9 Dec
2004
9 Dec
'04
1:28 p.m.
considering that the ATA-1 spec included 8 bit transfers from the harddrive, I don't
understand why no IDE drives that supported ATA-1 would not work on a PC/XT (unless the
DMA speed was the problem).
-----Original Message-----
From: Eric Smith <eric(a)brouhaha.com>
Sent: Dec 9, 2004 1:52 PM
To: "General Discussion: On-Topic and Off-Topic Posts"
<cctalk(a)classiccmp.org>
Subject: Re: Possible to speed up I/O subsystem of 5150?
Fred wrote:
Yes, there ARE some 8 bit IDE controllers,
but not all IDE drives will work with them.
To a first approximation, *NO* IDE drives will work with them.
8-bit support is not part of the official ATA specification. Even
back when 8-bit IDE adapters were being sold, not many drives worked
with them, and as soon as the adapters disappeared the drive makers
that had bothered to support them stopped doing so.
Eric
9 Dec
9 Dec
1:46 p.m.
Steve Thatcher wrote:
considering that the ATA-1 spec included 8 bit
transfers from the
harddrive, I don't understand why no IDE drives that supported ATA-1 would
not work on a PC/XT (unless the DMA speed was the problem).
I wasn't aware that 8-bit was included in ATA-1; I don't have a copy of
ATA-1 or ATA-2. Was the 8-bit support mandatory or optional? In any
case, it was certainly gone by ATA-3. I very much doubt that any ATA
drive manufactured in the last 15 years has 8-bit support.
Eric
2:03 p.m.
Am 9 Dec 2004 11:46 meinte Eric Smith:
Steve Thatcher wrote:
> considering that the ATA-1 spec included 8 bit transfers from the
> harddrive, I don't understand why no IDE drives that supported ATA-1 would
> not work on a PC/XT (unless the DMA speed was the problem).
I wasn't aware that 8-bit was included in ATA-1; I
don't have a copy of
ATA-1 or ATA-2. Was the 8-bit support mandatory or optional? In any
case, it was certainly gone by ATA-3. I very much doubt that any ATA
drive manufactured in the last 15 years has 8-bit support.
Well, original ATA was nothing more than a straight forward
adaption of the MFM controler - and that one was still good
for 8 Bit access. Also, there was never (at least in the beginning)
a formal Standard, so as usual everything was possible ...
just remember all the problems of Master/Slave combinations
from different vendors.
Gruss
H.
--
VCF Europa 6.0 am 30.April und 01.Mai 2005 in Muenchen
http://www.vcfe.org/
2:19 p.m.
Hans wrote:
Well, original ATA was nothing more than a straight
forward
adaption of the MFM controler -
It was a straightforward adaptation of the IBM *AT* MFM controller. The
standard IBM XT MFM controller was much different.
And the IBM AT MFM controller was basically an ISA-bus version of
the earlier WD task-file based controllers, which supported 8-bit
transfers only.
and that one was still good for 8 Bit access.
It's been so long since I've written any code for the original
IBM AT MFM controller that I don't recall whether it was actually
usable with 8-bit data register access or not (the control registers
are all only 8 bits). I don't have an AT Tech Ref handy to check.
Eric
7:30 p.m.
Well, original ATA was nothing more than a straight
forward
adaption of the MFM controler - and that one was still good
for 8 Bit access. Also, there was never (at least in the beginning)
You do realise that the XT disk controller and AT disk controller are
totally different things (and have a different software interface). AFAIK
the AT hard disk controller assumes a 16 bit bus, at least for the data
transfer register.
-tony
7332
days inactive
7332
days old
4 comments
4 participants
participants (4)
-
ard＠p850ug1.demon.co.uk -
eric＠brouhaha.com -
Hans.Franke＠siemens.com -
melamy＠earthlink.net