Comparing Sockets: Gaming Performance

DX9 and Media Encoding benchmarks confirm 939 is the fastest CPU at 2.2GHz.

Gaming Performance - DX9 and Encoding

2-pass Media Encoding was about 6% faster on 939 than 754 at the same speed. Direct X 9 games were closer in performance among the 3processors, but still showed a consistent 754-940-939 pattern from slowest to fastest.

Gaming Performance - DX8 and OpenGL

Quake 3 and other games based on the Quake Open GL engine are sensitive to memory bandwidth variations. So it was not a surprise to see Quake 3 increase in performance a bit over 6% from 754 to 939. Across all DX8 games 939 again came out as the top performer.

We did not expect large improvements in performance as Athlon 64 moved from 754 to 939. Since we have found the performance of the Dual-Channel Socket 940 and the Single-Channel 754 to be close when they ran the same speed with the same cache, it was already clear the Athlon 64 was not an architecture that was starved for memory bandwidth like the 'deep-pipes' Pentium 4 design. When P4 went dual-channel the performance improvement was dramatic. Athlon 64 shows more modest increases in performance, but that performance increase is still real and measurable. Dual-Channel 939 is the fastest Athlon 64 socket, followed closely by 940. It appears that the reduced latency of unbuffered memory actually does translate into slightly improved performance for 939. Socket 754 is slower than either of the DC solutions, but the difference between fastest and slowest among the 3 sockets is still relatively small.

Comparing Sockets: 939 vs. 940 vs. 754 Comparing CPUs: 3400+ and 3500+
Comments Locked

38 Comments

View All Comments

  • gherald - Tuesday, June 1, 2004 - link

    Search for 939 on newegg.com folks, you will be pleasantly surprised! Now if we could just find a motherboard...

    I had been hopping this launch would bring the 754 prices down a bit to, say, $220 for a 3200 and $350 for a 3400, but it doesn't seem like that's going to happen quite yet what with 939 starting out at $500 for the 3500 and $700 for the 3800... *sigh* ... more waiting... perhaps prices will come down by the time true PCIe motherboards are availeable.
  • amdfanboy - Tuesday, June 1, 2004 - link

    Go AMD !!
  • nserra - Tuesday, June 1, 2004 - link

    "Everything here is graphics card limited even at 1024x768, but perhaps in the near future when we upgrade the video card we use in CPU and motherboard tests..." i have said this at xbit site and I say it here again!
    What about lowering games resolution and details it may help! There is a detail level in almost every game, it may help. You are testing a processor not the video card! I don't want benches in a processor review that limit the differences between processors.
    fx53 35 fps vs xp3000+ 34fps, what a difference!
  • JGF - Tuesday, June 1, 2004 - link

    #22 while I certainly agree these new prices are too high I dont agree with the wholesale gutting of prices that you're recommending. AMD deserves some decent margins, they shouldnt have to give their product away all the time. Thats unhealthy business and has really hurt amd in the past. I want a healthy amd and for that they are going to need some fatter margins on some sku's.
  • RyanVM - Tuesday, June 1, 2004 - link

    Found a typo on page 13: "Winstones hare usually very static..." should be "Winstones are usually very static..."
  • XRaider - Tuesday, June 1, 2004 - link

    What the heat issues with these newer AMD cpu's? I believe I read on that French site that the new AMD's are putting out alot of thermal wattage! 100+ Can anyone confirm this? I believe the P4 3.4 puts out around 102 - 104 watts! That's alot.
  • AtaStrumf - Tuesday, June 1, 2004 - link

    I found a couple of typos if anyone would care to correct them:

    p.13

    Workstation performance is more sensitive to memory bandwidth, and we do see a wider range in variation among the --3.2GHz-- (THAT WOULD BE 2,2GHz) processors in SPECviewperf.

    The pattern was the generally the same fastest to slowest of 939-949-754 (THAT WOULD BE 939-940-754)

    p.14

    Since we have found the performance of the Dual-Channel Socket 940 and the Single-Channel 754 to be close when hey (THEY)ran...

    Anyway these new S939s suck a$$. Well actually their prices do. Here's what I suggest: retire Athlon XP socket A, introduce it as Socket 754 but only higher models, price 20% up, bring S754 down 33% and S939 down 50%, retire FXs. These prices now are insane.

    AMD are you listening? Of course not.
  • Viditor - Tuesday, June 1, 2004 - link

    "i'll be getting a Dual Channel for sure when the arrive. Pair that with a nice new X800 XT from ATI and i'll be sitting pretty with the flexibility of upgrading for awhile"

    Horses for courses I guess...I'll be getting one too, but I think I'm going nvidia because they appear to have a better handle on 64bit drivers...
    JMHO
  • vedin - Tuesday, June 1, 2004 - link

    Sure would be nice if they released something more along the lines of a 3000-3300+ for 939 this year. Maybe then my brother can finally be nice and justified with an upgrade without having to spend a lot.
  • Icewind - Tuesday, June 1, 2004 - link

    Hmmm, well i'll be getting a Dual Channel for sure when the arrive. Pair that with a nice new X800 XT from ATI and i'll be sitting pretty with the flexibility of upgrading for awhile.

    Thanks again AMD, im anxious to get out of this P4 setup I have now.

Log in

Don't have an account? Sign up now