26 May
1999
26 May
'99
9:49 p.m.
Woohoo! I got my vs3100 running with the hobbiest CD! Installed it on the
big hard disk too! Now I just have to find where I printed out the licence
before my PC ate itself.
--
Jim Strickland
jim(a)DIESPAMMERSCUMcalico.litterbox.com
-----------------------------------------------------------------------
Vote Meadocrat! Bill and Opus in 2000 - Who ELSE is there?
-----------------------------------------------------------------------
27 May
27 May
4:04 a.m.
Woohoo! I got my vs3100 running with the hobbiest CD!
Installed it on the
big hard disk too! Now I just have to find where I printed out the licence
before my PC ate itself.
My VS3100/m76 has been running it for two days... ;)
It's good ole VMS, and in my case an upgrade from 5.5 to 7.2... yeha I'm
modern again.
Allison
6:22 a.m.
Woohoo! I got my vs3100 running with the
hobbiest CD! Installed it on the
big hard disk too! Now I just have to find where I printed out the licence
before my PC ate itself.
My VS3100/m76 has been running it for two days... ;)
It's good ole VMS, and in my case an upgrade from 5.5 to 7.2... yeha I'm
modern again.
Allison 
6:42 a.m.
A non-text attachment was scrubbed...
Name: not available
Type: text
Size: 652 bytes
Desc: not available
Url :
http://www.classiccmp.org/pipermail/cctalk/attachments/19990527/7f8479fa/at…
7:34 a.m.
I'm hoping
to get the vs3100 up here. I wish DEC had put BasicPlus
or whatever the Basic from VMS was for hobbiest use so we could really
show that Basic code can be done quite nicely on the Vax (and PDP's).
C and Pascal are ok -- but the Visual Basic folks need to see a real
nice basic in action.
You do have VAX Basic on the VAX Hobbyist CD. However, the space that
should have gone to it on the Alpha CD is taken up with Java and Netscape
(which you can download, grrrr.) I was really hoping to get the latest and
greatest version of Alpha BASIC. Still, except for that, the CD's have the
main stuff I'm interested in.
Zane 
7:55 a.m.
New subject: Bringing up a CPM
Hi
I am having problems in bringing up a CPM on an old
IMSAI. Here are some of the issues I have:
1. Are the directory sectors interleaved like the file sectors are?
2. When I checked the size of the CPM image that was on the
unoffical CPM site, it was 60 sectors in size. When I looked
at the manual, it said that this would be loaded by a boot
loader that would load in locations 3400H to 4D80H. This is
only 51 sectors. This also matches the remaining track
zero plus track 1 of a single density disk. What am I suppose
to do?
Thanks
Dwight
10:17 a.m.
New subject: Bringing up a CPM
On Thu, 27 May 1999, Dwight Elvey wrote:
1. Are the directory sectors interleaved like the file sectors are?
The boot sectors (reserved tracks) are generally not, the rest of the disk
is. CPM it self does not care it's a BIOS issue..
2. When I checked the size of the CPM image that was
on the
unoffical CPM site, it was 60 sectors in size. When I looked
at the manual, it said that this would be loaded by a boot
loader that would load in locations 3400H to 4D80H. This is
only 51 sectors. This also matches the remaining track
zero plus track 1 of a single density disk. What am I suppose
to do?
that means the saved image has either dead space at the end or a bios
there. It could be that teh file was off a system with 4k allcoation
blocks so fine sized tend to round up.
The standard size for cpm a is:
CCP 800h 2048 bytes
Bdos E00h 3584 bytes
BIOS variable 8"sssd example is 1500bytes.
This is generally stuffed on to the sssd 8" as the first two tracks of the
disk, 52 sectors total or 6656 bytes for all three components.
Allison
10:48 a.m.
New subject: Bringing up a CPM
Its definitely up to the bios, when I wrote the BIOS for the Cromemco 16FDC
I ran out of space, so I took advantage of the fact that a) I could write
full tracks and b) that most SD disks could handle DD and wrote the first
two tracks of the disk double density, even if the disk was "formatted" for
single density. (I also wrote the formatter (in Turbo PASCAL no less!))
--Chuck
At 03:17 PM 5/27/99 -0400, you wrote:
On Thu, 27 May 1999, Dwight Elvey wrote:
1. Are the directory sectors interleaved like the file sectors are?
The boot sectors (reserved tracks) are generally not, the rest of the disk
is. CPM it self does not care it's a BIOS issue..
2. When I checked the size of the CPM image that
was on the
unoffical CPM site, it was 60 sectors in size. When I looked
at the manual, it said that this would be loaded by a boot
loader that would load in locations 3400H to 4D80H. This is
only 51 sectors. This also matches the remaining track
zero plus track 1 of a single density disk. What am I suppose
to do?
that means the saved image has either dead space at the end or a bios
there. It could be that teh file was off a system with 4k allcoation
blocks so fine sized tend to round up.
The standard size for cpm a is:
CCP 800h 2048 bytes
Bdos E00h 3584 bytes
BIOS variable 8"sssd example is 1500bytes.
This is generally stuffed on to the sssd 8" as the first two tracks of the
disk, 52 sectors total or 6656 bytes for all three components.
Allison
11:13 a.m.
New subject: Bringing up a CPM
Chuck McManis <cmcmanis(a)mcmanis.com> wrote:
Its definitely up to the bios, when I wrote the BIOS
for the Cromemco 16FDC
I ran out of space, so I took advantage of the fact that a) I could write
full tracks and b) that most SD disks could handle DD and wrote the first
two tracks of the disk double density, even if the disk was "formatted" for
single density. (I also wrote the formatter (in Turbo PASCAL no less!))
Hi Chuck
This isn't an option for me. I'm trying to get it onto a SSSD disk
system. I can't do formatting/reading in double density with the controller
I have. Since I am bringing this machine up from gound zero,
I had to write my own formatter in assembly and hand toggle it
in.
I am now to the point that I can download data through the serial
port from a PC. Still, I have to toggle in the start address.
I had a little space left on my boot sector so I wrote a simple
serial transfer ( no hand shake ).
I guess the issue is that the additional space is the BIOS and
since I'll be replacing this with my own BIOS, and I write efficient
code, I don't have to worry about it until I run out of space?
Dwight
11:37 a.m.
New subject: Bringing up a CPM
I guess the issue is that the additional space is the
BIOS and
since I'll be replacing this with my own BIOS, and I write efficient
code, I don't have to worry about it until I run out of space?
Dwight
???HUH??? Figure this if your running a SSSD 8" to be compatable with
media out there the system has two tracks to use for the CCP, BDOS and
bios. Thats 52 sectors!
The first sector DOES NOT HAVE TO BE THE BOOT.
The FIRST 1600h bytes are ccpand bdos (44 sectors).
The bios can be up to 8 sectors in length but the buffers can be placed
on the end so they dont have to be stored in the disk (uninitialized ram).
Infact the system can be in rom and copied to ram and mapped out.
The disk if your going to read SSSD disks must use the skew of 6, 2
tracks reserved with 63 directoriy entries(may be different , from
memory).
Allison
12:25 p.m.
New subject: Bringing up a CPM
allisonp(a)world.std.com wrote:
Hi
It does on my system unless I toggle in a loader or have
EPROMs to do it. I'd like to not have any EPROMs if I don't
need it. The controller automatically loads track 0 sector 1.
I start executing at address zero and it boot loads the
CPM system. The boot loader then jumps to the start of
the CPM and it over writes info at address zero with what
it requires there. As far as I know, I am following the examples
for a non-MDS800 type system and following exactly what is
shown on 6-14/6-15 of the CPM manual that I got from the
unofficial site. The mapping shows that the first sector
is a cold start loader and the CPM ( CCP ) part doesn't start until
the second sector.
It would seem that there was additional BOIS info loaded that
didn't fit into the 51 sectors. It sounds like I can ignore
this because I'll be replacing it anyway.
Let me say, no EPROM's, boots from disk.
Dwight
I guess the
issue is that the additional space is the BIOS and
since I'll be replacing this with my own BIOS, and I write efficient
code, I don't have to worry about it until I run out of space?
Dwight
???HUH??? Figure this if your running a SSSD 8" to be compatable with
media out there the system has two tracks to use for the CCP, BDOS and
bios. Thats 52 sectors!
The first sector DOES NOT HAVE TO BE THE BOOT.
1:08 p.m.
New subject: Bootstrapping CPM
At 02:25 PM 5/27/99 -0700, Dwight wrote:
... The controller automatically loads track 0 sector
1.
Presumably the controller accomplishes this with and EPROM of some sort :-)
Some controllers like the 16FDC have a monitor in ROM that is standalone.
Anyway, you needn't worry because you are almost there.
I start executing at address zero and it boot loads
the
CPM system.
When you say "I" here, do you mean the controller or your code?
The core of a disk loader is very simple (note i've lost how many sectors
the standard 8" 3740 format disk had on it (it was twenty something) I'm
using 'n' as a place holder)
Set track to 0
read track
copy sectors 1 - n to 0xE000
set track to 1
read track
copy sectors 0 - n to the next block of memory past
(0xE000+(n-1)*128)
jump to 0xE000
That was it! The system was booted.
Now there are some variables, if your bios was larger then the space
allocated by default you could "move" CPM (MOVCPM) to start on a different
address other than E000, this would affect your transient program area
(TPA) available for programs (and note that Turbo PASCAL programs couldn't
run if they had been compiled on a system with a larger TPA!)
Also some BIOSes (like mine) coded the disk geometry into the boot block so
that a single boot program would boot multiple disk types. I had an 8 bit
value that encoded drive size, # of sides, dd/sd, and "pertec" style
seeking. Other BIOSes actually wrote different boot blocks depending on the
disk they were written on.
--Chuck
2:48 p.m.
New subject: Bringing up a CPM
On Thu, 27 May 1999, Dwight Elvey wrote:
It does on my system unless I toggle in a loader or
have
EPROMs to do it. I'd like to not have any EPROMs if I don't
need it. The controller automatically loads track 0 sector 1.
I start executing at address zero and it boot loads the
CPM system. The boot loader then jumps to the start of
the CPM and it over writes info at address zero with what
it requires there. As far as I know, I am following the examples
for a non-MDS800 type system and following exactly what is
shown on 6-14/6-15 of the CPM manual that I got from the
unofficial site. The mapping shows that the first sector
is a cold start loader and the CPM ( CCP ) part doesn't start until
the second sector.
It would seem that there was additional BOIS info loaded that
didn't fit into the 51 sectors. It sounds like I can ignore
this because I'll be replacing it anyway.
Let me say, no EPROM's, boots from disk.
Dwight, in looking at an Imsai disk that I have here - which has Imsai
CP/M v1.33 on it - the BIOS begins at sector 19 and ends on 24.
Presumably, you are working with v2.2 so Tim's quote from the DRI guide
certainly follows as 2.2 is most certainly larger than the prior versions.
The directory is skewed and the first file recorded is interleaved with
it.
- don
11 a.m.
New subject: Bringing up a CPM
allisonp(a)world.std.com wrote:
The standard size for cpm a is:
CCP 800h 2048 bytes
Bdos E00h 3584 bytes
BIOS variable 8"sssd example is 1500bytes.
This is generally stuffed on to the sssd 8" as the first two tracks of the
disk, 52 sectors total or 6656 bytes for all three components.
Hi Allison
This also doesn't make sense. Track 0 sector 1 is the boot loader.
That leaves 25 sectors in the first track. 26 in the second track
makes a total of 51. This is contrary to your example. You only have
6528 bytes or a little fewer for BIOS variables.
Still that doesn't explain the 7680 bytes in the file from
the site. I still don't know what to do.
Dwight
11:20 a.m.
New subject: Bringing up a CPM
On Thu, 27 May 1999, Dwight Elvey wrote:
The stuff you deleted and didn't copy is the stuff you should read.
The copy you have has an extended bios at the end and Nad the 128bytes for
the boot are not always there as the boot is specific to the MDS800
example. So you may be right and the first sector _may_ be the boot.
Also of the cpm system has a really large bios or the copy is a "snapshot
of the system in ram from DDT the length may be way over the top.
Simply count of from the start and allow for the boot if there. The
beginning of the bios is easy to find as there will be some 17 jumps
starting at xx00h.
I guess I can just wrap a copy of the CCP and the BDOS and send them to
you this weekend. What address do you want to start at?
FYI: cpm nominally is supplied as a 8"sssd disk with the MDS800 bios so
the manual reads right and it's sysgen'd for 20kb of ram. Other vendors
did different things that often for the 5.25" floppy case were all over
the map.
No matter what copy you use, you have to rewrite the bios as likely it
does not match your IO or disks.
Allison
allisonp(a)world.std.com wrote:
>
> The standard size for cpm a is:
> CCP 800h 2048 bytes
> Bdos E00h 3584 bytes
> BIOS variable 8"sssd example is 1500bytes.
The word variable indicated
the length not the BIOS variables. Bioses for
CPM system can be very variable in size, I have one system that has a 32k
bank dedicated to the bios itself. It can be as small a few hundred
bytes plus the tables.
This is generally stuffed on to the sssd 8" as the first two tracks of the
disk, 52 sectors total or 6656 bytes for all three components.
Hi Allison
This also doesn't make sense. Track 0 sector 1 is the boot loader.
That leaves 25 sectors in the first track. 26 in the second track
makes a total of 51. This is contrary to your example. You only have
6528 bytes or a little fewer for BIOS variables.
Still that doesn't explain the 7680 bytes in the file from
the site. I still don't know what to do.
Dwight
9:45 a.m.
A non-text attachment was scrubbed...
Name: not available
Type: text
Size: 721 bytes
Desc: not available
Url :
http://www.classiccmp.org/pipermail/cctalk/attachments/19990527/59a0170a/at…
9:56 a.m.
At 11:45 AM 5/27/99 -0700, Zane H. Healy wrote:
As a long shot, try to find someone in technical support
at Thursby Software <http://www.thursby.com/>. They made
a DECnet for the Mac. My company once licensed their code
to port it to the Amiga. I bet they still have someone
in tech support who might want to help.
I once had my Amiga talking to a DEC Pro 350 and a MicroVAX
over DECnet, for example.
- John
Well, I'm
set then. Now to install and try to build DECnet for Linux.
Not that difficult.
(Maybe find a PDT11/150 and run Async DECnet/RT
again...)
I've yet to get my Linux box to talk to the PDP-11 I've got running
DECnet/RSX-11M (Phase IV). As I understand it I'm the
only person to try and get Linux to talk to a PDP-11 via DECnet.
7:25 a.m.
I'm hoping to get the vs3100 up here. I wish DEC had put BasicPlus
or whatever the Basic from VMS was for hobbiest use so we could really
show that Basic code can be done quite nicely on the Vax (and PDP's).
C and Pascal are ok -- but the Visual Basic folks need to see a real
nice basic in action.
Bill
I'd have to check the SPD but I remember Basic being in there too!
If I remember right it was Pascal/Basic/C/Cobol and a bunch of other
things.
Allison
7:32 a.m.
I'm hoping to get the vs3100 up here. I wish DEC had put BasicPlus
or whatever the Basic from VMS was for hobbiest use so we could really
show that Basic code can be done quite nicely on the Vax (and PDP's).
C and Pascal are ok -- but the Visual Basic folks need to see a real
nice basic in action.
Bill
I'd have to check the SPD but I remember Basic being in there too!
If I remember right it was Pascal/Basic/C/Cobol and a bunch of other
things.
Allison
6:28 a.m.
A non-text attachment was scrubbed...
Name: not available
Type: text
Size: 1132 bytes
Desc: not available
Url :
http://www.classiccmp.org/pipermail/cctalk/attachments/19990527/99e45fd9/at…
7:34 a.m.
How well does it do with a 16Mb system? All I really need is the Base OS,
16mb is the biggest machine I have! Does real well. It will run well in
a 8mb box too. The system required I think 4 or 5mb as a minimum for
sensible operation. Unlike PC based OSs VMS is NOT a memory hog.
Clustering Software, and DECnet. The VS3100/20 is
currently running 6.1,
and really needs to be running a minimum of 6.2 to cluster in nicely (it's
already clustered in). If the 7.2 memory requirements are to severe, I
think I've got a copy of 6.2 on TK50, just not sure if it's any good.
If you can run 6.1 you can run 7.2.
I'm still clustering with 5.5/5.4! However ,in the past clustering has
to be done with all node running the same or similar major version level
and pretty close on the revision level.
BTW, anyone know of any real advantages to running
DECnet Phase V instead of
DECnet Phase IV? My cluster is currently running DECnet Phase V, but I'm
Phase V was to solve the problem of more than 65000 nodes in phase IV and
I don't know it adds anything more of any great value for smaller systems.
Allison
9:39 a.m.
A non-text attachment was scrubbed...
Name: not available
Type: text
Size: 1199 bytes
Desc: not available
Url :
http://www.classiccmp.org/pipermail/cctalk/attachments/19990527/724d8f9c/at…
9184
days inactive
9184
days old
22 comments
8 participants
participants (8)
-
allisonp＠world.std.com -
cmcmanis＠mcmanis.com -
donm＠cts.com -
elvey＠hal.com -
healyzh＠aracnet.com -
jfoust＠threedee.com -
jim＠calico.litterbox.com -
pechter＠pechter.dyndns.org