Results 1 to 15 of 15

Thread: Will 6 Cores help Multi-GPU Bottle Necking?

  1. #1
    Xtreme Member
    Join Date
    Mar 2005
    Location
    Pennsylvania
    Posts
    356

    Will 6 Cores help Multi-GPU Bottle Necking?

    Any thoughts on this? Right now there are not many convincing arguments to go with 6 cores due to the lack of software threading. However, with that said, as far as gamers are concerned, a huge benefit could be found in the extra cores if it pegs your GPU utilization to 100% across the board.

    Any info or thoughts?

    With this in mind, right now I get pretty close to 99% utilization with my quad at 3.8GHZ juicing my 3 4850s. However I suspect that if I a added a forth card that would drop. What are other people seeing? I use Riva Tuner Stat Server to get my utilization percentages.
    Daily Desktop Custom Built - Modified Data General Server Case

    Asus M4A79T Deluxe | 955BE C2 Rev | 4GB OCZ DDR 2000 @ 1600 6 5 6 1T | 4850x2 + 4850 w/MCW60 | CPU on H20 w/ Enzo Sapphire l Custom Built Pump and Res Combo 1/2 Gallon! | 4 Swiftech 320mm Rads with 6 All Aluminum Delta Fans with their own PSU - Finger Loppers - for real!

    Literally, CPU&GPUs are almost always at ambient therefore heat ain't limiting sh|t.

  2. #2
    Registered User
    Join Date
    Feb 2010
    Posts
    40
    Quote Originally Posted by chris.y2k.r1 View Post
    Any thoughts on this? Right now there are not many convincing arguments to go with 6 cores due to the lack of software threading. However, with that said, as far as gamers are concerned, a huge benefit could be found in the extra cores if it pegs your GPU utilization to 100% across the board.

    Any info or thoughts?
    Games are becoming more threaded. As shown in this review Dirt2, Crysis and Need for Speed all show significant gains at 6 versus 4 cores.

    Meanwhile, BFBC2 will generally bottleneck the GPU at any setting.
    Last edited by AMD-C; 04-30-2010 at 05:04 AM.

  3. #3
    Xtreme Addict
    Join Date
    Jan 2009
    Location
    SF
    Posts
    1,070
    Damn those are some nice gains with 2 extra cores, in those games.

  4. #4
    Xtreme Addict
    Join Date
    Aug 2007
    Location
    Toon
    Posts
    1,570
    My testing says no.

    4GHz x6 is out performed by i7 870 @ 3.5GHz when running a single 5970.



    With a pair of 5970s the situation gets worse.

    AMD x6 @ 4GHz:



    Scaling from 5970 to 2x5970 ~ +50%

    Intel x6 @ 3.4GHz, 12GB @ 1800MHz 8-8-8-24, 2x5970 (not OC'd):

    Last edited by initialised; 04-30-2010 at 04:41 AM.
    Intel i7 920 C0 @ 3.67GHz
    ASUS 6T Deluxe
    Powercolor 7970 @ 1050/1475
    12GB GSkill Ripjaws
    Antec 850W TruePower Quattro
    50" Full HD PDP
    Red Cosmos 1000

  5. #5
    Xtreme Cruncher
    Join Date
    Jun 2006
    Posts
    6,215
    Any chance you can test with actual games people actually play,instead of futuremark's "test" ?

  6. #6
    Xtreme Addict
    Join Date
    Jan 2009
    Location
    SF
    Posts
    1,070
    ^^^^

    I'm looking at the Test Scores at the right side. The Thuban gets higher FPS except the 1st two GPU tests. And the 860 still gets a higher score???

  7. #7
    Xtreme Addict
    Join Date
    Dec 2002
    Posts
    1,467
    Quote Originally Posted by initialised View Post
    My testing says no.

    4GHz x6 is out performed by i7 870 @ 3.5GHz when running a single 5970.



    With a pair of 5970s the situation gets worse.

    AMD x6 @ 4GHz:



    Scaling from 5970 to 2x5970 ~ +50%

    Intel x6 @ 3.4GHz, 12GB @ 1800MHz 8-8-8-24, 2x5970 (not OC'd):


    whats your gpu score in performance with one 5970? you have Xtreme running up there then show intel at performance. I get 22,350 stock 5970 and 26,600+ with it overclocked to 970/1200 vantage compare is in my sig
    Last edited by Titan7171; 04-30-2010 at 05:38 AM.
    AMD 1090T@4.0ghz
    Enzotech sapphire/Mo-Ra extreme rad
    Asus Crosshair IV Formula
    ht 2400mhz / nb 2400mhz
    12gb Gskill 1300mhz
    HIS HD5970
    Enermax Evo Galaxy 1250
    case: XCLIO A380PLUS-BK

    4.61ghz water

    4.5ghz superpi 1M 15.585
    http://i271.photobucket.com/albums/j...rpi4500mhz.jpg
    25,396 06 Phenom 965@ 4.4ghz HIS 5970@960/1260
    21,893 Vantage, Phenom 965 4.2ghz HIS 5970 @960/1260

    Phenom 2 125w 965 test results
    http://futuremark.yougamers.com/foru...d.php?t=117414
    Phenom 2 140w 965 test results
    http://futuremark.yougamers.com/foru...d.php?t=109214
    Phenom 2 AM2+ 940 cold air results
    http://futuremark.yougamers.com/foru...ad.php?t=97430

    If I dont get every single drop out of my cpu I feel like someone is stealing from me

  8. #8
    Xtreme Enthusiast
    Join Date
    Jun 2008
    Location
    Northern Ohio
    Posts
    664
    BFBC2 alone could make a case by itself in a SINGLE gpu setup, that game is a monster threaded game, I really hope it's engine gets used more and new engines are developed like it is. Keeps my Phenom II X4 @ 3.6Ghz in the 70%'s on all cores almost all of the time, with spikes towards 100% on some cores. I plan to eventually get another 5850 and would assume my quad would start crying at that point.


    Work/Game System - ~24/7 WCG
    ASUS P8P67 PRO / i7 2600k @ 4.1Ghz / Gigabyte Radeon HD5870 / 4x4GB Corsair Vengeance @ 1600Mhz 9-9-9

    HTPC -~24/7 WCG
    Gigabyte GA-Z68AP-D3 / i7 2600k @ 4.0Ghz / Sapphire Radeon HD5830 / 2x2GB Mushkin Enhanced Essentials @ 1333Mhz 9-9-9

    XS WCG Team Forum - http://www.worldcommunitygrid.org/

  9. #9
    Xtreme X.I.P. Particle's Avatar
    Join Date
    Apr 2008
    Location
    Kansas
    Posts
    3,219
    Quote Originally Posted by Tom128 View Post
    BFBC2 alone could make a case by itself in a SINGLE gpu setup, that game is a monster threaded game, I really hope it's engine gets used more and new engines are developed like it is. Keeps my Phenom II X4 @ 3.6Ghz in the 70%'s on all cores almost all of the time, with spikes towards 100% on some cores. I plan to eventually get another 5850 and would assume my quad would start crying at that point.
    I've observed about 50-60% loading across all cores on the 4.0 GHz 1055T system I play with. That's with two 5850s and playing in Eyefinity. The HUD shows that both GPUs load to 97% or higher in such a setup. Kind of interesting.
    Particle's First Rule of Online Technical Discussion:
    As a thread about any computer related subject has its length approach infinity, the likelihood and inevitability of a poorly constructed AMD vs. Intel fight also exponentially increases.

    Rule 1A:
    Likewise, the frequency of a car pseudoanalogy to explain a technical concept increases with thread length. This will make many people chuckle, as computer people are rarely knowledgeable about vehicular mechanics.

    Rule 2:
    When confronted with a post that is contrary to what a poster likes, believes, or most often wants to be correct, the poster will pick out only minor details that are largely irrelevant in an attempt to shut out the conflicting idea. The core of the post will be left alone since it isn't easy to contradict what the person is actually saying.

    Rule 2A:
    When a poster cannot properly refute a post they do not like (as described above), the poster will most likely invent fictitious counter-points and/or begin to attack the other's credibility in feeble ways that are dramatic but irrelevant. Do not underestimate this tactic, as in the online world this will sway many observers. Do not forget: Correctness is decided only by what is said last, the most loudly, or with greatest repetition.

    Rule 3:
    When it comes to computer news, 70% of Internet rumors are outright fabricated, 20% are inaccurate enough to simply be discarded, and about 10% are based in reality. Grains of salt--become familiar with them.

    Remember: When debating online, everyone else is ALWAYS wrong if they do not agree with you!

    Random Tip o' the Whatever
    You just can't win. If your product offers feature A instead of B, people will moan how A is stupid and it didn't offer B. If your product offers B instead of A, they'll likewise complain and rant about how anyone's retarded cousin could figure out A is what the market wants.

  10. #10
    NooB MOD
    Join Date
    Jan 2006
    Location
    South Africa
    Posts
    5,799
    It wasn't even five years ago we were saying dual cores aren't needed
    Xtreme SUPERCOMPUTER
    Nov 1 - Nov 8 Join Now!


    Quote Originally Posted by Jowy Atreides View Post
    Intel is about to get athlon'd
    Athlon64 3700+ KACAE 0605APAW @ 3455MHz 314x11 1.92v/Vapochill || Core 2 Duo E8500 Q807 @ 6060MHz 638x9.5 1.95v LN2 @ -120'c || Athlon64 FX-55 CABCE 0516WPMW @ 3916MHz 261x15 1.802v/LN2 @ -40c || DFI LP UT CFX3200-DR || DFI LP UT NF4 SLI-DR || DFI LP UT NF4 Ultra D || Sapphire X1950XT || 2x256MB Kingston HyperX BH-5 @ 290MHz 2-2-2-5 3.94v || 2x256MB G.Skill TCCD @ 350MHz 3-4-4-8 3.1v || 2x256MB Kingston HyperX BH-5 @ 294MHz 2-2-2-5 3.94v

  11. #11
    Xtreme X.I.P. Particle's Avatar
    Join Date
    Apr 2008
    Location
    Kansas
    Posts
    3,219
    People will eventually come to understand that multi-core is very useful. It takes time for software to catch up to hardware, but it does happen.
    Particle's First Rule of Online Technical Discussion:
    As a thread about any computer related subject has its length approach infinity, the likelihood and inevitability of a poorly constructed AMD vs. Intel fight also exponentially increases.

    Rule 1A:
    Likewise, the frequency of a car pseudoanalogy to explain a technical concept increases with thread length. This will make many people chuckle, as computer people are rarely knowledgeable about vehicular mechanics.

    Rule 2:
    When confronted with a post that is contrary to what a poster likes, believes, or most often wants to be correct, the poster will pick out only minor details that are largely irrelevant in an attempt to shut out the conflicting idea. The core of the post will be left alone since it isn't easy to contradict what the person is actually saying.

    Rule 2A:
    When a poster cannot properly refute a post they do not like (as described above), the poster will most likely invent fictitious counter-points and/or begin to attack the other's credibility in feeble ways that are dramatic but irrelevant. Do not underestimate this tactic, as in the online world this will sway many observers. Do not forget: Correctness is decided only by what is said last, the most loudly, or with greatest repetition.

    Rule 3:
    When it comes to computer news, 70% of Internet rumors are outright fabricated, 20% are inaccurate enough to simply be discarded, and about 10% are based in reality. Grains of salt--become familiar with them.

    Remember: When debating online, everyone else is ALWAYS wrong if they do not agree with you!

    Random Tip o' the Whatever
    You just can't win. If your product offers feature A instead of B, people will moan how A is stupid and it didn't offer B. If your product offers B instead of A, they'll likewise complain and rant about how anyone's retarded cousin could figure out A is what the market wants.

  12. #12
    Registered User
    Join Date
    Feb 2010
    Posts
    40
    Quote Originally Posted by informal View Post
    Any chance you can test with actual games people actually play,instead of futuremark's "test" ?
    3DMark is very broken. I'm kind of amazed anyone puts much stake in it, much less pays for it.

    Quote Originally Posted by Oj101 View Post
    It wasn't even five years ago we were saying dual cores aren't needed
    The argument that you should buy for what's available right now or even yesterday is a popular one even though it seems like most people keep their setups for 3 years or more.
    Last edited by AMD-C; 04-30-2010 at 08:04 AM.

  13. #13
    Xtreme Member
    Join Date
    Jan 2010
    Posts
    323
    Quote Originally Posted by Oj101 View Post
    It wasn't even five years ago we were saying dual cores aren't needed
    There's still ignorant people out there saying it.

  14. #14
    Registered User
    Join Date
    Aug 2005
    Posts
    7
    Quote Originally Posted by Oj101 View Post
    It wasn't even five years ago we were saying dual cores aren't needed
    A milestone to be sure. Something tells me that these processors have very flat IHS finish and will not benefit greatly from lapping.

    Won't stop me though.

  15. #15
    NooB MOD
    Join Date
    Jan 2006
    Location
    South Africa
    Posts
    5,799
    Lap? I've never been bothered, removing the IHS has greater gains and is quicker and easier. All I do is steal the rubber pads from a dead AthlonXP, stick them around the core and slap on the cooler
    Xtreme SUPERCOMPUTER
    Nov 1 - Nov 8 Join Now!


    Quote Originally Posted by Jowy Atreides View Post
    Intel is about to get athlon'd
    Athlon64 3700+ KACAE 0605APAW @ 3455MHz 314x11 1.92v/Vapochill || Core 2 Duo E8500 Q807 @ 6060MHz 638x9.5 1.95v LN2 @ -120'c || Athlon64 FX-55 CABCE 0516WPMW @ 3916MHz 261x15 1.802v/LN2 @ -40c || DFI LP UT CFX3200-DR || DFI LP UT NF4 SLI-DR || DFI LP UT NF4 Ultra D || Sapphire X1950XT || 2x256MB Kingston HyperX BH-5 @ 290MHz 2-2-2-5 3.94v || 2x256MB G.Skill TCCD @ 350MHz 3-4-4-8 3.1v || 2x256MB Kingston HyperX BH-5 @ 294MHz 2-2-2-5 3.94v

Bookmarks

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •