GeForce GTX 285 Gets 2 GB: Gigabyte's GV-N285OC-2GI

Test Settings

Swipe to scroll horizontally
Test System Configuration
CPUIntel Core i7-950 (3.07 GHz, 8 MB Cache) Overclocked to 3.83 GHz (23 x 166.6 MHz)
MotherboardMSI X58 Pro BIOS 7.3 (04-17-2009) Intel X58/ICH10R Chipset, LGA 1366
RAMMushkin 6.0 GB DDR3-1333 CAS 9 at DDR3-1333 CAS 8-8-8-19 (1T)
GeForce GTX 285 2 GBGigabyte GeForce GTX 285 2 GB 660/1505 MHz GPU/Shader, GDDR3-2400
GeForce GTX 285 1 GBXFX GeForce GTX 285 1 GB Clocked to Gigabyte 2.0 GB Specs (above)
GeForce GTX 295 1.8 GBXFX GeForce GTX 295 1.79 GB 576/1242MHz GPU/Shader, GDDR3-2016
Hard DriveWestern Digital Caviar Black 1 TB 7,200 RPM, 32 MB Cache SATA 3.0 Gb/s
SoundIntegrated HD Audio
NetworkIntegrated Gigabit Networking
Software
OSWindows Vista Home Premium x64 SP1
GraphicsNVidia Forceware 186.18
ChipsetIntel INF 9.1.0.1007

Digital Storm’s Gaming Dominator was still on our bench from its review (which you'll see later this week), its 3.8 GHz overclocked processor already set to the target frequency we wanted to use for today’s graphics comparison. Using it allowed us to reuse some of its test data while providing more information on its graphics capabilities.

Swipe to scroll horizontally
Benchmark Configuration
CrysisPatch 1.2.1, DirectX 10, 64-bit executable, benchmark tool Test Set 1: High Quality, No AA Test Set 2: Very High Quality, No AA Test Set 3: Very High Quality, 8x AA
Far Cry 2Far Cry 2 Benchmark Tool 1.0.0.1 Test Set 1: High Quality (DX 10), No AA Test Set 2: Ultra High Quality (DX 10), No AA Test Set 3: Ultra High Quality (DX 10), 8x AA
Tom Clancy's H.A.W.XGame Version 1.00, in-game performance test Test Set 1: Highest Details, DX10, SSAO, No AA Test Set 2: Highest Details, DX10, SSAO, 4x AA Test Set 3: Highest Details, DX10, SSAO, 8x AA
Left 4 DeadGame Version 1.0.1.4, custom timedemo Test Set 1: Highest settings, No AA, No AF Test Set 2: Highest settings, 4x MSAA, 8x AF Test Set 3: Highest settings, 8x MSAA, 16x AF
S.T.A.L.K.E.R.Clear SkyClear Sky Benchmark File Version 1.5.8.1 Test Set 1: High preset, DX10, No AA Test Set 2: Ultra preset, DX10, No AA Test Set 3: Ultra preset, DX10, 4x MSAA
World in ConflictPatch 1009, DirectX 10, in-game performance test Test 1: High Details, No AA / No AF Test 2: Very High Details No AA / No AF Test 3: Very High Details 4x AA / 16x AF
Synthetics
3D mark VantageBuild 1.01 (3DMark, GPU, CPU Scores) Performance, High, Extreme Presets Disable PPU in Options table
Thomas Soderstrom
Thomas Soderstrom is a Senior Staff Editor at Tom's Hardware US. He tests and reviews cases, cooling, memory and motherboards.
  • twisted politiks
    ive had EVGA's GTX 285 for about two months now, nothing new in the memory department
    Reply
  • rambo117
    wow... that was quite pointless, i was really expecting a good article but it was the same numbers basically, why not just have a single page with the average gains with 2gb vs 1gb (which was completely nonexistent anyways)
    Reply
  • astrodudepsu
    Well, all I can say is good try. Not some of your best work, but worth exploring nonetheless.
    Reply
  • Crashman
    astrodudepsuWell, all I can say is good try. Not some of your best work, but worth exploring nonetheless.
    Tom's Hardware was hoping to find more 2560x1600 scenarios where the 2GB advantage would play out. When very few advantages were found, Tom's did the honest thing and published the numbers anyway.

    I think you can take a lot from this article. I just spoke to a guy who asked "2GB or water cooling?" when looking at cards of the same price. He has a powerful water cooling loop, so the answer was easy.
    Reply
  • one-shot
    I think it is a great article. Too often people approach me thinking a larger frame buffer means extra performance. Actually, just recently a co-worker wanted to get a GTX 285 with 2GB of VRAM. Great article, keep it up!
    Reply
  • SpadeM
    Just a quick question: Since the GTX285 is a high range card, those the same "1GB is enough" rule apply to mid range/low end cards?

    PS: I'm thinking slower GPU might benefit from more memory
    Reply
  • astrodudepsu
    CrashmanTom's Hardware was hoping to find more 2560x1600 scenarios where the 2GB advantage would play out. When very few advantages were found, Tom's did the honest thing and published the numbers anyway.
    Which is why I said it was worth exploring. I realize you wouldn't do all this work and NOT publish your results, as mundane as they may be.
    Reply
  • rambo117
    it was quite informative and worth exploring like astrodude said. just wish that it was a little more 'exciting'.. idk, maybe an sli 2gb vs 1gb will prove to be possibley more interesting.
    Reply
  • Hellbound
    If a card is going to cost $350-$400, and perform as much as %50 less than the car that costs $500....get the $500 card. You are already spending an insane amount for a card, might as well go all the way. As for me, I'm waiting for the DX11 cards to come out.
    Reply
  • Crashman
    rambo117it was quite informative and worth exploring like astrodude said. just wish that it was a little more 'exciting'.. idk, maybe an sli 2gb vs 1gb will prove to be possibley more interesting.
    Three-way would have been best, but there's just not enough samples to go around.
    Reply