The Notebook Review forums were hosted by TechTarget, who shut down them down on January 31, 2022. This static read-only archive was pulled by NBR forum users between January 20 and January 31, 2022, in an effort to make sure that the valuable technical information that had been posted on the forums is preserved. For current discussions, many NBR forum users moved over to NotebookTalk.net after the shutdown.
Problems? See this thread at archive.org.

    Strange Cinebench results on Vostro 1400

    Discussion in 'Dell' started by Zebulunite, Aug 20, 2007.

  1. Zebulunite

    Zebulunite Notebook Enthusiast

    Reputations:
    2
    Messages:
    38
    Likes Received:
    0
    Trophy Points:
    15
    I previously ran Cinebench 9.5 on my friend's Lenovo, which has a Core 2 running at 1.67 GHz on a 667MHz bus, and got a single CPU score of 274, and a multi-core score of 504. On my Vostro 1400, with a Core 2 running at 1.6GHz on a 800MHz bus, I got a slightly higher single-cpu score of 282, but a lower multi-core score of 482.

    I'm confused about these results: should my system gain more from using both cores than his, seeing as how mine has a faster bus speed (which would be more important when both cores are doing something than if only one is stressed)? Initially I thought Vista might be the problem (he has XP on his Lenovo), yet after installing XP on my Vostro I have about the same results.

    Does anyone know whats going on here? He hasn't formatted his computer since buying it, so he is all but certainly has more processes running the background, and likewise my Vostro has double the memory (1GB on the Lenovo vs 2GB on the Vostro).
     
  2. Zebulunite

    Zebulunite Notebook Enthusiast

    Reputations:
    2
    Messages:
    38
    Likes Received:
    0
    Trophy Points:
    15
    Actually I think I may have figured this out. I forgot that the Santa Rosa platform features Intel Dynamic Acceleration. chuck232 wrote of good summary of what it does in his recent XPS 1330 review, which I quoted below:


    "While software is beginning to take advantage of parallelization, many applications are still capable of using only a single core. As a result, the potential of multi-core processors is lost. Intel has come up with an innovative way to boost performance in single-threaded applications, while staying within the same power and thermal envelope, by overclocking the core being stressed. In this manner, single threaded applications can take advantage of the higher frequency, without compromising thermals."


    To test that, I ran CPU-Z while running a single-cpu Cinebench test. At first CPU-Z didn't exceed 1.6GHz (the default speed of my Vostro's CPU), but after realizing that CPU-Z only reports Core-0's speed, I used the task manager to set Cinebench to only run on Core 0. After doing this, when running the test I saw the CPU speed fluctuate between 1.6GHz, 1.8GHz, and 2GHz.

    So I guess that explains the performance discrepancy. Pretty neat feature on Intel's part (this coming from a long-time AMD fan),
     
  3. chuck232

    chuck232 Notebook Deity NBR Reviewer

    Reputations:
    274
    Messages:
    1,736
    Likes Received:
    1
    Trophy Points:
    55
    Hrrmmmm, I was thinking of replying to your original post, but I honestly wasn't certain what the reason was at the time. It looks like you (literally) took the words out of my mouth! ;)
     
  4. Zebulunite

    Zebulunite Notebook Enthusiast

    Reputations:
    2
    Messages:
    38
    Likes Received:
    0
    Trophy Points:
    15
    Heh, yeah. Thanks in any case for reminding me about that.

    On that note though, have you noticed on your laptop the cpu speed fluctuating when it goes into Intel Dynamic Acceleration mode? As I said in my other post, when running Cinebench the cpu speed would alternate between 1.6, 1.8, and 2 GHz.