Bad 6600GT performance...I think

Archived from groups: alt.comp.periphs.videocards.nvidia (More info?)

Maybe I am wrong here but from other benchmark scores I have
seen, mine seem kind of low. I can't understand why either. I thought
for sure with a CPU running like a 3650+ with over a 200 fsb I would
have better scores then these.

NF7-S
Turbo Memory Settings
Athlon Mobile Barton 208 fsb x 12 = 2.5 Ghz = 3650+
512 Megs

66.93 Nvidia Drivers

Default 6600GT speed 500/900

3DMark 2001 = 16,600
3DMark 2003 = 7,630
3DMark 2005 = 3,030


6600GT overclocked 560/1.13

3DMark 2001 = 17,500
3DMark 2003 = 8,675
3DMark 2005 = 3,400

Any ideas here...............Thanks?
26 answers Last reply
More about 6600gt performance think
  1. Archived from groups: alt.comp.periphs.videocards.nvidia (More info?)

    PRIVATE1964 wrote:
    > Maybe I am wrong here but from other benchmark scores I have
    > seen, mine seem kind of low. I can't understand why either. I thought
    > for sure with a CPU running like a 3650+ with over a 200 fsb I would
    > have better scores then these.
    >

    FWIW those are (slightly) higher scores than I get on my Athlon XP 3200+
    512MB PC2700 system. People can make these cards go faster, but I can
    get >50fps in Doom 3 at the resolution and detail level I want so why
    bother?

    Do you have any other processes running that are eating memory or CPU
    time? Otherwise, probably the best bang for buck performance increase
    is upgrade RAM to 1GB and make sure it is running synchronously with the
    FSB.

    Ian
  2. Archived from groups: alt.comp.periphs.videocards.nvidia (More info?)

    PRIVATE1964 wrote:
    > Maybe I am wrong here but from other benchmark scores I have
    > seen, mine seem kind of low. I can't understand why either. I thought
    > for sure with a CPU running like a 3650+ with over a 200 fsb I would
    > have better scores then these.
    >
    > NF7-S
    > Turbo Memory Settings
    > Athlon Mobile Barton 208 fsb x 12 = 2.5 Ghz = 3650+
    > 512 Megs
    >
    > 66.93 Nvidia Drivers
    >
    > Default 6600GT speed 500/900
    >
    > 3DMark 2001 = 16,600
    > 3DMark 2003 = 7,630
    > 3DMark 2005 = 3,030
    >
    >
    > 6600GT overclocked 560/1.13
    >
    > 3DMark 2001 = 17,500
    > 3DMark 2003 = 8,675
    > 3DMark 2005 = 3,400
    >
    > Any ideas here...............Thanks?
    >
    Those're the right scores for a 6600GT.
  3. Archived from groups: alt.comp.periphs.videocards.nvidia (More info?)

    Thanks for the reply.

    Nope nothing running in background and I always set ram to run
    synchronously. I might try shutting down as many windows prcosseses as
    I can just to be sure though.

    I have fast writes off in the bios, seems to lower my scores. I also
    find running at 4X instead of 8X I get better scores... really weird
    there. AGP aperture doesn't make a difference what I set it to so I
    usually have it at 64 or 128.

    I am only running 512Megs, but would increasing that to 1Gig help with
    the benchmarks? That is something I am truly wondering.

    Would a different version chipset driver make a difference?
  4. Archived from groups: alt.comp.periphs.videocards.nvidia (More info?)

    What are you running again and what are your system specs?

    Please post your scores again I forgot what they were.

    I really don't understand these benchmarks.

    I went from a 2500+ (166fsb) to a 3650+ (208 fsb) and running the
    6600GT at stock the score for;

    3DMark 2005 stayed around the same. = 3,100 and 3,030

    The score for 3DMark 2003 went up around 450 points.= 7,200 and 7,630

    The score for 3Dmark 2001 went up around 3,000 = 13,500 and 16,600


    Overclocked at 560GPU / 1.13 Memory here is my Aquamark3 score
    Aquamark3 = 53.217
    GFX = 7926
    CPU = 8,100
  5. Archived from groups: alt.comp.periphs.videocards.nvidia (More info?)

    I kill u in all benchmarks but 2001 lol WTF???


    "PRIVATE1964" <private1964@aol.com> wrote in message
    news:1108184786.441845.22920@c13g2000cwb.googlegroups.com...
    > Maybe I am wrong here but from other benchmark scores I have
    > seen, mine seem kind of low. I can't understand why either. I thought
    > for sure with a CPU running like a 3650+ with over a 200 fsb I would
    > have better scores then these.
    >
    > NF7-S
    > Turbo Memory Settings
    > Athlon Mobile Barton 208 fsb x 12 = 2.5 Ghz = 3650+
    > 512 Megs
    >
    > 66.93 Nvidia Drivers
    >
    > Default 6600GT speed 500/900
    >
    > 3DMark 2001 = 16,600
    > 3DMark 2003 = 7,630
    > 3DMark 2005 = 3,030
    >
    >
    > 6600GT overclocked 560/1.13
    >
    > 3DMark 2001 = 17,500
    > 3DMark 2003 = 8,675
    > 3DMark 2005 = 3,400
    >
    > Any ideas here...............Thanks?
    >
  6. Archived from groups: alt.comp.periphs.videocards.nvidia (More info?)

    Thanks.

    I think its time to try a few more different sets of drivers.
  7. Archived from groups: alt.comp.periphs.videocards.nvidia (More info?)

    PRIVATE1964 wrote:
    > Thanks for the reply.
    >
    > Nope nothing running in background and I always set ram to run
    > synchronously. I might try shutting down as many windows prcosseses as
    > I can just to be sure though.
    >
    > I have fast writes off in the bios, seems to lower my scores. I also
    > find running at 4X instead of 8X I get better scores... really weird
    > there. AGP aperture doesn't make a difference what I set it to so I
    > usually have it at 64 or 128.
    >
    > I am only running 512Megs, but would increasing that to 1Gig help with
    > the benchmarks? That is something I am truly wondering.

    Not sure about benchmarks, but makes a difference to current-gen games
    like Half-life 2 and Doom 3.

    > Would a different version chipset driver make a difference?
    >

    Worth trying I suppose, but I didn't notice any difference on my
    VIA-based system.

    Ian
  8. Archived from groups: alt.comp.periphs.videocards.nvidia (More info?)

    Did you happen to benchmark any games with the old cpu in? I'd be curious
    to see that comparison.

    DaveL


    "PRIVATE1964" <private1964@aol.com> wrote in message
    news:1108274458.546020.167030@l41g2000cwc.googlegroups.com...
    > What are you running again and what are your system specs?
    >
    > Please post your scores again I forgot what they were.
    >
    > I really don't understand these benchmarks.
    >
    > I went from a 2500+ (166fsb) to a 3650+ (208 fsb) and running the
    > 6600GT at stock the score for;
    >
    > 3DMark 2005 stayed around the same. = 3,100 and 3,030
    >
    > The score for 3DMark 2003 went up around 450 points.= 7,200 and 7,630
    >
    > The score for 3Dmark 2001 went up around 3,000 = 13,500 and 16,600
    >
    >
    > Overclocked at 560GPU / 1.13 Memory here is my Aquamark3 score
    > Aquamark3 = 53.217
    > GFX = 7926
    > CPU = 8,100
    >
  9. Archived from groups: alt.comp.periphs.videocards.nvidia (More info?)

    Nope sorry no games. I have found that even though they are only
    benchmarks they do still give a good idea about performance.

    When I was running my 4200, if the score in 3DMark dropped because of
    drivers by only 500 points I could notice that in actual games.
  10. Archived from groups: alt.comp.periphs.videocards.nvidia (More info?)

    I have checked on Futuremark, that's were I am saw a lot of really high
    scores. Then I check out the systems and they are running same bus
    speed, same CPU speed but their scores are really really high.

    I have tried alot of drivers and so far the best ones seem to be the
    71.21's for me.
    This is with my MSI 6600 overclocked 565/1.13

    3DMark 2001 = 17, 075
    3DMark 2003 = 8,650
    3Dmark 2005 = 3,737
  11. Archived from groups: alt.comp.periphs.videocards.nvidia (More info?)

    That's interesting how your score jumped with no CD-ROM.
    Even if you get a slightly higher score with fast writes, I don't think
    it is worth the trouble it might cause IMO.

    The fastest drivers I found with the least amount of problems are the
    71.21's. They give you clear water in HL2, that most of the other 70+
    drivers do not. Plus there is a good boost in the 2005 score.

    I know your using a 500 Watt supply, but is it a quality supply?
    Because you saw in improvement without the CD-ROM drive. I would
    suggest testing the card with as much devices disconnected as possible.
    That would help to rule out a power supply issue.

    With your system specs you should be hitting at least 3000 with no
    problems. What is the bus speed for that P4.
    Your 2703 score might not be too bad also. Depending on all your bios
    settings. I score around 3000 with the 66.93 drivers.
  12. Archived from groups: alt.comp.periphs.videocards.nvidia (More info?)

    With the drivers your using I think you should be hitting at least 3000
    not overclocked on 3DMark 2005. I get 3450 with no overclock, I think
    your system should get the same or close to it.

    That power supply looks nice, but having a quality supply matters a
    lot. There is no brand name on the supply so I have no way to tell if
    it is probably a good supply. A lot of power supplies will list wattage
    that they really can not do, or do it well enough.

    So I still suggest you unplug as much as possible and then test the
    video card again. I would also suggest making sure your memory is
    running at the proper speed and the fastest timings in the bios.
  13. Archived from groups: alt.comp.periphs.videocards.nvidia (More info?)

    "PRIVATE1964" <private1964@aol.com> wrote in message
    news:1108184786.441845.22920@c13g2000cwb.googlegroups.com...
    > Default 6600GT speed 500/900
    >
    > 3DMark 2001 = 16,600
    > 3DMark 2003 = 7,630
    > 3DMark 2005 = 3,030

    I can tell you haven't used project comparisons on the Futuremark site.

    My best 3DM05 score so far:

    http://tinyurl.com/4ku7x

    (that's 3809 from an overclocked BFG 6600GT AGP @ 567/1134 in a S939 3000+
    with a gig of ram and latest drivers for those who don't open TinyURL docs).

    standard 3dmarks from the BFG card were 3536, though BFG pre-OC their card
    to 526/1053.

    my old 9700 radeon np got 1835 non-oc'ed in the same rig, and that was
    higher end than a GF4Ti4600 at the time, so figure a standard 6600GT is
    twice as fast as a GF4Ti in 3dMark05, or about the same speed as a 6800nu.

    IOW- with the exception of having too little RAM for current games- you have
    a decent setup. Have fun.
  14. Archived from groups: alt.comp.periphs.videocards.nvidia (More info?)

    OK heres some benchmark results

    3DMark05

    1st Score 2216 (fast write on)
    2nd Score 2018 (fast write off)
    3rd Score 2703 (power direct from PSU - no cdrom)

    Thats with 67.66 beta drivers
  15. Archived from groups: alt.comp.periphs.videocards.nvidia (More info?)

    Thats intresting to hear as I have no one to compare scores with.... I'm new
    to beachmarking.

    My FSB on my P4 3.2 is 533

    If I over clock to 550 MHz core clock frequency and memory clock frequency
    1.00Ghz
    Then my score goes to 3118

    But how save is over clocking? I noticed my graphics card temp rise about
    10c

    I'm using drivers 71.20b

    This is the PSU I got
    http://cgi.ebay.co.uk/ws/eBayISAPI.dll?ViewItem&rd=1&item=6737912118&ssPageName=STRK:MEWN:IT

    I would be happy hitting 3000 with out Overclocking.

    Like I say, I'm new to benchmarking and Overclocking, when I saw the Frame
    rate in 3DMark05 - I would have expected 30 FPS not 6 FPS.... And at the LAN
    party I went to over weekend, everyones systems were so smooth in Farcry -
    no clitchs...


    "PRIVATE1964" <private1964@aol.com> wrote in message
    news:1108412637.186077.185290@f14g2000cwb.googlegroups.com...
    > That's interesting how your score jumped with no CD-ROM.
    > Even if you get a slightly higher score with fast writes, I don't think
    > it is worth the trouble it might cause IMO.
    >
    > The fastest drivers I found with the least amount of problems are the
    > 71.21's. They give you clear water in HL2, that most of the other 70+
    > drivers do not. Plus there is a good boost in the 2005 score.
    >
    > I know your using a 500 Watt supply, but is it a quality supply?
    > Because you saw in improvement without the CD-ROM drive. I would
    > suggest testing the card with as much devices disconnected as possible.
    > That would help to rule out a power supply issue.
    >
    > With your system specs you should be hitting at least 3000 with no
    > problems. What is the bus speed for that P4.
    > Your 2703 score might not be too bad also. Depending on all your bios
    > settings. I score around 3000 with the 66.93 drivers.
    >
  16. Archived from groups: alt.comp.periphs.videocards.nvidia (More info?)

    Thanks for all ur help, I'm gonna try a WinXP format, u never know! Watch
    this space......


    "PRIVATE1964" <private1964@aol.com> wrote in message
    news:1108414323.258876.50070@o13g2000cwo.googlegroups.com...
    > With the drivers your using I think you should be hitting at least 3000
    > not overclocked on 3DMark 2005. I get 3450 with no overclock, I think
    > your system should get the same or close to it.
    >
    > That power supply looks nice, but having a quality supply matters a
    > lot. There is no brand name on the supply so I have no way to tell if
    > it is probably a good supply. A lot of power supplies will list wattage
    > that they really can not do, or do it well enough.
    >
    > So I still suggest you unplug as much as possible and then test the
    > video card again. I would also suggest making sure your memory is
    > running at the proper speed and the fastest timings in the bios.
    >
  17. Archived from groups: alt.comp.periphs.videocards.nvidia (More info?)

    I wasn't aware SLI scores are listed there. Does it say anywhere on
    them they are SLI systems? I guess what I am asking is do the
    benchmarks detect an SLI setup and list it as such?
  18. Archived from groups: alt.comp.periphs.videocards.nvidia (More info?)

    "PRIVATE1964" <private1964@aol.com> wrote in message
    news:1108403768.810483.324360@z14g2000cwz.googlegroups.com...
    >I have checked on Futuremark, that's were I am saw a lot of really high
    > scores. Then I check out the systems and they are running same bus
    > speed, same CPU speed but their scores are really really high.
    >
    > I have tried alot of drivers and so far the best ones seem to be the
    > 71.21's for me.
    > This is with my MSI 6600 overclocked 565/1.13
    >
    > 3DMark 2001 = 17, 075
    > 3DMark 2003 = 8,650
    > 3Dmark 2005 = 3,737
    >

    yeah, but the first few pages on Futuremark are clearly from SLI systems,
    they are much too high for single 6600gts. On THGs VGA chart, the
    PCI-Express 6600GT gets 3304 in 3dMark 05.
  19. Archived from groups: alt.comp.periphs.videocards.nvidia (More info?)

    That's a low score for 3DMark 2001 with a 5600. I would think you
    should get at least 9,000+. My Geforce 3 Ti200 could hit almost 9,000
    with a 2100+.

    You score is not really all that bad for 2005, but it should be higher
    IMO. I think at least a minimum of 3,000.

    Mobile or desktop CPU shouldn't really make a difference in performance
    if that is the only difference. The Mobile usually has a lower wattage
    rating. It can also usually stand up to more heat then a desktop CPU
    also.

    Do you know if your P4 has automatic speed slowdown from getting too
    hot? Some of Pentiums do so that is why I am asking. If it is getting
    too hot maybe it is slowing down it's speed.
  20. Archived from groups: alt.comp.periphs.videocards.nvidia (More info?)

    Still seems low to me on 2003. I get around 7,600 on 2003 without
    overclocking the video card.

    There is something in your system that is slowing your graphics down. I
    don't think it's the video card because you had a low score with the
    5600 also.
  21. Archived from groups: alt.comp.periphs.videocards.nvidia (More info?)

    I've been trying to work out why my system gives me low benchmarks

    on my FX 5600 256ram on 3DMark2001SE I get a score of 5074
    on my 6600 GT 128ram on 3DMark2005 the most i've got is 2892

    Heres my results
    http://service.futuremark.com/servlet/Index?pageid=/orb/projectdetails&projectType=6&projectId=8430113

    Is 2892 / 3DMark05 normal on my systerm on a 6600 GT card? I'm not sure if
    it's my processor - I'm running a P4 3.2gigHT chip but it's the Mobile
    version.... would that make a big diffrence to the same spec processor if it
    was the standard version?


    500 Watt PSU
    P4 3.2gig HT (mobile version)
    6600 GT running 66.93 driver (I've also tried my FX 5600 256ram and still
    getting low benchmarks)
    Maxtor 160gig DiamondMax Plus 9 IDE Hard-Drive (system drive)
    Matxor 180gig IDE Hard Drive (storage)
    WinXP with SP2 and all current updates
    Pionner 106 DVD
    Liton DVD
    C-Media 5.1 onboard sound
    Very sexy case with 7 fans http://www.casetek.com.tw/midi/ck-1020-2a.htm
    CPU temp 83f - Graphic card temp 94f
    2x 256/133Mhz DDR266 (dual channel memory mode)
    ARock P4v88 http://www.asrock.com/product/product_p4v88.htm
  22. Archived from groups: alt.comp.periphs.videocards.nvidia (More info?)

    I know how you feel. I have all this power and I have nothing to use
    with it. I can't watch the nVidia Dawn demo anymore, I just get too
    excited!
  23. Archived from groups: alt.comp.periphs.videocards.nvidia (More info?)

    I've been messing around with diffrent RAM and Graphics cards for the last
    few hours. I've ordered a new P4 3.2gigHT with 800fsb - It should be a bit
    better than my 533fsb on my mobile chip.

    With 3DMark03

    The best result I could get was 6506 on my 6600GT Card with VStudioExtremeG
    71.20 Driver - 2x 256RAM DDR 133 Duel Channel Memory - Temp goes up about 8c
    when benchmarking

    This fell too 5359 with 2x 512 DDR 400 - strange uh?

    More tweaking to come.... Your comment are welcome.... Cheers


    > That's a low score for 3DMark 2001 with a 5600. I would think you
    > should get at least 9,000+. My Geforce 3 Ti200 could hit almost 9,000
    > with a 2100+.
    >
    > You score is not really all that bad for 2005, but it should be higher
    > IMO. I think at least a minimum of 3,000.
    >
    > Mobile or desktop CPU shouldn't really make a difference in performance
    > if that is the only difference. The Mobile usually has a lower wattage
    > rating. It can also usually stand up to more heat then a desktop CPU
    > also.
    >
    > Do you know if your P4 has automatic speed slowdown from getting too
    > hot? Some of Pentiums do so that is why I am asking. If it is getting
    > too hot maybe it is slowing down it's speed.
    >
  24. Archived from groups: alt.comp.periphs.videocards.nvidia (More info?)

    > There is something in your system that is slowing your graphics down

    I do agree with you 100% - both the FX5600 and the 6600GT have low scores

    My next step is to swap the CPU which should be Friday. I've got a feeling
    that the Mobile chip doesn't work too well with its 533fsb, if that dosen't
    improve things then I'm gonna have to start looking at my RAM or mobo....

    Thanks for replying - the nights seem to get longer and the sleep
    shorter!!!! Maybe I should just get a Xbox
  25. Archived from groups: alt.comp.periphs.videocards.nvidia (More info?)

    Martin Francis wrote:
    > "PRIVATE1964" <private1964@aol.com> wrote in message
    > news:1108470373.899741.274820@c13g2000cwb.googlegroups.com...
    > >I wasn't aware SLI scores are listed there. Does it say anywhere on
    > > them they are SLI systems? I guess what I am asking is do the
    > > benchmarks detect an SLI setup and list it as such?
    > >
    >
    > Don't think so- it's up to the person to note it. Some do, some
    don't- but I
    > think it's safe to assume any 6600GT 3dM05 scores over about
    4500-5000 are
    > from an SLI rig.


    That's good to know Thanks. I didn't think there were that many people
    running SLI yet.
  26. Archived from groups: alt.comp.periphs.videocards.nvidia (More info?)

    "PRIVATE1964" <private1964@aol.com> wrote in message
    news:1108470373.899741.274820@c13g2000cwb.googlegroups.com...
    >I wasn't aware SLI scores are listed there. Does it say anywhere on
    > them they are SLI systems? I guess what I am asking is do the
    > benchmarks detect an SLI setup and list it as such?
    >

    Don't think so- it's up to the person to note it. Some do, some don't- but I
    think it's safe to assume any 6600GT 3dM05 scores over about 4500-5000 are
    from an SLI rig.
Ask a new question

Read More

Nvidia Performance Graphics