12 Jan
2006
12 Jan
'06
1:29 p.m.
OK, so I found the Pertec stuff on bitsavers, but I haven't read
through their docs yet.
I purchased a Fujitsu 2444 tape drive that has a Pertec interface.
($0.99 on ebay with local pickup!) Apparently all the newer tape
drives have Pertec + SCSI or just SCSI.
Is Pertec similar enough to SCSI that I could rig up a dongle?
If not, how hard is it to find a Pertec interface for:
- VAXserver 4000
- PDP-11 Q-bus
- Sun 3 w/expansion chassis
- PCI/ISA
So that I could get the drive into a usable state?
I found a place that had Pertec interfaces for ISA based systems, but
it would not work with Windows 2000 or later OS, the cards were out of
stock and were pretty pricey (like $500+).
So what would y'all recommend as the easiest/cheapest way to get this
baby up and running?
Also -- the power cord has a non-standard 3-prong plug. I'll try to
get photos of this up somewhere for reference, but perhaps someone
knows the kind of cord this thing needs so I can know what to look
for?
Fujitsu M244X Streaming Tape Drive Customer Engineering Manual
<http://www.bitsavers.org/pdf/fujitsu/B03P-5325-0100A_244X_Jun87.pdf>
Pertec
<http://www.bitsavers.org/pdf/pertec/>
--
"The Direct3D Graphics Pipeline"-- code samples, sample chapter, FAQ:
<http://www.xmission.com/~legalize/book/>
Pilgrimage: Utah's annual demoparty
<http://pilgrimage.scene.org>
12 Jan
12 Jan
1:47 p.m.
Richard wrote:
OK, so I found the Pertec stuff on bitsavers, but I
haven't read
through their docs yet.
I purchased a Fujitsu 2444 tape drive that has a Pertec interface.
($0.99 on ebay with local pickup!) Apparently all the newer tape
drives have Pertec + SCSI or just SCSI.
Is Pertec similar enough to SCSI that I could rig up a dongle?
If not, how hard is it to find a Pertec interface for:
- VAXserver 4000
- PDP-11 Q-bus
- Sun 3 w/expansion chassis
- PCI/ISA
So that I could get the drive into a usable state?
The Pertec interface is an interface which exports control of the tape
drive to a
controller, and leaves much of the formatting up to the controller which
is attached.
There is no intelligence assumed on the part of the tape drive, and it
performs
only motion control, and read and write functions. Detection of file
marks, and
creation of file marks is assisted in the hardware design but little else.
SCSI, on the other hand is a general interface which assumes minimal
intelligence
and capability, and therefore cost on the computer (initiator end
usually) and puts
the load on the peripheral to do most of the work. It is an interface
which is
generalized to allow the same interface bus to handle tape, disk and all
manner
of devices if need be.
There are scsi to pertec interfaces, but were rare, and therefore
probably if you
found one, would also come with someone wanting to sell the tape drive it is
attached to.
There are PDP-11 pertec controllers. Same with Sun3 multibus - VME. I
had a fujitsu 2444 which came from a sun 3/280, and it had a multibus in the
sun VME adapter frame.
PCI/ISA includes Overland Data if you can find the controller. I had
trouble
running the 2444 on the Overland TXI, so be warned on that.
Jim
2:15 p.m.
In article <43C6CE61.1090205 at msm.umr.edu>,
jim stephens <jwstephens at msm.umr.edu> writes:
There are scsi to pertec interfaces, but were rare,
and therefore
probably if you
found one, would also come with someone wanting to sell the tape drive it is
attached to.
There are PDP-11 pertec controllers. Same with Sun3 multibus - VME. I
had a fujitsu 2444 which came from a sun 3/280, and it had a multibus in the
sun VME adapter frame.
PCI/ISA includes Overland Data if you can find the controller. I had
trouble
running the 2444 on the Overland TXI, so be warned on that.
So it sounds like the best bet is to try and find a controller for the
Sun 3/110 that I have. (Particularly since this is a drive that was
sold for Sun systems.) Does anyone have a part/model number for the
Sun 3 VME controller?
--
"The Direct3D Graphics Pipeline"-- code samples, sample chapter, FAQ:
<http://www.xmission.com/~legalize/book/>
Pilgrimage: Utah's annual demoparty
<http://pilgrimage.scene.org>
2:24 p.m.
On Thursday 12 January 2006 17:15, Richard wrote:
In article <43C6CE61.1090205 at msm.umr.edu>,
jim stephens <jwstephens at msm.umr.edu> writes:
I believe it was the Xylogics 472 (I've got one at home in my 3/180):
501-1155 Xylogics 472 1/2" Tape Controller Assembly, 9U VMEbus
501-1187 - 9U VME Adapter Assembly for Xylogics 472
370-1067 - Xylogics 472 Multibus 1/2" Tape controller
The bottom two part numbers are the parts that make up the whole kit
(the top line), if I remember correctly.
Pat
--
Purdue University Research Computing --- http://www.rcac.purdue.edu/
The Computer Refuge --- http://computer-refuge.org
There are scsi to pertec interfaces, but were
rare, and therefore
probably if you
found one, would also come with someone wanting to sell the tape
drive it is attached to.
There are PDP-11 pertec controllers. Same with Sun3 multibus -
VME. I had a fujitsu 2444 which came from a sun 3/280, and it had
a multibus in the sun VME adapter frame.
PCI/ISA includes Overland Data if you can find the controller. I
had trouble
running the 2444 on the Overland TXI, so be warned on that.
So it sounds like the best bet is to try and find a controller for
the Sun 3/110 that I have. (Particularly since this is a drive that
was sold for Sun systems.) Does anyone have a part/model number for
the Sun 3 VME controller? 
2:22 p.m.
On 1/12/2006 at 1:47 PM jim stephens wrote:
PCI/ISA includes Overland Data if you can find the
controller. I had
trouble running the 2444 on the Overland TXI, so be warned on that.
The M2444AC has a bit more intelligence that you'd otherwise expect. For
one, it does a lot of buffering, so the data transfer rate (from first byte
to the last) is exactly the same for 1600 as it is for 6250. And that's
the maximum 1.1MHz that the Pertec interface is rated for. Many simple ISA
interface cards have trouble doing a sustained transfer at this rate. The
high-end ones with local buffering should handle things a bit better.
When a Pertec-interface drive is reading, there's no handshake for data
transfer to the host--the drive simply puts the data on the read lines and
pulses a strobe line--you're expected to be there to catch it. Other than
the data transfer issues, a Pertec interface is a very simple thing and you
could probably work up your own interface without too much difficulty.
Cheers,
Chuck
2:46 p.m.
Chuck Guzis wrote:
On 1/12/2006 at 1:47 PM jim stephens wrote:
All of the Overland controllers had a good news / bad news way of
handling the transfer rate. They
had memory buffers on the board. The only one that did not have direct
access was the PC-50, which
would not work that well for anything but older 800 bpi drives.
The problem that I saw was on a PCI board, which had on board buffers,
and a microprocessor
that Overland included to assist in processing. It was not used for the
software i had, nor do I know
what software used it (TXI). I believe that the data was buffered into
a dual port memory before
reading it out.
I'll have to check the specs to see what it was rated for.
The problem I had was that most times there was no reaction from the
drive after the first record
was transfered. There seemed to be motion from the BOT over the first
record, then nothing.
I don't recall the signals to name what was missing, but there appeared
to be a lack of reaction
by the drive after the first record completed. I did not have any luck
figuring it out as I didn't have
that detailed a manual for the drive at the time to use to troubleshoot.
Jim
PCI/ISA includes Overland Data if you can find the
controller. I had
trouble running the 2444 on the Overland TXI, so be warned on that.
The M2444AC has a bit more intelligence that you'd otherwise expect. For
one, it does a lot of buffering, so the data transfer rate (from first byte
to the last) is exactly the same for 1600 as it is for 6250. And that's
the maximum 1.1MHz that the Pertec interface is rated for. Many simple ISA
interface cards have trouble doing a sustained transfer at this rate. The
high-end ones with local buffering should handle things a bit better.
When a Pertec-interface drive is reading, there's no handshake for data
transfer to the host--the drive simply puts the data on the read lines and
pulses a strobe line--you're expected to be there to catch it. Other than
the data transfer issues, a Pertec interface is a very simple thing and you
could probably work up your own interface without too much difficulty.
Cheers,
Chuck
3:50 p.m.
On 1/12/2006 at 2:46 PM jim stephens wrote:
I don't recall the signals to name what was
missing, but there appeared
to be a lack of reaction by the drive after the first record completed.
I did not have any luck figuring it out as I didn't have
that detailed a manual for the drive at the time to use to troubleshoot.
Well, if the drive is behaving correctly, a 1 usec pulse on IGO (J1, pin
8) when IRDY (J2, pin 28) is asserted and IDBY (J2, pin 38) is un-asserted
should begin a read operation. A logic analyzer would tell the whole
story.
Cheers,
Chuck
2:54 p.m.
New subject: Fujitsu 2444 data rate setting (was Re: Pertec interface)
I remember reading somewhere that this data rate was
configurable by poking the right combination of
buttons on the front panel of the 2444. I've been have
problems with VMS backups hanging part way through,
and the drive just stopping in its tracks (no pun
intended) It occurred to me it might be related to
this data rate setting but I can't find the writeup
any more. Does Chuck or anyone happen to know how this
is done ?
Dave
--- Chuck Guzis <cclist at sydex.com> wrote:
one, it does a lot of buffering, so the data
transfer rate (from first byte
to the last) is exactly the same for 1600 as it is
for 6250.
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
4:37 p.m.
New subject: Fujitsu 2444 data rate setting (was Re: Pertec interface)
David Comley wrote:
I remember reading somewhere that this data rate was
configurable by poking the right combination of
buttons on the front panel of the 2444. I've been have
problems with VMS backups hanging part way through,
and the drive just stopping in its tracks (no pun
intended) It occurred to me it might be related to
this data rate setting but I can't find the writeup
any more. Does Chuck or anyone happen to know how this
is done ?
The service manual in Richard's link from Bitsavers has a section which
describes a buffer adapter board that is placed between the deck formatter
and the host adapter, which contains buffering and rate control It
claims to
be able to do from 60k / sec up to the 1.2mb / sec.
I believe that page 4-52 of the manual, physical page 77 of the document
describes the paramter P0 which is the buffer rate select. It refers to
"Section 10.7" to see how to set the paramters. Unfortunately, there is
only section 1 thru 6, plus appendicies A and B in the manual. I would
imagine that to be refering to a user manual. I did not see that yet.
However it appears that this may be the paramter you are refering to.
I don't recall my tape drive having such a hookup, so that may be why
the problems occurred for me. Chuck is probably right about the transfer
rate from what he is saying, and that would have been a problem. If the
transfer clock was too fast, my host adapter would have not transfered
the data, and probably didn't get the transfer performed correctly, so it
just hung there.
I had suspected the PE burst was the culprit, as it is at the front of any
PE or GCR tape (ID burst is what it is also called IIRC). so I thought
it may have messed things up. We had had a problem on another deck,
a cipher 100x which did strange things between 800bpi (NRZI) and
1600bpi (PE) and the problem turned out to be the burst causing a
major malfunction in our somwhat dumb tape controller.
Jim
manual
http://www.bitsavers.org/pdf/fujitsu/B03P-5325-0100A_244X_Jun87.pdf
11:31 p.m.
New subject: Fujitsu 2444 data rate setting (was Re: Pertec interface)
On 1/12/2006 at 4:37 PM jim stephens wrote:
I believe that page 4-52 of the manual, physical page
77 of the document
describes the paramter P0 which is the buffer rate select. It refers to
"Section 10.7" to see how to set the paramters. Unfortunately, there is
only section 1 thru 6, plus appendicies A and B in the manual. I would
imagine that to be refering to a user manual. I did not see that yet.
Hmmm, take a look at:
http://sunsolve.sun.com/handbook_pub/Systems/Sun3/TAPE_10_Fujitsu-M2444AC.ht
ml
And scroll down to the section titled " IFC NVRAM Configuration
Procedures". Might the right-hand column of settings from P0-P7 have
something to do with this?
Cheers,
Chuck
1:53 p.m.
I've got one of these drives--good heavy iron. Mine was free and NOS (from
Intel) in original factory packaging. It was labeled "bad" because the
factory had shipped it jumpered for 240v with a 120vac line cord. A jumper
change later, I was up and running.
Pertec isn't even remotely close to SCSI; your best bet is to find an
interface card.
Personally, I'd love to find a reasonably-priced protocol converter (Pertec
to SCSI).
Can't help on the power cord--mine had one already.
Cheers,
Chuck
2:13 p.m.
In article <200601121353210979.05D782E6 at 10.0.0.252>,
"Chuck Guzis" <cclist at sydex.com> writes:
Pertec isn't even remotely close to SCSI; your
best bet is to find an
interface card.
Personally, I'd love to find a reasonably-priced protocol converter (Pertec
to SCSI).
I wonder how hard it would be to kludge this up in an FPGA? I saw
such Pertec->SCSI dongles for sale, but again they were multiple
hundreds of dollars.
Can't help on the power cord--mine had one
already.
OK, I found a diagram and parts number in the M244X PDF:
<http://www.xmission.com/~legalize/vintage/fujitsu/M2444-power-cord.png>
Most power cords are a rectangle with a triangular hat and this one was
a round plug, with what looks like a screw-on ring for strain relief.
Is this something fairly standard that I could find?
I mean, its just a power cord. I can always strip the standard
connector of a standard cord and solder it into the inside of the
unit, but it would be nice to have a cord that just plugged into it.
--
"The Direct3D Graphics Pipeline"-- code samples, sample chapter, FAQ:
<http://www.xmission.com/~legalize/book/>
Pilgrimage: Utah's annual demoparty
<http://pilgrimage.scene.org>
13 Jan
13 Jan
2:35 a.m.
On Thu, 12 Jan 2006 14:29:07 -0700
Richard <legalize at xmission.com> wrote:
If not, how hard is it to find a Pertec interface
for:
- VAXserver 4000
- PDP-11 Q-bus
Usually somthing like an Emulex or Dilog QBus to Pertec
controller isn't
that hard to find. Have a look at the pdp11-field-guide.txt
--
tsch??,
Jochen
Homepage: http://www.unixag-kl.fh-kl.de/~jkunz/
6921
days inactive
6921
days old
13 comments
6 participants
participants (6)
-
cclist＠sydex.com -
david_comley＠yahoo.com -
jkunz＠unixag-kl.fh-kl.de -
jwstephens＠msm.umr.edu -
legalize＠xmission.com -
pat＠computer-refuge.org