AMD's SB750: Enabling Higher Phenom Overclocks?
by Gary Key on July 23, 2008 12:00 AM EST- Posted in
- CPUs
Higher Overclocks in Vista 32-bit vs. 64-bit?
In preparation for this article we also wanted to test one more theory: that Phenom overclocks better under 32-bit Vista than 64-bit Vista. The basis for this theory is that the number of active registers increase in 64-bit mode, which creates additional workload routines that leads to increased transistor switching that ends up increasing thermals and loads on the IMC. Well, you get the picture, not exactly a cycle of events that would lead one to believe an already clock starved core design has a chance when moving to a 64-bit environment, well at least Vista 64-bit. In case you're wondering, Intel CPUs seem to overclock the same regardless of OS but just for kicks we tried our 9850 under the two OSes to see how things differed:
| Phenom 9850BE - Highest Core Speed / HT Ref Clock Vista 64 versus Vista 32 |
||||||
| Core Speed | HT Ref Clock | HT Link Speed | North Bridge Speed | Memory Speed | CPU VID | |
| Foxconn A79A-S ACC On - Vista 32 | 3382 | 205 | 2050 | 2255 | 1093 | 1.4750V |
| Foxconn A79A-S ACC On - Vista 64 | 3280 | 205 | 2050 | 2050 | 1093 | 1.4625V |
| . | ||||||
Strangely enough, our highest HT overclock with the 9850BE was bested by 102MHz in core clock speed and a 205MHz improvement in NB speed under 32-bit Vista. It's not a huge advantage, but something we were definitely able to identify and confirm on a variety of settings with several different processors.
| Phenom 9850BE - Highest Core Speed / HT Ref Clock Vista 64 versus Vista 32 |
||||||
| Core Speed | HT Ref Clock | HT Link Speed | North Bridge Speed | Memory Speed | CPU VID | |
| Foxconn A79A-S ACC On - Vista 64 | 3000 | 200 | 2000 | 2200 | 1066 | 1.3500V |
| Foxconn A79A-S ACC On - Vista 32 | 3150 | 210 | 2100 | 2520 | 1120 | 1.3500V |
| . | ||||||
The biggest difference we noticed between the two operating systems and overclocking occurred when we significantly raised our NB speeds and trying to increase HT ref clock with this particular CPU. We just could not run high NB speeds under Vista 64 in the sweet spot for this processor, which is around 3.0GHz at fairly reasonable voltages. We are still testing this phenomenon (pun not intended) as our 9950BE clocks at nearly the same rates under each OS.
Facebook
Twitter
RSS
48 Comments
View All Comments
wingless - Wednesday, July 23, 2008 - link
"In the end, we know after a good 60 hours of testing, over 600 different settings, 400+ screen shots, and countless reboots that Advanced Clock Calibration works, yet we do not know anymore about ACC at this point from AMD than SpongeBob SquarePants."AMD does nothing but right their wrongs these days. The SB750 is gonna be big for them. I'm amazed that Anandtech went through so many combinations trying to bring this thing down. They have proven without a doubt that this technology works.
I also can't wait until next week. They have given us confirmation that 790GX and 790FX boards with SB750 will be coming out NEXT WEEK! Also they keep mentioning some part that they "dare not mention". WTF is that going to be? I thought 790GX+SB750 was going to be the big news. Theres something else? Next week will be exciting for AMD in all the GOOD ways.
ZootyGray - Thursday, July 24, 2008 - link
Yup - I come here to hear about the results of what I call "brutal testing" rather than read somebody's hype or bias or sloppy inconclusive testing. I am still fairly new here but I believe Anandtech is the real thing.I want that to be true, esp. after wasting my time at tom's bubblegum guide with there trash bias garbage misleading bullschtein socalled reviews.
These guys are working hard to get the real goods - and in contact with AMD, delaying reports promised 'tomorrow' - who cares - they are willing to go to amazing lengths to output accurate info after deeper investigation.
This is no kiddy's website - it's a hard read to get through it; and I know I am not the only one who will reread it to better comprehend what I missed the first time through. This testing is consistent with the articles on 'the 3 amigos' and the one about the recent additions to the AMD line, including the 9950. Take it apart, what makes it squeal.
Also, it seems there is something different about the 9950, and it's not just an oclok 9850.
Looking forward to more next week - and the ?? mystery ?? item????
Big change on the AMD scene - wow.
Calin - Thursday, July 24, 2008 - link
I've quit reading Tom's Hardware long time ago - when I think it was a good site. However, Anandtech was better overall, and one site was enough :)As for in depth reviews... this should qualify as an easy reading, look for the technology behind NVidia's 280 and 260 series GPU for a "transistor-oriented" article - that's a hard read indeed
Comdrpopnfresh - Wednesday, July 23, 2008 - link
I am really interested to see how an ASUS or DFI board with the new sb do in overclocking and performance. As mentioned this board is more of a feature-rich pick. And just looking at the charts of ACC off on the foxconn vs the ASUS with sb600, I hope there will be more of an improvement with a fresh sb750 variant.Interesting though about the pci-e 1.1 and lan choices by amd. Is it possible having those two remedied would hinder the improvement by ACC?
Really looking forward to new comparisons to Intel chips with the new change. Hopefully with 45nm parts and ACC, amd can have a good go.
One thing I would like to know though: What are the specs on the sb750 foxconn vs the asus sb600 in terms of the number of phases in power delivery?
Goty - Wednesday, July 23, 2008 - link
I think that there must be some underlying issue with either the chipset drivers or windows itself when it comes to differing final overclocks, as the CPU shouldn't necessarily be aware of what kind of software its running, other than the fact that a few more registers are being used. IS there any evidence that this is a Phenom-specific issue, or might it affect Athlon64's as well?Locutus465 - Thursday, July 24, 2008 - link
Quite honestly, either way I would run 64bit over 32bit any day of the week period... I'm sorry but the little bit of extra performance you get OCing really doesn't out weigh the many many advantages 64b vista has over the 32b variant (and over XP). Simply put, 32bit operating systems are all but done in my home... I do run 32bit vista on my laptop but only for the following reasons:HP put 32bit on here, and it seems to me all their custom drivers and software (which seem to be required to keep the system from crashing) are all 32bit... so oh well.
This is a budget laptop to begin with and it maxes out at 2GB RAM... The extra overhead of 64b would just be more of a performance killer than anything else.
No gaming will be done on this laptop (so no need to worry about future games needing more that 2GB for a process) and I have my desktop at home if I need to manipulate large flat files.
Now my desktop? Yeah...
Phenom 9850BE
4GB OCz DDR2
Radeon 3870...
It's getting the 64bit treatment all the way :)
CyberHawk - Wednesday, July 23, 2008 - link
... boy, I was waiting for this one :DFireTech - Wednesday, July 23, 2008 - link
So was I ;)It leaves me with more questions than answers though. Time to re-read and await further developements/articles.