13 Jan
2007
13 Jan
'07
9:51 a.m.
I have a few 16/4 Token Ring ISA cards and an EISA 100Mb one that I want to test out in a
few 486 systems I have to compare it to 10Mb Ethernet that I currently run. Anybody have a
super cheap hub for sale I can use (these are the RJ45 type cables)?
Located in the US.
Also what is a good way to bridge Token Ring to Ethernet so that these 486 machines can
use my router/cable modem?
13 Jan
13 Jan
10:07 p.m.
New subject: desperately seeking pertec cables
I have a TS05 tape drive with cables for a TSV05, i.e. 50 pin edge to
50 pin high density (i.e. scsi-2 connectors).
I thought I had a way to connect the cables to 50 pin .1" headers (which
the controller board wants) but it turns out the way the DEC TVS05
cables uses the "high density/a.k.a scsi-2" connector is not the same as
the way SCSI-2 uses it - the pins don't map correctly if you use an off
the shelf scsi-2/.1"header converter.
This has me frustrated and an in need of some standard "pertec cables",
i.e. 50 pin edge to 50 pin .1" header.
(i was all set to boot off the ts05 or tm11 controller but no joy :-(
Does anyone have a pair of these, at least say 10' that they don't need
and would sell?
-brad
i'm going to go drown my sorrows in a martini :-)
14 Jan
14 Jan
1:15 a.m.
New subject: desperately seeking pertec cables
On 1/13/07, Brad Parker <brad at heeltoe.com> wrote:
This has me frustrated and an in need of some standard "pertec cables",
i.e. 50 pin edge to 50 pin .1" header.
Does anyone have a pair of these, at least say 10' that they don't need
and would sell?
Let us know if you find a cheap source. I'd like to try getting an M4
9914 running with either a DQ132 or a DQ142 but I don't have any
cables either. Going by single quantity prices for a section of 10ft
flat cable + socket connector + card edge connector adds up to about
$25 a cable or $50 for the pair on Digi-Key part prices. That might
be ok if I knew I would be able to get everything to work, but would
be an expensive experiment if I couldn't get it all to work.
-Glen
8:19 a.m.
New subject: desperately seeking pertec cables
In article <1e1fc3e90701131715ncd01037g957c65f619f2821e at mail.gmail.com>,
"Glen Slick" <glen.slick at gmail.com> writes:
Let us know if you find a cheap source.
Definately; I'm in the same boat. I have a Fujitsu M2444 9-track
drive with a pertect interface and a Q-bus pertec interface card but
no appropriate cable to connect the two.
--
"The Direct3D Graphics Pipeline" -- DirectX 9 draft available for download
<http://www.xmission.com/~legalize/book/download/index.html>
Legalize Adulthood! <http://blogs.xmission.com/legalize/>
20 Jan
20 Jan
3:57 a.m.
New subject: desperately seeking pertec cables
On 1/14/07, Richard <legalize at xmission.com> wrote:
I went ahead and ordered some 50-conductor flat ribbon cable and I'll
pick up some IDC socket and card edge connectors and see if I can
build up a set of working pertec interface cables.
If it works I'll have extra flat ribbon cable to build another set or
two. Twisted ribbon cable might be better for longer runs, but I
couldn't find a cheap supply of that yet.
-Glen
Let us know if you find a cheap source.
Definately; I'm in the same boat. I have a Fujitsu M2444 9-track
drive with a pertect interface and a Q-bus pertec interface card but
no appropriate cable to connect the two.
-- 
2:12 p.m.
New subject: pertec cables- 50 conductor twisted available
FWIW, I have available various lengths of 50 conductor twisted pair
cable. Pay for postage and packing (I'll just drop it off at a UPS
store, and let them pack it) and you (plural) can have it. I prefer to
spread it around if possible.
For all cables listed, every 2 feet the wires untwist so an IDC
connector can be used.
15' Spectra Strip cable, T&B/Ansley "Centronics" style female connector
on one end, 25x2 (.100"?) header connector on other. Came from "LEX"
system.
8' LL31941 CSA AWM T2 cable, DB-50 (non-idc) on one end, 25x2 (.100"?)
header connector on other. Labeled Prime PR 3/88, CBL10218-001 Rev. B
10' Spectra Strip Cable, 25x2 (.100"?) header connector on one end.
Other connector was removed.
6' Spectra Strip Cable, see previous 10' description.
10' Spectra Strip Cable, 25x2 (.100"?) header connector on one end,
"Centronics" style male connector on other.
10' Spectra Strip Cable, "Centronics" style male connector on one end,
other end has been cut up.
22' 60 CONDUCTOR Spectra Strip cable, Edge connector on one end, Berg
2x30 (.100"?) header connector on other. Labeled Prime CBL3715-902
Rev. D. Only one I have, so don't take it unless you can use it.
I also have dozens of 50 conductor regular cables, various lengths and
connectors, if they will help anybody's project.
I've been meaning to clean out some more, and maybe this will start it.
Joe Heck
Glen Slick wrote:
On 1/14/07, Richard <legalize at xmission.com>
wrote:
I went ahead and ordered some 50-conductor flat ribbon cable and I'll
pick up some IDC socket and card edge connectors and see if I can
build up a set of working pertec interface cables.
If it works I'll have extra flat ribbon cable to build another set or
two. Twisted ribbon cable might be better for longer runs, but I
couldn't find a cheap supply of that yet.
-Glen
Let us know if you find a cheap source.
Definately; I'm in the same boat. I have a Fujitsu M2444 9-track
drive with a pertect interface and a Q-bus pertec interface card but
no appropriate cable to connect the two.
-- 
14 Jan
14 Jan
12:18 p.m.
Teo Zenios wrote:
Also what is a good way to bridge Token Ring to
Ethernet so that
these 486 machines can use my router/cable modem?
Last time I did that was quite a few years ago, using a spare Linux box. I
think the TR side of things was a Madge PCI board of some flavour, with the
Ethernet side probably being a 3Com 905 PCI (100mbit).
As I recall, Linux supported the IBM ISA boards back then, along with Madge
cards using Madge's own drivers. I don't think there was any support around at
all for Olicom cards. Things may well be better these days... people have had
about ten years to write a few more drivers :-)
Our (small-ish, maybe 150 machines) network was token ring, but there'd always
be a few projects on the go which used server machines from the likes of Sun,
SGI, HP etc. and which really wanted to be on Ethernet - hence the need for
the bridge.
cheers
Jules
16 Jan
16 Jan
1:51 a.m.
Jules Richardson wrote:
Teo Zenios wrote:
I know for a fact that the IBM TROPIC-based ISA, MCA and PCMCIA token
ring cards work well under NetBSD.
Peace... Sridhar
Also what is a good way to bridge Token Ring to
Ethernet so that
these 486 machines can use my router/cable modem?
Last time I did that was quite a few years ago, using a spare Linux box.
I think the TR side of things was a Madge PCI board of some flavour,
with the Ethernet side probably being a 3Com 905 PCI (100mbit).
As I recall, Linux supported the IBM ISA boards back then, along with
Madge cards using Madge's own drivers. I don't think there was any
support around at all for Olicom cards. Things may well be better these
days... people have had about ten years to write a few more drivers :-) 
14 Jan
14 Jan
4:34 p.m.
Teo Zenios wrote:
Also what is a good way to bridge Token Ring to Ethernet so that these 486 machines can
use my router/cable modem?
I purchased some 3Com 3c359's off ebay, and built up a linux box to
bridge them.
There is a 3c509 on the ethernet side. There is a 3174 on the token
ring. It is a standard MAU, no rj45 baluns. Don't forget on ancient
MAU's you have to have a token ring starter upper, to charge up
the hub. I had a years old one, and was informed to use the supplied
IBM starter to set up mine. I might point out that the MAU had no
active power, as some of the ones which use rj45's do, and that
may have had to do with the "starter upper" gizmo.
Anyone who can educate me on that one would do me a favor, I
just did it and it worked (after replacing the 9v battery in the IBM
gizmo)
Jim
6:12 p.m.
jim wrote:
Anyone who can educate me on that one would do me a
favor, I
just did it and it worked (after replacing the 9v battery in the IBM
gizmo)
There are relays in the MAU that bypass a dead or disconnected port, as
far as I know. You need the starter thingy to switch at least one on so
the others come up.
At least, that's how it was for store loop. Real token ring may be
different.
Gordon
10:18 p.m.
New subject: the Cambridge Ring / was Re: Token Ring Hub
All this talk of token rings has reminded me of the original Cambridge ring.
In the early 80's a project in the CS department where I was working was going
to set up a Cambridge ring. A bunch of hardware was ordered and received from
England: as I recall the network nodes were a yellow metal box around
12"*16"*4" with a hinged door and PCB, etc. inside.
At the same time, Ethernet (thickwire, of course) was becoming current and
wire was pulled through the CS building to accommodate both Ethernet and
Cambridge ring (I think the decision-makers were hedging bets as to which
technology was going to predominate). The Cambridge ring nodes sat on a shelf
for years and were never used, while various machines (primarily a VAX780 and
750) went on the ethernet. I think a lot of the issue was availability of
processor interface hardware/drivers.
Anybody recall or have some of these nodes? Is there a Cambridge ring running
anywhere? (Bletchley perhaps.)
11:48 p.m.
New subject: the Cambridge Ring / was Re: Token Ring Hub
Brent Hilpert wrote:
Anybody recall or have some of these nodes? Is there a
Cambridge ring running
anywhere? (Bletchley perhaps.)
No we don't, sadly. I'm not even aware that we have *any* CR stuff, actually.
I've raised it before that we really should if we get the opportunity as it's
a good example (along with Econet and Research Machines' efforts) of UK
networking technology.
Cambridge university do still maintain a collection - although not in a
running / viewable condition. At least that does mean that someone still owns
a pile of the hardware, at least.
There's a chap called David Greaves who had some really nice CR / CFR pages up at:
http://koo.corpus.cam.ac.uk/projects/earlyatm/earlyatm.html
... however the server seems broken (at least for me) at present. Bashing
"corpus.cam.ac.uk early atm" into Google should get you a link to that page
and allow you to see the cached text of the pages though, but it's not as much
fun without the images :)
[it seems you can't put a link into Google and expect their entry for that
link back in order to reach the cached version, which is a bit annoying!]
cheers
Jules
15 Jan
15 Jan
12:13 a.m.
New subject: the Cambridge Ring / was Re: Token Ring Hub
Jules Richardson wrote:
There's a chap called David Greaves who had some
really nice CR / CFR
pages up at:
http://koo.corpus.cam.ac.uk/projects/earlyatm/earlyatm.html
... however the server seems broken (at least for me) at present.
Bashing "corpus.cam.ac.uk early atm" into Google should get you a
link to that page
and allow you to see the cached text of the pages though, but it's
not as much fun without the images :)
http://web.archive.org/web/20021211223436/http://koo.corpus.cam.ac.uk/pr
ojects/earlyatm/earlyatm.html
The thumbnails don't show up for me but if you click on them they do
seem to be there.
Antonio
6552
days inactive
6559
days old
12 comments
11 participants
participants (11)
-
arcarlini＠iee.org -
brad＠heeltoe.com -
glen.slick＠gmail.com -
gordon＠gjcp.net -
hilpert＠cs.ubc.ca -
julesrichardsonuk＠yahoo.co.uk -
jwstephens＠msm.umr.edu -
legalize＠xmission.com -
ploopster＠gmail.com -
teoz＠neo.rr.com -
trash3＠splab.cas.neu.edu