facebook rss twitter

Review: NVIDIA GeForce 7800 GS AGP shootout

by James Smith on 26 July 2006, 08:21

Tags: NVIDIA (NASDAQ:NVDA)

Quick Link: HEXUS.net/qafda

Add to My Vault: x

System Setup and Notes

Hardware

  • NVIDIA GeForce 7800 GS AGP Test Platform
Processor(s) AMD Athlon 64 FX-53
2.4GHz, single-core, 1MiB L2, Sledgehammer, S940
Mainboard(s) Gigabyte GA-K8NNXP-940
NVIDIA nForce3 PRO 150
Memory 2 x 512MiB Corsair CMX512RE-3200LL
2 x 512MiB
Memory Timings 3-3-3-8 @ 400MHz, 2T
BIOS Version F4d Beta
Disk Drive Seagate Cheetah ST336753LW
Adaptec 39160 Controller
Graphics Card(s) BFG Tech GeForce 7800 GS OC™ (400/625)
eVGA e-GeForce 7800 GS (460/675)
XFX GeForce 7800 GS Extreme Edition (440/650)

GeCube Radeon X1600 PRO (500/405)
NVIDIA GeForce 6800 (335/350)
MSI NX6800 ULTRA (400/550)
NVIDIA GeForce 5900 Ultra (450/425)
Graphics Driver NVIDIA ForceWare 83.40
ATI CATALYST 6.3
Operating System Windows XP Professional, SP2, 32-bit
Core Logic Driver(s) NVIDIA 4.36 (AGP)

Software

Test Software
  • Splinter Cell: Chaos Theory
  • Far Cry
  • Quake 4

Notes

The GeForce 7800 GS AGP boards get to fight each other in the first instance, then NVIDIA's last AGP champion (6800 Ultra), their big mid-range 6800-series seller (plain 6800) and the much older GeForce 5900 Ultra. Add the Radeon X1600 AGP into the mix and you get analysis to help the older board owners on AGP decide what's best for them, as well as a ranking of the three 7800 GS AGP boards on test.

An eclectic mix of products but one which makes good sense. Don't worry, we'll sum it all up for you in the conclusion.

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 1280x1024 and 1600x1200.

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.