facebook rss twitter

NVIDIA GeForce 7900 GTX, 7900 GT and 7600 GT Preview

by Ryszard Sommefeldt on 9 March 2006, 14:05

Tags: Nvidia Geforce 7900 GTX, NVIDIA (NASDAQ:NVDA)

Quick Link: HEXUS.net/qaeyi

Add to My Vault: x

System Setup and Notes

The same FX-60 test platforms that we used to test ATI Radeon X1900 XT and XTX recently was used to test the three new NVIDIA products. While more powerful than a 7600 GT will ever see, the test system is nonetheless useful for reference evaluation of that product.

Time pressures for your author due to our yearly Cebit expedition means that Radeon X1600 XT, GeForce 6600 GT and GeForce 6800 GS numbers to compare the 7600 GT to are absent until Cebit is done and dusted. We'll update you when that happens. Therefore, with Radeon X1800 XT, Radeon X1900 XT and XTX to fight the 7900s, and GeForce 7800 GTX and GTX 512 thrown in to boot, the performance gap upwards from 7600 GT is what's explored.

That comparison might be more useful than you think, keep reading to see why. NVIDIA helpfully supplied a driver to press when the boards were sampled, however that was soon replaced by a newer driver to fix an issue with F.E.A.R. That driver is what was used to test all of the GeForce products (7600, 7800 and 7900, regardless of SKU) and that driver (84.17) is what'll appear as the WHQL release in short order, at the time of writing.

The F.E.A.R. issue concerns antialiasing performance and assists NVIDIA's products significantly, so enjoy that driver if you're an NVIDIA product owner that enjoys Monolith's pant-wetting gun-fest.

GTX 256 was only used in the game tests, not the theoretical and architectural tests.

Hardware

  • Test Platforms
Processor(s) AMD Athlon 64 FX-60
2.6GHz, dual-core, 1MiB L2 per core, Toledo
Mainboard(s) ASUS A8R-MVP
ATI RD480 Crossfire
ULi M1575
ASUS A8N-SLI Deluxe
NVIDIA nForce4 SLI
Memory 2 x 512MiB G.Skill F1-3200DSU2-1GBLE
Memory Timings 2-2-2-5 @ 400MHz, 1T
BIOS Version 0402 1016.001
Disk Drive 80GB Western Digital PATA
Graphics Card(s) ATI Radeon X1900 XTX (650/775)
ATI Radeon X1900 XT (650/750)
NVIDIA GeForce 7900 GTX (650/800)
NVIDIA GeForce 7900 GT (450/660)
NVIDIA GeForce 7600 GT (560/700)

NVIDIA GeForce 7800 GTX 512 (550/850)
NVIDIA GeForce 7800 GTX (430/550)
Graphics Driver CATALYST 6.2 BETA ForceWare 84.17
Operating System Windows XP Professional, SP2, 32-bit
Core Logic Driver(s) ULi M1575 Chipset Driver V1.0.5.2a nForce4 AMD Edition 6.70

Software

Software
  • Futuremark 3DMark06
  • MDolenc's Fillrate Tester
  • GZeasy Fillrate V0.9
  • D3D Rightmark 1.0.5.0
  • Humus' HSRBench
  • SeriousMagic v1.3
  • HEXUS Texture Test
  • HEXUS Instruction Issuer
  • Splinter Cell: Chaos Theory
  • F.E.A.R.
  • Quake 4

Notes

We use separate test platforms for ATI and NVIDIA graphics boards to let each IHV's products work best on their own core logic. While the differences are slight, it's a worthwhile pursuit. The testing is architecturally focussed, our evaluation with a range of top-tier games.

Driver defaults were used throughout, and stock clocks were used for all cards unless noted. If in-game controls could be used for both antialiasing and anisotropic texture filtering, they were, otherwise the driver was used to force the required levels (if applicable and the game allowed it without rendering errors). Tested resolutions were 1024x768, 1280x1024, 1600x1200 and 1920x1200.

Game tests were run a minimum of three times at each setting, and the median value reported. In the case of manual 'run-through' testing with FRAPS, three consecutive runs that produced repeatable results, after further analysis, were used. If values weren't part of a repeatable set, they were discarded and obtained again.

If you have any questions or comments about our testing methods, please feel free to drop by any time and share in the HEXUS.community.

Miscellaneous Images