Gf2 Mx400 Oc???

Albreck · Mar 5, 2003

I just want to know how much may I oc my card is a gf2 mx400 chaintech, with stock fan, so if someone may help me.... thanks...

Shade00 · Mar 5, 2003

The only way to find out is to test it yourself. Start slow, go up in 5mhz increments, and when it starts to artifact, pull it back down. I usually use 3dmark2k1's lobby test as a good test for artifacts.

john240sx · Mar 5, 2003

when i overclocked my MX400 the Dragothic (sp?) test showed artifacts before the other tests.

the highest i was able to hit with my card was 230Mhz core and 190Mhz ram. most people seem to get higher on the ram then i did but my core overclocked higher then most people's cores. my card was a Visiontek Extasy MX400.

Gecedion · Mar 5, 2003

Check out this post about the GeForce line. Go to the GF2 section of the spread sheet.

Overclocker550 · Mar 5, 2003

dont overclock the core, it doesnt help. memory bandwith limited. also try the sprites test, it usually artifacts badly on that test

Shade00 · Mar 5, 2003

Yes, these cards are memory bandwidth limited, but the core is not running fast enough to take advantage of extra bandwidth. It has half the texturing capability of a regular GF2... i.e., not very much, so the faster the core goes, the more power you get. Overclock the core. It'll provide some increase.

KewlKid2000+ · Mar 5, 2003

... i wouldnt oc it if i were u man... i fried mine doing it.

Overclocker550 · Mar 5, 2003

dude, I underclocked the core of my mx440 from 270 to 180 and lost 4 fps went from 88 to 84. this shows how inefficient ocing core is

avzay · Mar 5, 2003

actually overclocking the core also does matter....at least it did with my regular old GF2

Shade00 · Mar 6, 2003

dude, I underclocked the core of my mx440 from 270 to 180 and lost 4 fps went from 88 to 84. this shows how inefficient ocing core is

In what bench? QIII? QIII is pretty seriously CPU bound at this point. I really can't take any of your proof objectively anymore OC550... there are so many variables that contribute to incorrect results.

I say try OCing the core yourself. If it works for you, great. If it doesn't, nothing lost but a bit of time. OCing the core on my Ti4200 produces gains in 3dmark and real world gaming.

... i wouldnt oc it if i were u man... i fried mine doing it.

I guess that's a plausible worry. Just go slow with it and don't start out by doubling your RAM speed or anything crazy like that.

Overclocker550 · Mar 6, 2003

I already oced the core on another card that was mem bandwith limited-not a single point increase. in fact its a bad idea to oc the cores on most video cards, this goes doubly for the MX series!

KILLorBE · Mar 6, 2003

My GF2MX (Stock @ 175/166 core/mem) running 3DMark2K1:
...Core/Mem.
@ 195/195 --> 2938
@ 200/200 --> 3014
@ 205/205 --> 3050
@ 205/210 --> 3112 (Artifacts)
@ 210/205 --> 3048
@ 215/205 --> 3070
@ 220/205 --> 3076
@ 225/205 --> 3093
@ 230/205 --> 3118
@ 235/205 --> 3110, 2nd run 3123, 3rd run 3143
@ 240/205 --> DNF

As you can see you will gain a few points by OC'ing the core, but not as much as by OC'ing the memory.

**NOTE** I should run 3DMark2K1 a few more times to get more accurate results.

Maxvla · Mar 6, 2003

got my gf2mx400 to 260/210

good luck on yours.

Shade00 · Mar 6, 2003

I already oced the core on another card that was mem bandwith limited-not a single point increase. in fact its a bad idea to oc the cores on most video cards, this goes doubly for the MX series!

Umm... why exactly is it a bad idea, and how is it twice as bad on MXes?

Gf2 Mx400 Oc???

Albreck

Member

Shade00

Member

john240sx

Member

Gecedion

Member

Overclocker550

Member

Shade00

Member

KewlKid2000+

Member

Overclocker550

Member

avzay

Member

Shade00

Member

Overclocker550

Member

KILLorBE

Member

Maxvla

Senior Member

Shade00

Member

Similar threads