At 05:48 PM 1/10/2012, Terry Stewart wrote:
>I've recently had some discussions with someone who collects a particular
>brand of classic computer from the early 1980s. He has a few of these but
>they are never, ever switched on or tested. He doesn't want to test them
>as he feels doing so may damage the old electronics.
Aldo Leopold said "The first prerequisite of intelligent tinkering is
to save all the pieces."
- John
A few days ago I announced that this year's VCF East 8.0 Day 1 keynote is Tom Kurtz, the co-inventor of BASIC. Now I'm happy to announce that the Day 2 keynote is Dan Kottke. Dan was a college buddy of Steve Jobs. They went to India together and sought enlightenment. Dan assembled Apple 1 boards in Jobs' parent's garage, and was officially employee #12, although in reality he was in the first five or so. He speaks on Sunday, May 6, at 12:30. Details to be posted at http://www.vintage.org/2012/east and http://www.facebook.com/vcfeast8.
Pontus writes:
>On Tue, Jan 10, 2012 at 04:14:25PM +0000, Mark Benson wrote:
>> Begin forwarded message:
>>
>> > Hi folks - I thought this was interesting and at least *somewhat* OT...
>> >
>> > http://hackaday.com/2012/01/10/help-chris-boot-his-cray-1-supercomputer/
>>
>> Dunno if anyone saw this - looks VERY interesting - can anyone help this guy??
> Doesn't the Cray-1 run of an Eclipse front end. So he is really asking
> how to boot an Eclipse?
Other brand front ends were used too... Cray themselves used an Eclipse, but some
customers used DEC or (ack!) IBM front ends.
If someone wants to understand booting RDOS on a Nova or Eclipse, this is the best start:
http://bitsavers.org/pdf/dg/software/rdos/093-000188-02_Load_and_Generate_R…
----- Original Message:
Date: Tue, 10 Jan 2012 14:50:29 -0000
From: "James Attfield" <james at attfield.co.uk>
To: <cctech at classiccmp.org>
Subject: RE: Cromemco FDC
> Date: Mon, 09 Jan 2012 14:53:19 -0500
> From: "Bill Sudbrink" <wh.sudbrink at verizon.net>
> Subject: RE: Cromemco FDC
>
> I have always found it interesting how quickly people will go chopping
> up circuit boards. Before you do this, consider chopping up a ribbon
> cable instead. Ribbon cables are cheap and easy to replace if you
> later change to a different make of drive or if you simply screw up
> making the modification. I run SA-851's on my 16FDC with a simply
> modified ribbon cable.
Bill, the Tandon 848's look like they have a pretty similar interface to the
SA-851. I have two TM-848-2's I would like to hook up - what chance do you
think that your modified cable might work (I'm with you on this one rather
than board cut'n'shut) and if so could you kindly email me the layout
off-list?
Jim
------------------------------- Reply:----------------------------
I'd be interested as well!
Quinnteam used to make and sell something called a Top Hat which consisted
of a pair of TM-848's mounted side-by-side horizontally across the top of a
Z-2; this was before the 64FDC came out so they made a number of mods to the
*drives * in order to work correctly with the 16FDC.
I've still got TM-848s out of scrapped Top Hats and they work well with the
16FDC, as well as unmodified TM-848s out of scrapped System 3's which work
with the 64FDC, but I've never taken the time to investigate just what's
involved in getting either drive type to work with the other FDC; this
thread has kind of reawakened my curiosity.
Maybe some folks on the Cromemco mail list have more information...
mike
At 3:54 am +0000 2012/01/09, Evan Koblentz wrote:
>Perhaps quite a lot, but not $800,000. :)
>
>Ebay #190404561375
I still think Intel should issue a commemorative 4004 in 22nm. Or failing
that, easter-egg one (or a few thousand) into Haswell somewhere.
--
Kevin Schoedel <schoedel at kw.igs.net> VA3TCS
----- Original Message:
Date: Mon, 09 Jan 2012 22:29:10 -0800
From: Josh Dersch <derschjo at mail.msu.edu>
To: General Discussion: On-Topic and Off-Topic Posts
...
> I've created a few floppies from my PC on a 5.25" drive using ImageDisk, I
> have been unable to get the FDC to read them (or boot, obviously). I
> wasn't 100% sure it wasn't just the setup I have on my PC getting in the
> way, hence my latest attempt to use the FDC itself to write the disks.
You don't say which images and whether you're talking about a DD drive or an
HD drive emulating an 8" disk.
Let me say it again: very few PCs will reliably write the Cromemco 360K DD
format, but 8" images written to a 1.2M HD drive (on 1.2MB disks!) are
usually quite reliable. See below.
>> You are aware that using a 5.25HD drive as an 8" requires a jumper to
>> connect /READY to the FDC's 5.25" interface?
> I was not aware, but the 16/64FDC has connectors for both 5.25" and 8"
> drives, I've been using the 5.25" connector and the controller seems to be
> able to communicate with it OK -- it has no trouble seeking and formatting
> tracks... it just can't read or write them at the moment...
Umm... if you can't read it how do you know it was formatted correctly?
The 5.25" connector normally only supports DD drives/formats (which, as I
said, are notoriously unreliable reading disks made on a PC) because the
/READY signal needed for 8" drives is not available on that connector. To
use the more reliable 1.2M HD drives (360RPM with 1.2M HD disks) (or
3.5" HD drives for that matter) connected to the 5.25" port as 8" you need
to connect pin 34 to /READY (and jumper the drive to supply /RY instead of
/DC). Incidentally, that also avoids the 8" drive issues that Bill and
Amardeep are arguing about.
So, frankly, I'm very skeptical that your 1.2M HD drive is working at all
*as an 8" drive*. BTW, your 300RPM HD drive is also going to be troublesome
if not completely useless.
And of course you have to select the correct drive type in RDOS with one,
two or three semicolons.
I'd be surprised if the 1793 were the problem, especially both of them. The
4FDC had a bad reputation for that, but out of the 20+ Cromemco systems that
I supported (and still have most of) only ONE ever developed problems
reading disks and I never did investigate what the cause actually was.
No offence, but most likely you're doing something wrong; image/disk
incompatibility, wrong type selected, etc. etc.
Good luck!
> Date: Mon, 09 Jan 2012 14:53:19 -0500
> From: "Bill Sudbrink" <wh.sudbrink at verizon.net>
> Subject: RE: Cromemco FDC
>
> I have always found it interesting how quickly people will go chopping
> up circuit boards. Before you do this, consider chopping up a ribbon
> cable instead. Ribbon cables are cheap and easy to replace if you
> later change to a different make of drive or if you simply screw up
> making the modification. I run SA-851's on my 16FDC with a simply
> modified ribbon cable.
Bill, the Tandon 848's look like they have a pretty similar interface to the
SA-851. I have two TM-848-2's I would like to hook up - what chance do you
think that your modified cable might work (I'm with you on this one rather
than board cut'n'shut) and if so could you kindly email me the layout
off-list?
Jim
On Mon, 9 Jan 2012, Bill Sudbrink wrote:
> Bill Sudbrink wrote:
>>
>> I have always found it interesting how quickly people will go chopping
>> up circuit boards. Before you do this, consider chopping up a ribbon
>> cable instead. Ribbon cables are cheap and easy to replace if you
>> later change to a different make of drive or if you simply screw up
>> making the modification. I run SA-851's on my 16FDC with a simply
>> modified ribbon cable.
>>
> By the way, based on your description of the symptoms, I still suspect
> the 1793 is going bad.
Unicorn Electronics may have stock on the 1793 -
http://198.170.117.30/IC/CRT.html it's listed for $4.99.
g.
--
Proud owner of F-15C 80-0007
http://www.f15sim.com - The only one of its kind.
http://www.diy-cockpits.org/coll - Go Collimated or Go Home.
Some people collect things for a hobby. Geeks collect hobbies.
ScarletDME - The red hot Data Management Environment
A Multi-Value database for the masses, not the classes.
http://www.scarletdme.org - Get it _today_!
Political correctness is a doctrine, fostered by a delusional, illogical
minority, and rabidly promoted by an unscrupulous mainstream media, which
holds forth the proposition that it is entirely possible to pick up a turd
by the clean end.
----- Original Message:
Date: Mon, 09 Jan 2012 08:13:10 -0800
From: Josh Dersch <derschjo at mail.msu.edu>
...
Right now I have both a 16FDC and a 64FDC to play with (one is on
loan). I'm using Dave Dunfield's RDOS transfer utilities to upload the
"INIT" software into RAM from a PC.
...
I'm wondering if there's some fundamental problem with my current set-up
that's causing issues here...
- Josh
----- Reply:
Have you tried creating a bootable Cromemco format disk (CDOS or CP/M) from
an image? Even if it's a configuration that doesn't match yours, you could
at least run some diagnostics on the diskette using RDOS.
The 5.25DD format can be troublesome to create, but generally an 8" image on
a 5.25" HD disk & drive is no problem, especially if you can use the same
physical drive in the PC and the Cromemco and avoid any alignment issues.
You are aware that using a 5.25HD drive as an 8" requires a jumper to
connect /READY to the FDC's 5.25" interface?
mike
>
> > (Now I just have to figure out why I can't get the FDC to write
floppies correctly...)
Can you describe the problem in detail? I recall that you're using a 16FDC
with 8" drives that are not Persci brand. Are you aware that the 16FDC 8"
bus is wired differently than what is required for Shugart compatible
drives? The Persci drives did not have a TG43 signal so the write current
may be too high on the inner tracks of a non-Persci drive mated to a 16FDC
-- causing errors.
Fortunately, it is possible to modify a 16FDC to be compatible with a
couple of cuts and jumpers. I have some photos and text that describe the
mod.
Should this happen to be the problem you're experiencing, I'll upload the
mod info to a site and send the link. If not, please describe what problem
you are having in more detail.
Amardeep
"Bill Sudbrink" <wh.sudbrink at verizon.net> wrote:
> I have always found it interesting how quickly people will go chopping
> up circuit boards. Before you do this, consider chopping up a ribbon
> cable instead. Ribbon cables are cheap and easy to replace if you
> later change to a different make of drive or if you simply screw up
> making the modification. I run SA-851's on my 16FDC with a simply
> modified ribbon cable.
I have always found it interesting how quickly people comment on issues
without first studying them in depth. :-)
The TG43 signal is entirely absent from the 16FDC drive connector and no
cable mod alone can correct that. The board modification brings the signal
>from the FDC1793 through a driver to the connector so that a cable can
carry it to the drive. If you choose to operate the SA-851 out of spec, it
is certainly your right. I chose to make my 16FDC compatible with the
drive's operating requirements.
Incidentally, the Shugart SA-800/801 doesn't require TG43 and will work
with just a cable mod (or maybe no mod at all since that drive is single
sided). Almost every newer drive, including SA-851, however, will have
higher error rates on the inner cylinders without the reduced write current
controlled by TG43.
I checked out a number of vintage items that were recently found at Weirdstuff. If you're interested in any of these items, contact "Jim" at Weirdstuff (408-743-5650). If you have any technical questions regarding these items, you can contact me as I've looked them over (but didn't test)...
1. (3) New in box Hayes JT Dual Port FAX modems 14400B w/manuals and software
2. (1) Toshiba P321SL "3 in 1" Printer (missing plastic cover)
3. (1) ThinkJet printer w/GPIB interface
4. (4) Sun 68 pin SCSI Disk UniPack enclosures with auto-termination
5. (1) Macintosh Portable, Model M5120, complete with case, power supply and unused ID cards.
6. (1) New in box DEC VMS 5.4 manual set dated 26 Jul 90
7. (1) Kaypro II, S/N 61085, clean keyboard w/cable, (2) 5.25" FDD
8. (1) Manual Set, HP83480A (Digital Communications Analyzer) & HP54750A (Digitizing Oscilloscope)
9. (1) HP 85025 A/B/E Detector
10. (1) Tektronix 577 D1 or D2 Curve Tracer Service/Instructional Manual
11. (1) Intermac Trakker Antares 241X Hand Held Terminal Manual
12. (1) Heath Appliance Control, Model SL-6166-RX
I am not affiliated with Weirdstuff and receive no financial renumeration to list these items - I'd just like to see vintage stuff get a good home...
Cheers,
Lyle
--
Lyle Bickley, AF6WS
Bickley Consulting West Inc.
http://bickleywest.com
"Black holes are where God is dividing by zero"
Josh Dersch <derschjo at mail.msu.edu> wrote:
> I'm wondering if there's some fundamental problem with my current set-up
> that's causing issues here...
Are you terminating the drive bus correctly?
Hi,
Any TRS 80 gurus that can point me in the right direction with the
disk/video interface box? I recently picked one up and have it connected
to my model 100. Everything appears to be working, but when I start the
interface box, it requests a system disk. Is this required to use the
interface box, or is there a way to go right to the mode to display 80
columns and skip the boot disk in the interface box? I have the owner's
manual, unfortunately the disk is missing and it appears the disk is
required to get the interface box to do anything? Any pointers where I
might find the software?
Thanks, Win
------ Original Message:
Date: Mon, 9 Jan 2012 12:04:41 -0500
From: Win Heagy <wheagy at gmail.com>
To: cctalk at classiccmp.org
Subject: TRS-80 Model 100 disk/video interface box
Hi,
Any TRS 80 gurus that can point me in the right direction with the
disk/video interface box? I recently picked one up and have it connected
to my model 100. Everything appears to be working, but when I start the
interface box, it requests a system disk. Is this required to use the
interface box, or is there a way to go right to the mode to display 80
columns and skip the boot disk in the interface box? I have the owner's
manual, unfortunately the disk is missing and it appears the disk is
required to get the interface box to do anything? Any pointers where I
might find the software?
Thanks, Win
-------Reply:
No, unfortunately you need the disk to load the required firmware into both
the DVI and the M100 and I don't believe anyone has taken the time yet to
figure out how to image/restore that format.
Send me your address off-list and I'll send you a copy.
You might also want to join the 'Model T' mailing list:
http://www.bitchin100.com/
And of course there's *THE* place for everything M100/102/200-related:
http://www.club100.org/
mike
Time to thin the herd - I have a number of DEC machines that are free to
a good home, though beer fund donations will be accepted. Available for
pickup in central Berkeley by appointment, or I may be able to deliver
to some parts of the East or South Bay at my convenience. I'll also take
them to a Mailboxes Etc equivalent to be packed and shipped if you
really want to pay for that...
Here's what's on offer:
DEC 3000/300LX - Alpha 21064, can get to PROM monitor on all of these
#1 - 32MB RAM, 1 x RZ25L HDD (~500MB)
#2 - 256MB RAM, 1 x RZ25L HDD (~500MB)
#3 - 160MB RAM, no HDD but bracket is included
DECstation 3100 - MIPS R2000, don't have MMJ or video cables, can't verify
#1 - Can see VSIMM and 4 SIMMs, no HDD
DECstation 5000/133 - MIPS R3000?, no graphics, no response on the DB25
serial ports
#1 - 32MB RAM, 1 x RZ25L HDD
VAXstation 3100 m38 - don't have MMJ or video cables, can't verify
#1 - Can see 2 RAM daughterboards, no HDD, 8 bit graphics board?
#2 - At least 1 RAM board, 1 x RZ23E HDD (100MB?), 1 x RX23 floppy,
graphics board
First come, first served based on email.
--/Steve/.
<smj (a) crash (d) com>
On 1/9/2012 6:31 AM, Bill Sudbrink wrote:
> Josh Dersch wrote:
>> (Now I just have to figure out why I can't get the FDC to write
>> floppies correctly...)
> Which FDC? I've had trouble with the original 1793 controller chips
> (the early programmed ones with the painted over window on top) going
> bad on a couple of 16FDCs I have. Replacing with a newer hard programmed
> plastic package chip and then recalibrating the data sep circuit solved
> a bunch of problems. The 4FDCs tended to just be "twitchy". I think
> there was a known problem with the controller chip used on them.
Right now I have both a 16FDC and a 64FDC to play with (one is on
loan). I'm using Dave Dunfield's RDOS transfer utilities to upload the
"INIT" software into RAM from a PC. From there I can initialize
floppies (single-sided, single-density), and this appears to work -- it
goes through all 40 sectors (5.25" drive) and the drive steps and no
errors are reported. Then INIT attempts to write the filesystem to the
drive and it fails (it's a "Home Error" on Track 0, Surface 0, Sector 8,
if I recall).
Ignoring that and using the RDOS utilities to attempt to write a disk
image fail similarly ("Err-H 34" IIRC).
I've tried a pile of different disks and three different floppy drives
-- a Tandon TM-100 (300 RPM), a modified 1.2M 5.25" drive that runs at
300RPM, and a standard 1.2M drive (360RPM) and all three fail in exactly
the same way. (I don't currently have an 8" drive I can wire up, I need
to build a cable...) I've also tried swapping the cable out, and both
the 16FDC and the 64FDC fail identically.
I'm wondering if there's some fundamental problem with my current set-up
that's causing issues here...
- Josh
On 1/4/2012 8:20 PM, Bill Sudbrink wrote:
> Josh Dersch wrote:
>> How does the "phantom" line work? I see references to this in some of
>> my S-100 manuals but I'm not clear how it works (who raises/lowers it,
>> when, etc...)
> It can be a little bit "implementation dependent" but usually there are
> cards that assert phantom (it is active low, there has to be a pull up
> on it somewhere (the CPU card I think)) and cards that "honor" phantom.
> When an address is presented to the bus, any phantom asserting card that
> whishes to override that address pulls the phantom line low. A phantom
> honoring card that would normally respond to the address will instead
> suppress any activity. Note that this system does not prevent multiple
> phantom asserting or phantom honoring cards from stepping on each other.
> Careful system configuration is still the responsibility of the builder.
> In the case of the Cromemco FDC cards, the FDC becomes a phantom asserter.
> Any RAM card that the system has that resides in the same address space
> ($C000-$CFFF in the case of a 4FDC) must honor phantom. The CPU presents
> an address to the bus, the FDC card does an address decode, determines that
> it is an address it wants to respond to and activates the ROM (the CE (chip
> enable) pin on the ROM is also active low) asserting PHANTOM at the same
> time, disabling the RAM card. When you deactivate the RDOS ROM with the
> poke to the I/O address, then the FDC card will no longer activate the ROM,
> PHANTOM will no longer be asserted and the RAM card will respond to the
> address previously occupied by RDOS. I run CDOS in a full 64K
> configuration this way.
>
> Bill S.
>
Thanks for the clarification! I do have a 32K Static RAM card that
seems to support the Phantom line, I'll have to play around with it once
I find something to fill the last 16K.
(Now I just have to figure out why I can't get the FDC to write floppies
correctly...)
- Josh
On Jan 4, 2012, at 11:20 PM, Bill Sudbrink wrote:
> Josh Dersch wrote:
>> How does the "phantom" line work? I see references to this in some of
>> my S-100 manuals but I'm not clear how it works (who raises/lowers it,
>> when, etc...)
>
> It can be a little bit "implementation dependent" but usually there are
> cards that assert phantom (it is active low, there has to be a pull up
> on it somewhere (the CPU card I think)) and cards that "honor" phantom.
> When an address is presented to the bus, any phantom asserting card that
> whishes to override that address pulls the phantom line low. A phantom
> honoring card that would normally respond to the address will instead
> suppress any activity. Note that this system does not prevent multiple
> phantom asserting or phantom honoring cards from stepping on each other.
This sounds a lot like the INHIBIT line on the Apple II bus. Is that roughly analogous?
- Dave
I have a "Thinker Toys Universal Interface Board" here, and I've been trying to figure it out. This is an early version of the Morrow Switchboard. It has a copyright date of 1978. The manuals available online are for the more common Morrow Switchboard, Rev 2. Comparing this board to a Switchboard reveals some subtle differences, but by and large, the boards are the same. This older version I have has no solder mask, while the Rev 2 has a blue solder mask. There are minor component placement differences.
The most drastic difference seems to be the comparators that read the switches at the top of the board. As found, the board had a LOT of trace cuts and jumps, and some traces just plain peeled off - all around those four LS266's at the upper right, directly under the SW5 and SW6. Attempting to figure out what was done is nearly impossible. The switches themselves are still connected to the 266's with factory traces, at least, most of them - but the rest of the inputs were a mess of wires with bad solder joints. Of course... it doesn't work, and the wiring of the switches doesn't even come close to matching the schematics. Not only that, the factory traces that remain don't match the schematics. I mean, the switches wouldn't even be in the same order. It's a mess.
So... does anyone have the schematics for this thing?
-Ian