6 Nov
2009
6 Nov
'09
1:42 p.m.
Hi! My good friend John and I are working on a joint www.S100Computers.com &
N8VEM <http://groups.google.com/group/n8vem> S-100 IDE project. It is based
on the following design by Peter Faasse.
http://www.pjrc.com/tech/8051/ide/wesley.html
John built a working prototype and some software and I am making a PCB. The
PCB is in trace route optimizing phase now. I can't say exactly when a PCB
will be available but if you think you might be interested please let me
know. PCBs will probably cost *about* $22 give or take some plus shipping.
If you do want a PCB it is important you tell me beforehand so I can include
it in the manufacturing order.
The board will connect to regular IDE devices and includes mounting areas
and interfaces for laptop (2.5" with the high density 44 pin connector) and
CF adapters. John has his prototype working with a CF adapter mounted on the
PCB and it seems to work fine.
Thanks and have a nice day!
Andrew Lynch
6 Nov
6 Nov
2:16 p.m.
How are you getting around the 8meg max drive size common with CPM ?
Are you splitting it into many small drives in hardware on in the driver
like Morows dod for the M26 ?
I have an old MPM system that had an M26 on it I would like to bring back to life.
The other Bob
On Fri, 6 Nov 2009 16:42:28 -0500, Andrew Lynch wrote:
Hi! My good friend John and I are working on a joint
www.S100Computers.com &
N8VEM <http://groups.google.com/group/n8vem> S-100 IDE project. It is based
on the following design by Peter Faasse.
John built a working prototype and some software and I
am making a PCB. The
PCB is in trace route optimizing phase now. I can't say exactly when a PCB
will be available but if you think you might be interested please let me
know. PCBs will probably cost *about* $22 give or take some plus shipping.
If you do want a PCB it is important you tell me beforehand so I can include
it in the manufacturing order.
The board will connect to regular IDE devices and
includes mounting areas
and interfaces for laptop (2.5" with the high density 44 pin connector) and
CF adapters. John has his prototype working with a CF adapter mounted on the
PCB and it seems to work fine.
Thanks and have a nice day!
Andrew Lynch
7 Nov
7 Nov
7:35 p.m.
Bob's right.
The CP/M 2 logical drive size limit is 8mb, however there are up to 16
possible logical drives making the physical drive size 128mb. Now that
is a
basic limit. There is nothing to say you can't make partitions mountable
either, though the BIOS would have to be written to suitably support that.
The CP/M 3 logical drive size limit is 32mb, however there are up to 16
possible logical drives making the drive size 512mb.
If you run any of the CP/M 2 improved versions (Zsystem, P2Dos, Novados
and friends)
that limit is 1gb for a logical drive(note that means the allocation
size is 32K and a
bit on the large granular side). You still can address 16 drives.
Allison
Bob Bradlee wrote:
How are you getting around the 8meg max drive size
common with CPM ?
Are you splitting it into many small drives in hardware on in the driver
like Morows dod for the M26 ?
I have an old MPM system that had an M26 on it I would like to bring back to life.
The other Bob
On Fri, 6 Nov 2009 16:42:28 -0500, Andrew Lynch wrote:
Hi! My good friend John and I are working on a
joint www.S100Computers.com &
N8VEM <http://groups.google.com/group/n8vem> S-100 IDE project. It is based
on the following design by Peter Faasse.
John built a working prototype and some software
and I am making a PCB. The
PCB is in trace route optimizing phase now. I can't say exactly when a PCB
will be available but if you think you might be interested please let me
know. PCBs will probably cost *about* $22 give or take some plus shipping.
If you do want a PCB it is important you tell me beforehand so I can include
it in the manufacturing order.
The board will connect to regular IDE devices and
includes mounting areas
and interfaces for laptop (2.5" with the high density 44 pin connector) and
CF adapters. John has his prototype working with a CF adapter mounted on the
PCB and it seems to work fine.
Thanks and have a nice day!
Andrew Lynch
7:52 p.m.
On Sat, 7 Nov 2009 08:12:28 am Andrew Lynch wrote:
Hi! My good friend John and I are working on a joint
www.S100Computers.com
& N8VEM <http://groups.google.com/group/n8vem> S-100 IDE project. It is
based on the following design by Peter Faasse.
http://www.pjrc.com/tech/8051/ide/wesley.html
I'm using this design on a single board 8088 micro I've built. The bottleneck
is horrific, but I can see how it could be useful when I/O space is limited.
Later, when I made the IDE interface for my 8080 computer, I wired it directly
to the bus with some glue logic, and I ignored the upper eight bits of the 16-
bit data bus. Considering I'm using a 120MB hard drive and CP/M has a limit of
8MB I didn't think the wasted space would be much of an issue. 8MB is lots of
space for CP/M, especially after I was used to using 320kB floppies.
You're welcome to look at the CBIOS too.
http://kaput.homeunix.org/~thrashbarg/cbios.asm
Alexis.
8:35 p.m.
While I have nothing but fondness for recollections of CP/M on x80
systems, I've wondered that since memory is cheap, why not resurrect
some of the Unix-cousin attempts? I think Morrow was advertising one
around 1980 and there were others.
I found one here:
http://www.dougbraun.com/uzi.html
It would seem that if the inode arithmetic could be extended to 24 or
32 bits, you could mount filesystems in the GB range. It might be
kind of interesting.
FWIW,
Chuck
8 Nov
8 Nov
5:30 a.m.
There is a version ported to Z180 of Uzi. that would be the
best and most complete starting point.
Allison
Chuck Guzis wrote:
While I have nothing but fondness for recollections of
CP/M on x80
systems, I've wondered that since memory is cheap, why not resurrect
some of the Unix-cousin attempts? I think Morrow was advertising one
around 1980 and there were others.
I found one here:
http://www.dougbraun.com/uzi.html
It would seem that if the inode arithmetic could be extended to 24 or
32 bits, you could mount filesystems in the GB range. It might be
kind of interesting.
FWIW,
Chuck
5:36 a.m.
Alexis wrote:
On Sat, 7 Nov 2009 08:12:28 am Andrew Lynch wrote:
Don't forget that limit is per Logical Drive. You can partition that
120 MB into multiple drives.
I have 8 CP/M systems 3 of which are S100 using hard disks minimally
31mb is size. A few are IDE and the latest two are CF (64mb or 8
logical drives).
8Mb does make it far nicer, but a whole drive for C and another for
Assembly programming (and so on) is very useful.
Allison
Hi! My good friend John and I are working on a
joint www.S100Computers.com
& N8VEM <http://groups.google.com/group/n8vem> S-100 IDE project. It is
based on the following design by Peter Faasse.
http://www.pjrc.com/tech/8051/ide/wesley.html
I'm using this design on a single board 8088 micro I've built. The bottleneck
is horrific, but I can see how it could be useful when I/O space is limited.
Later, when I made the IDE interface for my 8080 computer, I wired it directly
to the bus with some glue logic, and I ignored the upper eight bits of the 16-
bit data bus. Considering I'm using a 120MB hard drive and CP/M has a limit of
8MB I didn't think the wasted space would be much of an issue. 8MB is lots of
space for CP/M, especially after I was used to using 320kB floppies.
You're welcome to look at the CBIOS too.
http://kaput.homeunix.org/~thrashbarg/cbios.asm
Alexis.
6 a.m.
Hi!
2009/11/7 Andrew Lynch <lynchaj at yahoo.com>:
Hi! My good friend John and I are working on a joint
www.S100Computers.com &
N8VEM <http://groups.google.com/group/n8vem> ?S-100 IDE project. It is based
on the following design by Peter Faasse.
This 8255 schematik is damn slow since requires about 20..30 commands,
time delay loops and massive I/O. The resulting speed is about 10..20 K/s
There is better schematic floatind around, which use "shadow" 8-bit latch
(273/573) for HI and LO bytes to mux/demux 16-bit IDE bus to 8-bit
system bus. It uses _system_ R/W, does not require loops and thus
much faster.
One example (with lot of links):
http://www.hanssummers.com/computers/cpcng/ide/index.htm
Another example, with ASCII-ART schematic and CP/M BIOS sources
(with Russian comments):
http://sensi.org/~svo/scalar/media/w/hdd.zip
This scematic gives full I/O bus speed divided by 2 for reading but
requres some "pointer magic" for writing and writing is a bit slower.
(the LO register at 50H, the HI "shadow" register at 58H)
;write sector
SAVE: INR L ; select an ADDR+1 i.e. HI byte
MOV A,M
OUT 58H ; prepare HI byte _before_ IDE bus transaction
DCR L ; return to ADDR i.e. LO byte
MOV A,M
OUT 50H ; write LO byte and initiate IDE bus transaction
INR L
INR L
JNZ SAVE
There is also VCI XTIDE project. They seems use similar schematic
with 573 lathes but they does not publish schematic and sources
http://wiki.vintage-computer.com/index.php?title=XTIDE_project
P.S. Buildind IDE controller for 8-bit is a pandemy like a swine flu... ;)
--
-=AV=-
5300
days inactive
5302
days old
7 comments
6 participants
participants (6)
-
ajp166＠verizon.net -
alec＠sensi.org -
caveguy＠sbcglobal.net -
cclist＠sydex.com -
lynchaj＠yahoo.com -
thrashbarg＠kaput.homeunix.org