29.42/28.32 comparison results

[SolrFlare24]

OK here are my specs:

Windows XP
P4 1.6a @2136mhz (yeah yeah I'm waiting for my new HSF to crank it more )
512mb DDR @ ~ 333
Abit IT7 based on 845e chipset
GeForce 4 ti4600

28.32 3D2001se marks: 10578
29.42 3D2001se marks: 10795

So a 223 point increase. Not too shabby...but here is the real interesting stuff...my low detail game test FPSs stayed about the same in both tests....but check the high detail game FPS differences:

Game 1 High 28.32 - 51.7
Game 1 High 29.42 - 53.3

Game 2 High 28.32 - 105.4
Game 2 High 29.42 - 112.3 !!!!!!

Game 3 High 28.32 - 155.9
Game 3 High 29.42 - 157.1

Game 4 28.32 - 43.3
Game 4 29.42 - 44.6

Low detail test saw anywhere from a 1 to 2 FPS increase. But the interesting one to note is the jump in Game 2's high detail. Other tests in other games like Quake 3, Dungeon Siege, etc also seem to indicate an overall increase in High detail performance and minor adjustments in low detail. Not too shabby.

Also worth noting is the image quality is in fact slightly improved over the 28.32s. A definate must have for GF4 owners at least.

 

[Gonad the Barbarian]

You sure those Game 3 scores aren't for low detail?

 

[Acanthus]

its a geforce 4, i get nearly those scores with my geforce 3 in lobby sounds about right to me.

 

[Gonad the Barbarian]

Right at 60 FPS in Game 3- Lobby high detail. Total score is ~9400.

Athlon XP 1700+ (1.47Ghz)
Epox 8KHAL (VIA 266A)
Crucial 512 MB PC2100
MSI GF4 TI4400 128MB
Hercules Fortissimo II
D-Link 10/100 NIC
Lucent V.92 56K
Seagate 80 GB Barracuda IV HD
Cyberdrive 36x12x48 CD-RW

 

[THUGSROOK]

the 28.32 are really the slowest drivers out there. (for 98 at least)
your comparison falls in line with mine tho - the 29.42s are faster. they are also very stable too.
i ran 3dmark2001 twice and i actually got the exact same score :Q

 

[motojeff]

You have to qualify this statement by the video card... My Geforce 256 slowed down with the new
drivers (only a little though)...

Originally posted by: THUGSROOK
the 28.32 are really the slowest drivers out there. (for 98 at least)
your comparison falls in line with mine tho - the 29.42s are faster. they are also very stable too.
i ran 3dmark2001 twice and i actually got the exact same score :Q

 

[THUGSROOK]

im on a ti4400 @ ti4600 speeds ...comparable to the posters card.

 

[Uclagamer_99]

yep...i highly doubt they optimize for anything under a geforce 3 nowadays and specifically the gf4's

 

[BD2003]

Test with anisotropic filtering please.

 

[OneOfTheseDays]

umm they better be optimizing for the gf3 cuz the gf4 is brand new and there is still a massive userbase of gf3 users out there. They don't just stop all together and quit optimizing for something just cuz they have a new chipset out there. Besides when they optimize for one chipset it usually improves performance on all of them.

 

[BD2003]

bump

 

[dude]

GF3 and GF4 Ti are basically the same thing except the GF4 Ti has programmable pixels, better FSAA implementation, new and faster memory enhancement sub system, a kitchen sink, and other small changes that take up another 25 Gazillion transistors on the GPU.

 

[dougp]

Originally posted by: dude
GF3 and GF4 Ti are basically the same thing except the GF4 Ti has programmable pixels, better FSAA implementation, new and faster memory enhancement sub system, a kitchen sink, and other small changes that take up another 25 Gazillion transistors on the GPU.

I think I saw a bathtub on the back of my Siluro Ti4400 somewhere... God damn that thing is a monster and it's freakin heavy!!!

I noticed about a 300 point increase in 3DMark2001, maybe the next set of drivers will let me break 11,000... Or maybe I should just give in and o/c my 4400 to 4600 speeds when I get my new fans (Panaflo M1A's baby!!)... I also noticed a speed increase in Dark Age of Camelot, it doesn't lag as bad when I walk into the higher poly areas.

And dammit, tab won't work on this thing