The Story of Phenom's Erratic Performance

A few months ago I called AMD with a problem. In testing for AMD's Phenom re-launch, I encountered a major issue: Phenom performance actually degraded since I first tested it. There were two benchmarks in particular that saw performance go down: SYSMark 2007 and Adobe Photoshop CS3. SYSMark gave me scores that were around 10% lower than what they were when Phenom launched, despite these being B3-stepping parts. Photoshop was far worse, with performance being around half of what it was at the Phenom launch.

I originally attributed the changes to something strange that happened with the move to Vista SP1. I theorized that somehow the TLB fix was being applied to B3 stepping parts and negatively impacting performance, but WinRAR and memory tests didn't support the thought. AMD couldn't figure out what was happening so I chalked it up to a problem with my testbed or testing methodology, something I'd have to revisit at a later time.

The SYSMark issue actually went away on its own; I swapped from my 780G motherboard to a 790FX without reinstalling Windows to see if it was a 780G/integrated graphics issue, the performance problems remained. But upon swapping back, once again without a Vista reinstall, my SYSMark scores magically jumped around 10%. The "fix" lasted long enough for me to finish the benchmarks for the Phenom re-launch review, but sure enough the problem reappeared when I tried to re-run one test after I'd gotten everything I needed for the review. I never did figure out what was causing my Photoshop performance issues however.

The Culprit: Cool'n'Quiet

When I started testing for today's review, I ran into the same issue again. I always start by benchmarking SYSMark first, since the suite takes forever to complete on a single CPU. As soon as I got my first results, I realized my problem was back. Determined to find the cause I tried everything...again. The one thing I didn't change however was the Cool'n'Quiet setting in the BIOS, but I did try it this time.

Cool 'n Quiet is the marketing name for AMD's on-the-fly clock speed/voltage adjustment. Depending on the software load on the CPU, AMD's Cool'n'Quiet will adjust the p-state of the CPU cores - which includes adjusting core voltage and clock speed. If you're running a processor intensive game or application, CnQ will make sure your CPU runs at full speed, but if you're just typing a text document it will underclock/undervolt the CPU.

Phenom's CnQ is a more advanced version of what was in the Athlon 64 X2, it not only allows for individual core clock speed adjustment but is able to transition between states faster than previous versions of CnQ (at least that's what AMD implies).

SYSMark 2007 Preview Overall Score CnQ/EIST Enabled CnQ EIST/Disabled Performance Increase from Disabling CnQ/EIST
AMD Phenom X4 9350e 101 113 11.8%
AMD Athlon X2 6400+ 121 123 1.7%
Intel Core 2 Quad Q9450 153 156 2.0%

 

 

Disabling CnQ increased my SYSMark scores by around 12% and cut my Photoshop CS3 render times in half (58.7s with CnQ enabled, 35.2s with CnQ disabled); enabling CnQ had the opposite effect. Gary ran similar numbers using PCMark Vantage and found a 5% difference. AMD originally insisted that the problem was because SYSMark introduces unrealistic pauses into its benchmark (so called "think" times or periods of time while the system is waiting for user input), but since we found the same issue in other benchmarks (PCM Vantage and our Photoshop test), we believe this is more than just a SYSMark issue.

The SYSMark problem was mostly repeatable, it would consistently produce lower scores with CnQ enabled on the Phenom CPUs. The Photoshop scores were a bit more erratic - the problem went away for a little while but has since returned and won't go away again, even with CnQ disabled. It is worth mentioning that the majority of our benchmarks wasn't impacted by the problem, but that doesn't mean that it won't impact your daily usage.

Note that the same problem doesn't plague the Athlon X2, this appears to be a Phenom/K10 issue only. As a reference we ran some numbers with Intel's SpeedStep enabled vs. disabled and didn't see any similar behavior.

I had found the source of my problems, but I didn't understand why it caused them.

Intel Makes 45nm Affordable The Mystery of the Missing Performance
POST A COMMENT

36 Comments

View All Comments

  • Sylvanas - Tuesday, July 1, 2008 - link

    Wheres the 9950BE overclocking results? It is an unlocked CPU so what about Overclocking the NB? What performance difference does that bring? I doubt people that buy IGP's are going to overclocking much anyway since they are usually silent HTPC rigs... Reply
  • Gary Key - Tuesday, July 1, 2008 - link

    The 9950BE overclocking results are coming in a different article. Unfortunately, our 790FX boards (they have been beat on for six months) were not exactly up to speed and we thought it would be better to not show anything instead of a 2.8GHz clock that obviously is not representative of the processor at this point.

    Also, most of our previous results were run on the 780G, a chipset that when tuned correctly and on a good board will outclock the 790FX with a discreet graphics card by the way. Jetway just released a fairly comprehensive BIOS for their new 780G we ended up using after the others started failing. We just received BIOS updates for the 780a boards and have a new 790FX/SB750 arriving shortly for a CF/SLI update on AMD (gaming is not that bad by the way on the Phenom for the mid-range market).

    Increasing the NB core (IMC) clock (in Phenom it runs async from the Core Speed unlike Athlon which is Sync) drops latencies (especially L3) and increases memory performance/throughput, which in turn improves system performance. The Phenom starts to come to life when you hit a 2.6GHz core speed with a NB core clock at 2200MHz+. Depending on the application and CPU, increasing NB core speeds (getting up to 2200MHz+) can result in performance differences from 3%~12% in most cases.

    Almost as important is increasing HT speed for further optimizing the pipeline links (CPU/Memory/PCIe,etc). Our 9950BE follow up will have an overclocking guide along with optimization details.
    Reply
  • Sylvanas - Wednesday, July 2, 2008 - link

    Excellent, thanks for the info Gary- I look forward to the follow up 9950BE overclocking article. If there is some info on the SB750 aswell that's even better :) Reply
  • DigitalFreak - Tuesday, July 1, 2008 - link

    AMD post X2 = ROFLMAO

    The C&Q thing is probably another respin waiting to happen. What a bunch of boobs.
    Reply
  • acejj26 - Tuesday, July 1, 2008 - link

    what's a seccond?
    why didn't you include the 9950 in the first page of benchmarks?
    is the 9960 a new processor from AMD?

    i've come to expect these errors from other staff writers, but not you Anand.
    Reply
  • skiboysteve - Tuesday, July 1, 2008 - link

    why are you using 780G to overclock and check stability on the same article you say how someone else wrote an article about how that is a bad idea because of power...

    you even say at the bottom of your overclocking page, a mere footnote, that you got higher clocks on a different platform
    Reply
  • js01 - Tuesday, July 1, 2008 - link

    I think they scale much better then that hothardware got the 9950be to 3.1ghz barely even trying and the 9350e to 2.7ghz.
    http://www.hothardware.com/Articles/AMD_Phenom_X4_...">http://www.hothardware.com/Articles/AMD...nom_X4_9...
    Reply
  • Gary Key - Tuesday, July 1, 2008 - link

    It depends on the board and CPU actually. We have a retail 9850BE that will do 3.3, but three others struggle to make it to 2.8. Until we see some consistency in the retail parts, we would rather play it safe with the comments. A separate overclocking article is on its way though with the new lineup. :) Reply
  • woofermazing - Tuesday, July 1, 2008 - link

    Odd that you guys couldn't get any OC out of the 9950. Results from other sites have been pretty impressive using the stock cooler. 3.6ghz is the highest of seen so far. Reply
  • Clauzii - Wednesday, July 2, 2008 - link

    I second that!

    PS: And why does the comment page keep looking like pre-95 internet :O (I'm on FF3)
    Reply

Log in

Don't have an account? Sign up now