PDA

View Full Version : 70.16 PS2 increase by 60% (share your thoughts)


Pages : [1] 2 3

PeterJensen
10-31-04, 03:49 AM
Ok, so i was over at Guru3d when i saw this post in the driver forum.
http://forums.guru3d.com/showthread.php?s=&threadid=112093

What is your thoughts? I mean 60% is allot :nanahump:

wEEt
10-31-04, 03:55 AM
Proving that with 1 Screenshot is a bit stupid :lame:
But if that's true it would be great that nVIDIA continues supporting their old cards. But we'll see...

theultimo
10-31-04, 04:17 AM
Seems a bit bogus, but we will never know until 70.16 are leeked.

Dazz
10-31-04, 04:20 AM
You should of noted that was with the FX5700Ultra don't get the 6800 owners hopes up.

rewt
10-31-04, 04:37 AM
Pixel shader 2 is pixel shader 2. Whats the difference if its on a 5700 or a 6800?

Anyways I don't believe it. If this type of performance improvement was actually possible, nV would have done it a long time ago.

And that screenshot doesn't even look like a 70 series driver. It was probably edited with Photoshop or something...

RanCorX2
10-31-04, 04:47 AM
they do sound like a good set, but theres not enough proof :(

rewt
10-31-04, 04:49 AM
Theres more proof that says its false rather than true.

1) no 70.16 on NVD
2) driver CPL looks like 6x.xx
3) 70.41 doesn't have this 60% SM2 performance increase

RanCorX2
10-31-04, 04:57 AM
ok i c your point....f a k e.

wEEt
10-31-04, 07:43 AM
I think that these fps shown by fraps are faked. Looks like :lame:

JRd1st
10-31-04, 08:03 AM
...
3) 70.41 doesn't have this 60% SM2 performance increase

I have no idea if these are real or not but the fact that one set doesn't have something doesn't mean beans to me because by the way these sets are leaked it seems that they have different teams working on different sets. And besides the 70.41 are for Quadro not series 6.

So where does that leave us? We'll have to wait and see if something materializes.

Is this fake? http://pees.neostrada.pl/Futuremark%20-%20ORB%20-%20Project%20Details.htm

Looks real to me.

MikeC
10-31-04, 08:17 AM
Theres more proof that says its false rather than true.

1) no 70.16 on NVD
2) driver CPL looks like 6x.xx
3) 70.41 doesn't have this 60% SM2 performance increase

4) In FRAPS, the frame rate counter is left justified against the edge of the screen. In their Painkiller screenshots, there is a gap of about 30 pixels between the edge of the screen and the frame rate counter.

Edit: Right after this post was made, I noticed the images were removed.

ush
10-31-04, 08:18 AM
I've made a new thread on GURU3D with ORB proof... if you want to chek it:

ORB
login: opson@interia.pl
pass: asdf

Guru3D thread (http://forums.guru3d.com/showthread.php?s=&postid=967081#post967081)

rewt
10-31-04, 08:22 AM
Is this fake? http://pees.neostrada.pl/Futuremark%20-%20ORB%20-%20Project%20Details.htm

Looks real to me.

Not to me.

"The checkbox is not visible if a problem was detected in the benchmark project. Possible reasons for this are many, but most are due to an unknown component in your system, or inconsistensies during your benchmark run."

wEEt
10-31-04, 08:28 AM
Fake (xxx)

C'Mon...No one knows anything about those drivers but some strange polish guys...

digitalis
10-31-04, 08:49 AM
lol

ush
10-31-04, 08:55 AM
Hmm ok I can be strange... if you willing to check this -

ORB
login: opson@interia.pl
pass: asdf

and say if it's fake or not I would be really grateful. And I'm sceptic too but this sounds interesting (if the ORB resaults are true).

Skinner
10-31-04, 09:21 AM
Well, the 70.41 gives a good gain in the latest Painkillerdemo.

rewt
10-31-04, 09:43 AM
Hmm ok I can be strange... if you willing to check this -

ORB
login: opson@interia.pl
pass: asdf

and say if it's fake or not I would be really grateful.

"The checkbox is not visible if a problem was detected in the benchmark project. Possible reasons for this are many, but most are due to an unknown component in your system, or inconsistensies during your benchmark run."

PeterJensen
10-31-04, 11:42 AM
look at this tread. It sounds a little be true there :) one can only hope :wong:

http://forums.guru3d.com/showthread.php?s=&threadid=112204

Gaco
10-31-04, 11:49 AM
It might give some performance, but not a 60% increase. Thats simply unrealistic...

Kev1
10-31-04, 11:52 AM
If its true, WOOT :drooling: :drooling: :drooling:

Wonder what flow control is tho? And what it has to do with PS2? And why disabling it gives such huge performance boost?

Well, in any event I hope this is true. Would be very worth the long wait for new drivers :) Im sure it will boost the 6xxx series, but it would be an AWESOME boost for the FX 5xxx series :thumbsup:

Graphicmaniac
10-31-04, 11:59 AM
Why we always have to read new urban legends about next forceware drivers?

why we always have to read about new beta that resolve old problems and create new ones, and are 1% faster than the old???


why can't we finally get a new official release???

uffffffffffffffffffffffffff

Zelda_fan
10-31-04, 11:59 AM
flow control is a PS3.0 feature.

Graphicmaniac
10-31-04, 12:17 PM
but if was really possible get 60%+ performance with ps2 i think Nvidia should have already done a loot of time ago.

with the poor ps2 performance of the gffx serie they have lost a lot of market share% and a lot of money ...

then : or these 70.16 are true and Nvidia has been nvidiot

or are fake and the fake man is the idiot lol

JRd1st
10-31-04, 12:45 PM
It might give some performance, but not a 60% increase. Thats simply unrealistic...

No one said it was a 60% overall performance increase. What I read was a 60% increase in PS2 performance. Pixel shading is not the only thing that GPU's do.

To me, it seems that it's possible that there has always been a coding bug(s) in the drivers that affects PS2 is such a way that it slows it down. Who knows? Maybe some new guy started working at nVidia and he has a different way of looking at code. Anything's possible.