Comparing Sockets: 939 vs. 940 vs. 754

To get a clearer idea of exactly what Socket 939 brings to the performance table, we brought some real test results to some of the speculation that has been brewing on the web as to whether 939 is really faster than 754. We suspect our objective tests of Socket 939/940/754 will fly in the face of some of the absurd speculation and sloppy test results that are being posted about the new Socket, but the truth is rarely as exciting as controversies created to stand out from the crowd.

The comparison is simple - there are the 3 sockets that all have processors that can run at 2.2Ghz. To keep the comparison as fair as possible we tested the 3 sockets -754, 940, and 939 with 3 processors with 1MB of on-chip cache all running the same 2.2GHz speed . This gave a head-to-head comparison of the single-channel memory controller of Socket 754 to the Dual-Channel Registered Memory of 940 to the latest Dual-Channel Unbuffered Socket 939.

Sockets were compared using the standard motherboard test suite to give a broad comparison of performance. General Performance was compared using Veritest Multimedia Content Creation 2004 and Business Winstone 2004. results were also compared in PCMark 2004.

General Performance - 2.2GHz & 1MB Cache

General Performance - 2.2GHz & 1MB Cache

Winstones are usually very static at a given CPU speed on a processor. Even wide variations in memory bandwidth and graphics performance rarely have much impact on the Winstone scores. The increases in Winstone scores were only 2.6 to 3% from 754 to 939, but the pattern was very consistent with 939 the fastest, 940 close to 939 and 754 slower than either socket for dual-channel memory. PCMark 2004 was an even more modest spread of 1.4% from slowest to fastest 2.2GHz.

High End Workstation Performance - SPECViewperf 7.1.1

Workstation performance is more sensitive to memory bandwidth, and we do see a wider range in variation among the 2.2GHz processors in SPECviewperf. 754 to 939 ranged from +6.5% in UGS to an 11.4% increase in DX. Considering the CPU's are all the same core at 2.2GHz this is a wide variation just from different memory controllers. The pattern was the generally the same fastest to slowest of 939-940-754, except 940 did outperform 939 in a couple of the SPECviewperf benchmarks.

Development Workstation Performance Comparing Sockets: Gaming Performance


View All Comments

  • Icewind - Tuesday, June 1, 2004 - link

    Slight gramatical problem here......

    As we have mentioned in previous news articles, these new CPUs will run at the 3500+ will run at 2.2GHz while the 3800+ and FX-53 will run at 2.4 GHz each.

    I think you wanted to say "These new CPU"S will be starting at the 3500+ model running at 2.2ghz to the 3800+ running at 2.4ghz."

  • Viditor - Tuesday, June 1, 2004 - link

    "I have updated the article to reflect the fact that we did indeed run our tests with 1T timings on the MSI K8T800 Pro 939 board"

    Thanks Derek...that's why I always read you guys!
  • WBurton - Tuesday, June 1, 2004 - link

    I'm getting a bit frustrated with the Sponsored Links constantly crashing my Opera 7.x. It'd be nice to review an article without having to reboot all the time. Reply
  • MIDIman - Tuesday, June 1, 2004 - link

    Is it possible that the release of a 64-bit OS will change all of these numbers and conlcusions? Reply
  • Lonyo - Tuesday, June 1, 2004 - link

    1066MHz HT bus?
    I thought the HT bus ram 200xmult
    So isn't it 200x5, or 1000MHz?
    (Typo on the first page?)
  • boban10 - Tuesday, June 1, 2004 - link

    what i dont like about many review sites that they encoding always the same codec and then say p4 is faster.
    well look here how they test it:,1558,1603531...

    i hope that your and anandtech next reviews will be more in depth about encoding, cause if people read your reviews about encoding they will buy p4.
    but p4 is not faster in all encoding and that is important to say and test. and i like this site, but if next time i see again only one test in encoding (and that where is know that p4 win) then i will not read your page anymore. and no im not amd fan, im performance fan.
  • mechBgon - Tuesday, June 1, 2004 - link

    Jeff7181, you asked "Is running four unbuffered DIMMs really that necessary?"

    I was running three 512MB modules on my A7N8X Deluxe before replacing it with my K8V Deluxe. That was working out nicely for what I was using it for. When I installed my A64 and K8V Deluxe, I stepped *down* to 1.0GB because if I used all three modules, it would want to run them at DDR200/PC1600 speeds. If I could add a fourth module for 2.0GB total, that would be a welcome improvement. Yeah, I could invest in two 1GB DIMMs, I guess...

    Intel's i865 and i875 families have brought 4 DDR400 DIMMs to Pentium4 owners, and that capability, along with CSA Gigabit, are two places where I have to admit Intel trumped AMD & Co. nicely, and has kept them trumped for quite a while too. So it would be nice to see AMD get their mojo working here.
  • DerekWilson - Tuesday, June 1, 2004 - link

    I have updated the article to reflect the fact that we did indeed run our tests with 1T timings on the MSI K8T800 Pro 939 board.

    I appologize for the omission.
  • Viditor - Tuesday, June 1, 2004 - link

    I wonder if Derek caught the bios setting tweak that Aces found.

    "An incredible difference: with a faster bus turnaround, the memory subsystem is able to serve up to 24% more bandwidth, and the latency goes down from 51 (21.25 ns) to 47 cycles (19.6 ns). This results in measurable real world performance gains:

    In 3DS Max 5.1, we gained 3% of performance
    In Medieval War, Comanche we also gained 3%
    In Wolfenstein: Enemy Territory, we gained 5.5%
    In WinRAR and Plasma, the performance advantage was no less than 9%"
  • nycxandy - Tuesday, June 1, 2004 - link

    Which motherboard was used for the 939 processors? Reply

Log in

Don't have an account? Sign up now