patches/drivers to make the OS function properly with the newer CPU/mobo architecture being released. I also realize the design of the CPU is more of a direct conversion from the Opteron than a tweaked Opteron. So we shall see what Piledriver and the next FX is about.
Posted on Reply
#5
trickson
OH, I have such a headache
by: eidairaman1
In the past both companies have had MS release patches/drivers to make the OS function properly with the newer CPU/mobo architecture being released. I also realize the design of the CPU is more of a direct conversion from the Opteron than a tweaked Opteron. So we shall see what Piledriver and the next FX is about.
Yeah it does happen . I just hope AMD gets this all worked out soon .
Posted on Reply
#6
eidairaman1
Rest assured they will, I mean supposedly the 2900XT was a debacle and they recovered from that in a big way
Posted on Reply
#7
theoneandonlymrk
by: John Doe
Grow up.
youve come on here to shit on AMD and thats all, and not to learn anything or advise anyone and he should grow up, funny guy:laugh:

it does seem to be the same people continuously on here, in every AMD thread ,essentially just repeatedly sayin "AMD's shit haha" , they should just copy and paste that then repost it as and when

and this is the only relevant comment

by: LAN_deRf_HA
Quote:
We've spoken with an industry source familiar with this situation, and it appears the release of this hotfix was either inadvertent, premature, or both. There is indeed a Bulldozer threading patch for Windows in the works, but it should come in two parts, not just one. The patch that was briefly released is only one portion of the total solution, and it may very well reduce performance if used on its own. We're hearing the full Windows update for Bulldozer performance optimization is scheduled for release in Q1 of 2012. For now, Bulldozer owners, the best thing to do is to sit tight and wait.

This is why no site of substance has run these tests; it's not indicative of squat.
Posted on Reply
#8
AthlonX2
HyperVtX™
+1 We all know mr Doe is a troll,every thread he post's in turns into something like this. Thats why he hides behind his proxy.
Posted on Reply
#9
Poisonsnak
I think the patch is supposed to have this kind of effect:

http://techreport.com/articles.x/21865/2

when you're running between 2 and 6 threads on an 8-module bulldozer CPU, it seems to be best to spread them out as much as you can (mask 55 in the graphs) as opposing to having 2 threads on 1 module and 1 module sitting idle.

Take it with a grain of salt though, they may be cherry-picked benchmarks
Posted on Reply
#11
qubit
Overclocked quantum bit
by: btarunr
The results are less than impressive. Perhaps, that's why the patch was redacted.
Indeed it was why. MS explained that the patch is supposed to be in two parts and that running the incomplete patch can actually decrease performance, so to me, those benchies don't mean much as they stand. Of course, it's gonna be very interesting to see how well the full patch works and how differently it works from what SweClockers posted.
Posted on Reply
#12
pantherx12
by: qubit
Indeed it was why. MS explained that the patch is supposed to be in two parts and that running the incomplete patch can actually decrease performance, so to me, those benchies don't mean much as they stand. Of course, it's gonna be very interesting to see how well the full patch works and how differently it works from what SweClockers posted.
I would hope it brings the BD cpu up to scratch with a theoretical Phenom II x 8 ( well except at floating point, but gpus/apus will do all the floating point math eventually)

I don't expect actual proper improvements until piledriver.
Posted on Reply
#13
AphexDreamer
Well I'm extremely happy with MY BD chip and super excited at the thought of a Windows Patch exclusively designed to boost BD chips even further.

I had a feeling that the first patch would get botched. Things of that nature always get pulled or have some sort of fault upon first release that requires patching.



I await patiently.
Posted on Reply
#14
qubit
Overclocked quantum bit
by: pantherx12
I would hope it brings the BD cpu up to scratch with a theoretical Phenom II x 8 ( well except at floating point, but gpus/apus will do all the floating point math eventually)

I don't expect actual proper improvements until piledriver.
Yeah, it might deliver 10% or so (speculation) but still better than nothing and as these chips overclock well, 10% of that boosted performance can be noticeable.

by: AphexDreamer
I await patiently.
I await impatiently! :toast: I really wanna see what this can do and it's quite a cliffhanger.
Posted on Reply
#16
AphexDreamer
by: qubit


I await impatiently! :toast: I really wanna see what this can do and it's quite a cliffhanger.
Yeah lol but we don't want any more rushed releases from MS that could further decrease performance either :p

They just had to tease us with an improper release. :banghead:
Posted on Reply
#17
pantherx12
by: AphexDreamer
Well I'm extremely happy with MY BD chip and super excited at the thought of a Windows Patch exclusively designed to boost BD chips even further.

I had a feeling that the first patch would get botched. Things of that nature always get pulled or have some sort of fault upon first release that requires patching.



I await patiently.
Yeah I'm pretty happy with it as well, at the speed I have it it's the equivalent of my 1055t @ 3.9 ghz at most tasks, but tasks that take advantage of the new instructions/architecture the BD chip is faster.

( decided to step down to 4.3ghz as it can be done on stock volts even with this rubbish bios I'm using)


If I had better cooling I could easily clock this chip to 5ghz and beyond so will be equivalent to a 4.5ghz 6 core phenom at most tasks :)

It's not bad when you bare in mind that it's supposed to have higher clocks.
Posted on Reply
#18
xenocide
by: pantherx12
Yeah I'm pretty happy with it as well, at the speed I have it it's the equivalent of my 1055t @ 3.9 ghz at most tasks, but tasks that take advantage of the new instructions/architecture the BD chip is faster.
That is my major issue with BD. It's only as good as Thuban, and is usually slower clock-for-clock. I was hoping at least for the per-thread performance of Phenom II, but that didn't happen.
Posted on Reply
#19
ensabrenoir
Kinda sad really

Amd aint a baf co. bd aint...needs work . These public ...punking of amd aint helping though. Love the gpus but spirite was wrpng. Image do count for something. Even if bd is fixed tommorrow...the jokes will continue. Amd..so advanced that by the time u figure out how to use it properly....its,obsolete. Now microsoft inadvertly punks them.... Though year
Posted on Reply
#20
pantherx12
by: xenocide
That is my major issue with BD. It's only as good as Thuban, and is usually slower clock-for-clock. I was hoping at least for the per-thread performance of Phenom II, but that didn't happen.
If this patch works as rumoured ( + 10% ) then it will be like Phenom x 8.

Cept it over clocks 1ghz more ;)

It really isn't all that bad, it's £30 more for an fx8120 over a 1100t here.

It's definitely worth the £30 difference for me anyway, and that's before any patches.

I mean some programs do really suck on it, but in general it's pretty nice.

+ I like to run lots of programs at once because hitting the close button/cancel button is to much effort.
Posted on Reply
#21
Super XP
by: xenocide
That is my major issue with BD. It's only as good as Thuban, and is usually slower clock-for-clock. I was hoping at least for the per-thread performance of Phenom II, but that didn't happen.
AMD has a lot of work ahead of themselves. Definitely something is wrong with Bulldozer and how the OS schedules due to the nature of AMD’s CPU modular concept.
IMO, I can see this 2-part patch giving today’s Bulldozer a 10% on average performance boost.
Piledriver getting another 20% to 30% performance boost over Bulldozer.
Excavator getting another 10% to 15% performance boost over Piledriver.
Posted on Reply
#22
xenocide
by: Super XP
AMD has a lot of work ahead of themselves. Definitely something is wrong with Bulldozer and how the OS schedules due to the nature of AMD’s CPU modular concept.
IMO, I can see this 2-part patch giving today’s Bulldozer a 10% on average performance boost.
Piledriver getting another 20% to 30% performance boost over Bulldozer.
Excavator getting another 10% to 15% performance boost over Piledriver.
I think you're deluding yourself by expecting 10% on average. I saw some Linux Benchmarks and W8 Benchmarks with a "fixed" scheduler, and it was MAYBE a 5% gain in some situations, and in other upowards of a 5% loss in performance. This patch may offer a slight performance gain, but I don't expect it to really change much.
Posted on Reply
#23
trickson
OH, I have such a headache
by: ensabrenoir
Amd aint a baf co. bd aint...needs work . These public ...punking of amd aint helping though. Love the gpus but spirite was wrpng. Image do count for something. Even if bd is fixed tommorrow...the jokes will continue. Amd..so advanced that by the time u figure out how to use it properly....its,obsolete. Now microsoft inadvertly punks them.... Though year
Can you say this in English please ? AMD screwed up plane and simple . Sucks to ! We waited 4 years for this ? A launch filled with fluff and when it got into the hands of the consumers it was a total mess . We find out all kinds of things like they did not even know the transistor count ? and that it is out performed by it's older line the Phenom ! REALLY ??? Really you need another hot fix for another CPU line AGAIN AMD ??? WTH ? They put more work and more R&D into there GPU line than there CPU line !
Posted on Reply
#24
pantherx12
by: xenocide
I think you're deluding yourself by expecting 10% on average. I saw some Linux Benchmarks and W8 Benchmarks with a "fixed" scheduler, and it was MAYBE a 5% gain in some situations, and in other upowards of a 5% loss in performance. This patch may offer a slight performance gain, but I don't expect it to really change much.
Wait if you have a fixed scheduler why don't you share it man? :laugh:
Posted on Reply
#25
trickson
OH, I have such a headache
by: pantherx12
Wait if you have a fixed scheduler why don't you share it man? :laugh:
:roll:
Posted on Reply
Add your own comment