Microsoft Hotfix For AMD "Bulldozer" Based Processors

What do you mean? the links do not work. They don't have the full URL. Using firefox if that matters.

Nah, they're sites that are blocked by [H], that's why they don't work. Any reference to them is automatically removed.
 
And those are blocked because... ? Ah well, doesn't really matter, they'll be out soon enough. I don't even have a BD yet (or decided if I want one) so I really have no idea why I'm so damn interested. Damn genes built into me that make we wanna tweak stuff!!
 
I expect the patches to not be needed then. I mean when AMD fixes the slow high latency cache the impact from threads switching physical cores will be lessened.

Perhaps that will be the case, but seeing it'll still use modules shouldn't this kind of patch keep future revisions of the chip optimized regardless? I mean if it's better to use two separate modules instead of one sharing resources, even with improved latency it still makes sense to have these optimizations. Unless they put better instructions right on the chip, why not?
 
Thanks AMD...I put it on my I7 and smp scores went up in the few multithread tests I have run.
 
I swear, it's like they work for Intel. I don't even bother clicking on the Intel forum, I don't own one so why should I care about it?
How could you waste so much time and energy on a forum that is completely irrelevant to you? I'm not on the Ferrari forums telling
people they are stupid because they could have bought a Corvette ZR1 for less money and similar performance. I just don't understand the mentality.


OMG What an awesome analogy (I LOVE the ZR1)

iobqtirt.jpeg


BTW does anyone have any information on why the update isn't downloadable? It seems like a catch 22.

You can only download 2646060, but the notes in 6060 say you should not apply this patch w/o 2645594. And you can't download 5594! :confused:
 
my, this is so ridiculous. you still can't get that frikken patch anywhere. every page just copies the news over and over ("Patch is out here ya know, I show where you can dl it") with the same stupid M$ support link in it where you can't get that damned thing...
 
Thanks AMD...I put it on my I7 and smp scores went up in the few multithread tests I have run.

That is probably due in part to you disabling core parking. Disabling core parking has been a known fix for better multithreaded performance for quite a while.

Just do a google search and you will see what I mean.

That being said, I installed the original leaked patch a while ago and am going to install this new patch as well.
 
In three years Piledriver will be released and on point to compete directly with Ivy Bridge. By then Intel will have released "Don't Need No Bridge" architecture which will be years ahead of Piledriver.

AMD needs some serious help.
 
8% is the maximum increase but what was the AVERAGE performance delta? Somebody please tell me!
Also, does this hotfix shine particularly well in only certain areas of use? If so where is it worst and where is it best?

average was 1-2% according to AMD. http://www.xbitlabs.com/news/other/...pdate_Brings_1_2_Performance_Improvement.html

It was never going to save the chip. It only addressed how to neatly throw the threads at it and maximize turbo core functionality. Those of you OC'ing with turbo core off will likely see an even smaller performance gain. An increase in performance is an increase in performance, so I'm glad its out.
 
Sounds like the patch does very minor work on the scheduler (no surprise) and it will take Windows 8 to fully realize BD module scheduling.

The same thing happened with HT on Windows 2000 back when Northwood was released. It's just a risk of making a large change that some already released OSs aren't expecting. It usually leaves people disappointed, but that's life. It's going to happen again in the future.
 
I never understood this AMD versus Intel thing. If you want the fastest CPU you can get and don't care too much about the cost, then buy Intel. If you are on a tight budget and want to get the most bang for your buck, then get AMD. I just bought a 3.6Ghz FX4100 for $110 and overclocked it to 4.5Ghz (air cooled). And yes, it is stable at that speed. Right now that's just about the most computing power per $ that you can buy. Does everything I need it to do with cycles to spare.
 
Not all motherboards overclock. So that's more money to spend. Honestly I think Intel has captured low, mid and high end right now.
 
its sadly true. amd needs to buck up fast. I'm still rocking the green though.. supportin the underdogs
 
Forget the AMD vs. Intel ordeal, that's never going to end. It's the problem with people running their mouths about products they don't even know much about, use, much less own. I engage in the BD discussions because I own one, use it, and I know quite a bit about what they can do, yet I would like to learn more about it. If I didn't own one or seek interest in buying one, I wouldn't waste my time posting shit about it, much less even care about it. Forums are supposed to be a useful database of recorded conversations and shared ideas for people to search through. NOBODY wants to sift through all your worthless opinions plastered all over the threads. If you don't have something constructive or informative to add to the thread, keep your reckless little fingers off the keyboard and just read. I haven't even participated in this forum for one month and i'm about to GTFO, there is AMD vs. Intel flame wars being started daily. I don't see people starting shit in the Intel forum, they must get bored and start poking around in the AMD section starting flame wars and shit. If you think Intel is SO much better for whatever reason, go start a fucking blog, shit.... go make some god damn T-Shirts I don't give a fuck. Better yet, if you don't own a Intel rig, go stroke someone's E-Dick that does, there seems to be plenty of that going on. Have a nice day. ALTF4
 
Last edited:
^^ this

again +1000 from me ;)

btw: finally got the patches from guru3d

am now finally trying them out.
 
Here I was hoping I could find comments about people's experience with the hotfix. Instead its moreintel vs amd crap. Too bad. It would be nice to know what people's experience with the hotfix was before I installed it.

Chris
 
Here I was hoping I could find comments about people's experience with the hotfix. Instead its moreintel vs amd crap. Too bad. It would be nice to know what people's experience with the hotfix was before I installed it.

Chris

Not on [H] forums :p
 
well, we didn't start the crap. thank those guys who did.

hey and sorry we expressed our opinions about how effed up this constant babble is... wasn't aware that could annoy ppl oO[/irony]

To your question: still benching some things but from the first looks of it: not much of a difference.
 
Question. Shouldnt improvements for the scheluder in Windows 7 also yield improvements for Intel processors? Thats the impressions I got from some Windows 8 posts some time back. Or are these Win7 hotfixes for AMD only?
 
Yeah if the scheduler improvement is in windows 8 then this should benefit intel cpus. These hotfixes install on Intel systems but does it improve performance for intel? Not sure.
 
I have an FX 6100 processor and will be installing the patch. I had to enter my e-mail address in the MS kb site to get the file sent. From what I have read, I don't expect much improvement, maybe 1-2%. Also, since my overclock is not the extreme turbo, but the mid overclock to 3.8 Ghz, will that influence the performance results.
Paul
 
Here I was hoping I could find comments about people's experience with the hotfix. Instead its moreintel vs amd crap. Too bad. It would be nice to know what people's experience with the hotfix was before I installed it.

Chris
Benchmarks I've seen show no real improvement. The hotfixes were only to address low cpu utilization scenarios to prevent core parking, not benchmarks which max the cores. Moving to Windows 8 is going to be the 'fix' BD users are going to want to make. Those benefits are going to work for Intel though too, so AMD still won't 'gain ground' on the other guys.
 
My understanding, perhaps flawed, is that the current scheduler puts a thread to each physical core, then comes back and starts assigning to multiple "virtual" processors ala hyper-threading. That makes sense to the intel and classic AMD architecture.

The FX chips lose out in this scenario. The correct scenario for FX has two threads assigned to each "module" before moving to the next one so that the module can go turbo if the other module is idle.

I can see that this could help a little.. but not as much as one would hope unless, I suppose, you have a single or double thread application but how many of those really exist anymore?

Mostly I was worried this hotfix would make things worse, so was hoping more motivated individuals would try it :)

Chris
 
In three years Piledriver will be released and on point to compete directly with Ivy Bridge. By then Intel will have released "Don't Need No Bridge" architecture which will be years ahead of Piledriver.

AMD needs some serious help.

Haha! What makes you say that? Just because Bulldozer was originally meant to compete with the i7-920? :D
 
Just kinda a followup to what I said earlier for matlab

comparing to Tom's Hardware dozer tests since there isn't exactly a lot of people doing matlab tests with it

10 runs for me, error in parenthesis besides for 3d/2d since it slows done when ran in a loop
3d 230.5 ms (first 196.5 ms, last 275.3 ms) fft 132.0 ms (+/- 3.7ms) LU 123.5 ms (+/- 0.7ms) 2d 347.3 ms (first 298.6 last 363.9ms) ODE 122.2 ms (+/- 1.3ms) Sparse 251.4 (+/- 8.3ms)
 
I saw improvements with AIDA64 benchmark with memory speeds, they went up which affected the AES and Photoworxx part of the benchmark in a positive way. Also 3dMark11 physics score saw a very nice improvement. If anyone is interested I could post the graphs of the AIDA64 benches with the before and after.

Is there an efficient sure way to remove the hotfix? Will uninstalling it effectively return the scheduler back to previous configuration? I could do more general tests.

There are factors which can probably make this hotfix more noticeable. For example upping the northbridge speed (increasing memory and L3 cache efficiency or speed). 1-2% doesn't really mean much since for some applications it could be more then 10% while others nothing and maybe a lost.

Read the TomsHardware review, my thoughts where these guys are idiots :D, concluding supplying less voltage to a cpu will make it more efficient (no shit!). Concluding about a whole host of items based upon their failures in BD. They never investigated HT speeds effects, NB effects, fsb effects, memory and so on. Anyways it was so long winded and almost pointless in their whole disussion without even touching a drop out of a bucket of water of options.
 
Removing the hotfix works fine to getting the old scheduler back, never tried it with the core parking update however.
 
Removing the hotfix works fine to getting the old scheduler back, never tried it with the core parking update however.

Right now I am running the 2 hotfixes on my 8120.

I can't tell that much of a difference, but it seems boot times are a bit faster, and games are tad bit smoother.
 
[RIP]Zeus;1038269796 said:
Right now I am running the 2 hotfixes on my 8120.

I can't tell that much of a difference, but it seems boot times are a bit faster, and games are tad bit smoother.

I'm experiencing the similar results, the benchmarks aren't throwing out anything special but some games just seem smoother, reduced twitching. Probably due to the better load scheduling. BF3 performs amazing, super smooth, even in 64 player servers. Now when those MSI R7970 Lightning cards come out..... I may have to pick up a pair of those to replace my R6970's.
 
Do these updates improve all multicore CPU's, regardless of brand, or do they only work for AMD chips?

I also noticed these fixes are listed under release "SP2".
 
Do these updates improve all multicore CPU's, regardless of brand, or do they only work for AMD chips?

I also noticed these fixes are listed under release "SP2".

These hotfix's are for AMD 4000, 6000, and FX series chips.

They do not improve any of the Phenom's or lower.
 
Back
Top