Wow, Price vs Perf right now in PC's is nice...

Discussion in 'PC Hardware, Software and Displays' started by Albuquerque, Dec 4, 2007.

  1. Putas

    Regular Newcomer

    Joined:
    Nov 7, 2004
    Messages:
    397
    Likes Received:
    62
    Your fantasy went too far. I just wanted to show that statement like:
    A pair of DDR-800 modules accomplishes exactly jack squat on a system that doesn't operate at 1600FSB..
    is a lie as all benchmarks shown.
     
  2. Albuquerque

    Albuquerque Red-headed step child
    Veteran

    Joined:
    Jun 17, 2004
    Messages:
    3,845
    Likes Received:
    329
    Location:
    35.1415,-90.056
    Really? Care to show me those benchmarks again? Because 667->800 is a 20% increase, yet all the benchmarks you provided failed to show so much as a 2% gain in all "real" applications tested.

    And since that's well within the margin of error, I say bullocks.
     
  3. I.S.T.

    Veteran

    Joined:
    Feb 21, 2004
    Messages:
    3,174
    Likes Received:
    389
    *Is being devil's advocate*

    He might be referring to the better timings and whatnot you can put on a DDR2-800 stick running at 667 MHZ compared to a normal DDR2-667 stick.
     
  4. Putas

    Regular Newcomer

    Joined:
    Nov 7, 2004
    Messages:
    397
    Likes Received:
    62
    You are still missing/ignoring my whole and only point- faster ram gives better performance. Even if you theoretically breach fsb cap, which should hold it in Albuquerque's opinion.
    If you want to defend this with margin of error differences although you get that results hundred times in a row and the trend does not stop at way higher clocks, maybe I can take this winrar test as a proof?
    http://www.xbitlabs.com/articles/memory/display/core2duo-memory-guide_7.html
     
  5. Skrying

    Skrying S K R Y I N G
    Veteran

    Joined:
    Jul 8, 2005
    Messages:
    4,815
    Likes Received:
    61
    You realize how bad that article makes your argument look though, Putas? One application on that page shows a improvement that is even worth looking at! How often are you unraring files that would take that much time? Hardly ever I would guess. Therefore the advantages are INCREDIBLY small... as PROVED by your own link that is suppose to "prove" the completely opposite...
     
  6. Putas

    Regular Newcomer

    Joined:
    Nov 7, 2004
    Messages:
    397
    Likes Received:
    62
    You totally failed to comprehend what I am trying to prove although I explicitly wrote it in simple post you are replying to. Now how bad are you looking?
     
  7. Albuquerque

    Albuquerque Red-headed step child
    Veteran

    Joined:
    Jun 17, 2004
    Messages:
    3,845
    Likes Received:
    329
    Location:
    35.1415,-90.056
    Now you're just trolling.

    Funny how a thread about cost versus performance could result in someone trying to prove such abusdity as this. And how such a thread could warrant the only two times in this forum that I've had to hit the "Ignore this user" button.

    Your special brand of "fantasy" will not be missed.
     
  8. Putas

    Regular Newcomer

    Joined:
    Nov 7, 2004
    Messages:
    397
    Likes Received:
    62
    I don't know what abusdity is but whatever. If you have no more arguments (or trolling as you call it) I hope we can finally close it with result: don't take fsb bandwith into account when buying memory.
     
  9. Davros

    Legend

    Joined:
    Jun 7, 2004
    Messages:
    15,026
    Likes Received:
    2,373
    Hey Albuquerque the other time wasnt me was it :D
     
  10. Albuquerque

    Albuquerque Red-headed step child
    Veteran

    Joined:
    Jun 17, 2004
    Messages:
    3,845
    Likes Received:
    329
    Location:
    35.1415,-90.056
    LOL, no. :)

    It still surprises me how even basic math can't be acknowledged by someone who's too fanatical to pay attention. If you have a 5cm pipe that flows water between two buckets, then no matter HOW fast you pump water into Bucket A, it will not go any faster into Bucket B.

    Ultimately it's their money to waste on their own systems for some "less than rounding error" amount of difference, and more than rounding error amount of power consumption and heat. But trying to convince me to tell my boss to spend another $10 on something he will never use, will never need, and will never notice the difference is delusional at best.

    I guess there are people who put Z-rated racing compound tires on Kias too, because they go faster! :D
     
  11. Skrying

    Skrying S K R Y I N G
    Veteran

    Joined:
    Jul 8, 2005
    Messages:
    4,815
    Likes Received:
    61
    Funny... You've proved your point faster RAM = faster... by next to nothing. Again, you've failed to realize that 1 or 2 seconds in the vast majority of instances doesn't matter at all. The WinRAR benchmark is incredibly pointless when all other instances show besides that one that the higher clocked memory does next to nothing for you. You're failing to see the point that when the term faster is used it means something that is noticeable, tangible, not 1 or 2 second difference in 900 or seconds! And to FINALLY shut you up here is a wonderful quote from the very article you posted.

     
  12. Davros

    Legend

    Joined:
    Jun 7, 2004
    Messages:
    15,026
    Likes Received:
    2,373
    hang on though using ddr2-800 instead of ddr2-533 that my cpu requires allows me to o/c to 3.6 without o/c-ing the ram so there is a legitimate reason to buy faster ram
     
  13. Skrying

    Skrying S K R Y I N G
    Veteran

    Joined:
    Jul 8, 2005
    Messages:
    4,815
    Likes Received:
    61
    Which is not at all what he was arguing...
     
  14. BRiT

    BRiT (╯°□°)╯
    Moderator Legend Alpha

    Joined:
    Feb 7, 2002
    Messages:
    12,792
    Likes Received:
    9,122
    Location:
    Cleveland
    Wow, this discussion is still raging on over the same minor tangental point?

    This almost reminds me of some Madden-isms -- the more teams score the more points they put on the board ... or some other silliness.
     
  15. Albuquerque

    Albuquerque Red-headed step child
    Veteran

    Joined:
    Jun 17, 2004
    Messages:
    3,845
    Likes Received:
    329
    Location:
    35.1415,-90.056
    I'll never disagree with that, and in fact if you go back a page or two, you'll see me mention the exact same thing. In that same paragraph, you'd also see me saying that I too would purchase something faster (DDR2-1066 probably) because it would give me the overclocking headroom. Basically, buying ram that's "too fast" because you simply don't know where you'll end up is a perfectly valid reason to do so.

    But running the memory faster than 1:1 is, for lack of any other really pertinent term, useless. Just as that website points out -- a 150% increase in memory speed combined with a 50% increase in FSB netted a 54% increase in performance.

    It doesn't take a rocket scientist to figure out where the bottleneck truly lies...
     
  16. Davros

    Legend

    Joined:
    Jun 7, 2004
    Messages:
    15,026
    Likes Received:
    2,373
    Thankfully I am a rocket scientist ;)
     
  17. Putas

    Regular Newcomer

    Joined:
    Nov 7, 2004
    Messages:
    397
    Likes Received:
    62
    Still surprise me you don't acknowledge there are more pipes and other equipment then that. If it was like how you imagine you could not get any increase at all.

    I did not fail, since I have the same opinion from very beginning. The argue is about why it is so small difference.
    For the xbit labs conclusion- I say they are wrong. Why would be the winrar test fall out of the rule?

    But you don't think it is superlinear scaling, right? And hey, that was just 266 MHz FSB. Too bad I cannot find any memory benchmarks with smaller cache then 4 MB.
     
  18. Arun

    Arun Unknown.
    Moderator Legend Veteran

    Joined:
    Aug 28, 2002
    Messages:
    5,023
    Likes Received:
    299
    Location:
    UK
    Modern northbridges prefetch aggressively instead of idling and that easily explains the small (or very small) performance benefits of faster memory. Is there really anything else to say here?
     
  19. Davros

    Legend

    Joined:
    Jun 7, 2004
    Messages:
    15,026
    Likes Received:
    2,373
    Actually i was thinking why has no one made a board with quad channel memory instead of dual ?
     
  20. 3dilettante

    Legend Alpha

    Joined:
    Sep 15, 2003
    Messages:
    8,137
    Likes Received:
    2,939
    Location:
    Well within 3d
    Doubling the number of channels would need a lot of motherboard traces.
     
Loading...

Share This Page

  • About Us

    Beyond3D has been around for over a decade and prides itself on being the best place on the web for in-depth, technically-driven discussion and analysis of 3D graphics hardware. If you love pixels and transistors, you've come to the right place!

    Beyond3D is proudly published by GPU Tools Ltd.
Loading...