Little Parhelia Preview at Aceshardware

[KeiFront]

http://www.aceshardware.com/#60000384

[Jazzz]

I was just there to see if they had posted up a review, those are some really really disapointing results, i hope the drivers performance improves as the matrox card is being outpeformed by the Geforce4 in alot of games and professional applications.

[Cobos]

As was said in the article and I beleive has been hinted at here or the matrox boards...together with the 256MB version there will be a completely redone OpenGL driver set. Prepared for professional apps, though if this will also improve the results in OpenGL games is still up to debate

Cobos

[Pace]

Hmm...differen drivers for 128MB and 256MB?

[Cobos]

Possible, though the impression I had was that Matrox has hired someone to write pro OpenGL drivers and they are not finished yet, but they will be released with the 256MB P as this is aimed towards the pro market. Though nothing is stopping Matrox from making sure they don't work on the "little P" (the 128 and 64MB versions ). Or I might simply be so off I'm not even close to the barn

Cobos

[CaineTanathos]

well that would be total bullshit not releasing the new opengl driver for the 64/128 mb card

if there even will be a new one (driver I mean)

matrox always put stability on the first placen , then features and then speed.

[Stringy]

Personally any comments on Future products, even if alluded to by Matrox is complete rumour...
We all remember all the things the P was supposed to be, and this reeks of someone'e UN-Informed imagination...

Seperate drivers?!?! Umm doesn't Matrox use Unified Driver sets? Isn't Matrox quite proud of their unified driver set??

complete rubish...

Anyway, Ace's numbers really show how long Matrox has to go before they have well performing OGL drivers... Atleast these work out of the box - a feat in its own right...


Craig

[KvHagedorn]

Those marks are really, awfully embarassing. I hope the powers at M have the good sense to be truly embarassed though..

[Wombat]

Doubt it.

[isochar]

Ouch, that's just a plain ole' GF4 too. :O

[Jazzz]

Originally posted by isochar
Ouch, that's just a plain ole' GF4 too. :O

to make it even worse its a Ti 4400

[CaineTanathos]

with current games the parhelia won't be faster than a GF4
not even with unreal2

why , one word : pixelfillrate

the most important thing for those games is fillrate, because they don't use anything else (maybe the vertex shader)

All games will run fast enough on the parhelia though

and maybe it will be faster if a game will use 4 textures like kreed (the russian game )

[Jon P. Inghram]

Hey, maybe the Parhelia will tear up the competition as soon as all the new wireframe only games come out!

[Stringy]

Originally posted by CaineTanathos
with current games the parhelia won't be faster than a GF4
not even with unreal2

why , one word : pixelfillrate

the most important thing for those games is fillrate, because they don't use anything else (maybe the vertex shader)

So then the O/C'ing results are all the more important on these chips.. if they O/C well they could close the gap, if they don't well the same average performance on games that don't take advantage of the P's specail features...

anyone find a way to O/C??


Craig

[CaineTanathos]

if they start using the pixelshaders and vertexshaders then the parhelia will be much much faster.

take a look at the sharkmark from matrox