Friday, January 13th 2012

AMD FX-8150 Tested with Latest Windows Hotfixes, Still No Improvement

German tech website TweakPC did a before-after comparison of applying Microsoft's recently-released KB2645594 + KB2646060 Windows updates, which intend to improve performance of systems running AMD FX processors, by improving the way in which the OS deals with Bulldozer cores, using a top-of-the-line FX-8150 processor. The reviewer put FX-8150 through synthetic tests such as AIDA64 (CPU benchmarks, FPU benchmarks), Cinebench 11.5, MaxxPi (multi-threaded PI calculations), WPrime, Twofish AES, 3DMark (Vantage and 11), ComputeMark; and some real-world tests such as WinRAR, Resident Evil 5, and Battleforge. Barring Resident Evil 5, where the patched FX-8150 produced 4% higher performance and WinRAR, where it produced 3% higher performance, there were no significant performance gains noticed. The review can be accessed at the source.
Source: TweakPC.de
Add your own comment

165 Comments on AMD FX-8150 Tested with Latest Windows Hotfixes, Still No Improvement

#1
Damn_Smooth
So you really can't make a turd a rose. Who would've figured?
Posted on Reply
#2
mtosev
more fail for amd. scrap bulldozer and start working on a new cpu.
Posted on Reply
#3
LDNL
Why wont they just put this horse down and focus more on the next best thing so they wont f it up aswell.
Posted on Reply
#4
Suhidu
The patch does what it was supposed to do. It was never meant to speed up all applications a significant percent in benchmark scenarios, but it does remove a core-scheduling handicap of BD.
Posted on Reply
#5
Sihastru
Well, that's because that's not what the hotfixes... hmmm... fix.

You will not see an improvement when all 8 threads are used, you will not see a great improvement when 4 threads are used (but there should be some, for cases when the modules were not loaded correctly, although most of the times they were), and you will not see an improvement when just 1 thread is used.

People read the title of the previous article, they saw that "10%", and automatically assumed that it would automagically make the CPU faster. That's not the case. There are just a few situations where the hotfixes will work.
Posted on Reply
#6
NC37
lol, my 945 beats an FX in the 4T Cinebench...ha!
Posted on Reply
#7
RejZoR
Resident Evil 5 ? That's like testing a Ferrari on a construction site... I'm not saying this CPU has certain problems but i don't think they could pick any worse game to set an example...
Posted on Reply
#8
entropy13
RejZoRResident Evil 5 ? That's like testing a Ferrari on a construction site...
There's a 4% improvement in the game. So if your analogy holds through, testing a Ferrari on a construction site would also entail some improvement?
Posted on Reply
#9
seronx
NC37lol, my 945 beats an FX in the 4T Cinebench...ha!
Well no duh SSE2 performance dropped with Bulldozer



But don't expect it to do better in real world applications that use SSE4.1 which your CPU can't really do for feasible performance gain

It will be fixed with Vishera or by 40h-4Fh Steamroller the SSE2 performance
Posted on Reply
#10
RejZoR
entropy13There's a 4% improvement in the game. So if your analogy holds through, testing a Ferrari on a construction site would also entail some improvement?
It did gain tiny bit on those few meters of brand new asphalt. In most cases the perfromance even degraded. Like it would when using Ferrari on a construction site... not exactly impressed.
Posted on Reply
#11
meirb111
a song for amd :

"I'm just a soul whose intentions are good

Oh Lord please don't let me be misunderstood"
Posted on Reply
#12
qubit
Overclocked quantum bit
Oh well. Makes you wonder why Microsoft bothered, doesn't it?
Posted on Reply
#13
air_ii
mtosevmore fail for amd. scrap bulldozer and start working on a new cpu.
While BD may not be what everyone wanted it to be, maybe further iterations of the architecture will bring improvements. Maybe there is a potential in the architecture. Look what became of R600 -> RV770.
Posted on Reply
#14
mtosev
air_iiWhile BD may not be what everyone wanted it to be, maybe further iterations of the architecture will bring improvements. Maybe there is a potential in the architecture. Look what became of R600 -> RV770.
by the time they get it right intel will still be ahead of amd. at least AMD err... ATI knows how to make graphic cards
Posted on Reply
#15
qubit
Overclocked quantum bit
air_iiWhile BD may not be what everyone wanted it to be, maybe further iterations of the architecture will bring improvements. Maybe there is a potential in the architecture. Look what became of R600 -> RV770.
That wasn't a framerate performance improvement (not much). It simply made the chip smaller and use less power.
Posted on Reply
#16
laszlo
mtosevmore fail for amd. scrap bulldozer and start working on a new cpu.
why fail? there are improvements and they are free;maybe we'll see a better patch which more improvements
Posted on Reply
#17
Dent1
What people have to remember is, they only tested a few applications and games. This review is a very small sample and does not reflect the patch as a whole. If people look back at the other Bulldozer "patch" thread you would see that members in this very forum are reporting upto 7% performance increase. So it's really application dependant, and no review can cater for every application.


Below are quotes from another thread, trusted people from this community whom are saying the patch DID improve things.
omagicWell i only had time to check Crysis CPU Benchmark
Before 47 fps
After 56 fps

So quite nice boost. Ill try some more games after the work

FX-8120 8GB RAM HD6870
1680x1050 all maxed
TweetyWindows CPU score went from 7.4 to 7.5... for free :)
screamer980Here's my result. Before Patch 7.15. After 7.23
pantherx12Got the same 7.6% increase in cinebench single core as the leaked patch, this time with no hit to multicore performance : ]
n0tiertMy Futuremark 3DMark11 test on FX-8150 / 6990

as you can see , it slightly raised the PhysX score a bit and the Total score is arround 50 points higher
Link: www.techpowerup.com/forums/showthread.php?p=2514812#post2514812
Posted on Reply
#18
moonlord
"there were no significant performance gains noticed" Well.. i know that , those hotfixes are a waste of time. Bad investment :banghead:
Posted on Reply
#19
Dent1
moonlord"there were no significant performance gains noticed" Well.. i know that , those hotfixes are a waste of time. Bad investment :banghead:
Read my post above.
Posted on Reply
#20
moonlord
Dent1Read my post above.
My WEI is still 7.7, maybe this patch improves other configurations, but for my software and games makes no difference.
Posted on Reply
#21
Zubasa
moonlordMy WEI is still 7.7, maybe this patch improves other configurations, but for my software and games makes no difference.
LOL like WEI matters :shadedshu
Posted on Reply
#22
mtosev
laszlowhy fail? there are improvements and they are free;maybe we'll see a better patch which more improvements
fail because an i7 2600/K costs less and is still faster
Posted on Reply
#23
pantherx12
I got a single core and dual core performance boost.

Nothing at 10% thus far but still a wee boost for free aint bad : ]

I wouldn't say it was no improvement due to that website only testing multi-core stuff.

The patch was designed to fix threads jumping from one core to another which should fix up turbocore and the second patch designed to stop processors going into sleep mode whilst a thread was active on them. ( Caused stuttering like effect in games)

That it comes with a lil boost as well is nothing but a good thing.
Posted on Reply
#24
Horrux
mtosevfail because an i7 2600/K costs less and is still faster
Not in SSE 4.1.

I have no idea what that is, except any code heavy with these kinds of functions in it will fly on the BD compared to a 2600k. Maybe in the future there will be more such code?
Posted on Reply
#25
seronx
HorruxNot in SSE 4.1.

I have no idea what that is, except any code heavy with these kinds of functions in it will fly on the BD compared to a 2600k. Maybe in the future there will be more such code?
SSE4.1 is for transcoding

Bulldozer prefers the traditional ideology Integer does Transcoding and it would make sense because there is 2 Integer Pipes per Core(Ignoring the AGLUs which can make it 4) and 3 Integer Pipes per Floating Point Unit

Sandy Bridge aims for floating point to do it....but ultimately it is easier to implement Bulldozer as the best transcoding beast in the world because it is easy to make it transcode and it is very fast at transcoding

Encoding and Decoding <3s Bulldozer as it is pretty much a big DSP

Also FMA can replace a majority of MAC instructions making Bulldozer faster in those areas

forum.doom9.org/showpost.php?p=1533501&postcount=1732

I'll just leave this link here...
Posted on Reply
Add your own comment
Apr 16th, 2024 17:18 EDT change timezone

New Forum Posts

Popular Reviews

Controversial News Posts