Standard Performance Test Configuration: 3200+ with ATI 9800 PRO

We first tested the MSI K8N Neo using the same 3200+ and ATI 9800 PRO, which has been the standard for past motherboard tests. In this series of tests, we compared the performance of the MSI K8N Neo to other Athlon 64 boards that we have tested. We also included performance results from a 3200+ Athlon XP for comparison.

We have dropped P4 comparisons from our A64 test results. Since the Athlon 64 is single-channel and we have tested Socket 754 boards with a 3200+ (2.0GHz), we have also removed test results with the Dual-Channel 2.2GHz FX51. This should make charts comparing Socket 754 motherboards much easier to understand.

If you are interested in more information comparing the Athlon 64, Athlon 64 FX, Prescott, P4, and P4EE, please see our in-depth comparisons in the recent Prescott/3400+/3000+ launch reviews:

Intel's Pentium 4 E: Prescott Arrives with Luggage
Athlon64 3400+: Part 2
AMD's Athlon 64 3400+: Death of the FX-51
Athlon64 3000+: 64-bit at Half the Price

 Performance Test Configuration
Processor(s): AMD Athlon 64 3200+ (2.0GHz)
AMD Athlon XP 3200+ (2.2GHz, 400MHz FSB)
RAM: 2 x 512MB Mushkin PC3500 Level II OR
2 x 512MB OCZ PC3500 Platinum Ltd
Hard Drive(s): Seagate 80GB or 120GB 7200 RPM (8MB Buffer)
Video AGP & IDE Bus Master Drivers: SiS AGP 1.17 (10-07-2003)
VIA 4in1 Hyperion 4.51 (12-02-2003)
NVIDIA nForce version 4.12 Beta
Video Card(s): ATI Radeon 9800 PRO 128MB (AGP 8X)
Video Drivers: ATI Catalyst 4.4
Operating System(s): Windows XP Professional SP1
Motherboards: MSI K8N Neo (nVidia nForce3-250Gb)
AOpen AK89 Max (nVidia nForce3-150)
nVidia nForce3-250 Reference Board
AOpen AK86-L (VIA K8T800 3200+)
ECS 755-A2 (SiS 755 A64 3200+)
SiS 755 Reference Board (A64 3200+)
Abit KV8-MAX3 (VIA K8T800 A64 3200+)
Chaintech ZNF3-150 (nForce3-150 A64 3200+)
MSI K8T Neo (VIA K8T800 A64 3200+)
DFI NFII Ultra (nForce2 U400 Athlon XP 3200+)

Tests on Socket 754 Athlon 64 motherboards and Socket A Athlon XP used either Mushkin PC3500 Level II or OCZ PC3500 Platinum Ltd memory modules. Both memories use Winbond BH5 chips and perform virtually the same in our benchmarks.

All performance tests were run with the ATI 9800 PRO 128MB video card with AGP Aperture set to 128MB with Fast Write enabled. Resolution in all benchmarks is 1024x768x32 unless otherwise noted.

Additions to Performance Tests

We have updated to Veritest Multimedia Content Creation Winstone 2004 and Veritest Business Winstone 200 for system benchmarking. Some earlier board tests used the earlier Winstones 2003/2002, which are not results comparable to Winstone 2004. We have, therefore, dropped those results from our graphs.

Game Benchmarks

We have added several benchmarks to our standard Gaming tests. These include Halo, Microsoft's Direct X 9.0b game; Splinter Cell, a DX9 game; X2 Benchmark, a DX 8.1 game that includes Transform and Lighting effects; the DX9 Aquamark 3; and the DX 8.1 Comanche 4 benchmark using the 4X anti-aliasing setting to differentiate system performance better using our standard ATI Radeon 9800 PRO video card. We will be reporting more results at 1280 x 1024 resolution in future system and motherboard benchmarking where that resolution provides useful performance data. We have included results of these tests wherever they were available. Older benchmarks did not include some of the newer tests.

Overclocking and Stress Testing: MSI K8N Neo Platinum Standard Content Creation, General Usage and Encoding
Comments Locked

26 Comments

View All Comments

  • Starstream - Thursday, September 16, 2004 - link

    Well, I currently have an MSI board and, frankly, I would have to really have my arm twisted before I bought another. Zillions of issues with their support and documentation. I don't know...this is impressive. The "cross mount" memory slots are an interesting twist.
  • ksherman - Sunday, August 22, 2004 - link

    this is probably a dump question, but here goes:
    since the 3200 OCs to the level of the 3800, does that mean i am essentially (if i can reach those speeds of course) getting a 3800 in term of performance? the 3800 are clocked at 2.4GHz, and the only real difference in the processors is the clock speed correct? so a $600+ card for around $200+ seemd like a good deal to me...
  • Klaasman - Saturday, May 8, 2004 - link

    When are these due to be available in the USA for purchase??
  • l3ored - Saturday, May 1, 2004 - link

    i'd really like to know weather the 300fsb outperformed the max oc, then i can make memory purchase decisions, perhaps a follow up review is in order?
  • MadAd - Wednesday, April 28, 2004 - link

    Raid Questions: a) I run a fasttrack tx2000 with 4 IDE drives and am wondering would I be able to use the 2 IDE channels for raid 0+1 and get a SATA to IDE converter to run an optical drive or two?

    b) The last raid review at AT was Jun 2001. With new provison for raid in nf250 plus older high end boards offering variations of highpoint and promise on board plus sata raid vs ide becoming a factor, wouldnt this be a good time for another one?
  • AtaStrumf - Tuesday, April 27, 2004 - link

    Same as ATi X800Pro. Well, well don't we have a lot of 8's on the 6th. Too bad I got an exam on that very same day. Hope I get an 8 as well :)
  • Wesley Fink - Tuesday, April 27, 2004 - link

    #19 -

    Several nF3-250 boards are expected at Anandtech in the next couple of weeks. They are starting to show up from many manufacturers.

    Also, VIA's update to the K8T800 Pro will launch on May 6.
  • hifisoftware - Tuesday, April 27, 2004 - link

    Very, very nice review.

    OC bench would not hurt, but otherwise OC is very well covered.

    I belive RAID was covered in the chipset review.

    One thing that I would've really liked is to know when other mobos are coming out. It would seem that even without overclocking this chipset should not be slower then previous. So maybe another motherboard is gogin to be a bit faster.

  • mikeymasta - Tuesday, April 27, 2004 - link

    Also how 'real' is the hardware raid on the nforce3?
    I mean can I setup a simple 2 hd based mirror raid have install FreeBSD 5.2.1 and expect it to see only 1 drive like 'true' hardware based raid or is this just software based crap that boil down to just extra IDE ports that only do so called 'raid' by software drivers?

    Personally if I was in charge of making standards I would make rule #1 of the standard of raid being that you cant put the word 'raid' on your software product unless its true hardware based raid!
    Bit like the true standard of PC133 ram where it couldn't be classed as 'PC133' memory unless it had a clearly printed label on it with letters 'PC133' for a start, (I remember anand pointing that out a long time ago)
  • AtaStrumf - Tuesday, April 27, 2004 - link

    Yep, we want OC benchies!

Log in

Don't have an account? Sign up now