GF4 4200 vs GF-FX 5600 ( Confusing!! )

Status
Not open for further replies.

vilex

Posts: 19   +0
Confusing!!

I have a Geforce 4 TI 4200 64MB. My friend gave me his Geforce FX 5600 Ultra 128. I ran a benchmark test between the two in 3DMark03 and here are the figures...

2917 marks - Geforce FX 5600

1334 marks - Geforce 4 TI 4200

The FX smoked the TI in the benchmark...however in almost every game I've played (and I've tried many) the 4200 is faster! I don't understand this, seeing that in the 3DMark benchmark the 4200 was barely chugging along and the FX was just flying through. Can someone please enlighten me?

Win XP Pro
Athlon XP 2400+
512 2100 DDR
80GB WD HD
Soundblaster Audigy 2

- Title edited
Didou
 
3DM03 scoring depends heavily on DX9.....
The Ti4200 has no DX9 hardware support, so it will score much lower.

However, although the FX5600 has DX9 hardware support, it's very poor at running DX9 app's in true DX9, so the FX5600U will run most DX9 app's in DX8 mode, just as the Ti4200. Virtually no difference in IQ (Image Quality), while the Ti4200 is quite comparable to the performance of an FX5600U.....many will prefer the Ti4200 over the FX5600U, especially if you get one that overclocks well:grinthumb


If you run 3dM01 w/ both cards (DX8 benchmark), they will perform very close.....
 
Hmm, that makes sense. About overclocking, I've read the 5600 is quite the overclocker as well. Just wish I knew how to do this kind of stuff.
 
go to the "CPUs, Chipsets and Mobos" forum and check out the thread labled "Read before asking "How do i overclock". that should give you some basic insight.
 
Originally posted by vilex
Hmm, that makes sense. About overclocking, I've read the 5600 is quite the overclocker as well. Just wish I knew how to do this kind of stuff.

Depends on the revision..the 1st revision FX5600U's don't overclock so well....the second revision does indeed:grinthumb
 
I do notice a difference in image quality though. When playing Painkiller demo 1 with the 4200, the darker walls have that 16-bit gasoline puddle look to them, which I didn't notice with the 5600. I used the exact same graphics settings for both cards.
 
After lots of testing this and that, I ended up with these figures in 3DMark01 :

------Geforce 4 TI 4200 64MB------

9011 Marks - with default settings
4997 Marks - with 4 sample AA
9511 Marks - overclocked with default settings ( 301 / 530 )

------Geforce FX 5600 Ultra 128MB------

9221 Marks - with default settings
7064 Marks - with 4 sample AA
9575 Marks - overclocked with default settings ( 447 / 860 )

Though the 5600 outperforms the 4200 in these tests, it still runs worse in the majority of games I have tested, except for Painkiller and Max Payne 2 which both ran like a dream on the 5600. The card seems to have much higher quality visuals and runs faster at extreme settings than the 4200.

So far, choosing between these two cards isn't unlike deciding whether you want speed or quality.
 
Well if there's one thing that is proven in these tests, it's that 3DMark01-03 isn't very accurate. I think it would only be useful if one card completely stomped the other, but if there's just a 100 mark or so difference, that means nothing. This program alone is not enough to make an accurate decision of which card is better.
 
Originally posted by vilex
Well if there's one thing that is proven in these tests, it's that 3DMark01-03 isn't very accurate. I think it would only be useful if one card completely stomped the other, but if there's just a 100 mark or so difference, that means nothing. This program alone is not enough to make an accurate decision of which card is better.

Well, if nVidia didn't have application specific "optimizations" in their drivers that gave inflated benchmark scores, you might have a clearer picture...but that's a different thread alltogether:blackeye:
 
Status
Not open for further replies.
Back