ASRock 939SLI32-eSATA2: ULi Dual x16 SLI
by Gary Key on March 2, 2006 12:15 PM EST- Posted in
- Motherboards
Firewire and USB Performance
After looking at many options for Firewire and USB testing, we finally determined that an external USB 2.0, Firewire 400, and Firewire 800 hard disk would be a sensible way to look at USB and Firewire throughput.
Our first efforts at testing with an IDE or SATA drive as the "server" yielded very inconsistent results, since Windows XP sets up cache schemes to improve performance. Finally, we decided to try a RAM disk as our "server", since memory removed almost all overhead from the serving end. We also managed to turn off disk caching on the USB and Firewire side by setting up the drives for "quick disconnect" and our results were then consistent over many test runs.
We used 1GB of fast 2-2-2-5 system memory set up as a 450MB RAM disk and 550MB of system memory. Our standard file is the SPECviewPerf install file, which measures 432,533,504 bytes (412.4961MB). After copying this file to our RAM disk, we measured the time for writing from the RAM disk to our external USB 2.0, Firewire 400, or Firewire 800 drive using our Windows bases timing program. The copy times in seconds were then converted into Megabits per second (Mb) to provide a convenient means of comparing throughput. Higher Rates therefore mean better performance in this particular test.
The ASRock board offers a TI based IEEE 1394 Firewire option with performance equal to that of other TI solutions. The USB 2.0 performance is consistent with other ULi based controllers and continues to lag behind the NVIDIA nForce4 chipset solutions in throughput.
After looking at many options for Firewire and USB testing, we finally determined that an external USB 2.0, Firewire 400, and Firewire 800 hard disk would be a sensible way to look at USB and Firewire throughput.
Our first efforts at testing with an IDE or SATA drive as the "server" yielded very inconsistent results, since Windows XP sets up cache schemes to improve performance. Finally, we decided to try a RAM disk as our "server", since memory removed almost all overhead from the serving end. We also managed to turn off disk caching on the USB and Firewire side by setting up the drives for "quick disconnect" and our results were then consistent over many test runs.
We used 1GB of fast 2-2-2-5 system memory set up as a 450MB RAM disk and 550MB of system memory. Our standard file is the SPECviewPerf install file, which measures 432,533,504 bytes (412.4961MB). After copying this file to our RAM disk, we measured the time for writing from the RAM disk to our external USB 2.0, Firewire 400, or Firewire 800 drive using our Windows bases timing program. The copy times in seconds were then converted into Megabits per second (Mb) to provide a convenient means of comparing throughput. Higher Rates therefore mean better performance in this particular test.
The ASRock board offers a TI based IEEE 1394 Firewire option with performance equal to that of other TI solutions. The USB 2.0 performance is consistent with other ULi based controllers and continues to lag behind the NVIDIA nForce4 chipset solutions in throughput.
Facebook
Twitter
RSS
46 Comments
View All Comments
Gary Key - Saturday, March 4, 2006 - link
It works very well on this board with an Opteron. ;-)Kiste - Thursday, March 2, 2006 - link
In the past, Realtek has solved performance issues by reducing the number of buffers and disabling EAX effects without telling anyone. I hope someone will have a really good look at these integrated audio solutions one day, one that isn't just FPS and CPU utilization.
Gary Key - Thursday, March 2, 2006 - link
To date with the HD Codecs, Realtek has not decreased the buffers nor have we noticed changes in EAX2 output except for the positive. It is a very subjective area to comment on but in BF2 I like the sound quality of the 1.29 drivers better but Serious Sam II seems to have improved a great deal in the 1.30 and 1.31 drivers. While the on-board audio solutions are still lacking in most areas compared to the add-in solutions at least the manufacturers (board/chipset) are trying to improve quality in this area instead of pumping out the same old AC97 codecs. If it were not for the cost I think we would have seen several additional Audigy-SE solutions on the motherboards by now and even possibly an improved offering from Creative. While the A-SE is an older chipset it does offer better performance and audio quality in gaming than the current HD codecs from Realtek, C-Media, ADI, and SigmaTel. I really thought the VIA Envy24 series would have had greater market penetration on the motherboards but one can only speculate as to why this never happened. ;->
tjpark1111 - Thursday, March 2, 2006 - link
is it just me, or does this thing have a 20-pin power connector. ?!?!?!??!???kelim - Thursday, March 2, 2006 - link
Yes, the board does in fact use a 20-pin power connection. If you have a 24-pin PSU, this shouldn't be a problem, since most are 20+4-pin, meaning you can detach the extra 4 pins and only use the ones you need. Either that, or the uneeded pins will hang off the side.Gary Key - Thursday, March 2, 2006 - link
You are correct, it is a 20-pin ATC connector as stated in the article. I was surprised also and did a double take after removing it from the box. :)dab - Thursday, March 2, 2006 - link
It says that this motherboard supports ddr2,"(4) x DIMM, max. 4GB, DDR2 400/333/266, non-ECC, un-buffered memory..."
Is this a joke? There's no 240pin slots on that board, only on the riser (which is not included). s939's IMC doesn't support this either..
Am I mistaken?
Gary Key - Thursday, March 2, 2006 - link
I was just changing that to DDR, sorry about the mistake. The AM2 daughter card will have slots for DDR2 memory.
Puddleglum - Thursday, March 2, 2006 - link
The figure is noteworthy, but it's actually the only Firewire 800 test result I can find in AT's reviews.Gary Key - Thursday, March 2, 2006 - link
I know we only have it in a recent review and probably should drop it. I keep hoping Firewire 800 will be implemented on the higher end boards or those designed for HTPC life. :) However, it appears to be a dead-end option at this point and the technology itself seems to be going away without a fight.