AquaMark3@Sep.15

[Rookie]

Quote from Aquamark3.com :

"We are proud to announce the release of our long-awaited benchmark AquaMark3 – The Reality Benchmark for September 15, 3:00 pm GMT. And before you ask: yes, the swatch beat countdown has been set to indicate the time to go.

AquaMark3 is the first commercial benchmark offering both: a real world DirectX 9 game engine and free access to an online database, the AquaMark Result Comparator ARC. It is based on Massive Development’s multi platform krass™ Engine, a real life game engine used by AquaMark2.3 and games like subwater action shooter AquaNox, AquaNox2: Revelation and the upcoming strategy hit Spellforce.

The basic application will be available as a free download on a large number of mirror sites around the world. This version supports the AquaMark TRISCOREâ„¢ benchmark test and free online access to the ARC.

The AquaMark online shop will offer a range of extended versions for private as well as for commercial use. All of these licenses contain the capability to check the customer’s system extensively. They also comprise a wide range of unique features like pixel rate tests and pixel shader complexity tests. Prices start at $9.95 (€ 9.95 for EURO countries) for the fully featured Professional Version."

Their slogan:AquaMark3 - The Reality Benchmark

 

[g__day]

I wonder if major review sites get their copies about now so they can write in-depth articles to appear by the time we get our copies online?

I asked if MD have seen Detonator 50.xx yet - but they are being non commital in their answers, just talking about developer code drops.

http://arc.aquamark3.com/forum/showthread.php?s=&threadid=88

 

[mboeller]

You can already find a few benchmarks online. Looks good for the FX5900Ultra. Only ~15% slower than a R9800Pro :

# GF FX 5900 Ultra
AM3
640x480, No AA, No AF

XP 3200+: 44,65 fps
P4 3.2: 51,51 fps

++++++++++++++++++++++++++++

1.024x768, No AA, No AF

XP 3200+: 41,06 fps
P4 3.2: 45,85 fps

++++++++++++++++++++++++++++

1.024x768, 2X FSAA, 4:1 AF

XP 3200+: 32,68 fps
P4 3.2: 33,78 fps

-----------------------------------------------------------

# 9800 Pro

640x480, No AA, No AF

XP 3200+: 51,20 fps
P4 3.2: 59,64 fps

++++++++++++++++++++++++++++

1.024x768, No AA, No AF

XP 3200+: 44,40 fps
P4 3.2: 49,53 fps

++++++++++++++++++++++++++++

1.024x768, 2X FSAA, 4:1 AF

XP 3200+: 38,60 fps
P4 3.2: 39,7 fps

_______________

Source : 3DCenter.de posted from the PC-Games-Hardware journalist (?) Thilo

A complete benchmark writeup can be found in the new edition of PCGH already

 

[Joe DeFuria]

Hmmm...

Looking at the differences in the scores for "Athlon vs. Intel" CPUs leads me to believe this benchmark has some "not insigificant" CPU limitations.

At 1024x768...no AA and no AF, there's roughly 10% difference between an Athlon 3200+ and a P4 3.2?

There's only about a 14% difference between 640x480 and 1024x768.

I wish they benched some more aggressive resolution / AA settings though....

 

[Klux]

I read that the Professional version is $9.95 only. That's a sure buy for me

 

[digitalwanderer]

Cool beans, for $9.95 I'll be picking up a pro copy too.

 

[parhelia]

Can't wait till Sep 15

 

[Ante P]

From what I've read the scores (from Aquanox 2) published up to now don't have any PS2.0 in them at all.
A future patch will add PS2.0 support. (And Aquamark3 will of course ship with the support).

Just read at NFI that Det50 might launch on the 15th..

 

[PatrickL]

Before using Aquanox 3 as a benchmark, would be nice to know what it tests exactly and how it handles the differents card. Same path for all cards, PP hints and so on no ?

 

[g__day]

PatrickL - agreed, it takes 2 minutes to join their forums and ask them - hint, hint!

 

[PatrickL]

I read your posts on MD forums and that s why i said we may need to wait to know more about AM3.
They are not answering with any precision to how many PS2 are in, how much exactly PP hint are used and so on.

 

[g__day]

PatrickL - yes I know, I get the feeling they openned their forums and didn't expect to receive any really meaty questions or issues and aren't quite sure how to respond.

Normally posts there get around 30-50 views then die. Mine are in the hundreds or thousands and loaded with as much unbiased facts as I can locate, so they have better chance of being answered.

My theory is if more people join - and add their voice to my questions - or ask their own - there will be a better chance of a substantive reply. Its easy to ignore one or two Australians, harder to ignore 100 folk from an international community all asking insightful questions.

On the up side - its management and technical people answering my questions - not PR mouth pieces so occassionally something juicy comes out.

 

[g__day]

Will ATi and NVidia cards execute exactly the same shader code?

http://arc.aquamark3.com/forum/showthread.php?s=&threadid=101

Answer from Alexander Jorias / Managing Director / Massive Development

the code path are not optimized in a vendor-specific way. the capabilites of the gfx card are considered when the engine determines which set of shaders is used for a spefic material / effect. so if two cards report the same capability flags for a given situation exactly the same code / vs / ps path are used.

... we are comparing apples vs apples

 

[digitalwanderer]

Will ATi and NVidia cards execute exactly the same shader code?

http://arc.aquamark3.com/forum/showthread.php?s=&threadid=101

Answer from Alexander Jorias / Managing Director / Massive Development

the code path are not optimized in a vendor-specific way. the capabilites of the gfx card are considered when the engine determines which set of shaders is used for a spefic material / effect. so if two cards report the same capability flags for a given situation exactly the same code / vs / ps path are used.

... we are comparing apples vs apples

What if the two cards DON'T report the same capability flags for a given situation...I thought that was the root of the problem.

 

[g__day]

Once again their MD replies:

"then the engine decides which fallback is to be used. i.e. if there is no 8x multitexture caps it falls back to 4x multitexture with multipass or even 2x multitexture with multipass. "

So asking abit more Alexander replied

"

quote g__day:
--------------------------------------------------------------------------------
1) Detect card
2) Check capabilities - assign DX9 shaders based on capabilities detected
3) user accepts or modifies visual settings defaults
4) run benchmark with assigned shaders and requested visual settings
--------------------------------------------------------------------------------

Alexander

kind of that. if you use advanced settings you are able to tweak am3's visual settings and by this you might overrule am3 default choice for your specific gfx card.

however if you do a TRISCORE benchmark run (wich results in an overall score, a gpu score and a cpu score) the decision cascade of am3 can not be overruled (if you could do so the scores wouldn't be comparable at all)."

PS

If you visit the Massive Development forums for AM3 - the developers are really starting to answer questions well now. So if you have a buring question - join up and ask - like I have, and please share the answers here.