From testing low end systems over the last few months with 3DMark2001SE, I have started to question the program's ability to accuratly test those computers. Is there a better way to test systems slower than 1 GHz, running on a 100 MHz FSB, systems like one of the computers I have that is running on a K6-2+ CPU and a 2X AGP?
It seems like 3DMark is designed for newer computers with better 3D cards, faster CPUs, and faster ram.
What really brought this issue to light was when my pci GF2 beat my agp GF2 by about 70 points, both MX 400 64 MB. But when playing 3D games with the card, the pci card sucked. The agp card does a much better job in a real situation. So can these scores be trusted when changing video cards, ram, the fsb, ect.
<font color=red><b>To reign is worth ambition though in Hell:
Better to reign in Hell, than serve in Heav'n.</b></font color=red>
John Milton, <i>Paradise Lost</i>, II 262-263
It seems like 3DMark is designed for newer computers with better 3D cards, faster CPUs, and faster ram.
What really brought this issue to light was when my pci GF2 beat my agp GF2 by about 70 points, both MX 400 64 MB. But when playing 3D games with the card, the pci card sucked. The agp card does a much better job in a real situation. So can these scores be trusted when changing video cards, ram, the fsb, ect.
<font color=red><b>To reign is worth ambition though in Hell:
Better to reign in Hell, than serve in Heav'n.</b></font color=red>
John Milton, <i>Paradise Lost</i>, II 262-263