New harddrive question

I'll try updating my bios although nothing in the release notes for the new bios says anything about the IDE controllers. I am bit confused about the difference between burst rate and sustained transfer rate. All the rates I have been quoting before have been sustained transfer rates. I'll include the dskbech results for the fireball below and maybe if you notice anything weird you can comment.

Chris

DskBench 2.11
(c) 1998, SESA, J.M.Catena (admin@sesa.es, www.sesa.es)
Timer Check = 991 (should be near 1000)
CPU Check = 50.93 % (should be near 50.00 %)
CPU index (relative to Pro 200 MHz) = 2.763487
Open = 0 ms
Write = 14733 ms, 17.38 MB/s, CPU = 3.19 %
Flush = 10 ms
Rewin = 0 ms
Read = 16101 ms, 15.90 MB/s, CPU = 2.93 %
Close = 14 ms
BlockSize = 131072, MB/s = 7.61, Tracks = 90.45, CPU = 1.96 %
BlockSize = 65536, MB/s = 5.38, Tracks = 63.91, CPU = 1.00 %
BlockSize = 32768, MB/s = 8.06, Tracks = 95.81, CPU = 1.72 %
BlockSize = 16384, MB/s = 7.18, Tracks = 85.37, CPU = 2.79 %
BlockSize = 8192, MB/s = 6.36, Tracks = 75.64, CPU = 4.98 %
BlockSize = 4096, MB/s = 5.80, Tracks = 69.01, CPU = 8.75 %
 
does it work?

After being in hardware hell for a couple of weeks and exhausted many many suggestions just short of reinstalling windows I had my maxtor drive replaced (under warranty) and all of my drop outs etc. went away. This was my second in about 8 months.....if its working for you now forget about bench tests and have fun because there will be plenty of real problems I'm sure will crop up in your adventures in DAW land.

buenas suerte'
 
just another suggestion...recheck the cable..
with the Ultra ATA cable I.m using, the blue
connector has to go to the Motherboard...
the grey...has to be the slave..this is the
middle connector..and the other at the
farthest end of the cable goes to the
Master Hard drive..and the jumpers should
be checked accordling..
 
Yeah my cables are colour coded the same way. I have tried switching them and no difference. I guess I am thinking about getting a PCI ATA 100 controller. Has anyone had any experiences with them? Is anyone confident it would be the answer?

Regardless I am back making music again so its really just a peace of mind issue now.

Thanks again.
 
I have the Maxtor ATA100 controller which is simply a Promise controller with the Maxtor name put on it. It uses it's own on-board BIOS. The bad side of this is that it only works in Windows, after its 32 bit Windows driver has been loaded. The good side is that Windows treats it much like a SCSI card - it only uses a single IRQ even though it has two IDE controllers (supporting up to 4 devices)
 
I did a little checking for you and it seems that this is a not so uncommon problem, and might be resolved by downloading and installing the latest IDE busmastering drivers directly from VIA. You might also look for similar drivers directly from Gigabyte and compare.

Good luck.

Slackmaster 2000
 
final chapter

Wow..this has been quite the ordeal but I think I have finally reached the end of the road. Here is the latest:

I did find DMA bus master drivers on the Gigabyte site and they actually came with a little app that allowed to specify the speed on the ide channels. The max I could get was Ultra DMA3(ATA 66) which was giving the numbers I have posted above. If i tried lower speeds PI0 0, 1,2 and regular DMA etc. etc. then the results did drop. So I basically came to the conclusion that everything was working as it should but that my mobo was just not fast enough to really take advantage of my new drives.

So the next chapter begins. i found a promise ATA100 IDE controller card for approx $30 U.S. so I figured I would give that a shot. Inititially I was overjoyed to find that I was getting read and write speeds of over 30mb/s. Then I got the speeds for the smaller block reads. With block sizes smaller than 128KB I was getting transfer rates of below 5mb/s which was a little disturbing. Anyway I fired up Cakewalk and tried playing back my project (my file buffer size is set to 128KB) and no more performance problems. So I guess everything is working. Its just a little unsettling that the small block transfer rates were lower than they were before. I have included the latest dskbench report below. Thanks again everyone for all the help. I am pretty sure I'm good to go now.

DskBench 2.11
(c) 1998, SESA, J.M.Catena (admin@sesa.es, www.sesa.es)
Timer Check = 1040 (should be near 1000)
CPU Check = 50.00 % (should be near 50.00 %)
CPU index (relative to Pro 200 MHz) = 2.781498
Open = 0 ms
Write = 7836 ms, 32.67 MB/s, CPU = 7.23 %
Flush = 11 ms
Rewin = 0 ms
Read = 7556 ms, 33.88 MB/s, CPU = 6.87 %
Close = 15 ms
BlockSize = 131072, MB/s = 11.00, Tracks = 130.76, CPU = 3.11 %
BlockSize = 65536, MB/s = 5.79, Tracks = 68.88, CPU = 1.33 %
BlockSize = 32768, MB/s = 3.98, Tracks = 47.26, CPU = 1.12 %
BlockSize = 16384, MB/s = 2.13, Tracks = 25.31, CPU = 1.11 %
BlockSize = 8192, MB/s = 6.53, Tracks = 77.67, CPU = 5.89 %
BlockSize = 4096, MB/s = 6.61, Tracks = 78.63, CPU = 11.50 %
 
You must log in or register to reply here.
Back
Top