So I tested my Riva TNT2 M64 with 8.05 and my MX440 with 30.82 and tested with 3dmark99, 2000, and 2001, the riva TNT2 M64 is faster in 3dmark99, and only 900 slower in 2000. My question is because I don't have a GF2 is it possible to edit the 8.05 drivers to support the MX 440, they are basically geforce2 with geforce3 memory controller, is it possible to backport support into the GF3 drivers
Not fully. You may port some DLLs from older drivers, but how far it can go, until everything will just break, is another question. It could work with DLL which is aware of Lightspeed Memory Architecture, i.e. early GeForce 3 driver parts.
I must be some kind of standard: the anonymous gangbanger of the 21st century.
candle_86 wrote on 2022-10-27, 22:47:So I tested my Riva TNT2 M64 with 8.05 and my MX440 with 30.82 and tested with 3dmark99, 2000, and 2001, the riva TNT2 M64 is faster in 3dmark99, and only 900 slower in 2000. My question is because I don't have a GF2 is it possible to edit the 8.05 drivers to support the MX 440, they are basically geforce2 with geforce3 memory controller, is it possible to backport support into the GF3 drivers
nVidia has a so-called "Unified Driver Architecture", but I'm not sure if it's the case for drivers as old as 8.05.
It doesn't hurt to actually try it, as GF4 MX440 is of GF2's lineage (NV17), and the 8.05 driver appears to support up to GF3 (NV20).
Architecture-wise it looks like GF4 doesn't bring too many new features and was mostly based on the last two generations (with MX being GF2 and Ti being GF3).
NOTE: This site contains some useful info about nVidia GPUs.
EDIT: Took a look at the thread about MX440 drivers. It was mostly about relatively newer drivers (20s to 40s) so I'm not really sure if such an old driver would really work...
I would still prefer GeForce 4 MX440. It should be plenty fast for very old games and faster for games from 2000-2001.
Pentium III 900E, ECS P6BXT-A+, 384MB RAM, NVIDIA GeForce FX 5600 128MB, Voodoo 2 12MB, 80GB HDD, Yamaha SM718 ISA, 19" AOC 9GlrA
Athlon 64 3400+, MSI K8T Neo V, 1GB RAM, NVIDIA GeForce 7600GT 512MB, 250GB HDD, Sound Blaster Audigy 2 ZS
AlexZ wrote on 2022-10-28, 19:59:I would still prefer GeForce 4 MX440. It should be plenty fast for very old games and faster for games from 2000-2001.
The 128bit mx440 and the mx460 are decent cards. I have a mx460 in my socket A system and its a great performer.
AlexZ wrote on 2022-10-28, 19:59:I would still prefer GeForce 4 MX440. It should be plenty fast for very old games and faster for games from 2000-2001.
I agree but with a Pentium III 500 i wont be running a game that would stress that kind of card
28.32 was used in reviews so at least that should be good
asus tx97-e, 233mmx, voodoo1, s3 virge ,sb16
asus p5a, k6-3+ @ 550mhz, voodoo2 12mb sli, gf2 gts, awe32
asus p3b-f, p3-700, voodoo3 3500TV agp, awe64
asus tusl2-c, p3-S 1,4ghz, voodoo5 5500, live!
asus a7n8x DL, barton cpu, 6800ultra, Voodoo3 pci, audigy1
candle_86 wrote on 2022-10-27, 22:47:tested with 3dmark99, 2000, and 2001, the riva TNT2 M64 is faster in 3dmark99, and only 900 slower in 2000.
Who cares? What do the games say? Drivers do matter a lot, but I don't put all that much stock in 3Dmark scores, they don't always represent well real gaming performance. There are only 2 scenarios when I care about them. When I compare the card to itself on different drivers and when I want to stress test a card after I did maintenance on it.
candle_86 wrote on 2022-10-28, 23:54:I agree but with a Pentium III 500 i wont be running a game that would stress that kind of card
Also having a roughly 3x as fast card (compared to the full TNT2) held back by CPU limit puts that 3DMark score difference into question. How high those scores were?
RandomStranger wrote on 2022-10-29, 20:58:Who cares? What do the games say? Drivers do matter a lot, but I don't put all that much stock in 3Dmark scores, they don't alwa […]
candle_86 wrote on 2022-10-27, 22:47:tested with 3dmark99, 2000, and 2001, the riva TNT2 M64 is faster in 3dmark99, and only 900 slower in 2000.
Who cares? What do the games say? Drivers do matter a lot, but I don't put all that much stock in 3Dmark scores, they don't always represent well real gaming performance. There are only 2 scenarios when I care about them. When I compare the card to itself on different drivers and when I want to stress test a card after I did maintenance on it.
candle_86 wrote on 2022-10-28, 23:54:I agree but with a Pentium III 500 i wont be running a game that would stress that kind of card
Also having a roughly 3x as fast card (compared to the full TNT2) held back by CPU limit puts that 3DMark score difference into question. How high those scores were?
So i tested some games between my Quadro2 MXR I grabbed for $5 and my MX440, the Quadro2 is running on 6.23 and the MX440 on 30.82
In 3dmark99 and 2000 the Quadro wins by 300pts roughly, in games I tested, Star Trek Elite Force Quadro2 43FPS 10x7 16bit, MX440 39FPS 10x7 16bit. What I will say is I think the bottleneck is currently the Pentium III 500
