Yeah i now but wy have my winrar score nearly halved

Sorry but I thought you guys aready know this?

For example, in testing by AMD with the AMD FX-8150, we are seeing up to 10% uplift on a number of games with the Windows 8 Developer Preview compared to Windows® 7. Of course, results do vary.
We are also working with Microsoft on a scheduler update for Windows 7 that will be available soon.

Tuesday , November 29, 2011
www.hardocp.com/article/2011/11/29/hardocp_readers_ask_amd_bulldozer_questions/2

Good news for our Bulldozer Pals :toast:

nt300Sorry but I thought you guys aready know this?


Tuesday , November 29, 2011
www.hardocp.com/article/2011/11/29/hardocp_readers_ask_amd_bulldozer_questions/2

Good news for our Bulldozer Pals :toast:

"Available soon" is not what this is about. Its about the patch actually coming out. But thank you for the link!

nt300Sorry but I thought you guys aready know this?


Tuesday , November 29, 2011
www.hardocp.com/article/2011/11/29/hardocp_readers_ask_amd_bulldozer_questions/2

Good news for our Bulldozer Pals :toast:

Yeah, that was two weeks ago, and now the patch is here. The patch itself was released yesterday.

dezzThis fix is for improving less-threaded application performance (like f.ex. games)!
No wonder it has no effect on CineBench@8 threads.

I'm not surprised that multi-threaded apps son't show much if any benefit. As far as I understand it, the behavior noticed in the FX Overclocknig thread here on TPU, with not all cores scaling up in frequency, is Bulldozer's default behavior. This behavior, without the scheduler update, had some threads running on the lower-speed cores instead of forcing them on the high-speed threads, affecting end results. The patch should have apps like games, that are not using the CPU fully, running on the faster cores when possible.

must be something wrong with my install, ran my benchmarks, applied hotfix and windows crashed hard, safemode, and restore fixed it , redownloaded hot fix same again.

Before anyone asks, Retail copy win 7 x64 ultimate, fully updated.

As for this new performance increase I suspect its only for situational issues.

I will install after it has been fully tested.

Right so my multithreaded performance has gone down, how ever I've had a 7.6% increase in cinebench single core performance.

Will try more benchmarks.

riskaYeah i now but wy have my winrar score nearly halved

And what was the score after?

By this time, AMD could have made a driver like they did for the Athlon 64 x2, when Windows XP SP2, at the time, couldn't use the two cores effectively.

So it looks like the mythical Windows speedboost for BD was real after all. If the boost is something worthwhile, then this is excellent news all round, as Intel is gonna get some much needed competition - it doesn't matter which brand one prefers to buy.

TheMailMan78Sometimes.......just sometimes WE ARE THE ARTICLE YO!

+1 and welcome back. :toast:

qubitSo it looks like the mythical Windows speedboost for BD was real after all. If the boost is something worthwhile, then this is excellent news all round, as Intel is gonna get some much needed competition - it doesn't matter which brand one prefers to buy.



+1 and welcome back. :toast:

It was never mythical. Some people just love to post negativity and don't have a real grasp on the facts.

And thank you.

I seem to be getting the ~same/similar score for Cinebench 11.5 (multi cpu) after updating to the hotfix.

Orange is after updating the hot fix... Blue is before:

PolRogerI seem to be getting the ~same/similar score for Cinebench 11.5 (multi cpu) after updating to the hotfix.

Orange is after updating the hot fix... Blue is before:
i695.photobucket.com/albums/vv315/PolRoger/AMD%20FX/Cinebench11546GHzW7hotfixforBD.png

Can you try running cinebench on four threads ?

This patch fixes scheduling ( I.E when using 4 threads it will use one core per module rather tahn the first two modules)

TheMailMan78It was never mythical. Some people just love to post negativity and don't have a real grasp on the facts.

And thank you.

When I frist heard about it I dismissed it, "some group had found it was a scheduler issue, found a fix that improved some benchmarks by upto 10%"

No more details were given, sounded a little mythical to me.

I will find out myself and share but need to get my install fixed as it keeps crashing out on the update.

The desktop is almost done re installing and then...updates then later tonight I will post as many benchmarks as I can find.

This changes almost nothing. You cant affect buldys weak IPC. Yes, multithreaded apps will benefit from this, but most important thing remains - with 4 threads active, buldy wont beat similarly clocked X4 phenom. So these are good news for server farm owners and other pros, gamers wont feel a thing.

roll on the 21st i was ponderin a 960T but this'll swing me in the 8120's direction , good news by the way, a very good first show their MM

If anyone does spot any noticeable improvements please let me know.

Turbo still doesn't work properly. With my set up.

pantherx12Can you try running cinebench on four threads ?

This patch fixes scheduling ( I.E when using 4 threads it will use one core per module rather tahn the first two modules)

I could... but I don't have any Cinebench runs saved from before the hot fix (4 threads) to compare it with?

PolRogerI could... but I don't have any Cinebench runs saved from before the hot fix (4 threads) to compare it with?

Ahhh not to worry then, but yeah this shouldn't really affect things that use 8 threads ( although it has, and negatively for me) will mostly improve single/dual/tri and quad threaded applications.

Or general system smoothness.

shb-This changes almost nothing. You cant affect buldys weak IPC. Yes, multithreaded apps will benefit from this, but most important thing remains - with 4 threads active, buldy wont beat similarly clocked X4 phenom. So these are good news for server farm owners and other pros, gamers wont feel a thing.

You don't seems to understand what this patch is about. :slap:
This patch increase exactly what gamers care about, which is lightly thread performance.

Right I've rerun cinebench a whole bunch of times now (8 threads)

And I have definitely lost performance since this patch, it's only a 1.1% difference but that's a difference of

7.35 prepatch to 7.21 after patch.

Could disappoint bench markers.

qubitSo it looks like the mythical Windows speedboost for BD was real after all. If the boost is something worthwhile, then this is excellent news all round, as Intel is gonna get some much needed competition - it doesn't matter which brand one prefers to buy.

+1 and welcome back. :toast:

You can all thank me for spreading this wonderful rumour which I knew was true, but many did not believe me including the Knight Rider Moderator :p

Anyhow expect a 12% to 20% CPU usage increase via Skyrim, though not sure about performance yet until I get my H100 back from the shop.

Super XPYou can all thank me for spreading this wonderful rumour which I knew was true, but many did not believe me including the Knight Rider Moderator :p

Anyhow expect a 12% to 20% CPU usage increase via Skyrim, though not sure about performance yet until I get my H100 back from the shop.

Lol, show me results. :rolleyes: Update or not, I'll still take the less expensive, yet more powerful, yet less power consuming chip over BD.

Hopefully this patch will at least help those who are suffering from random, unwarranted BSOD's.