http://forums.anandtech.com/showthread.php?t=1723791
I put my core I7 to rest and going to analyze amd BD
Now most aps use up to 4 cores, well lets the game begins.
PD: my research is to find hidden performance, not for marketing.
.....there has already been dozens of locked bd threads already here.
Pretty sure most people here will not remember a post from 7 years ago.
And if i were you, i wouldn't waste your time analyzing Bulldozer.....there has already been dozens of BD threads already here.
We all know what it can and cannot do.
Stop the jokes.
The trick is modify the os and apps for the bd, a 4 tread app may gain a lot if is spreaded the right way.
Try to make numa work for bd.
Dude 12 life time posts and 2 here and 10 of them made in you're first thread from 2005?? Each one is about AMD performance?? Epic plan.
Are you some kind of War Hammer 40k Space Marine that spends his time forzen in suspended animation only to be awoken when AMD needs some muscle?
Yes, but wouldn't that statement count for every piece of software in regards to every piece of hardware. Any piece of software that specifically engineered to work specific hardware will likely take better advantage of its capabilities. That's not called hidden performance. That's called optimization.
Intel and AMD have to develop CPUs that not only push the line forward, but also work well with current software. From what I understand, the problem with Bulldozer was it's lackluster performance with present day software, contrary to overall expectations were. I'm sure if you engineer software specifically for Bulldozer you could make some magic happen, but all that does is reaffirm that AMD made some bad judgement calls on what the market needed at the time it was developed.
The beat thing you can do with your BD to make apps run better is to sell it and buy Intel. Sorry, but even if BD is a gem of a chip that can't shine due to software, it doesn't matter. Hardware is nothing without software and the vast majority of software works better on Intels archetecture.
Anandtech said:The first update simply tells Windows 7 to schedule all threads on empty modules first, then on shared cores. The second hotfix increases Windows 7's core parking latency if there are threads that need scheduling. There's a performance penalty you pay to sleep/wake a module, so if there are threads waiting to be scheduled they'll have a better chance to be scheduled on an unused module after this update.
Note that neither hotfix enables the most optimal scheduling on Bulldozer. Rather than being thread aware and scheduling dependent threads on the same module and independent threads across separate modules, the updates simply move to a better default cause of scheduling on modules first. This should improve performance in most cases but there's a chance that some workloads will see a performance reduction. AMD tells me that it's still working with OS vendors (read: Microsoft) to better optimize for Bulldozer. If I had to guess I'd say that we may see the next big step forward with Windows 8.
AMD was pretty honest when it described the performance gains FX owners can expect to see from this update. In its own blog post on the topic AMD tells users to expect a 1 - 2% gain on average across most applications. Without any big promises I wasn't expecting the Bulldozer vs. Sandy Bridge standings to change post-update, but I wanted to run some tests just to be sure.