Menu
Epox M762A and Gigabyte A-7DPXDW in the test: Dual-AMD motherboards in comparison

Epox M762A and Gigabyte A-7DPXDW in the test: Dual-AMD motherboards in comparison

Waver 2.66

Now we come to the most interesting tests. First of all, converting WAV files to MP3 format. Since the MP3 codecs do not allow the computation to be split between more than one CPU, Waver took a different approach. Waver always calculates two WAV files at once. A WAV file is encoded on each CPU. If one is finished before the other, which is the case if both WAV files are not of the same length, the system automatically continues with the next WAV file. This cuts the time it takes to encode WAV files in half the time when more than one file needs to be converted. For the test, we used the same WAV file with a size of 200 MB, which we converted to MP3 format at the same time.

Waver 2.66
Unit : seconds
  • Frauenhofer:
    • MSI
      178
    • Gigabyte
      188
    • Epox
      196
    • Asus
      197
    • Tyan
      200
    • Gigabyte 1 CPU
      378
  • Blade 0.94:
    • Asus
      70
    • Tyan
      71
    • Epox
      72
    • Gigabyte
      72
    • MSI
      76
    • Gigabyte 1 CPU
      142
  • Lame 3.87:
    • MSI
      47
    • Epox
      58
    • Gigabyte
      58
    • Tyan
      67
    • Asus
      67
    • Gigabyte 1 CPU
      90

Nice to see how two processors (can) pay off. The data is converted at almost twice the speed. While the Frauenhofer and Lame Codec copes best with the MSI board, the Asus board again dominates the 'Blade' codec.

Xmpeg 4.5

With the popular conversion of a DVD into a DivX-AVI file, we want to check whether a dual-processor mainboard is particularly worthwhile. We used a normal DVD, here the movie Star Trek V, and converted it directly into the new DivX5 format. Our goal was that the final file size does not exceed 700 MB.

Therefore we have the following settingsmade:

  • Output format :
    • Video: DivX 5 Pro, 1-pass, 800 kbps, Key: 12 frames
    • Audio: MPEG Layer-3 , 128 Kbit/s, 48 ​​KHz, Stereo
  • Project options:
    • Format 720x576, Optimized MMC iDCT, RGB, MMX Bilinear, Trim to 720x352
  • File size:
    • Target: max. 700 MByte
Xmpeg 4.5
  • DVD (Star Trek V) to DivX5:
    • Asus
      37,91
    • Tyan
      37,32
    • MSI
      37.05
    • Gigabyte
      36.12
    • Epox
      35,98
    • Gigabyte 1 CPU
      29.76

Unfortunately, the two new mainboards couldn't do that well here Don't do as the competition did in the previous tests. The differences are over 2%.

SETI @ home

For all of our Setians who are eager to get a WU in half the time with a dual processor mainboard ( Workunit) to crunch (calculate), let it be said, Seti is not dual-processor capable. In our test, we had the same standard WU calculated twice with AR 0.417. The result of the time used relates to two calculated WUs. That means that two WUs are calculated in the time of one.

SETI @ home
Unit: hours, minutes
  • Standard WU AR 0.417:
    • Asus
      3:34
    • MSI
      3:36
    • Epox
      3:43
    • Gigabyte
      3:47
    • Tyan
      4:06
    • Gigabyte 1 CPU
      6:47

Here, too, the Epox and the Gigabyte are roughly in the middle. The fact that the structure with only one CPU should take twice as long for a work unit is a bit confusing. Since we calculated two work units at the same time with the mainboards that ran with two processors, we had to run both work units one after the other on the Gigabyte with only one CPU. Viewed per processor, the dual-board mainboard with one CPU is even a little faster in processing a work unit. But seen per mainboard, the mainoards with two processors are of course faster, as they do double the work, i.e. work units, in the same time.

On the next page: Overclocking