Application Performance using PC WorldBench 5

General Performance - WorldBench 5


Switching over to WorldBench 5, all of the scores become much closer. The spread between the fastest and slowest tested processor is only 38%, and overclocking of the E6300 and E6400 by 39% and 35% results in a 20% and 17% performance increase, respectively. Given the number of applications being tested in WorldBench 5, the overall results are not too surprising. Some tests are CPU limited while others are bottlenecked by hard drive performance. Athlon 64 X2 is more competitive in this benchmark, and the truth is that any of these systems would be more than fast enough for typical home/office use. If you want the fastest current CPU architecture, however, that title clearly belongs to Intel's Core 2.

WorldBench 5's applications are a bit older than those used in SYSMark 2004, and the data sets not as large - meaning that the smaller cache of the E6300 and E6400 has less of a negative impact. The result is that the 2.88GHz E6400 performs very close to the 2.93GHz X6800 and the 2.59GHz E6300 performs very close to the 2.66GHz E6700. Compared to AMD, the overclocked E6300 does quite well - at 2.592GHz the E6300 is already faster than AMD's Athlon 64 FX-62 - and we're talking $183.

Application Performance using SYSMark 2004 SE Application Performance using Winstone 2004
Comments Locked

137 Comments

View All Comments

  • goinginstyle - Wednesday, July 26, 2006 - link

    quote:

    You might want to pick up the new stepping 6 (mass produces ones) A lot of people over at xteamesystesm are complaining that the stepping 6 doesn’t over clock nearly as well as the stepping 5 and that the temperatures are staring to go though the roof.


    Read the entire post and see what the outcome is before posting this kind of information.
  • Kiijibari - Wednesday, July 26, 2006 - link

    quote:

    You might want to pick up the new stepping 6 (mass produces ones) A lot of people over at xteamesystesm are complaining that the stepping 6 doesn’t over clock nearly as well as the stepping 5 and that the temperatures are staring to go though the roof.


    That's the reason why he was using stepping 5 cores, if he would have used stepping 6, no article, no clicks, no advertising money (from Intel(?) ;-) )

    I mean an overclocking article itself is nonsense, exspecially if you only have 1 kind of a specific CPU and that one is directly from Intel...

    Just wait until the first guys bought E6300s because of anandtech and then stuck around ~2.0/2.2 GHz. Guess who is angry then ...

    Sadly but it looks like anandtech does not care too much about that :(

    regards

    Kiijibari
  • Gary Key - Saturday, July 29, 2006 - link

    quote:

    ust wait until the first guys bought E6300s because of anandtech and then stuck around ~2.0/2.2 GHz. Guess who is angry then ...


    The retail chips are overclocking just as well as the ES chips from all indications the past few days. I know my retail E6400 is 150MHz higher than the ES sample. ;-) Here is an example at XS.......

    http://www.xtremesystems.org/forums/showthread.php...">E6700 Retail
  • Anand Lal Shimpi - Wednesday, July 26, 2006 - link

    As soon as we can get our hands on something other than B1 stepping CPUs we'll include those results. As far as I know, there's nothing that has been changed in current silicon revisions to severely limit overclocking. I haven't run into the issues myself but I will do my best to follow up once I can get later silicon.

    Take care,
    Anand
  • mkruer - Wednesday, July 26, 2006 - link

    Thanks Anand,

    I don’t know if this is true any more but I seem to recall that Intel has a small scale FAB just for engineering samples, and I think that they tend to use it both as a test to validate the new FAB process as well as the CPU design. Thus the engineering samples tend to be better then the mass production chips. Remember the 5 GHz Prescott Intel showed off. I don’t think that anyone go a 5 GHz Prescott running from production chips without having to use liquid nitrogen to keep the chip cool.

    This is something to keep in mind when benching the ES ability. The real production chip might be totally different from a thermal and OC aspect.
  • mine - Wednesday, July 26, 2006 - link

    its been reported that there is one change in b2 retail stepping .
    based on 3 6800 retail versions (b2) @ (xs) b1 stepping seem to oc better than the retail versions

    so right now it seems to early to prejudge ..
  • PetNorth - Wednesday, July 26, 2006 - link

    Anand, your X2 4200+ $215 is wrong. It's EE version. Normal version is $187

    http://www.amd.com/us-en/Processors/ProductInforma...">http://www.amd.com/us-en/Processors/Pro...ion/0,,3...
  • aldamon - Wednesday, July 26, 2006 - link

    Get a better mobo AT and you'll see what the E6300 can really do. The $150 Gigabyte DS3 goes well over 400 FSB with the right RAM.
  • goinginstyle - Thursday, July 27, 2006 - link

    His motherboard is fine, using stock cooling is what limited the overclocks.
  • rjm55 - Wednesday, July 26, 2006 - link

    It's been interesting to watch as AT has paid more and more attention to overclocking. Fortunately for me, the overclocking on good air cooling has been a perfect match for what I'm looking for. I just checked and the monster Tuniq cooler we saw in the Conroe launch sells for just $49. People using it say it is silent because of the huge fan that turns slow inside the core. The point is I can likely go even further with the Tuniq than Anand found here - which was impressive enough.

    Now that we see the headroom on even the cheapest Conroe CPUs it is hard to understnad why anyone can consider an AM2 for anything but the low-end. Until Am2 drops to 65nm the Conroe OC blows away anything I can do with my AM2 chip.

    Thanks for showing us what great overclockers Conroe is turning out to be! The E6600 with 4MB cache for $312 is looking mighty sweet for me. If I remeber you got yours to 4GHz.

Log in

Don't have an account? Sign up now