Yet Another UDMA/P2B issue (WDAC33200)
richa...@usa.net
P2B.
However, none of them quite described my exact problem.
The Goods:
-P2B rev. 1.02
-Bios 1005b15
-WD33100 (non-UDMA) as master on primary interface
-WD33200 (UDMA-capable) as slave on primary interface
-<18" IDE cable
The Problem:
Occurs when copying multi-megabyte files from the master to the slave drive.
About every fifth copy, the system will come to an almost complete halt and
the HDD LED will blink rapidly (3 times a second) for about 7-8 seconds.
When this little wig-out is complete, the file will have been copied without
errors.
I *think* this problem started after flashing the BIOS from 1002 to 1005.15
(beta).
The DMA box in Win98 is checked for both drives.
Am I an idiot for slaving a UDMA drive to a non-UDMA drive, or is something
else at work here? I may try flashing the BIOS back to 1002.
Thoughts? (besides for "you're an idiot for using a beta BIOS!")
-----== Posted via Deja News, The Leader in Internet Discussion ==-----
http://www.dejanews.com/rg_mkgrp.xp Create Your Own Free Member Forum
Steve Hull
This has nothing to do with the problems you're having, but have you
considered moving one of your drives to the secondary IDE port? This
will require a 2nd IDE cable, but they're available for less than $15
at most computer "Super Stores". Once you've done that, you can move
your swap file to a different drive than the one your O/S is installed
on. It's a relatively inexpensive tweak that yields a very noticeable
performance improvement. Who knows, it may fix your problem, too.
- Steve
----------------------------------------------------------------
In an effort to thwart Spam, my return address has been altered.
To reply, remove "_NOSPAM" from my return address.
richa...@usa.net
dr_baud...@ibm.net (Steve Hull) wrote:
> This has nothing to do with the problems you're having, but have you
> considered moving one of your drives to the secondary IDE port?
Yes. I have a CD-ROM over there, though. I'm leery of chaining an HDD to a
CD- ROM. The second HDD is earmarked for a second machine that I'm building
so it doesn't really matter anymore, anyway...
RM
PS. One thing I forgot to mention is that disabling UDMA support in CMOS fixed
the problem...
Hilton Travis
>
>-P2B rev. 1.02
>-Bios 1005b15
>-WD33100 (non-UDMA) as master on primary interface
>-WD33200 (UDMA-capable) as slave on primary interface
>-<18" IDE cable
>
>The Problem:
>
>Occurs when copying multi-megabyte files from the master to the slave
drive.
>About every fifth copy, the system will come to an almost complete halt and
>the HDD LED will blink rapidly (3 times a second) for about 7-8 seconds.
>When this little wig-out is complete, the file will have been copied
without
>errors.
For starters, I'd split the drives for speed reasons. It is generally
quicker to copy between channels than over the same channel. This will also
stop any UDMA/non-UDMA problems that MAY be occuring.
>I *think* this problem started after flashing the BIOS from 1002 to 1005.15
>(beta).
If you go to ftp.asuscom.de and look for the BXLS1005 BIOS, you will find it
is the release version (or at least it seems to be). Try using this BIOS
and see if that helps solve your woes.
>The DMA box in Win98 is checked for both drives.
This is good.
>Am I an idiot for slaving a UDMA drive to a non-UDMA drive, or is something
>else at work here? I may try flashing the BIOS back to 1002.
The UDMA rules are the same that applied for DMA drives, as far as I am
aware. If you wish to enable DMA for a drive, then all drives on that
channel must support DMA and be configured for DMA transfers. If a DMA and
PIO drive are on the same channel, data corruption can occur. A UDMA drive
can be connected with a DMA drive with no problems, I was running this for
quite a while, but should NOT be used on the same channel as an older PIO
drive.
Flashing back to 1002, or up to a later 1005 BIOS may solve the trick.
ASUStek have some serious problems with all the 100x BIOSes for the P2B
series (and P2L97-series) boards, and this may just be another one to add to
the ever-expanding list.
>Thoughts? (besides for "you're an idiot for using a beta BIOS!")
You are not an idiot for using a beta BIOS, especially for a P2B board - the
one that came on the board contains a number of bugs, and the later betas
should help your system to be more stable. I'd try the 1005 from the German
site and see if that works, though.
-Hilton Travis