From: Pete Turnbull <pete(a)dunnington.u-net.com>
>Which leads me to think about a tangentially-related problem -- I recall
>that some FDCs insist upon the index mark at the start of the track, and
>some don't, but I can't remember which FDCs do/don't (I think WD 37C65
etc
>need it but NatSemi DP8473 and WD 1770/1772 don't care).
They all do save for the WD. However without index you may have
other problems.
>Also, I recall that some FDCs are more likely to work with 8" drives
than
>others, and some are more likely to work at single density than others
(at
WD had a few that were speed limited but all the 765 and later based
on that core cared less what siase save for many PC implmentations
didn't provide the adaquately fast clocks needed to run 8". that was
not a chip problem but was a board (controller) problem.
>least, some PC ISA controllers are unlikely to work at SD because the
>relevant components are not there). I want to set up a PC to use with
8"
Some yes, other would easily do 8" single but not double or the reverse
case. The older 765 PC controller designs were often PC centric,
minimalist and limited. The 37C65 if it had the 9.6 and 16mhz external
clocks would do all modes if programmed properly.
>drives, including single density, mostly with 22DISK. I have a choice
of
>386/486/Pentium motherboards, and a selection of ISA controllers, with
an
>assortment of 765, 8272, 37C65, and 8473 FDCs (I haven't checked, but
I'm
>pretty sure I have at least one of each). What is most likely to work?
I'd go with the 37C65 board IF it has the clock logic for both 9.6 mhz
(1.2mb) and 16 mhz (all other 3.5, 5.25 and 8 inch formats). FYI the
8272 is a 765 so no difference at all. The 8743 is a 765 core with
some of the external logic the older 765 didnt contain on the same
silicon.
SMC did some 765 core with clock and data sep like WD37c65. In the
end you'll find most PC controllers do not support the full 8" singnal
set,
missing are TG43, ready/fault and a few other lesser ones.
Allison
From: ip500 <ip500(a)home.com>
> Would a "modern" switching supply ie: a cheap, generic PC supply
>have enough output to run a vintage machine? I had considered doing that
At best the answer is maybe. Reason is I have no idea of the specific
vintage machines power needs.
For example the average PC power supply can run from 145-350 watts
and that alone covers a lot of ground. Ther other problem is most of the
PC power supplies outout is +5 and some +12, some of the vintage
machines may have heavier needs for +12 or maybe -12. Another
example is the PDP-8/f I have, a 250W PC supply would easily
satisfy the +5v needs but would fail badly as the PDP-8 wants
+15 and -15 something a PC supply cannot do.
>faulty memory] the HP drew about 20 amps or so on the +5 line. Shouldn't
>be too hard to pull a couple of the faulty cards to make room for the
>much smaller [physically] supply. Might be a problem with lack of decent
>air circulation for cooling though.
> Again from memory ... even the CE's really didn't want to fool with
>repairing the PS units on these. When I bought a bunch of HP gear there
>were 8 or 10 dead supplies that had been stashed away to maybe fix
>"someday'.
Many people are scared of PS units. Switch mode PSUs are harder
to fix than linear. Then again I fixed a BA123 supply by replacing a
failed internal fuse! So first testing and deciding how bad the failed
unit is likely first order of the day.
Allison
Does anyone have docs on a DRV-11? (16 bit Parallel I/O). I bought one on
Ebay (perhaps it will help in my quest to kludge a cheap SCSI controller
for Q-bus vaxen :-) and now I need some docs for it... I'll check the
handbook when I get home too.
--Chuck
From: Chuck McManis <cmcmanis(a)mcmanis.com>
>things into memory where the BAR register is pointing. Changing the BAR
>from the edge connector is going to be a challenge unless you had two
and
>the second one was set up to DMA into the BAR of the other :-)
Since the DEC chipset used made getting at the DMA address and
DMA counters impossible from the card edge (it was a bus interface)
it's awkward at best.
Allison
From: Chuck McManis <cmcmanis(a)mcmanis.com>
>The LSI-11 is "out of the loop" in this case. The DRV11-WA can take over
>the Q-bus and hold off the processor while it reads or writes memory
>locations. That gives you the ability to deposit and examine stuff in
Hard way to go. ;)
>What we're missing is a way to load the PC on the LSI 11 externally, but
we
>can do some of that using an interrupt vector and by manipulating one of
>the vector locations to have the PC where you want the system to start.
two locations 04 and 24 are candidates. If memory serves 24 is
powerfail restart.
>Yes, its a kludge on a major scale, but its blinkenlights :-)
Yes, blinkin lights are cool but even in 1977 they were considered a
PITA rather than an aid save for special cases. I busted enough
fingers on 8s, 11s and altairs/imsais to keep me unhappy.
I see putting blinkin lights on machines that never had them
to be a guilding the Lilly pointlessness.
Allison
From: joe <rigdonj(a)intellistar.net>
> I swapped my 1.2mb drive for a Teac 360k drive. No difference. I
>tried both BIOS and Direct settings with no difference. I'm running
MS_DOS
>and rebooted with NO drivers or TSRs but nothing helps. The one odd
thing
That will not work. RX50 is 96tpi and ALL 360k drivs are 48TPI, a non
starter.
The TEAC FD55GFV is the same drive DEC used (RX33) which is 1.2mb
and RX50 (400k single sided) compatable. The trick is getting the
controller
and drive to interface right and the software should follow.
>know it seems to be working for nearly everyone else but Teledisk simply
>will not work for me! My only guess at this point is that perhaps it's
>incompatible with my FD controller IC. But BIOS should isolate it from
>that, so who knows.
Some FDCs cannot do the 10 sector format, end game. Actually most
765 based designs are not best suited as the GAPs and IRGs are very
small (below reasonable). The DEC designs Pro, Rainbow, Decmate,
VT180 used WD1793 (not used on PCs) and only the PDP-11/VAX
designs used a 765 biased (SMC9224) design for the MSCP controller
(RQDX) which could format and read/write the RX50.
Hoep that helps some. Me I'd try a different (meaning secondary)
disk controller in the same 486/100 box.
From: joe <rigdonj(a)intellistar.net>
>>What your havent tried is differnt controllers and different drives.
>
>
> I have tried it with two 1.2Mb drives and one 360k drive. I
haven't
>changed the controller since my video card, HD controller and ports are
all
>on the same card. I would have to rebuild the system to replace that
card.
Which 1.2mb drive? First of RX50 is not 1.2 compatable, it's 96tpi (as
most
1.2mb drives are) and would corrspond to a 360k formatted 5.25 (single
sided).
A 360k drive is NOT useable for RX50 format. The usual 360k media is
fine however.
> Absolutely! I run everything that I can in DOS mode. Also I'm
using
>a 486-100. FWIW I've run lots of other weird file conversion SW
should
>as the HP LIF utilities with no problems. I also used a lot of odd
>interfaces (on occasion) in it and I've had no problem with them.
Doesnt rule out bios/hardware though.
> I have a couple more old PCs laying around for parts. I need to
try
>Teledisk on them but I'm so discusted with it that I haven't bothered.
maybe. I found that the standard teac FD55GFR in my system had
to be jumpered correctly to work right. The likely reason was the TEAC
I used came from a DEC system (also called RX33) so I know TEAC
are compatable at the drive level for the media (rx50).
Allison
I'm trying to write some RX50 disk images
using Teledisk. I can use 22DISK and PUTR
to format the floppies, but Teledisk will
not write to them. I get an error that
the drive or controller is returning an
unsupported status.
Since Teledisk has been acquired by
the Ministry of Disinformation I
can't buy a commercial copy (I've
tried 2.12, which locked up my
system, and 2.15, which just give
the errors an quits). Is there
any other program that will write
Teledisk image files?
--
Eric Dittman
dittman(a)dittman.net
From: joe <rigdonj(a)intellistar.net>
>the same kind of errors that you're getting. I've tried Teledisk with
at
>least 50 different disks (most KNOWN good) and at least 6 different data
>formats but the ONLY thing it will copy is MS_DOS disks. I've given up
on
>it. I've already tried everything that everyone has suggested and I'm
fed
What your havent tried is differnt controllers and different drives.
I found an old 486 or 386 machine to be less troublesome and also
has fewer preconceived notions of the world. Also I ran dos not
windows(any version)!
I was able to make Venix disks for my Pro many years ago and also
OS278 (also rx50). What I had to do however was to play with the
jumpers on a TEAC FD55GFV (the verseion with many jumpers)
to lock out the 1.2mb mode.
I may add My experience is with the older (must be at least 6 years old)
copy of the freeware version of teledisk not the full bore commercial.
Allison
On May 11, Chuck McManis wrote:
> However, one of the other things was an Emulex board that I'm trying to
> identify. It has these numbers:
> K12S12/500W
> SN: AAG6933 CM
> Assy #: QS0910201-02
>
> Then four high density 36 pin honda connectors on an Sbox handle. There is
> also a small three position switch labeled 'norm', 'reset', and 'setup.'
> I'm guessing it is some sort of async port. Anyone know for sure? Does
> anyone know the actual Emulex model number and if cables are available? If
> its a 32 port serial card it would be pretty cool.
Emulex QS09: 16 port async mux.
-Dave McGuire
