3D Rendering Performance using 3dsmax 7 & CineBench 9.5

We're looking at 3D rendering performance using two different applications: 3D Studio Max and Cinebench 9.5. Cinebench is a free performance testing utility based off of the CINEMA 4D R8 rendering package. Our scores from 3D Studio Max are a composite score from four rendering tests: CBalls2, SinglePipe2, UnderWater, and 3dsmax5 Rays.

General Performance - 3D Rendering

General Performance - 3D Rendering

General Performance - 3D Rendering


3D rendering relies almost entirely on CPU performance, and cache sizes have very little impact. The end result is that our overclocked E6300 and E6400 place very near the top of the charts, and the overclocked E6400 actually manages to take the lead over the X6800 in the Cinebench multi-CPU rendering test. Clock for clock, Core 2 Duo holds about a 9-11% performance advantage in 3D rendering over the AMD X2 processors. The difference between the fastest and slowest systems tested here is roughly 60%-70%, and due to the time-consuming nature of 3D rendering even small performance increases are very welcome.

Once again we see that while the Core 2 Duo E6300 is slightly faster than the Athlon 64 X2 4200+, and once overclocked it's out of reach of even an FX-62. The E6400 is also an impressive little chip, offering performance around the X2 4600+ and X2 5000+ levels.

Application Performance using Winstone 2004 Encoding Performance using DivX 6.1, WME9, Quicktime (H.264) & iTunes
Comments Locked

137 Comments

View All Comments

  • getbush - Wednesday, July 26, 2006 - link

    There is a for that should be four and you start the oblivion page with will instead of we'll.
  • JarredWalton - Wednesday, July 26, 2006 - link

    Thanks - I gave the document a final proofing now that I'm a bit more coherent and squashed several more "typos" (speech-recognition-os?) I helped Anand fill in a bunch of the text, but it was late and my eyes weren't cooperating. LOL
  • yacoub - Wednesday, July 26, 2006 - link

    What I see here is that the E6400 is easily the way to go for folks who aren't interested in overclocking but want the best bang-for-the-buck.

    For very little more $$ than the E6300, you get a chip that rides quite a bit higher up on the charts in many tests.

    Now the question: What affordable motherboard is recommended for stable, reliable non-overclocked C2D Conroe performance? Perhaps the Intel P965 board?

    There's no reason to drop $200-250 for a motherboard when you aren't going to utilize its overclocking functionality. I believe that opens up the user to the more affordable P965 boards, right? They tend to be more around $150 and if it's made by Intel it should be plenty stable, right?

    Also most boards now are passively-cooled which is excellent since the dinky fans on older motherboards were always noisy and died quickly. Avoiding those is another benefit as I believe the Intel P965 board is passively cooled as well.

    Thoughts?
  • anandtechrocks - Wednesday, July 26, 2006 - link

    Check out the Gigabyte DS3. It uses the P965 chipset and costs ~$144. It overclocks just as well as the $250 Asus motherboard in this article and it uses very high quality solid capacitors. Only drawback is that no SLI or Crossfire.
  • yacoub - Wednesday, July 26, 2006 - link

    I don't know, looks like there's some cause for concern about currently available 965 boards now...
    http://forums.anandtech.com/messageview.aspx?catid...">http://forums.anandtech.com/messageview...amp;thre...
  • anandtechrocks - Wednesday, July 26, 2006 - link

    Very good article, I really enjoyed it. I think there is an error on page 4, on the 3rd graph from the bottom. The E6300 and E6400 bars are miss-labled.
  • JarredWalton - Wednesday, July 26, 2006 - link

    Fixed - thanks.
  • code65536 - Wednesday, July 26, 2006 - link

    How do the OC'ed chips do with power consumption. Does a 6400 @ 2.88 use more or less power than a 6800, for example?
  • supremelaw - Wednesday, July 26, 2006 - link

    ... and heat.

    I assume that the stock Intel HSF hasn't changed:

    http://www.supremelaw.org/systems/heatsinks/warnin...">http://www.supremelaw.org/systems/heatsinks/warnin...

    and that a superior HSF with proper backing plate
    is still recommended for Conroe CPUs, even though
    they run cooler in general.


    Sincerely yours,
    /s/ Paul Andrew Mitchell
    Webmaster, Supreme Law Library
    http://www.supremelaw.org/">http://www.supremelaw.org/
  • houe - Wednesday, July 26, 2006 - link

    fp

Log in

Don't have an account? Sign up now