Aquamark Preview at Guru3D

[TheGreatGrapeApe]

TheGURUof3D has a <A HREF="http://www.guru3d.com/article.php?cat=article&id=76" target="_new">review of the Auqamark3 Benchmarking tool</A>.

Just thought those of us who;ve been waiting would like to know. The rest of you can... well read and maybe learn something. Looks nice sofar and I like the division of performance indicators, and there are so many options. Yipee. I hope they are all in the widely distributed version. We shall see. And we shall also see how nV tries to kill this like they did 3Dmk03. Or at least that's what I expect, and was concerned about in our last thread.

<A HREF="http://www.guru3d.com/article.php?cat=article&id=76" target="_new">http://www.guru3d.com/article.php?cat=article&id=76</A>

Edit: There was another post I just noticed that had a <A HREF="http://www.driverheaven.net/articles/aquamark3/index.htm" target="_new">review from DriverHeaven</A> also.


- You need a licence to buy a gun, but they'll sell anyone a stamp <i>(or internet account)</i> ! <A HREF="http://www.redgreen.com" target="_new"><font color=green>RED</font color=green> <font color=red>GREEN</font color=red></A> GA to SK <P ID="edit"><FONT SIZE=-1><EM>Edited by TheGreatGrapeApe on 09/13/03 02:40 PM.</EM></FONT></P>

 

[ufo_warviper]

Interesting that the GeForce FX cards scored the simlarly from the old and new Detonator versions but the new Detonator has a definite but small performance boost. but I was especially surprised that the GeForce FX was ahead of the Radeon 9800 Pro in a DX9 game with <b><font color=red>old</b></font color=red> Detonator drivers! <b><font color=red>WTF!?!?!?!?!?</b></font color=red>


Somebody explain this.

My OS features preemptive multitasking, a fully interactive command line, & support for 640K of RAM!

 

[TheGreatGrapeApe]

Well the question is how much DX9 elements are employed, how it works, and also whether there are portions where the two cards differ in their approach (precision levels etc). Definitely a more in-depth disection will be needed to understand the benchmark as a whole. Thiis is especially needed as a benchmarking utility versus a bunch of in-gmae tests. However I'm sure like we've ben saying ther is only small facets where the FX lags (hey rememeber it does VERY well in alot of areas) and the question will be what exactly is emplyed and how does each engine/driver tackle the benchmark. I'm sure more info will come out in due time, it's not as 'shocking' as the HL2 stuff though, and I suspect that it's just that there are some portions of the benchmark that the nV hardware just does better, optimized or not.

Gotta run.


- You need a licence to buy a gun, but they'll sell anyone a stamp <i>(or internet account)</i> ! <A HREF="http://www.redgreen.com" target="_new"><font color=green>RED</font color=green> <font color=red>GREEN</font color=red></A> GA to SK

 

[bandikoot]

Since it is my understanding that the aquamark test can default back to the older versions, is it possible that we are seeing something like that here, or is the fx series forced to run dx9, ps 2.0 etc? I suppose there really isn't an answer for that yet. *sigh*

 

[HardWareBoss]

Have you downloaded an ran this yet?

 

[TheGreatGrapeApe]

NO, it is not availible to the public until tomorrow.

<A HREF="http://" target="_new">Here's their site</A> and the CLOCK is ticking (literally!). 119500 seconds to go or so it seems (weird clock).

However I will likely download it before going to bed tomorrow, or before the Sunday Night Football game.


- You need a licence to buy a gun, but they'll sell anyone a stamp <i>(or internet account)</i> ! <A HREF="http://www.redgreen.com" target="_new"><font color=green>RED</font color=green> <font color=red>GREEN</font color=red></A> GA to SK

 

[barnettgs]

Another 3D benchmarking program? huh?

Believe me, I think you would be way better off with frame per sec rating based on any real games running which reflects the real world testing.

I don't know how you would want to know about another brenchmarking programs because i have lost faith completely in any 3D brenchmarking programs since Nvidia was found to be cheating! I think any 3d brenchmarking like Aquamark is for guidelines only and should not be taken for granted when buying a card.

 

[TheGreatGrapeApe]

i have lost faith completely in any 3D brenchmarking programs since Nvidia was found to be cheating!

And this is different than taking FPS in a game in what way?

Have you heard of an obscure game called UT2K3? Yeah? Well it has optimizations up the ying-yang, so how is that different. People USED to think that synthetic benchmarks were more suceptible to fl-optimizations, but it turned out whatever the current 'buyer's indicator' is will get it. Some people argued, 'well that's good then it's optimized for the game which is good for players', except that nV (Hmm pattern?) once again fl-optimized and took away user preference to get better scores, and lowered IQ with no way for the average (and even many of the most advanced) gamers to disable it.

So you are left withthe reality that WHATEVER you bench with, once it becomes a 'standard' then is a target for fl-optimizations. SUX Hard!

I like benchmarks as the more there are it seems the more info we have. Hey, without the HL2 'benchmark' we wouldn't have our most recent controversy to occupy our empty lives.

And ALL benchmarks should help YOU determine the stability and performance increases offered by your various setups and driver combinations, and that should help you tweak YOUR personal performance. Yes you could use fraps, but you would never know exactly where you affected optimizations if you were only left with an FPS.

Well of course that's just my 2 frames worth.



- You need a licence to buy a gun, but they'll sell anyone a stamp <i>(or internet account)</i> ! <A HREF="http://www.redgreen.com" target="_new"><font color=green>RED</font color=green> <font color=red>GREEN</font color=red></A> GA to SK