Tooth and nail video card battle (put yer shades on)

[ragejg]

:rant: :lame: :banghead: :fu:
http://www.nvnews.net/vbulletin/attachment.php?s=&attachmentid=1584


Yes the 5200 is mine... and I'll tell you what stinks... gainward duped one of their non-tier one distros (the vibe with the reps I spoke to dang near sounds like it is a mistake that they're all laughing about)... (one guy thought he was buying 14 GFFX5200U's from Compugate... i pity him)...

I'm ups'ing this card back to them as soon as I fully observe all this card has to offer(painfully), in some sort of review-ish format (painfully)...

wait, I dunno if it's even worth a review...

For the record:

XP2000+ @ 133fsb
GA-7VTXE+ kt266a mobo
512mb DDR @ PC2100 (one stick samsung pc2700, one stick Buffalo PC2100)
Temps: 48idle, 53loaded down...

Right before I sold my ti4200 (and thus had to run a gf2ti for a couple days... a card I for some reason always have liked), I culled some benchies from it as well... and they were all normal, but not too impressive @9.8k 3dmarks... The hd's not loaded down... it only has winxp, via 440's, unreal warfare engine games/demos, and nvidia drivers on it...

All driver installs/uninstalls went fine... and the system runs very smooth...

BUT WHAT A POOPY VIDCARD... for *anyone* at this price...

This 5200 will benifit someone else other than me... maybe...:(

[RobHague]

Im confused :confused:

The 5200 isnt as bad as a GF2 Ti :S well strange.

[jAkUp]

yipes... i dont see how thats possible... what drivers is it using? maybe you got a bad card there

[digitalwanderer]

...at least it skoolz the GF2 on fillrate.... :rolleyes:


That's pretty sad, even for a $79 card. :(

[ragejg]

43.45 both...

GF2ti set to Quality... no af/aa

GFFX set to performance... no af/aa

this card is a POS... absolutely no benifit from the OC... at all.

:(

[SurfMonkey]

I don't get the single texturing score at all. That doesn't add up at all... The multi-texturing rate is almost spot on though.

[marcocom]

ya 3dmark...totally the authority. it says i have a 100mhz fsb too.

theres actually a few carnations of the 5200 chip. check this out -
http://www.evga.com/articles/public.asp?AID=133 (specs listed)

[ragejg]

GF2ti @ perf... @ 300/480...

Single tex: 412.7
MultiTex: 785.2

2yr old thumps the new baby...

Well, I now have a credit with the place I bought it from... and I had extra $ anyway for a newer vidcard...

So I guess tonite I'll compile as much info as possible, to save and remember how pitiful this card was at the beginning of it's shelf life...

http://www.nvnews.net/vbulletin/attachment.php?s=&attachmentid=1585

[ragejg]

box pic...

http://www.nvnews.net/vbulletin/attachment.php?s=&attachmentid=1586

[Sazar]

Originally posted by ragejg
:rant: :lame: :banghead: :fu:
http://www.nvnews.net/vbulletin/attachment.php?s=&attachmentid=1584


yes it's mine and gainward duped their distro's... and it's goin back AFTER I review it (painfully)...

wait, I dunno if it's even worth a review...

am i right in understanding per that pick that you could not run game test 4 with your 5200 ?

thats odd considering it is supposed to be a dx9 card and a dx9 test...

care to elaborate ?

[ragejg]

I didn't run nature on that set of tests... It was terrible anyway...
Like 17-24fps...

In my current personal areas of interest, nature scores do not fit in...

btw here's a situation where they did, on the 4200, right before I removed it...

http://www.nvnews.net/vbulletin/attachment.php?s=&attachmentid=1587

[ragejg]

no tex sharpen, mip map to *poopy*, set @ perf...

14.9fps...

so this is what a Xabre feels like...

[digitalwanderer]

Could you post up some onion compare links for 3dm2k1se for me please? These scores are just blowing my mind, I wanna see how my "ancient" rig compares.... :evil:

[ragejg]

what settings?

remember, I'm using 43.45...

[digitalwanderer]

Go for it's best, the best possible settings you can have for the card without technically cheating at the benchmark. :)

No AA, no AF, no V-sync, if'n you wanna hork the mipmap level down go ahead, 1024x768 @ 32bit....pretty much the standard 3dm2k1se settings that gamers compare on. (But I never use the mipmap trick, I think it's silly to kill IQ for just a few more points. :rolleyes: I actually like to watch the benchmark run. :cool: )

[ragejg]

pitiful... even with EVERYTHING borked down AND OC'ed!!... 4079!! uhh... excuse me while I puke...

I'm gonna change my CL back to 2.5 and screw with some stuff...

but for now... http://service.futuremark.com/compare?2k1=6377060

:(

[Sazar]

Originally posted by ragejg
pitiful... even with EVERYTHING borked down...

I'm gonna change my CL back to 2.5 and screw with some stuff...

but for now... http://service.futuremark.com/compare?2k1=6377060

:(

those scores are quite low IMO...

dang.. I thought you would get around 6000 or so... at least...

[digitalwanderer]

Originally posted by ragejg
pitiful... even with EVERYTHING borked down...

I'm gonna change my CL back to 2.5 and screw with some stuff...

but for now... http://service.futuremark.com/compare?2k1=6377060

:(

OMMFG! Thanks Ragejg, that's a staggeringly awful score! My son's GF2 MX400 could whup that thing in the same rig! :eek:

That is TRULY embarressing, truly!

[StealthHawk]

the gfFX5200 is bad, but it shouldn't be that bad

[omghi2u]

Originally posted by ragejg
pitiful... even with EVERYTHING borked down AND OC'ed!!... 4079!! uhh... excuse me while I puke...

I'm gonna change my CL back to 2.5 and screw with some stuff...

but for now... http://service.futuremark.com/compare?2k1=6377060

:( OMFG, even my mx440 on a 1.4athlon got 8k+...that is truly embarassing.

[Sazar]

i am just wondering what fps one will get with dx9 effects...

and I wonder when the pr stating the card will be MUCH faster with dx9 games will roll out :D

[mbvgp]

Searching on the futuremark website shows scores for the 5200 (non ultra) in the 8700 range. Driver version is 43.45. The single texturing fill rate also comes out to about 770 and multitexturing to about 800. Something seems very wrong with the card you got.

[StealthHawk]

Originally posted by omghi2u
OMFG, even my mx440 on a 1.4athlon got 8k+...that is truly embarassing.

you must be kidding me. my gf3 on a 1.33 athlon got 6500

[SurfMonkey]

Your scores are really borked. The multi-texturing fillrate of the 5200 is never going to compare to the GF4 it has less TMU's. But it really shouldn't be that low.

I presume you have DX9 installed? Not that I can see it making too much difference.

With a clock speed of 320Mhz, and given a 4x1 architecture, then your theoretical single texturing fillrate should 1280Mpixels Even the multi-texturing should be capable of this figure even if it swaps to a 2x2 architecture.

Something is really not right there. Looks almost as if it running as 2x2 constantly. That would give you a theoretical max of 640Mpixels in single texturing and 1280Mpixels in multi-texturing. Which seems to go with your scores but not with other results.

[StrikerPro]

Yikes! Now THAT is pitiful...Feel sorry for you mate, seems that OC didnt help...AT ALL!

I suppose those benchmarks "5200 vs 9200" were true after all...this DX 9 card cant give you any DX 8 performance, let alone DX 9..