23 Jan
2021
23 Jan
'21
10:46 a.m.
The question was asked (I misplaced the note) if the OS images that are on the Xhomer
website work with David Gesswein's MFM emulator on real hardware.
You can turn these into emulator data files, provided you have the latest ext2emu tool
that's part of David's software (as of 1/19/2021). The format needed is
"Elektronika_85" which is the Pro format, called that because it was first added
to the emulator as the format found on the Russian bootleg copy of the Pro of that name.
Results, on my Pro 380:
1. P/OS 3.2: works.
2. Venix: panic at boot, "Bad access number". That doesn't mean anything to
me. I wonder if the issue is that Venix doesn't support the 380.
3. RT-11 5.3: works.
For grins I tried a smaller interleave factor than the default 5:1, since it seems
reasonable that this should make things go faster on the Pro 380, given that its CPU is
about 4x the speed of the one in the Pro 350. Somewhat to my surprise this isn't the
case; the performance indicates that at 4:1 or smaller interleave the software misses the
next sector and has to wait for another revolution.
paul
23 Jan
23 Jan
10:54 a.m.
On Sat, Jan 23, 2021, 11:47 AM Paul Koning via cctalk <cctalk at classiccmp.org>
wrote:
The question was asked (I misplaced the note) if the
OS images that are on
the Xhomer website work with David Gesswein's MFM emulator on real hardware.
You can turn these into emulator data files, provided you have the latest
ext2emu tool that's part of David's software (as of 1/19/2021). The format
needed is "Elektronika_85" which is the Pro format, called that because it
was first added to the emulator as the format found on the Russian bootleg
copy of the Pro of that name.
Results, on my Pro 380:
1. P/OS 3.2: works.
2. Venix: panic at boot, "Bad access number". That doesn't mean anything
to me. I wonder if the issue is that Venix doesn't support the 380.
The Venix image is special. Venix doesn't support more than 16 sectors per
track. You may need to put it on a disk whose geometry is 16spt. Maybe that
is why? Venix should support both the 350 and 380.
Warner
3. RT-11 5.3: works.
For grins I tried a smaller interleave factor than the default 5:1, since
it seems reasonable that this should make things go faster on the Pro 380,
given that its CPU is about 4x the speed of the one in the Pro 350.
Somewhat to my surprise this isn't the case; the performance indicates that
at 4:1 or smaller interleave the software misses the next sector and has to
wait for another revolution.
paul
11:05 a.m.
On Jan 23, 2021, at 1:54 PM, Warner Losh <imp at
bsdimp.com> wrote:
On Sat, Jan 23, 2021, 11:47 AM Paul Koning via cctalk <cctalk at classiccmp.org>
wrote:
The question was asked (I misplaced the note) if the OS images that are on the Xhomer
website work with David Gesswein's MFM emulator on real hardware.
You can turn these into emulator data files, provided you have the latest ext2emu tool
that's part of David's software (as of 1/19/2021). The format needed is
"Elektronika_85" which is the Pro format, called that because it was first added
to the emulator as the format found on the Russian bootleg copy of the Pro of that name.
Results, on my Pro 380:
1. P/OS 3.2: works.
2. Venix: panic at boot, "Bad access number". That doesn't mean anything
to me. I wonder if the issue is that Venix doesn't support the 380.
The Venix image is special. Venix doesn't support more than 16 sectors per track. You
may need to put it on a disk whose geometry is 16spt. Maybe that is why? Venix should
support both the 350 and 380.
That doesn't explain it, because all hard drives on the Pro by definition have 16
sectors per track.
paul
1 p.m.
2. Venix: panic at boot, "Bad access
number". That doesn't mean anything
to me. I wonder if the issue is that Venix doesn't support the 380.
The Venix image is special. Venix doesn't support more than 16 sectors
per track. You may need to put it on a disk whose geometry is 16spt. Maybe
that is why? Venix should support both the 350 and 380.
24 Jan
24 Jan
7:38 a.m.
On Jan 23, 2021, at 4:00 PM, Cameron Kaiser
<spectre at floodgap.com> wrote:
That doesn't explain it, because all hard drives on the Pro by definition
have 16 sectors per track.
Venix runs fine on my 380. In fact, it's the only OS installed on it. It has
the stock hard disk, though I'd like to convert to solid state, so I've been
reading this thread with interest.
Interesting. I wonder if it has to do with the peripherals I have installed. I should
try it again with the PC3XC and CNA pulled out.
What is your configuration? Do you have EBO? Memory size? Option cards?
paul
2. Venix: panic at boot, "Bad access
number". That doesn't mean anything
to me. I wonder if the issue is that Venix doesn't support the 380.
The Venix image is special. Venix doesn't support more than 16 sectors
per track. You may need to put it on a disk whose geometry is 16spt. Maybe
that is why? Venix should support both the 350 and 380. 
8:03 a.m.
On 1/24/21 10:38 AM, Paul Koning via cctalk wrote:
Listening to these conversations make me really wish I had not
given my 2 PRO's away.
bill
On Jan 23, 2021, at 4:00 PM, Cameron Kaiser
<spectre at floodgap.com> wrote:
Venix runs fine on my 380. In fact, it's the only OS installed on it. It has
the stock hard disk, though I'd like to convert to solid state, so I've been
reading this thread with interest.
Interesting. I wonder if it has to do with the peripherals I have installed. I should
try it again with the PC3XC and CNA pulled out.
What is your configuration? Do you have EBO? Memory size? Option cards?
> 2.
Venix: panic at boot, "Bad access number". That doesn't mean anything
> to me. I wonder if the issue is that Venix doesn't support the 380.
The Venix image is special. Venix doesn't support more than 16 sectors
per track. You may need to put it on a disk whose geometry is 16spt. Maybe
that is why? Venix should support both the 350 and 380.
That doesn't explain it, because all hard drives on the Pro by definition
have 16 sectors per track. 
8:40 a.m.
Listening to these conversations make me really wish I
had not
given my 2 PRO's away.
They are neat, but they are so god-damned slow. Compiles on the pro/380
with an RD52 are 1/4th the speed of an 11/83 with a 330mb ESDI disk.
It's to the point where I compile on the 83, then upload the results to
the Pro to run.
I wonder if compiling a true version of M+ could help: At least you
could dedicate memory as a disk cache to get around that slug of a
controller in there.
But the lack of CPU cache, 10mhz processor, extremely slow disk speed,
and the like really hurt the Pro.
CZ
8:48 a.m.
-----Original Message-----
From: cctalk <cctalk-bounces at classiccmp.org> On Behalf Of Chris Zach via
cctalk
Sent: 24 January 2021 16:40
To: cctalk at classiccmp.org
Subject: Re: DEC Pro operating systems and MFM emulator
> Listening to these conversations make me really wish I had not given
> my 2 PRO's away.
I have two but neither of them works. I have made a couple of attempts to find the fault
but failed so far. One of them is quite high on my list of the next thing to look at
(again).
Rob
They are neat, but they are so god-damned slow. Compiles on the pro/380
with an RD52 are 1/4th the speed of an 11/83 with a 330mb ESDI disk.
It's to the point where I compile on the 83, then upload the results to the Pro
to run.
I wonder if compiling a true version of M+ could help: At least you could
dedicate memory as a disk cache to get around that slug of a controller in
there.
But the lack of CPU cache, 10mhz processor, extremely slow disk speed, and
the like really hurt the Pro.
CZ
8:51 a.m.
On 1/24/21 11:40 AM, Chris Zach via cctalk wrote:
Well, my first LSI-11 was a Terak 8510. It had 28KW of memory and
4 8" RX01 compatible floppies and an LSI-11/02 CPU. And I developed
programs in Pascal, Fortran-77 and COBOL (yes, COBOL!!) on it. Now,
there is slow. I would gladly take a screamer like any Pro over doing
software development on the Terak.
bill
Listening to
these conversations make me really wish I had not
given my 2 PRO's away.
They are neat, but they are so god-damned slow. Compiles on the pro/380
with an RD52 are 1/4th the speed of an 11/83 with a 330mb ESDI disk.
It's to the point where I compile on the 83, then upload the results to
the Pro to run.
I wonder if compiling a true version of M+ could help: At least you
could dedicate memory as a disk cache to get around that slug of a
controller in there.
But the lack of CPU cache, 10mhz processor, extremely slow disk speed,
and the like really hurt the Pro.
8:53 a.m.
> > 2. Venix: panic at boot, "Bad access
number". That doesn't mean anything
> > to me. I wonder if the issue is that Venix doesn't support the 380.
[...]
Venix runs
fine on my 380. In fact, it's the only OS installed on it. It has
the stock hard disk, though I'd like to convert to solid state, so I've been
reading this thread with interest.
Interesting. I wonder if it has to do with the peripherals I have
installed. I should try it again with the PC3XC and CNA pulled out.
What is your configuration? Do you have EBO? Memory size? Option cards?
8:54 a.m.
On Sun, Jan 24, 2021 at 9:53 AM Cameron Kaiser via cctalk <
cctalk at classiccmp.org> wrote:
cards?
It's a pretty vanilla 380. I think it has just the 256K. No EBO, no option
cards. RD52-A and RX50-AA.
Note that I can't exclude that it has a 380-specific Venix installed; it
came
to me with it and I don't have floppies (part of why I'd like to convert
it).
But I don't recall Venix coming in a 380-specific version.
The Venix/Pro that's in the archives works on all Pros (1.0) but 2.0
requires the 380.
Warner
> > 2.
Venix: panic at boot, "Bad access number". That doesn't mean
anything
> > to me. I wonder if the issue is that
Venix doesn't support the 380.
[...]
> Venix runs fine on my 380. In fact, it's
the only OS installed on it.
It has
> the stock hard disk, though I'd like to
convert to solid state, so
I've been
reading
this thread with interest.
Interesting. I wonder if it has to do with the peripherals I have
installed. I should try it again with the PC3XC and CNA pulled out.
What is your configuration? Do you have EBO? Memory size? Option
8:57 a.m.
The Venix/Pro that's in the archives works on all
Pros (1.0) but 2.0
requires the 380.
I should have said what it was running. It is indeed 2.0.
--
------------------------------------ personal: http://www.cameronkaiser.com/ --
Cameron Kaiser * Floodgap Systems * www.floodgap.com * ckaiser at floodgap.com
-- The steady state of disks is full. -- Ken Thompson -------------------------
11:43 a.m.
On Jan 24, 2021, at 11:54 AM, Warner Losh via cctalk
<cctalk at classiccmp.org> wrote:
On Sun, Jan 24, 2021 at 9:53 AM Cameron Kaiser via cctalk <
cctalk at classiccmp.org> wrote:
cards?
It's a pretty vanilla 380. I think it has just the 256K. No EBO, no option
cards. RD52-A and RX50-AA.
Note that I can't exclude that it has a 380-specific Venix installed; it
came
to me with it and I don't have floppies (part of why I'd like to convert
it).
But I don't recall Venix coming in a 380-specific version.
The Venix/Pro that's in the archives works on all Pros (1.0) but 2.0
requires the 380.
That's odd. The image I tried is "venix2" from the Xhomer website. From
the name I'd guess that is V2.0, but it runs fine on Xhomer which is a 350 emulation.
On the other hand, it panics (as I reported) on my Pro 380.
paul
>>
2. Venix: panic at boot, "Bad access number". That doesn't mean
anything
>> to me. I wonder if the issue is that
Venix doesn't support the 380.
[...]
> Venix runs fine on my 380. In fact, it's
the only OS installed on it.
It has
> the stock hard disk, though I'd like to
convert to solid state, so
I've been
reading
this thread with interest.
Interesting. I wonder if it has to do with the peripherals I have
installed. I should try it again with the PC3XC and CNA pulled out.
What is your configuration? Do you have EBO? Memory size? Option
1427
days inactive
1428
days old
12 comments
6 participants
participants (6)
-
bill.gunshannon＠hotmail.com -
cz＠alembic.crystel.com -
imp＠bsdimp.com -
paulkoning＠comcast.net -
robert.jarratt＠ntlworld.com -
spectre＠floodgap.com