Bulldozer

Ankur

Ankur

Active Member
Quick question, compared to the Sandy Bridge CPUs does the Bulldozer perform less because there are no programs or benchmarks that couldn't utilize the 8 cores?

Other way around, is a program that needs assuming 8 cores is test on both Bulldozer and Sandy Bridge then will Bulldozer perform higher? :)
 
maroon1

maroon1

New Member
Ankur said:
Quick question, compared to the Sandy Bridge CPUs does the Bulldozer perform less because there are no programs or benchmarks that couldn't utilize the 8 cores?

Other way around, is a program that needs assuming 8 cores is test on both Bulldozer and Sandy Bridge then will Bulldozer perform higher? :)
Click to expand...

Cinebench 11.5 can take advantage of 8 cores, yet i7 2600 beats it in that benchmark

From what I seen i7 2600 is better than FX-8150 in most multi-threaded benchmarks. There is only few cases were FX-8150 wins

As for single-threaded performance, Bulldozer is very weak. In some cases it is weaker than Phenom II, despite the clock speed advantage of Bulldozer

AMD_FX-8150-202.jpg
 
Shane

Shane

Super Moderator
Staff member
maroon1 said:
As for single-threaded performance, Bulldozer is very weak. In some cases it is weaker than Phenom II, despite the clock speed advantage of Bulldozer
Click to expand...

I wonder why its performance is poor on single-threaded :confused:

Shame really because i was looking forward to Bulldozer for my next possible upgrade,Unless i can get a very good deal on a Sandybridge Cpu + motherboard i might just wait for Ivy Bridge.:cool:
 
jonnyp11

jonnyp11

New Member
Nevakonaza said:
I wonder why its performance is poor on single-threaded :confused:
Click to expand...

cuz the win7 scheduler is retarted when it comes to bd since it hasn't been updated, so it's throwing stuff everywhere so it can't use its turbo when it could. Also nothing is programmed to use the 2 or 3 new technologies they introduced with it, so there's another problem.


linkin said:
From what I've seen, it does. I'm ordering an FX-8120 from newegg this week :)
Click to expand...

Have fun man, and from what i saw, apparently it is fairly easy to hit 5 or so ghz on 4/6 cores :)D), just not on 8 :)(), so for gaming that would pull it up alot.
 
Last edited:
spynoodle

spynoodle

Active Member
linkin said:
FX doesn't natively support x87 is why SuperPi results suck, though I will happily do wprime :D
Click to expand...
I see.... are you running Windows 8 or Linux on any partition? If so, then it would be really interesting to see a single-thread WPrime benchmark (if it exists, I've never tried WPrime) on an OS that can utilize Bulldozer well. Or, better yet, a quad-thread WPrime benchmark.
 
StrangleHold

StrangleHold

Moderator
Staff member
linkin said:
http://quinetiam.com/?p=2356

Registry fix = 40% performance boost!!
Click to expand...

I've heard that AMD and Microsoft are talking about who will provide the patch. If it will be a microsoft patch or a new AMD processor driver. Wish they would make up there mind and just get it out. Windows just doesnt know what to do with a module. The pisser about it, this should have been took care of long before the release.
 
linkin

linkin

VIP Member
StrangleHold said:
I've heard that AMD and Microsoft are talking about who will provide the patch. If it will be a microsoft patch or a new AMD processor driver. Wish they would make up there mind and just get it out. Windows just doesnt know what to do with a module. The pisser about it, this should have been took care of long before the release.
Click to expand...

It should have been done, yeah. I guess they couldn't really delay release any more though...

Still, nice to know that when I order mine the driver/fix could be out already.

Slightly off topic: Did you know that Phenom II CPU's and windows vista/7 still have the TLB bug fix enabled by default, even when the Phenom II doesn't have the bug? I disabled it with a little program... I tested with wprime before and after and I got a lower time :D
 
StrangleHold

StrangleHold

Moderator
Staff member
linkin said:
It should have been done, yeah. I guess they couldn't really delay release any more though...

Still, nice to know that when I order mine the driver/fix could be out already.

Slightly off topic: Did you know that Phenom II CPU's and windows vista/7 still have the TLB bug fix enabled by default, even when the Phenom II doesn't have the bug? I disabled it with a little program... I tested with wprime before and after and I got a lower time :D
Click to expand...

Never knew that. I know at first it was a bios update, but killed Phenom I performance.
 
C

CrazyMike

New Member


Uploaded with ImageShack.us


Is the FX8150 the maximum processor that they are coming out with? I thought they were going to have a higher one?

If this is true, looks like i will be buying it sooner than what i thought :)
 
linkin

linkin

VIP Member
I see a lot of finger pointing and questioning...
Here is my current understanding:

In the past we have seen CPU drivers for windows (mostly XP, and mostly AMD.) This is NOT what is occurring here, its not some weird issue with the processor causing it (although it kind of is)

What is happening is this: The FX 8xxx series of processors has 4 cores, each core with TWO integer pipelines, effectively granting two processor cores for the price of one. (not "hyperthreading" effective, but almost literally additional core effective.)
This is all fine and dandy, except that when Windows (or Linux) goes to issue work to a core in the processor, it picks the core with the least load, and distributes things using a task scheduler. The problem here is that when a single pipeline on a bulldozer core is being utilized both cores are effectively "in use" even though Task Manager sees the cores appropriately, the task scheduler does not. This isn't really anybody's fault, but rather an oversight.
Standard CPU design from a by-the-book standpoint would never add a second integer pipeline to a single core. So nobody ever really thought to code a task scheduler that worked in that situation. So basically, AMD thought outside the box, and Windows got confused.

Oh not to mention that if the scheduler tries to run non-integer work on one of the integer only threads, the CPU returns an error and the command is re-queued and has to sit in line waiting to be executed again. Basically its a scheduler issue, and could definitely cause some HUGE impact in synthetic benchmarks and normal use. And even more issues in a cluttered server environment.
Click to expand...

Quote from OCN regarding the issue in the link I posted before.
 
jonnyp11

jonnyp11

New Member
CrazyMike said:


Uploaded with ImageShack.us


Is the FX8150 the maximum processor that they are coming out with? I thought they were going to have a higher one?

If this is true, looks like i will be buying it sooner than what i thought :)
Click to expand...

there is an 8170 and i think one or 2 others (not 8 cores) coming out in the first quarter of next year.

I just can't wait to see what Bigfella's response to this new info will be :D
 
linkin

linkin

VIP Member
My understanding is that BD has 8 INT and 8 128bit FP cores (I.e. 1 of each for each core). Does this mean 16 max instructions at once?

Each module however has 4 256bit FP cores.

So if you are using all 4 256bit cores then maximum instructions is 12.

Therefore BD should be able to run between min 12 and max 16 instructions per cycle.

I bet windows is only running max 8 and not correctly with all the other CACHE associated issues.

So 40% could very well be plausible.

What ya think?
Click to expand...

Another OCN quote.
 
claptonman

claptonman

New Member
I really hope a new update or BIOS update fixes this. I'm not getting my hopes up, but we'll see. Not believing anything without proof.
 
You must log in or register to reply here.
Top