+ Reply to Thread
Page 86 of 100 FirstFirst ... 36 76 82 83 84 85 86 87 88 89 90 96 ... LastLast
Results 1,276 to 1,290 of 1499
  1. #1276
    Member
    Join Date
    Dec 2001
    Location
    Texas
    Posts
    121
    Hello all, I've been out of the benchmarking game for a while now so need some help.

    System specs:
    Asus A7N8X-X
    AMD XP2500 Barton @ 166x12.5 or 2083Mhz
    512 Corsair XMS PC3000 C2 @ 6 3 3 2 ( I think anyway)
    Gainword FX5900 128MB non ultra at stock 400core 850mem
    WD 40GB 7200HDD

    3DMark2k1se 14600

    Now the reason I need help is because this doesnt look right to me. I say that because I just sold a system with these specs and scores..........

    System Specs:
    Asus A7N8X Deluxe
    AMD XP2500 Barton @ 200x10.5 or 2100Mhz (highest I could get stable)
    512 Samsung Twin PC3200 C3 @ 7 3 3 3
    Albatron GF4Ti4200 128MB at stock speed
    WD 60GB 7200HDD

    3DMark2k1 12860

    Obviously I've gone to 166Mhz on FSB instead of 200Mhz so I know I'm taking a hit there. But I would think I would get a bit more than 1800 extra marks with this video card. And it does have the 256bit bus and 8 pixel pipes.....

    Any thoughts or does this sound about right? I will be getting PC3200 mem in the next few days and hopefully this chip will be stable at higher speeds than the last one.

    Also the dual channel doesnt make that much difference either. I actualy got better Sandra memory scores with the X board than I did with the Deluxe and that is why I sold it.....


    Slapshot

  2. #1277
    Senior Member MrBurns's Avatar
    Join Date
    Sep 2003
    Location
    Vienna, Austria
    Posts
    642
    Your score is a little bit too low.

    What driver/OS/DirectX versiopn do you use?

    Make sure, that AntiAliasing and Ansitropic Filtering is turned off and all driver settings are turned to performance.
    Last edited by MrBurns; 11-22-2003 at 03:41 AM.

  3. #1278
    Member
    Join Date
    Dec 2001
    Location
    Texas
    Posts
    121
    All those were turned off.

    OS Win2KPro
    DX9

    Ok the 14600 score I got with the stock Gainward driver I think 44.03. I just installed the 53.03 driver and now I'm getting 13997. I am seeing the drop in nature but everything else goes up. With the 44.03 driver Nature test starts at 100fps and goes as high as 180fps. With the 55.03 nature starts at 80fps and peeks at about 130.
    I also just ran 3DMark2K3 and got 5130. I realy dont use 2K3 yet so I dont know if that is an O.K. score or not. Also there is a possibility that this memory is getting flakey. I got it about a year and a half ago and may have started to cook it from OC'ing it all the time.

    Slapshot
    Last edited by Slapshot; 11-22-2003 at 04:11 AM.

  4. #1279
    Senior Member MrBurns's Avatar
    Join Date
    Sep 2003
    Location
    Vienna, Austria
    Posts
    642
    Originally posted by Slapshot
    I also just ran 3DMark2K3 and got 5130. I realy dont use 2K3 yet so I dont know if that is an O.K. score or not.
    The 3DMark03 score is ok, and I think, the 01 score is also ok. 3DMark01 is a DirectX 8 benchmark and the Ti cards are very good in DirectX 8 games, so you cannot expect such a big difference when you dont use at least a FX5800. My scores w/ a Ti4600@stock are 13753 w/ my Athlon XP2700 @ stock (166x13=2166MHz) and 14371 w/ my CPU@178x13.5 (2403MHz). Edit: I thought, that you had a 5600, not a 5900. For a 5900 your 3dMark01 scores are also too low.

    Also 3DMark01 doesnt seem to like the 5x.xx detonators.
    Last edited by MrBurns; 11-22-2003 at 03:12 PM.

  5. #1280
    Member
    Join Date
    Sep 2001
    Location
    S. Williamsport, PA.,USA
    Posts
    50
    Originally posted by Slapshot
    Obviously I've gone to 166Mhz on FSB instead of 200Mhz so I know I'm taking a hit there. But I would think I would get a bit more than 1800 extra marks with this video card. And it does have the 256bit bus and 8 pixel pipes.....

    Any thoughts or does this sound about right? Slapshot
    Yeah I have a few thoughts now and then and one of them is that the 5900 series is a 4 Pixel pipe design, not 8. Nvidia spread a bunch of FUD about the FX series early in it's developement about it being an 8 pipe design but it turnsd out they were misinforming their customers as usual.

    4 Pipes... not 8

    Under certain conditions they can do 2 Pixel ops per clock per pipe, which Nvidia then twists into the fud that the FX has 8 pipes.

    From a thread at beyond3d about an Anand article on the FX5900...

    "OK, he starts off by mentioning how nVidia called 4 pixel pipelines 8,... A pretty blatant spin to start off with, and then a technical example leading up to a pretty dedicated obfuscation of the term "pixel" by such terms as "Z pixel", "texture/clk", "Stencil pixel", and "pixel shader ops", where "Color + Z pixel" is established as the special case that doesn't weigh as heavily (it's only listed once, after all, whereas the rest of the items present a count of "8" for them that are more numerous ). Nevermind that they are the special cases of splitting Color+Z up into its parts, and that the (4 Pipeline)Color+Z limitation is the most frequent one, even with pixel shading. "
    Last edited by benknobi; 11-22-2003 at 10:18 AM.

  6. #1281
    Ultimate Member Rugor's Avatar
    Join Date
    Dec 2001
    Location
    Pacific Northwest, Earth
    Posts
    2,694
    I tried to stay out of this bit, but Benknobi is right.

    The high end GfFX cards can operate as either a 8x0 or 4x2 card. Unfortunately, the so-called "8-pipe" configuration only works when the data (you can't really call it a pixel) will not be output on this pass. It can handle 8 parallel datastreams internally, in some cases, but can only ever output four pixels per pass. So no matter what Nvidia's marketing dept says it's a four pipe card when it counts.
    "Dude you're getting a Dell." Obscure curse from the early 21st Century, ascribed to a minor demon-spirit known as "Stephen?" [sp].

  7. #1282
    Member Beast_USA's Avatar
    Join Date
    Dec 2002
    Location
    CA USA
    Posts
    172
    My fx is fine the xt I bought wouldn't start up my apple display.

    3D2003 7492
    http://service.futuremark.com/compare?2k3=1541592

    Aquamark3 55979
    http://arc.aquamark3.com/arc/arc_view.php?run=809746628

    So no matter how nvidia calls it ....it works for me

  8. #1283
    Ultimate Member Rugor's Avatar
    Join Date
    Dec 2001
    Location
    Pacific Northwest, Earth
    Posts
    2,694
    Glad it does:

    From everything I've seen the second generation FX cards (5700 and 5900) are pretty decent pieces of hardware. They may not be as good at DX9 as ATI but they have other strengths.

    My biggest problem is that Nvidia's been marketing them as things they aren't not so much for what they are. They're really fast cards in most games, and when the drivers don't disable it they have great AF.
    "Dude you're getting a Dell." Obscure curse from the early 21st Century, ascribed to a minor demon-spirit known as "Stephen?" [sp].

  9. #1284
    Member
    Join Date
    Dec 2001
    Location
    Texas
    Posts
    121
    I have had a Ti4200 of some sort or another for the last 2 years(gone through a few of them). I recently sold my computer and started to build another and thought I'd try a Radeon 9600Pro. Talk about a performance hit! What a POS that thing is. So I sent it back and got an FX5600Ultra with flip chip running at 400/800. I mean with those kind of clocks you'd think it would beat the Ti4200. NOPE! So looking at newegg I saw the Gainward non-pro FX5900 for $229.00 and I figured what the hell. It is faster but not as fast as I thought it would be. I'm O.K. with it though. I still hate the ATI drivers and they're too **** expensive so it will do for now...... Thanks for the input though.

    Slapshot

  10. #1285
    Member
    Join Date
    Nov 2003
    Location
    SoCal, USA
    Posts
    66

    BFG Asylum FX 5900

    Please help! I just ran 3DMark2001SE Build 330 and it came up with a score of 11,275, but I think my system should be doing much better:

    Athlon XP 2000 (not oc'd)
    Epox 8K3A+ MB
    1 GB PC2700 RAM
    Samsung 160GB 2MB HDD
    BFG Asylum FX5900 (non-ultra, oc'd to 444/921)
    Windows XP Professional

    I made sure AA&AF were turned off, Vsync on auto select, Refresh Rate at 100hz, High Performance setting in Forceware, and nothing else running in the background.

    Should this be higher? When the card wasn't oc'd, I was getting just a hair over 11,000.

  11. #1286
    Senior Member MrBurns's Avatar
    Join Date
    Sep 2003
    Location
    Vienna, Austria
    Posts
    642
    Originally posted by Slapshot
    I have had a Ti4200 of some sort or another for the last 2 years(gone through a few of them). I recently sold my computer and started to build another and thought I'd try a Radeon 9600Pro. Talk about a performance hit! What a POS that thing is. So I sent it back and got an FX5600Ultra with flip chip running at 400/800. I mean with those kind of clocks you'd think it would beat the Ti4200. NOPE! So looking at newegg I saw the Gainward non-pro FX5900 for $229.00 and I figured what the hell. It is faster but not as fast as I thought it would be. I'm O.K. with it though. I still hate the ATI drivers and they're too **** expensive so it will do for now...... Thanks for the input though.
    The Radeon9600 and GeForce FX 5600 cards have a poorer performance in DirectX 8 than the GeForce 4 Ti series, but they have a major advantage: they are fully DX9 compliant. So if you decide between a FX 5600/Radoen 9600 and a Ti you have to make a decition between performance and DX9.

  12. #1287
    Ultimate Member Rugor's Avatar
    Join Date
    Dec 2001
    Location
    Pacific Northwest, Earth
    Posts
    2,694
    I have a Radeon 9500 Pro, which generally gets me a hair over 10K 3Dmarks in 2K1SE. That may not be a huge score but with a Pally 1800+ and only 256MB of RAM I'm pretty satisfied.

    What I like about it is not just the DX9 compatibility, but also that it has a much lower performance hit from AA and AF than the Ti series. The DX9 cards aren't so much about higher frame rates as improved image quality.

    That's one reason I went ATI, they have better AA than Nvidia.

    As to drivers, I've only ever had one game killer ever, and that was with Nvidia drivers. I had a Gf2MX at one point and going to the 40.xx detonators killed Freedom Force.
    "Dude you're getting a Dell." Obscure curse from the early 21st Century, ascribed to a minor demon-spirit known as "Stephen?" [sp].

  13. #1288
    Senior Member MrBurns's Avatar
    Join Date
    Sep 2003
    Location
    Vienna, Austria
    Posts
    642
    Originally posted by Rugor
    What I like about it is not just the DX9 compatibility, but also that it has a much lower performance hit from AA and AF than the Ti series. The DX9 cards aren't so much about higher frame rates as improved image quality.
    I thought, that the Radeon 9500 series is not DX9 compatible (or is that only the non-pro?). The Ti 4600 however outperformes the Radeon 9500. I get 14371 3DMarks in 2k1 w/ my Ti 4600@stock speed and my CPU@13.5x178=2403MHz. CPU is Athlon XP 2700+ w/ a deafult clock speed of 2166MHz. I can get a higher FSb, but I didnt benchmark the high FSBs yet.

  14. #1289
    Ultimate Member Rugor's Avatar
    Join Date
    Dec 2001
    Location
    Pacific Northwest, Earth
    Posts
    2,694
    All Radeons from the 9500 and up are fully DX9 compatible. My 9500 Pro differs from the 9700 Non-Pro only in the width of the memory bus, it has the same VPU running at the same clock. The 9500 NON-pro disabled half the pipes and Hyper-Z.

    I'm not surprised you're getting much higher 2K1 scores than I am. Even at default clocks the Ti4600 runs faster (300/325) than the 9500 Pro (275/270). So since the benchmark doesn't tax any DX9 functionality it really comes down to overall system performance, and my AXP at 1533MHz with a 133FSB is not going to match your system.

    You have about a 10% edge in GPU speed, about 20% in GPU bandwidth, 50% in CPU speed and about 33% in System bandwidth, all of which give you roughly 40% more 3DMarks, so it's not quite scaling with CPU speed.

    The advantage the 9500 Pro has is better AA, and generally less of a performance hit when that and AF are engaged. On a given system, it will normally about match a Ti4200 without AA/AF enabled, and beat a Ti4600 with AA/AF enabled. Then you add in the DX9 capability and you get a better idea of what it's aimed at.
    "Dude you're getting a Dell." Obscure curse from the early 21st Century, ascribed to a minor demon-spirit known as "Stephen?" [sp].

  15. #1290
    Senior Member MrBurns's Avatar
    Join Date
    Sep 2003
    Location
    Vienna, Austria
    Posts
    642
    I am sorry to disappoint you, but the Radoen 9500 series has no DirectX 9 support, only the 9600 series has:

    http://www.tech-report.com/onearticle.x/4122

    Facing the new competition from Nvidia, especially in the medium-range market, ATI Technologies announced that in October it will introduce its Radeon 9500 series, a simpler version of its top-end Radeon 9700 PRO. Though based on the R300 core and offering the AGP 8x specification, the Radeon 9500 series (Radeon 9500 and Radeon 9500 PRO) is designed with four pipelines and DirectX 8.1 support.
    Edit: I found out, that the 9500 sereis also supports DX9, that it is not supporting DX9 is only a rumor which comes from a press release at a time before the 9500 series was available and ATI didnt want to implement DX9 on the 9500 series at this time, but they changed their mind:

    http://www.hardwarecentral.com/hardw...eviews/4513/1/

    http://www.infosatellite.com/news/20...adeon9500.html
    Last edited by MrBurns; 11-23-2003 at 04:29 AM.

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts







New Security Features Planned for Firefox 4
Another Laptop Theft Exposes 21K Patients' Data
Oracle Hits to Road to Pitch Data Center Plans
Microsoft Preps Array of Windows Patches
Microsoft Nears IE9 Beta With Final Preview
Simplified Analytics Improve CRM, BI Tools
Android Passes RIM as Top Mobile OS in 2Q
VMware Updates Hyperic System Management
File Monitoring Key to Enterprise Security
LinkedIn Snaps Up SaaS Player mSpoke