"kony" said in :
Ummmm, that had nothing to do with the 2nd device, the 440BX chipset
only supports up to ATA33, there is no possible way for any device in
any possible configuration to run any faster than ATA33 from it's
onboard controller.

And hard drives started at UDMA-33, huh? Forgot about the older PIO
modes?

On Wed, 10 Mar 2004 05:39:04 -0600, "*Vanguard*"
wrote:

"kony" said in :
Ummmm, that had nothing to do with the 2nd device, the 440BX chipset
only supports up to ATA33, there is no possible way for any device in
any possible configuration to run any faster than ATA33 from it's
onboard controller.

And hard drives started at UDMA-33, huh? Forgot about the older PIO
modes?


Forget? No, I meant nothing more than I wrote... at least the way it read
it me, it seemed as though you were implying that this was causing the
hard drives to drop down to ATA33 mode, which was the highest mode they
could support anyway. If you're claiming that PIO mode ATAPI devices will
slow down the BC chipset, I won't refute that as I don't recall enough
first-hand experiences trying to run PIO mode as late as the BX chipset.

Thanks Kony, that wraps it up, its all there - a site worthy of bookmark
methinks:-)

Steve

"kony" wrote in message
...
On Wed, 10 Mar 2004 10:22:23 -0000, "Steve James"
wrote:

OK Thanks for info, though a few different views not sure I am certain
which
way to go

http://www.storagereview.com/guide20...onfTiming.html
http://www.storagereview.com/guide20...rformance.html

"kony" said in :
And hard drives started at UDMA-33, huh? Forgot about the older PIO
modes?


Forget? No, I meant nothing more than I wrote...

Okay, reread your reply and you're right that UDMA-33 was the highest
mode supported by Intel's 440BX chipset. I just wanted to make sure the
it was understood that all modes (PIO and UDMA) might mix okay, might
not, and I wasn't sure about the 440BX ... until today when I finally
decided to look.

I got rid of my last Intel 440BX box a little while ago. As I recall,
that one was configured with the ATA devices (hard drives) on IDE0 and
the ATAPI devices (CD-RW & DVD-ROM) on IDE1. Over the lifetime of this
mobo, it first started out with really old drives scavanged from even
older boxes and just ran DOS and then SCO and, I think, Solaris x86, so
it did have some old PIO-only mode hard drives. About when Windows got
put on it was when the drives were updated. In fact, because the mobo's
IDE ports only supported up to UDMA-33, I put in a Promise Ultra100 to
better support the UDMA-66 & -100 drives and so each had its own channel
(which also meant the ATAPI devices could each be placed on their own
channel on the mobo's IDE ports).

I'm not sure the ATA and ATAPI devices really had to be separated on
different channels for the Intel 440BX. I think that was how it was
first configured because it was unknown at the start if independent
timing was supported per channel; not everything was on the Internet
back then for ease in lookup. I had another coworker using the box who
kept claiming the ATA & ATAPI devices should be separated on different
channels and since it didn't hurt I just let him have his way. Although
we don't have that old 440BX box anymore, I finally decided to check on
my break. Went to Intel and dug around for awhile and found for the
82371AB PCI-to-ISA chip (southbridge that provides the IDE channels),
"Integrated IDE Controller: Independent timing of up to 4 drives." Aha!
I was right all along. Oh well, running in the
lowest-common-denominator hardware configuration didn't hurt, either.

Thanks for participating. Always something to learn in the newsgroups.

"kony" said in :
On Wed, 10 Mar 2004 10:22:23 -0000, "Steve James"
wrote:

OK Thanks for info, though a few different views not sure I am
certain which way to go

http://www.storagereview.com/guide20...onfTiming.html

http://www.storagereview.com/guide20...rformance.html

Thanks for the links. Ever visit www.pcguide.com who this link is
referencing (in a static copy rather than a link)? After a year or
more, it sure would be nice if www.pcguide.com someday fixes their
search function to it works again. Having to drill around through the
index takes a lot longer. I used to go there a lot to look up some info
but using Google is often faster going through the matches than having
to drill through pcguide.com. They/he really needs to fix his web site,
or move back to a web host provider that includes Frontpage extensions.
Unfortunately it doesn't look like it is being maintained to be kept up
to date. Alas, twas a good reference site.

kony wrote:
On Wed, 10 Mar 2004 10:22:23 -0000, "Steve James"
wrote:

OK Thanks for info, though a few different views not sure I am
certain which way to go

http://www.storagereview.com/guide20...onfTiming.html
http://www.storagereview.com/guide20...rformance.html

Good references. However, note the "Copyright 1997-2000......" at the bottom
of the page. So, what that page refers to as "Three years old" is in fact
seven years old. Also, it says "Most new motherboards......." You can read
that as "All new motherboards...." fairly safely.
--
~misfit~

On Wed, 10 Mar 2004 13:32:51 -0600, "*Vanguard*"
wrote:

snip

"Integrated IDE Controller: Independent timing of up to 4 drives." Aha!
I was right all along. Oh well, running in the
lowest-common-denominator hardware configuration didn't hurt, either.


Prior to the 440BX, for example early socket 7, boards supported IDT
(independent device timing) but what I'm less certain of is when, if ever,
the controllers were able to run one device in PIO mode while the other on
same channel is in DMA/UDMA mode... I was under the impression that both
modes couldn't be supported simultaneously per channel, but then I've
seen at least a couple of posts claiming IDT makes that possible... never
bothered testing that myself since I've never had any desire to run PIO
drives after there were UDMA alternatives.

On Thu, 11 Mar 2004 11:24:23 +1300, "~misfit~"
wrote:

kony wrote:
On Wed, 10 Mar 2004 10:22:23 -0000, "Steve James"
wrote:

OK Thanks for info, though a few different views not sure I am
certain which way to go

http://www.storagereview.com/guide20...onfTiming.html
http://www.storagereview.com/guide20...rformance.html

Good references. However, note the "Copyright 1997-2000......" at the bottom
of the page. So, what that page refers to as "Three years old" is in fact
seven years old. Also, it says "Most new motherboards......." You can read
that as "All new motherboards...." fairly safely.


Yep, it's pretty old by today's standards. Most PIO drives don't even
work anymore unless they were just sitting on a shelf/unused.

kony wrote:
On Wed, 10 Mar 2004 13:32:51 -0600, "*Vanguard*"
wrote:

snip

"Integrated IDE Controller: Independent timing of up to 4 drives."
Aha! I was right all along. Oh well, running in the
lowest-common-denominator hardware configuration didn't hurt, either.


Prior to the 440BX, for example early socket 7, boards supported IDT
(independent device timing) but what I'm less certain of is when, if
ever, the controllers were able to run one device in PIO mode while
the other on same channel is in DMA/UDMA mode... I was under the
impression that both modes couldn't be supported simultaneously per
channel, but then I've seen at least a couple of posts claiming IDT
makes that possible... never bothered testing that myself since I've
never had any desire to run PIO drives after there were UDMA
alternatives.

I've tested it, althought with my relatively new nForce2 Ultra 400 board,
and having an old 120MB drive in PIO mode running alongside a modern drive
running UltraDMA mode 5 on the same channel made virtually no difference to
access times of the modern drive.
--
~misfit~

What about sequential data transfer?

~misfit~ wrote:

I've tested it, althought with my relatively new nForce2 Ultra 400 board,
and having an old 120MB drive in PIO mode running alongside a modern drive
running UltraDMA mode 5 on the same channel made virtually no difference to
access times of the modern drive.
--
~misfit~

--
Mike Walsh
West Palm Beach, Florida, U.S.A.