Asus A7M266-D, MSI K7D Master-L and Tyan Tiger MPX in the test: First look at AMD's dual mainboards

Asus A7M266-D, MSI K7D Master-L and Tyan Tiger MPX in the test: First look at AMD's dual mainboards

The test structure

The Athlon MP processor is intended for the operation of a dual AMD mainboard. For our review, AMD provided us with the most powerful processor in the MP family, the Athlon MP2000 +. Many thanks for the generous provision of both processors at this point to AMD Technologies .

We also had the following hardware in use:

  • Mainboards
    • Asus A7M266-D
    • MSI K7D Master-L
    • Tyan Tiger MPX
  • Processors
    • AMD Athlon MP2000 +
    • AMD Athlon MP2000 +
  • CPU cooler
    • ARKUA 6228
  • Memory
    • Nanya 512 MB CL2
  • Graphics card
    • Asus v8200 DeLuxe
  • Hard drive
    • 40GB MaxtorD740X-6L
  • CD-Rom
    • Yamaha CRW2100E (16/10/40)
  • Power supply
    • Enermax EG465P-VE (431W)
  • Housing
    • Noname Midi-Tower additionally with two 80x80 mm fans (1, 92 W)
  • Mouse
    • Logitech
  • Monitor
    • Iiyama Vision Master 451

All boards were squeezed into a midi tower case. The tower was also equipped with two housing fans from YS-Tech in the 80x80mm version with 1.92 watts. One of the two fans provided fresh air to the front of the case. The second fan was responsible for extracting the heated air, directly behind the CPU cooler on the back of the tower.

We deliberately opted for a normal tower case for this test and not for an aluminum case with four or five fans, which has better cooling performance, but is also correspondingly expensive.

It has to be said that two processors produce a huge amount of waste heat. Of course, it needs to be brought out of the case first so that the CPU fans have cooler air to cool the CPU coolers. As described above, the minimum is two case fans. Either a second fan on the back or, if technically feasible, a fan in the top of the housing cover is advantageous. It should also be mentioned that when installing in a midi tower, a power supply unit with a second fan, which extracts the warm air directly above the two processors, is absolutely necessary.

We were in charge of cooling the CPUs the choice between three pairs of coolers. There was a choice of two Kanie Hedgehog 294M full copper coolers, two Thermaltake Volcano 7s, and two ARKUA 6228s, which a few months ago caused a sensation because their very good cooling performance set new records. After a quick test of all three fans, I decidedWe opted for the Arkuas, which offered the best cooling performance with pleasant noise development.

For the main memory, we had a 512 MB PC266 DDR-Ram CL2 module from Nanya. We also use an ATA133 hard drive from Maxtor, the D740X-6L with 40 GB of storage capacity. Not important for our test, but mentioned here for the sake of completeness, we used the 16/10/40 CD burner from Yamaha, the CRW-2100E, as the CD-ROM drive.

For the constant and reliable power supply we used the well-known 431-watt Enermax power supply, the EG465P-VE.

The Geforce3 graphics card from Asus, the V8200, was used.

Test setup - PSU
Test setup - graphics card
Test setup - memory
Test setup - hard disk

The test software

For testing the performance of the two processors We decided on four graphics benchmarks, three benchmarks for measuring the performance in the areas of integer and floating point calculation, as well as memory transfer. Furthermore, two application benchmarks and three applications were used.

The three graphics benchmarks were the well-known onesMadonion benchmarks, the 3DMark2000 v1.1 and the new 3DMark2001 SE, are used. The Aquamark v2.3 and the GLMark from Vulpine are also used.

To compare the computing power, we used the well-known Sandra benchmark in version 2002 Pro, the PCMark2002 from Madonion, and the Benchmark32 from Chip that mucks the computing power when loaded with one to five simultaneous operations.

To improve the performance with normal applications To test it, the Sysmark2002 was used, which simulates working with image processing software or publishing programs such as Macromedia Dreamweaver, Photoshop and Adobe Premiere, or the Microsoft Office package. With the CineBench2002 we tested the performance in the area of ​​3D rendering, which simulates the well-known animation and rendering program Cinema 4D XL.

Finally, we tested how the two processors perform in tasks such as mp3 -Conversion, DVD conversion, and the increasingly popular calculation of SETI @ home behavior. For the conversion of uncompressed audio files in WAV format to MP3 format, we decided on the relatively unknown program Waver 2.66. The advantage of Waver is that it can convert two WAV files at the same time. We are testing the recently increasing popularity of converting DVDs into DivX films with Xmpeg 4.5, a FlaskMPEG clone that can address both processors, with the video stream being converted on one processor and the audio stream on the other. However, since the conversion of the audio stream does not require as much computing power as that of the video stream, the total load on the system is only around 80-85 percent. Finally, SETI @ home is used, which calculates the data packets and ran separately on each processor, i.e. once SETI @ home per processor at the same time, since the program itself onlysupports one processor.

We use Windows XP Professional as the operating system, since two processors are only supported from the Professional version. The most used Home Edition supports only one processor. We had to use the 11/23 Detonator for the graphics cards, as this was the only way to run the Aquamark smoothly.

Here is a short list of the test software:

  • 3DMark 2000 v1.1
  • 3DMark 2001 SE
  • AquaMark v2.3
  • Vulpine GLMark v1.1p
  • PCMark2002
  • Sandra 2002 Pro
  • Chip Benchmark32
  • Sysmark2002
  • Cinebench2000
  • Waver 2.66
  • Xmpeg 4.5
  • SETI @ home

On the next page: 3DMark2000 v1.1