MMM
Page 3 of 4 FirstFirst 1234 LastLast
Results 51 to 75 of 76

Thread: GTX 280 voltmod thread

  1. #51
    Registered User
    Join Date
    Jan 2008
    Posts
    51
    Are there other places to solder for vgpu (not measure)? Thank's!

  2. #52
    Registered User
    Join Date
    Dec 2007
    Posts
    19
    anyone going with the GTX260 ?

  3. #53
    Xtreme Addict
    Join Date
    Mar 2006
    Location
    Sydney , Australia
    Posts
    1,600
    Well this is a bummer , I have a 4870x2 and was going to dump it for a vmodded 280 but the results are pretty poor , what a shame . My last vmod effort was a G92 gts that clocked to 910mhz on water , some dice/ln would have been good on that one.

    Bencher/Gamer(1) 4930K - Asus R4E - 2x R9 290x - G.skill Pi 2200c7 or Team 2400LV 4x4GB - EK Supreme HF - SR1-420 - Qnix 2560x1440
    Netbox AMD 5600K - Gigabyte mitx - Aten DVI/USB/120Hz KVM
    PB 1xTitan=16453(3D11), 1xGTX680=13343(3D11), 1x GTX580=8733(3D11)38000(3D06) 1x7970=12059(3D11)40000(vantage)395k(AM3) Folding for team 24

    AUSTRALIAN DRAG RACING http://www.youtube.com/watch?v=OFsbfEIy3Yw

  4. #54
    Registered User
    Join Date
    Jun 2008
    Posts
    86
    Did anybody of you a mod @ GTX280 and how are the results?

  5. #55
    Registered User
    Join Date
    Dec 2003
    Posts
    94
    I've done my two and the shader clock seems very unresponsive to voltage and is the main performance driver. I only went up to about 10% over volt tho, or about 1.3V. One of my cards could hit 1566 shader no problem, the other is stuck at 1458. The very next tick introduces instant artifacts, that can't be "dialed out" with more voltage as was possible on older cards.

    Also, I noticed if you pushed the shaders really hard, the failure mechanism was very strange. Instead of just dropping vertices or pixels it would sometimes loose portions of the frame buffer altogether. When I say lost, I literally mean it's pointed to the wrong address in local memory. You could get it to show a flat 2D source texture as the entire screen, like if you browsed through the games assets. It was very interesting, particularly because the game (crysis/bioshock/shader heavy game) wouldn't be locked up. You could still escape out and get back to the desktop.

    Based on that I think the 512bit architecture complexity is pushing the setup and hold time requirements for the bus to it's absolute limit. Particularly in keeping the shaders fed properly at increased clock rates. This would also explain how they are much more responsive to temperature than voltage and why 260s clock further, and looks to me like the typical limit to this current part. It certainly isn't starved for DC power.

    Going to a smaller process will certainly help, as would reducing the bus width to relax the timing requirements. The only benefit Vcore gives you is to match the core clock to half the shader clock without issue, but the cards performance is largely driven by the shader clock.
    Last edited by 101; 09-02-2008 at 11:11 AM.

  6. #56
    Registered User
    Join Date
    Jun 2008
    Posts
    86
    Quote Originally Posted by 101 View Post
    I've done my two and the shader clock seems very unresponsive to voltage and is the main performance driver. I only went up to about 10% over volt tho, or about 1.3V. One of my cards could hit 1566 shader no problem, the other is stuck at 1458. The very next tick introduces instant artifacts, that can't be "dialed out" with more voltage as was possible on older cards.

    Also, I noticed if you pushed the shaders really hard, the failure mechanism was very strange. Instead of just dropping vertices or pixels it would sometimes loose portions of the frame buffer altogether. When I say lost, I literally mean it's pointed to the wrong address in local memory. You could get it to show a flat 2D source texture as the entire screen, like if you browsed through the games assets. It was very interesting, particularly because the game (crysis/bioshock/shader heavy game) wouldn't be locked up. You could still escape out and get back to the desktop.

    Based on that I think the 512bit architecture complexity is pushing the setup and hold time requirements for the bus to it's absolute limit. Particularly in keeping the shaders fed properly at increased clock rates. This would also explain how they are much more responsive to temperature than voltage and why 260s clock further, and looks to me like the typical limit to this current part. It certainly isn't starved for DC power.

    Going to a smaller process will certainly help, as would reducing the bus width to relax the timing requirements. The only benefit Vcore gives you is to match the core clock to half the shader clock without issue, but the cards performance is largely driven by the shader clock.
    Yes, Shader goes not realy higher with more voltage.
    Without Mod 1620, with Mod 1620 too, but i test 1674 when i find any time
    Core is stable at 810MHz, more i can´t test, because you need more than 1620 Shader.

    Here somethnig in german about my Mod.

    ==================================================

    vGPU Mod GTX260/280

    So, nachdem es fast nix mehr zu testen gibt, musste ich mich natürlich an den vMod der GTX280 setzen.

    Ganz vorn weg, so einfach wie bei der GTX260/280 war ein vMod wohl noch nie bei einer Grafikkarte. Die Lötpunkte sind alle super zu erreichen und und es liegen auch keine benachbarten Teile in direkter Nähe, die man evlt. beschädigen könnte.

    Bei der Karte sind von Werk aus im Bios 1,18v GPU Spannung eingestellt, was bei meiner Karte reale 1,196v ergeben wenn man an den Messpunkten nachmisst. Diese kann man durch einlöten eines 100Ohm Potis langsam und auch sehr genau erhöhen.

    Aber bevor ich hier nun einen Roman schreibe, poste ich einfach ein paar Bilder, die die nötigen Infos liefern. Bilder sagen bekanntlich mehr als 1000 Worte.

    Trotzdem könnt ihr natürlich gern Fragen stellen, wenn es noch offene Punkte gibt oder ihr selbst den Mod anwenden wollt.

    Rückseite, vMod


    Vorderseite, Messpunkte


    Ergebnis

    (klick to enlarge)

    ==================================================


    Here the full review in german about the GTX280 ...
    http://www.forumdeluxx.de/forum/showthread.php?t=504260

  7. #57
    Xtreme Enthusiast
    Join Date
    Aug 2005
    Location
    Melbourne, Australia
    Posts
    942
    It looks like the heatspreader is easy to remove too... i can see the gap between the IHS and the substrate.

    http://www.nordichardware.com/news,8065.html

    The die looks just large enough to be covered by a water block but if i take it off i'm condemned to water cooling it forever
    Q9550 || DFI P45 Jr || 4x 2G generic ram || 4870X2 || Aerocool M40 case || 3TB storage


  8. #58
    Registered User
    Join Date
    Jun 2008
    Posts
    86
    Quote Originally Posted by oohms View Post
    It looks like the heatspreader is easy to remove too... i can see the gap between the IHS and the substrate.

    http://www.nordichardware.com/news,8065.html

    The die looks just large enough to be covered by a water block but if i take it off i'm condemned to water cooling it forever
    Yes, you can cut 5,5mm, ...

    Pix say more than 1000 words

    http://www.partypicssaarland.de/incl...ic.php&pid=541

    Last edited by OliverM84; 09-03-2008 at 01:28 AM.

  9. #59
    Registered User
    Join Date
    Jun 2008
    Posts
    86



  10. #60
    Xtreme Enthusiast
    Join Date
    Aug 2005
    Location
    Melbourne, Australia
    Posts
    942


    Is that with watercooling + removed IHS?

    If so, what block/wc gear do you use?
    Q9550 || DFI P45 Jr || 4x 2G generic ram || 4870X2 || Aerocool M40 case || 3TB storage


  11. #61
    Registered User
    Join Date
    Jun 2008
    Posts
    86
    Quote Originally Posted by oohms View Post


    Is that with watercooling + removed IHS?

    If so, what block/wc gear do you use?
    Hi,

    no, the IHS is not removed.
    Only 1,28v on GPU and the watercool Heatkiller on de card.

    1674 shader doesn´t run, so i can´t go higher than 810 core.
    highest possible coreclock with 1620 shader is 810mhz.

  12. #62
    Registered User
    Join Date
    Dec 2003
    Posts
    94
    Quote Originally Posted by OliverM84 View Post
    Hi,

    no, the IHS is not removed.
    Only 1,28v on GPU and the watercool Heatkiller on de card.

    1674 shader doesn´t run, so i can´t go higher than 810 core.
    highest possible coreclock with 1620 shader is 810mhz.
    The question is how much did the voltage increase improve your shader clocks, as the shader is the limiting factor for the part.

  13. #63
    Xtreme Enthusiast
    Join Date
    Jul 2007
    Posts
    668
    Why don't you run unlinked?
    Proud owner of an iPhone3G 16G White




    SpiTweaker by Monteboy,try it

  14. #64
    Registered User
    Join Date
    Jun 2008
    Posts
    86
    Quote Originally Posted by 101 View Post
    The question is how much did the voltage increase improve your shader clocks, as the shader is the limiting factor for the part.
    The voltage increase doesn´t improve your shader clocks. 1620 run without Mod too.

  15. #65
    Registered User
    Join Date
    Jun 2008
    Posts
    86
    Quote Originally Posted by DMH View Post
    Why don't you run unlinked?
    I run unlinkted, but both clocks have to be not to far distanced and not to close together. So 810 is the max. coreclock with 1620 shader.

  16. #66
    Registered User
    Join Date
    Dec 2003
    Posts
    94
    Quote Originally Posted by OliverM84 View Post
    The voltage increase doesn´t improve your shader clocks. 1620 run without Mod too.
    It also looks like you started with a highly OCed card, like an EVGA FTW. A good purchase as it seems the G200 are binned very heavily and have little OCing margin. Additionally, we have no way of reasonably increasing the shader performance without extreme cooling, core voltage mods do nothing. Maybe if we had some kind of mod for the memory IO interface of the part, similar to CPUs, it would help some.
    Last edited by 101; 09-08-2008 at 11:34 AM.

  17. #67
    Registered User
    Join Date
    Jun 2008
    Posts
    86
    Quote Originally Posted by 101 View Post
    It also looks like you started with a highly OCed card, like an EVGA FTW. A good purchase as it seems the G200 are binned very heavily and have little OCing margin. Additionally, we have no way of reasonably increasing the shader performance without extreme cooling, core voltage mods do nothing. Maybe if we had some kind of mod for the memory IO interface of the part, similar to CPUs, it would help some.
    No, its a POV with default 602/1296/1107 ...

  18. #68
    Registered User
    Join Date
    Dec 2003
    Posts
    94
    Ah, then you must have flashed the bios as it shows default clocks of 702/1458/1215

  19. #69
    Registered User
    Join Date
    Jun 2008
    Posts
    86
    Quote Originally Posted by 101 View Post
    Ah, then you must have flashed the bios as it shows default clocks of 702/1458/1215

  20. #70
    Xtreme Addict
    Join Date
    Mar 2006
    Location
    Sydney , Australia
    Posts
    1,600
    Has anyone else had good results with voltmodding a 280 ? Or not worth it ?

    Bencher/Gamer(1) 4930K - Asus R4E - 2x R9 290x - G.skill Pi 2200c7 or Team 2400LV 4x4GB - EK Supreme HF - SR1-420 - Qnix 2560x1440
    Netbox AMD 5600K - Gigabyte mitx - Aten DVI/USB/120Hz KVM
    PB 1xTitan=16453(3D11), 1xGTX680=13343(3D11), 1x GTX580=8733(3D11)38000(3D06) 1x7970=12059(3D11)40000(vantage)395k(AM3) Folding for team 24

    AUSTRALIAN DRAG RACING http://www.youtube.com/watch?v=OFsbfEIy3Yw

  21. #71
    Xtreme Addict
    Join Date
    Feb 2008
    Location
    London Ontario Canada
    Posts
    1,157
    Not sure if this is 100% accurate but apparently Fugger modded the gtx 280 and reached 1000+mhz on the core with the gtx280 LN2 of course, Im pretty sure its when he got the top scores for vantage.
    Case: Corsair 400R
    PSU: Corsair HX1000W
    mobo: Maximus IV Gene
    CPU: 2500K @ 4.2ghz 1.19 volts
    RAM: Gskill Ripjaws 1866mhz 2 x 4 gigs
    OS Drive: Kingston Hyper X ssd 120 gig
    Graphics: XFX HD5850
    Cooling: Corsair H100
    OS: Windows 7 Pro 64 bit







  22. #72
    Xtreme n00berclocker
    Join Date
    Mar 2006
    Location
    San Jose, CA
    Posts
    1,445
    1200MHz core is possible with ln2 and vmods.
    Quote Originally Posted by 3oh6
    damn you guys...am i in a three way and didn't know it again
    Quote Originally Posted by Brian y.
    Im exclusively benching ECS from this point forward

  23. #73
    Xtreme Addict
    Join Date
    Nov 2003
    Location
    Oslo, Norway
    Posts
    1,218
    anyone here with a good bios for the 280gtx..? 702Mhz+++?

  24. #74
    Xtreme Guru
    Join Date
    Dec 2002
    Posts
    4,046
    anyone posted an ocp mod for 280 ??

    thanks!
    Last edited by NapalmV5; 10-10-2009 at 10:53 AM.

  25. #75
    Registered User
    Join Date
    Nov 2008
    Location
    Czech Republic, Liberec
    Posts
    66
    Hi, Is any alternative soldering point on GTX 260 for GPU VMod?

Page 3 of 4 FirstFirst 1234 LastLast

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
  •