Thread: Help me choose my hardware

  1. #16
    Cat without Hat CornedBee's Avatar
    Join Date
    Apr 2003
    Posts
    8,895
    Under optimal conditions, I think the C3D may well have 10 times the computing power of that system. This assumes a perfectly multi-threaded application making use of advanced SSE capabilities if available - like 3d rendering.
    All the buzzt!
    CornedBee

    "There is not now, nor has there ever been, nor will there ever be, any programming language in which it is the least bit difficult to write bad code."
    - Flon's Law

  2. #17
    Internet Superhero
    Join Date
    Sep 2006
    Location
    Denmark
    Posts
    964
    Quote Originally Posted by CornedBee View Post
    Under optimal conditions, I think the C3D may well have 10 times the computing power of that system. This assumes a perfectly multi-threaded application making use of advanced SSE capabilities if available - like 3d rendering.
    Even in an application that isn't multi-threaded, that will be the case. Try to compare the 23-24 seconds that an E4500 uses to calculate a 1M test in SuperPI, against whatever a P3 @ 500mHz uses, it will be less than a tenth. And SuperPI is neither multi-threaded nor does it use advanced SSE - it was developed in 1996.
    How I need a drink, alcoholic in nature, after the heavy lectures involving quantum mechanics.

  3. #18
    Kernel hacker
    Join Date
    Jul 2007
    Location
    Farncombe, Surrey, England
    Posts
    15,677
    Quote Originally Posted by Neo1 View Post
    Even in an application that isn't multi-threaded, that will be the case. Try to compare the 23-24 seconds that an E4500 uses to calculate a 1M test in SuperPI, against whatever a P3 @ 500mHz uses, it will be less than a tenth. And SuperPI is neither multi-threaded nor does it use advanced SSE - it was developed in 1996.
    It does, however, use a fair amount of memory, and on a modern CPU, it probably fits within the cache (or a large portion gets placed in cache), whilst the older processors with smaller caches will struggle. Memory and related controller logic being 4x or so faster (peak speed) will also help a fair bit. And yes, of course, the processor running about 4x the clock-speed will also be beneficial.

    --
    Mats
    Compilers can produce warnings - make the compiler programmers happy: Use them!
    Please don't PM me for help - and no, I don't do help over instant messengers.

  4. #19
    (?<!re)tired Mario F.'s Avatar
    Join Date
    May 2006
    Location
    Ireland
    Posts
    8,446
    They have a Quad there for 599 Euros without monitor. But comes with 2Gb Ram only and an onboard graphics card. I think its perhaps too much anyways since I plan on using only XP on it and later migrate to Linux.

    To be honest I'm more excited about the Thinkpad. Ever since CornedBee mentioned it, it got stuck in the back of my head. I had heard good remarks about them before. But my immediate search revealed, as I said a few posts down, they are poorly represented in Portugal. But after getting annoyed at Asus and Toshiba, I decided "what the heck". Ordered it from the web today. It will take maybe 14 days to get here. And they are black!

    I'm going to get the desktop later today at around 8pm. I'm not going to risk getting annoyed again (because of that SATA drive) so I went the extra mile and decided to add a 160GB PATA from a local store. First I'll see if SATA is native or I can boot from an USB stick in which case I will install the SATA drivers on the stick (the desktop comes with no floppy drive). If both fail, I'll drive there quickly and get the PATA drive.

    You'd think a C2D ready mobo can boot from USB?
    Originally Posted by brewbuck:
    Reimplementing a large system in another language to get a 25% performance boost is nonsense. It would be cheaper to just get a computer which is 25% faster.

  5. #20
    Internet Superhero
    Join Date
    Sep 2006
    Location
    Denmark
    Posts
    964
    Quote Originally Posted by Mario F. View Post
    You'd think a C2D ready mobo can boot from USB?
    What motherboard is it? Shouldn't take more than a couple of minutes to figure out...

    And they are black!
    Oh now that's just racism!


    Edit - Matsp:

    All valid points, i do however believe that the performance jump will easily be 10 times, still. My new machine is atleast 4 times faster than my old Celeron D, and that one is still considerably faster than an old P3...
    How I need a drink, alcoholic in nature, after the heavy lectures involving quantum mechanics.

  6. #21
    (?<!re)tired Mario F.'s Avatar
    Join Date
    May 2006
    Location
    Ireland
    Posts
    8,446
    I don't know, bleh. I saw Asrock on the chipset on a picture in the box. I'm ridiculous, I know...
    Originally Posted by brewbuck:
    Reimplementing a large system in another language to get a 25% performance boost is nonsense. It would be cheaper to just get a computer which is 25% faster.

  7. #22
    (?<!re)tired Mario F.'s Avatar
    Join Date
    May 2006
    Location
    Ireland
    Posts
    8,446
    I think it was this one or something very similar. The image is almost a look-a-like.

    http://www.asrock.com/mb/overview.as...33-2.66G&s=775
    Originally Posted by brewbuck:
    Reimplementing a large system in another language to get a 25% performance boost is nonsense. It would be cheaper to just get a computer which is 25% faster.

  8. #23
    Internet Superhero
    Join Date
    Sep 2006
    Location
    Denmark
    Posts
    964
    Quote Originally Posted by Mario F. View Post
    I think it was this one or something very similar. The image is almost a look-a-like.

    http://www.asrock.com/mb/overview.as...33-2.66G&s=775
    RTFM then
    How I need a drink, alcoholic in nature, after the heavy lectures involving quantum mechanics.

  9. #24
    (?<!re)tired Mario F.'s Avatar
    Join Date
    May 2006
    Location
    Ireland
    Posts
    8,446
    I know I'm behaving like a kid, but these things have changed so much. I can't recognize half the new acronyms, abbreviations and terms...

    Anyways, is it what they call "Ready-to-Use" USB ports?

    EDIT: Ok. Nevermind. I think I got it. The manual speaks of user togable legacy USB support, which is I think what I need in order to have it available prior to the OS installation.
    Last edited by Mario F.; 06-02-2008 at 10:21 AM.
    Originally Posted by brewbuck:
    Reimplementing a large system in another language to get a 25% performance boost is nonsense. It would be cheaper to just get a computer which is 25% faster.

  10. #25
    Chinese pâté foxman's Avatar
    Join Date
    Jul 2007
    Location
    Canada
    Posts
    404
    I might be a little late but the E4xxx series is getting a bit old and is being replaced by the E7xxx series. The E7200, which has the same release price of a E4500, has a higher frequency, SSE4.1 support, more L2 cache, higher front side bus and may be a bit more energy efficient since it's a 45nm chips vs 65nm. The performance difference isn't huge between those two, but the E7200 is leading. Of course, this kind of information is more interesting if you buy your parts by yourself than if you buy a pre-assembled system.

    As for the SuperPi calculation time (1M digits) between a C2D vs a P3, here's some results on my own computers:

    • C2D (3.0 GHz, 6MB L2): 15 sec
    • P3 (1.0 GHz, 256KB L2): 2 min 25 sec

    The first system is 1 month-old and the second is... near to 9 years old.
    I hate real numbers.

  11. #26
    (?<!re)tired Mario F.'s Avatar
    Join Date
    May 2006
    Location
    Ireland
    Posts
    8,446
    Quote Originally Posted by foxman View Post
    C2D (3.0 GHz, 6MB L2): 15 sec
    P3 (1.0 GHz, 256KB L2): 2 min 25 sec
    10x faster. Quiet accurate the predictions in the past posts.

    Meanwhile yes, I could see E7xxx's. But precisely as you said it, being pre-assembled I couldn't get quiet the optimal configuration. If I remember correctly these machines where being sold as part of more featured-rich systems, consequently more expensive. And that just doesn't fit with the usage I want to give to this desktop.
    Originally Posted by brewbuck:
    Reimplementing a large system in another language to get a 25% performance boost is nonsense. It would be cheaper to just get a computer which is 25% faster.

  12. #27
    (?<!re)tired Mario F.'s Avatar
    Join Date
    May 2006
    Location
    Ireland
    Posts
    8,446
    Ok, I found all I need. To wrap it up (and thanks a bunch to everyone who posted. You people are awesome. Even though we may disagree on other issues, I love this little corner of the web and couldn't make it without you)

    - All AsRock boards support boot from USB device, which basically guarantees I can load the SATA drivers from a USB stick during XP installation.

    - More important though... AsRock boards have this:

    Attachment 8199

    Which, if it means what it seems to me, I will not even need to worry about the drivers.
    Originally Posted by brewbuck:
    Reimplementing a large system in another language to get a 25% performance boost is nonsense. It would be cheaper to just get a computer which is 25% faster.

  13. #28
    Chinese pâté foxman's Avatar
    Join Date
    Jul 2007
    Location
    Canada
    Posts
    404
    If some people are still interested, here's one article I remember reading, where they compare some of the C2D processors from different series.

    http://www.xbitlabs.com/articles/cpu...duo-e7200.html
    I hate real numbers.

  14. #29
    (?<!re)tired Mario F.'s Avatar
    Join Date
    May 2006
    Location
    Ireland
    Posts
    8,446
    I am thanks. I'm suddenly gaining a renewed interest in these matters.
    Originally Posted by brewbuck:
    Reimplementing a large system in another language to get a 25% performance boost is nonsense. It would be cheaper to just get a computer which is 25% faster.

Popular pages Recent additions subscribe to a feed

Similar Threads

  1. Replies: 21
    Last Post: 06-24-2009, 09:49 AM
  2. Hardware interrupts
    By Sentral in forum Tech Board
    Replies: 5
    Last Post: 02-23-2009, 06:46 PM
  3. N Choose R
    By .ZG. in forum C Programming
    Replies: 7
    Last Post: 05-18-2004, 02:43 AM
  4. How does hardware interpret data?
    By Silvercord in forum Tech Board
    Replies: 3
    Last Post: 01-29-2003, 01:46 PM
  5. Linux Hardware Handbook
    By Nutshell in forum A Brief History of Cprogramming.com
    Replies: 0
    Last Post: 04-02-2002, 06:06 AM

Tags for this Thread