LOSTCIRCUITS

ASUS V9280S vs. ASUS V8460 (Ti4600) vs. ASUS V8420 (Ti4200). The V9290S was run at either AGP 8X, or 4X (with 8X mode disabled in the BIOS) or with pin# A3 and A11 taped over to force AGP2.0 (4X) mode. The Ti 4600 was run in AGP 4X Mode and with AGP 8X disabled in the BIOS which results in AGP 2X mode. The Ti4200 was run in AGP 4X mode only. It is not a huge lead but the lead is there and can be reproduced from one run to the next. Furthermore, there does not appear to be any difference between AGP 2.0 and AGP 3.0 as long as the transfer frequency is the same. The sensitivity of 3DSMax to the AGP rate is probably shown best by the performance hit taken by the TI 4600 when running in AGP 2X mode.

DRV-08 sees AGP 8X increasing its lead over the 4 X mode with no sensitivity at all to the protocol. Again, note the huge performance delta between AGP 2X and 4X. If geometry transfers are the bottleneck here, then it is clear that there will be a ceiling effect somewhere with regard of how much bandwidth is actually needed. The numbers show nicely, though that AGP 8X is an improvement albeit with diminishing returns. A 5.7% increase is still remarkable.

DX-07 tells a similar story as DRV-08 even though the delta is only 4.5%.