AquaMark3 scores - Page 3

NaRyan · 09-18-2003, 05:52 AM

Using Nvidia's 44.03 drivers I get a score of 24,526 (any above that and I get a bad flickering on my tv capture prog, don't get it with the 51.75 drivers)
And with the 51.75 drivers I get 41,671

Now either they have made a massive improvement or there is a little bit of cheating going on!!!

Mind you my eyesight aint that good so I can't tell any difference between the two :confused:

Soulburner · 09-18-2003, 05:57 AM

They perform optimizations without changing image quality, same way its done to make games run faster and look the same.

It seems ATI was ready for it first, theirs already ran that fast.

Soulburner · 09-18-2003, 07:39 AM

Just scored 44,900 at daily settings:

250 5:4
Card stock

I'll do a max run some other time...

CtrlAltDel · 09-18-2003, 08:04 AM

Quote:

Originally Posted by Soulburner

They perform optimizations without changing image quality, same way its done to make games run faster and look the same.

It seems ATI was ready for it first, theirs already ran that fast.

Well that's not simply true. ATI cards just handle DX9 better than FX cards which simply must take DX8.1 (or 8.0) routes to perform reasonably. I'm not trying to start yet another NVidia vs. ATI war here, as I've owned a pile of NVidia products and I've been satisfied with every one of them, only the FX line blows as somethings gone terribly wrong with their DX9 implementations.

CtrlAltDel

Soulburner · 09-18-2003, 08:44 AM

Quote:

Originally Posted by CtrlAltDel

Well that's not simply true. ATI cards just handle DX9 better than FX cards which simply must take DX8.1 (or 8.0) routes to perform reasonably. I'm not trying to start yet another NVidia vs. ATI war here, as I've owned a pile of NVidia products and I've been satisfied with every one of them, only the FX line blows as somethings gone terribly wrong with their DX9 implementations.

CtrlAltDel

Then explain to me my Aquamark score? It tests all the latest DX9 shaders...if you ask me that's pretty good for a baseline run....

Read this thread....may be worth something.
http://www.nvplanet.com/forums/index...t=0&#entry2553

Aquamark 3 is supposed to be the benchmark that was going to make or break Nvidia and DX9 after the recent results with Half-Life 2. Looks like my NV35 passes with flying colors. :cool:

CtrlAltDel · 09-18-2003, 09:25 AM

Quote:

Originally Posted by Soulburner

Then explain to me my Aquamark score? It tests all the latest DX9 shaders...if you ask me that's pretty good for a baseline run....

What detonators were you using when you got that score? If you were using the 51.xx dets, it has already been discovered that the IQ with these drivers is actually lowered a lot, read NaRyan's post in this thread, first he got 24526 with the older drivers and then he suddenly got 41671 using the newer drivers. You seriously can't believe that that kind of improvement is done just by "optimizing without changing IQ". Besides, it has already been proved in numerous sites, that the 51.xx detonators do lower IQ, and I'm sure you've read it too and seen the numerous screenshots and captures etc, and I'm not talking about any biased fanboysites either, like for example rage3d.com or nvplanet.com.

I respect your choice of Nvidia hardware over ATI hardware, you might gain some compatibility that way. What I'm really trying to say is just that the DX9 features of FX cards are not on par with ATIs 9600-9800 cards, namely the pixel shader 2.0. And the fact is, that if PS 2.0 comes to a crawl, you need to apply a PS 1.4 codepath to attain reasonable speed, and this results in loss of image quality. This is why I chose ATI hardware this time around even though I've favored Nvidia cards in the past. I'm not saying the FX is a ****ty card, it most certainly isn't, it's just somewhat flawed and I'm not trying to put down people who chose differently than I did.

And I respect that you didn't get all medieval on me, as has happened many times when this subject has popped up, I prefer debating and discussion over "OMG taht card si teh suxx0rs" ranting.

CtrlAltDel

Soulburner · 09-18-2003, 10:00 AM

It looked exactly the same to me, both times...

There was just an enormous difference in speed.

Also on the subject of IQ......I tested them in 3DMark01.....my score was down about 800 points, and guess what....all the tests have improved image quality, especially the Nature test. They look better than ever and I actually lost a good 30fps in Nature...

DraGoN-BoY · 09-22-2003, 11:24 AM

P4 2.5 @ 2.6
Radeon 9800 PRO @ Clock Speeds Using Catalyst 3.4 Drivers

PrairieDawg · 09-22-2003, 01:53 PM

Found a way to increase my score from 4036 to 7775 and it only cost me $550 (Canadian) to do it. All I had to do was change out my Compaq board for a Gigabyte P4 Titan GA-8S648FX(-L), my Duron 950 for a Celeron 2.0, my 448 MB of SDRAM for 512 of DDR, and replace my MX440 with a FX5200 8X 128MB DDR video card! Thinking I can prolly do a bit better once I get brave enough to overclock it all.

dude281 · 09-25-2003, 10:18 AM

wow i havent ran it yet with my 44.03 dets, but with 51.75's i got 19,xxx....i forgot exactly what it was, 19 thousand something. i think once i roll back the drivers my score is gonna plummet since i only get 8000 on 3dmark2001se with 44.03 drivers and 8100 with 44.03 dets. we'll see...