1. Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

AquaMark questions

Discussion in 'Overclocking & Cooling' started by Alec§taar, Aug 21, 2006.

  1. Alec§taar New Member

    Joined:
    May 15, 2006
    Messages:
    4,677 (1.56/day)
    Thanks Received:
    94
    Location:
    Someone who's going to find NewTekie1 and teach hi
    When you run this, I need to know a few things:

    1.) Do you just run it in DEFAULT mode during its analysis test phase? If so, what menu options do you typically pick when navigating to run it initially...

    2.) What do you do to speed it up (i.e.-> Do you load test specific videodriver profiles, etc. &/or turn off AntiAliasing-aniIsoTropic Filtering etc.)??

    3.) What is a GOOD solid resultscore from it based on cards you have seen it run on (average, good, best etc. & what kind of vidcard was it run on + CPU combo'd w/ it)???

    :)

    * Unlike 3dMark06, I can run this one (3dMark 06 won't run here because I do not keep 1gb of RAM on my mobo (only 512mb here))... so, I need to ask these questions.

    (Thanks for answers)

    APK

    P.S.=> Oh, lastly: Has there been a "formal" benchmark test done here using this program already (long time forums goers would know I would think)???? apk
  2. Ketxxx

    Ketxxx Heedless Psychic

    Joined:
    Mar 4, 2006
    Messages:
    11,510 (3.75/day)
    Thanks Received:
    570
    Location:
    Kingdom of gods
    AM3 is an old tool now, none the less its nice to see a score of 84,000+ :p (which actually means your average FPS was around 84) the only changes to make to AM3 before running is resolution, 1024*768 is what everybody ran it at, no AA or AF. AM3 does still have one useful feature about it, its very CPU dependent, around 50%, and 50% graphics. meaning in the final score you will be able to note if your CPU is bottlenecking your graphics. Typically you want your CPU to be a good 500pts ahead of your graphics score, if its not, theres a good chance your CPU is bottlenecking your graphics.
  3. giorgos th.

    giorgos th. New Member

    Joined:
    Nov 10, 2005
    Messages:
    1,543 (0.48/day)
    Thanks Received:
    117
    Location:
    Athens - Hellas
    in my opinion
    1.in order to have a fair result you run it in the default mode,because everyone does it that way.
    2.to speed it up you can do these things (software based): turn off AF,use high performance in your cards driver and through riva use LOD +3.
    3.in fact,there is not a score you must bare in mind because AM3 is depending to a lot of stuff.vga-ram-cpu etc etc.
  4. Ketxxx

    Ketxxx Heedless Psychic

    Joined:
    Mar 4, 2006
    Messages:
    11,510 (3.75/day)
    Thanks Received:
    570
    Location:
    Kingdom of gods
    actually, am3 is dependent on graphics, ram and cpu, thats the only hardware involved with am3 :p and tell me the last person that ran am3 in 800*600 resolution, i was running 1024*768 way back when a FX5700 Ultra was considered pretty good :eek:
  5. giorgos th.

    giorgos th. New Member

    Joined:
    Nov 10, 2005
    Messages:
    1,543 (0.48/day)
    Thanks Received:
    117
    Location:
    Athens - Hellas
    the default test is running at 1024,4x AF.that`s what i mean by default.not the menu stuff.
  6. tigger

    tigger I'm the only one

    Joined:
    Mar 20, 2006
    Messages:
    10,183 (3.33/day)
    Thanks Received:
    1,399
    all my me and my mates run it at the default settings.the only way we use to up your score is by clocking your cpu or gfx card.(i suppose tweaking aa,af etc is ok,but everyone has to run the same setings).
  7. Alec§taar New Member

    Joined:
    May 15, 2006
    Messages:
    4,677 (1.56/day)
    Thanks Received:
    94
    Location:
    Someone who's going to find NewTekie1 and teach hi
    Well, then it sounds pretty simple to use, and all of the gains are MOSTLY driven by drivers tweaks to the vidcard (& CPU if possible) then I take it... cool, & which is HOW it ought to be then.

    I think once I update all of my ASUS A8N-SLI Premium drivers this week & my BIOS as well for this mobo?

    We're going to do another one of these for the heck of it here, IF such a test was already done here (hence, my question in my P.S. in the first post in this thread).

    :)

    * I just batted off an (iirc) HIGH 83000's mark, with a ton of things running besides this benchmark program itself browsers, mediaplayer (OASIS ROXX!), & Outlook Express, an instance of Explorer.exe filemgt. etc.

    (... & NOT "playing any tricks" like disabling services, running this program @ HIGH CPU PRIORITY etc. either)

    Just using I guess what is the default settings (never used it before, felt like I was navigating a game in fact thru its menu) & just clicked on the "start test/analysis" button (not direct quote of the name, but I think you get my point).

    I'd wager I can go higher, because I ran my NVidia drivers @ "Max Quality" settings, not "Max Performance" oriented ones (Saw profiles in its driver settings for 3dMark 05/06 out the wazoo, but none for AquaMark, so... I had to ask the questions I did).

    APK

    P.S.=> Anyhow, if I do such a test? It will be in another thread, I just had to do my prepwork here... gotta against some of the systems I see here & based on their results in other benchmark tests, you know?

    Thanks for info. all...

    Well, off to update my mobo driverset & then see what's-what more w/ this benchmarking program... apk
    Last edited: Aug 21, 2006
  8. Alec§taar New Member

    Joined:
    May 15, 2006
    Messages:
    4,677 (1.56/day)
    Thanks Received:
    94
    Location:
    Someone who's going to find NewTekie1 and teach hi
    Well, after putting on the "latest/greatest" chipset drivers & BIOS revision here, I got a score of:

    GFX 13,467
    CPU 11,574
    85,120

    :)

    * For whatever that all means...

    (I used a 251FSB x 11 multiplier, & can go MUCH higher there (up to 258FSB before instabilities result, on my OLD BIOS & DRIVERS @ least, possibly higher still NOW with updated ones, but... I will try it later with the faster settings!))

    APK

    P.S.=> It also doesn't LET you raise the CPU priority on it oddly via taskmgr.exe! apk
  9. Alec§taar New Member

    Joined:
    May 15, 2006
    Messages:
    4,677 (1.56/day)
    Thanks Received:
    94
    Location:
    Someone who's going to find NewTekie1 and teach hi
    Well, after turning up my FSB to 256 & 11 multiplier = 2815mhz

    (See GPU/NVidia GeForce 7900GTX OC settings in my signature for the o/c on them)

    My AquaMark 3 results were (w/ ALL DRIVERS SETTINGS TO ABSOLUTE HIGH QUALITY MAX):

    GFX = 13,485
    CPU = 11,569
    85,207

    &

    My AquaMark 3 results were (w/ ALL DRIVERS SETTINGS TO ABSOLUTE HIGH PERFORMANCE MAX):

    GFX = 20,466
    CPU = 11,723
    109,262

    :)

    * That's it for now, but... I don't KNOW if that's "Good" or "Bad" really!

    APK
    Last edited: Aug 21, 2006
  10. tigger

    tigger I'm the only one

    Joined:
    Mar 20, 2006
    Messages:
    10,183 (3.33/day)
    Thanks Received:
    1,399
    my best is-
    10,686 on gfx and
    13,103 on cpu.
    total=74,870

    your well kicking me on the gfx score and i'm kicking you on the cpu score.lol

    your 7900gtx whups my x800gto tho'

    nice score
  11. Alec§taar New Member

    Joined:
    May 15, 2006
    Messages:
    4,677 (1.56/day)
    Thanks Received:
    94
    Location:
    Someone who's going to find NewTekie1 and teach hi
    Ok, cool... I wasn't sure what a "STRONG SCORE" is in that test, or not, or if mine below (reposted here for reference) was good:

    My AquaMark 3 results were (w/ ALL DRIVERS SETTINGS TO ABSOLUTE HIGH QUALITY MAX):

    GFX = 13,485
    CPU = 11,569
    85,207

    &

    My AquaMark 3 results were (w/ ALL DRIVERS SETTINGS TO ABSOLUTE HIGH PERFORMANCE MAX):

    GFX = 20,466
    CPU = 11,723
    109,262

    :)

    * Thanks for the feedback Tigger69!

    (I may do a benchmark shootout on THIS tool, & I like the one in Fresh Diagnose too (Free, & much like SiSoft Sandra stuff))

    Still, I would like to see what a HIGH-END ATI can do on that test, paired up w/ a CPU of my type...

    * Made an assumption, need clarification/verification/feedback for those that are "in-the-know":

    I think the test LOOKS like DirectX coded gaming stuff!

    BUT is it?

    Or is it an OpenGL coded test??

    I am not sure... does anybody know?

    APK

    P.S.=> Your CPU doesn't "WHIP" mine bro! Heck: We battle for 10th's of points to get ahead of one another on the tests like ScienceMark 2.0 &/or Super-Pi!

    The margin's separating us on BOTH those tests were like 1%!!! LOL... no doubt about it that You & I are VERY close on the CPU account...

    (& imo? Your mobo & its chipset MAKE that CPU the best it can be combined w/ your o/c know-how, more than the CPU itself you have).

    I got that based on how far you kept ahead of others w/ CPU's like yours... apk
    Last edited: Aug 21, 2006
  12. b1lk1

    b1lk1 New Member

    Joined:
    May 4, 2005
    Messages:
    688 (0.20/day)
    Thanks Received:
    17
    Location:
    Ontario, Canada
    Dual cores always score at least 1K less in the CPU score than a single core CPU. While your score isn't bad, it is still at touch low for your hardware. I'd expect to see somewhere around 120K with a good overclock. Aquamark3 is heavily OpenGL dependant, that is why Nvidia cards shine and ATI X1x00 cards have problems scoring the same. It was desperately in need of a patch when they pulled the plug and now it is a novelty more than anything.
  13. Alec§taar New Member

    Joined:
    May 15, 2006
    Messages:
    4,677 (1.56/day)
    Thanks Received:
    94
    Location:
    Someone who's going to find NewTekie1 and teach hi
    Really? In what test are you meaning??

    (Curious on this one large, thanks for feedback!)

    On which test?

    I know that my RAM is "holding me back/down" from what the folks here that taught me about o/c'ing said... Tatty One, 33, giorgos.th, Ketxxx, POGE, & others.

    I think I can get near to that 120k mark on a COLDER day & certainly probably with better RAM than I have!

    However?

    I just noticed that folks that scored HIGHER than I did on ScienceMark 2.0 (giorgos.th in fact, in his sig)? Didn't score as high on AquaMark as I have...

    Ah, thank you for the feedback on its technologies used, & YES, the history trivia as well (love that stuff).

    APK
  14. Ketxxx

    Ketxxx Heedless Psychic

    Joined:
    Mar 4, 2006
    Messages:
    11,510 (3.75/day)
    Thanks Received:
    570
    Location:
    Kingdom of gods
    No, they made the patch before the plug was pulled enabling the newer cards to work with AM3 right

    Attached Files:

  15. b1lk1

    b1lk1 New Member

    Joined:
    May 4, 2005
    Messages:
    688 (0.20/day)
    Thanks Received:
    17
    Location:
    Ontario, Canada
    THat patch was only to read the clocks properly on X8x0 cards and the Nvidia 7800 series cards. The patch I am referring to would have taken away the Open GL advantages of Nvidia over ATI in the actual score for X1x00 cards VS 7900 series cards. It was always slightly Nvidia biased, but now it is almsot completely Nvidia biased. Look at my 3D Mark scores. The best AQ3 score I could drag out of my machine was 118K. My machine should easily do 130K and it never could.
  16. Ketxxx

    Ketxxx Heedless Psychic

    Joined:
    Mar 4, 2006
    Messages:
    11,510 (3.75/day)
    Thanks Received:
    570
    Location:
    Kingdom of gods
    patch wasnt just for that, it stopped the program from bombing out with a fake error with newer cards
  17. Alec§taar New Member

    Joined:
    May 15, 2006
    Messages:
    4,677 (1.56/day)
    Thanks Received:
    94
    Location:
    Someone who's going to find NewTekie1 and teach hi
    Putting up actual photograph of score

    [​IMG]

    :)

    * That's the photo of my highest score yet, for posterity's sake & verification...

    APK
  18. giorgos th.

    giorgos th. New Member

    Joined:
    Nov 10, 2005
    Messages:
    1,543 (0.48/day)
    Thanks Received:
    117
    Location:
    Athens - Hellas
    it doesnt mean anything.its a different test.you`ve got a better vga (GTX vs GT) and you`ve got the dual core..the strange thing would be to have a lower result from mine in AM3.
    if you had a better memory your sciencemark result would be over 1550 for sure..
  19. Alec§taar New Member

    Joined:
    May 15, 2006
    Messages:
    4,677 (1.56/day)
    Thanks Received:
    94
    Location:
    Someone who's going to find NewTekie1 and teach hi
    Agreed, but only to some extent - Ketxxx iirc, above stated this test is a mixture of CPU and GRAPHICS... the cpu part comes into play on BOTH tests (ScienceMark 2.0 & this one).

    Ah, so you're an NVidia man too? I wasn't aware of that...

    DualCore would help me on both tests, assuming BOTH are multithreaded & coded properly to take advantage of it. I know ScienceMark 2.0 uses multiple threads, but not sure if AquaMark 3 does.

    Based on a vidcard comparison alone, yes. Since the test is based in BOTH CPU & GRAPHICS, per Ketxxx's statement (iirc he stated that earlier in this thread), that may NOT be entirely true.

    You certainly "blew me away" in ScienceMark 2.0, a memory/cpu intensive test, & this makes me assume you actually have a better/more powerful CPU for that type of work, than I do, & that you know how to O/C better also (& I know you have "top-notch" RAM too).

    Yea, I know man... it's a matter of me doing the research & shopping around for better RAM... you guys've convinced me of THAT being the case for my SM2 scores being lower than they can be, no matter WHAT kind of memory timings alterations I do.

    (x800 has a CPU like mine, watercooled though, & he batted off 20 points higher than I did in ScienceMark 2.0 - between THAT & better RAM? I'd wager I might hit into higher scores than dj dn did using an Opteron (he scored 1461, I scored 1441)).

    :)

    * ANYHOW/Anyways - there's my latest result in this test, above...

    APK
    Last edited: Aug 23, 2006
  20. giorgos th.

    giorgos th. New Member

    Joined:
    Nov 10, 2005
    Messages:
    1,543 (0.48/day)
    Thanks Received:
    117
    Location:
    Athens - Hellas
    in gfx score in AM3 you`ve got 20K and i`ve got ~16K...
    i beat you in cpu score clearly because of my cpu mhz and especially of my memory.
  21. Alec§taar New Member

    Joined:
    May 15, 2006
    Messages:
    4,677 (1.56/day)
    Thanks Received:
    94
    Location:
    Someone who's going to find NewTekie1 and teach hi
    Then I made some correct assumptions above it appears, right?

    :)

    APK
  22. giorgos th.

    giorgos th. New Member

    Joined:
    Nov 10, 2005
    Messages:
    1,543 (0.48/day)
    Thanks Received:
    117
    Location:
    Athens - Hellas
    some correct yes.
    CHANGE YOUR MEMORY and you`ll have a whole different system..
  23. Alec§taar New Member

    Joined:
    May 15, 2006
    Messages:
    4,677 (1.56/day)
    Thanks Received:
    94
    Location:
    Someone who's going to find NewTekie1 and teach hi
    [​IMG]

    I thought so. This test, according to Ket's statements, is both a mix of CPU (where you have me, & in RAM too) & GRAPHICS (where I have you).

    Makes sense given our hardware makeup & scores.

    Again: I agree 110% here, & so do the folks that helped me on o/c'ing (Tatty One, POGE, & Ketxxx).

    It's just a matter of "zeroing-in" on the BEST maker/oem (Geil, OCZ, Mushkin, &/or Patriot are the contenders) for my system, stepping up from DDR-400 Corsair "ordinary/non-overclocking ready" RAM.

    APK
  24. giorgos th.

    giorgos th. New Member

    Joined:
    Nov 10, 2005
    Messages:
    1,543 (0.48/day)
    Thanks Received:
    117
    Location:
    Athens - Hellas
    all the ram makers you give are top.you`ll just need some luck to get the right chips on it.
    except you can hand-pick them..
  25. Alec§taar New Member

    Joined:
    May 15, 2006
    Messages:
    4,677 (1.56/day)
    Thanks Received:
    94
    Location:
    Someone who's going to find NewTekie1 and teach hi
    Well,

    Time to see how my machine "stacks up" vs. any of yours in a gaming oriented benchmark test examination!

    So... I decided to start one up using this program, here:

    http://forums.techpowerup.com/showthread.php?t=16136

    :)

    * REMEMBER: "ONLY THE FASTEST SYSTEMS NEED APPLY!"

    APK

Currently Active Users Viewing This Thread: 1 (0 members and 1 guest)

Share This Page