Chipset Question

BIOS Questions that don't belong in the other forums. Read them!
Post Reply
KachiWachi
The New Guy
Posts: 1451
Joined: Fri Mar 29, 2002 10:32 pm
Location: Pennsylvania, USA

Both my DFI and my PM-7900 (M520) use the i430VX chipset.

To the best of my knowledge, this chipset cannot provide independent device timings on the same channel...correct?

That is...if I place a PIO-4/MWDMA-2 drive as Master, and a PIO-0 drive as Slave, the transfer rate of the Master (to the bus) will be slowed to PIO-0...correct??

This should be able to be seen in Windows with a utility such as HDTach, and at the DOS level, be seen with a utility such as Seagate's UATA100, which shows the system setting and maximum possible device timing...correct?
KachiWachi
The New Guy
Posts: 1451
Joined: Fri Mar 29, 2002 10:32 pm
Location: Pennsylvania, USA

Hmmm...nobody knows huh...or is it that this hardware is just too old to bother with anymore...???
ajzchips
El cheapo dude
Posts: 3048
Joined: Sat Mar 23, 2002 12:41 pm
Location: Madrid, Spain

I think that used to hold true on older systems. Newer chipsets (VX and above) have IDE buses that can support asynchronous modes...
I'm not too sure about this, so you'd need a second opinion.
Denniss
BIOS Guru
Posts: 3153
Joined: Thu Mar 21, 2002 8:16 pm
Location: Near Hannover (CEBIT) Germany
Contact:

AFAIK independent timing is only possible with UDMA modes - maybe with DMA modes too but not with PIO .
KachiWachi
The New Guy
Posts: 1451
Joined: Fri Mar 29, 2002 10:32 pm
Location: Pennsylvania, USA

I have a reason for asking that I'm withholding for the moment.

As I said in the first post, from what I have seen/read in the past, it appeared the VX could not do independant timing at all, in any mode. I think I saw this either on a web page, or in the chipset documents themselves.

Anyone have a VX based board they can test this on (besides me...lol!!)??
soupy
Absentee administrator
Posts: 1086
Joined: Thu Mar 21, 2002 1:06 am
Location: Ontario, Canada
Contact:

http://www.pcguide.com/ref/mbsys/chip/p ... 0VX-c.html

No independent device timing for IDE/ATA devices.
Flash your BIOS at your own risk.
Denniss
BIOS Guru
Posts: 3153
Joined: Thu Mar 21, 2002 8:16 pm
Location: Near Hannover (CEBIT) Germany
Contact:

Sound a little strange .
IDE-Controller should be in chipset southbridge and VX and HX use the same southbridge .
KachiWachi
The New Guy
Posts: 1451
Joined: Fri Mar 29, 2002 10:32 pm
Location: Pennsylvania, USA

See...that's what I thought. Conflict of information???

Intel doc's show that a PIIX3 (82371SB) CAN DO independent IDE timings on the same channel (29055002.pdf and related documents).

Could it be revision related???

I have an SU093 (B-0 stepping) southbridge in both machines. I just got a 40GB Maxtor ATA100 drive (free...again!!) to replace the two very old PIO-3 drives that were installed on the DFI system. Before that, I decided to run HDTach on them. Both reported dismal burst speeds of 3500/4300 kps and average read speeds of 2400/2900 kps (ugh). (One drive is MWDMA1 capable, the other is only MWDMA0 - 4.75MB/s max!!) Because of the way things were setup, I copied the system over to the new Maxtor, and left the MWDMA0 drive still attached as its slave, where the swap file and other temporary files were located.

After rebooting, I just wanted to see if the old drive would drag down the new one, and by how much.

Surprise...Surprise!!

The Maxtor responded like the old drive wasn't even there...posting a burst speed of 16000 kps and an average read speed of 15000 kps...just as it did when I tested it alone on the channel. (Note that the DFI system can only run at PIO-4/MWDMA2 - 16.7MB/s.)

So the question is...can anybody else repeat this, and maybe see if it is related to the chipset revision level or not??

On a side note...the MWDMA1 drive never achieved the 13.3MB/s transfer rate (or even PIO-3's 11.1MB/s)...I'm guessing that just because the drive had those features (ATA-2), it still didn't have to meet them if it physically couldn't...like due to slower RPM or whatever...correct?
Denniss
BIOS Guru
Posts: 3153
Joined: Thu Mar 21, 2002 8:16 pm
Location: Near Hannover (CEBIT) Germany
Contact:

SUO93 is the standard PIIX3 Southbridge .
Only a few of earlier design were used - some BUGs with USB .
With your MWDMA1 drive : It's the same with UDMA133 drives and 40-60MB/sec transfer speed .
Physical limitations
Post Reply