agp speed among other things...

[Hammer]

Danr wrote:
They don't tell you this, but DirectX automatically sets AGP speed to 2x for AMD processors because older athlon motherboards weren't stable with 4x.  You get the same results because your running at 2x no matter what.  Guess they still havn't bothered to fix it.

My old VIA KT133 based mobo can operate @ AGP 4X.


Just a side note; My ASUS nForce 2 based mobo can operate @ AGP 4X.


IF a feature doesn’t work, most X86 PC manufactures removes the said feature from their marketing materials. At later date they may issue a revise design with the said feature enabled.

[Hammer]

ksk wrote:
FYI

Quake3 PPC vs x86: http://www.barefeats.com/p4game.html
PPC & Radeon7000 PCI vs AGP cards ... : http://www.barefeats.com/gr7000.html

btw. Radeon7000 PCI seems to be 66MHz PCI card... therefore perhaps the best card for future SharkPPC + AOS4 users.

Just a side note;

One could better the Athlon 2000+’s results by employing nForce 2 based chipset (with aggressive settings). For example, my Athlon 1800+/nForce2 test machine pumps Quake 3 @ ~235/240 fps** (at “Normal” settings), while the old Athlon1800+/VIA KT133 only pumps around 172 fps** (at “Normal settings”).  Both employs the same video cards (i.e. nVidia Geforce 4 TI).

** "timedemo 1", "demo demo001", "normal settings".

From mainstream VIA news, VIA is currently working on “hyper transport” based chipset. Since MAI has a close relationship with VIA, I wonder IF MAI could develop a product based on AMD’s hyper transport designs.

With today’s nForce 2 and the dual DDR400 setup, the bandwidth is pegged at 6.4Gb/s, which is a match with IBM’s 2003 H2 release of PPC 970’s chipset bandwidth.