Anybody home?
Sarr J. Blumson
things just really quiet, or has my nntp source lost its feed
for this group?
--
--------
Sarr Blumson sarr.b...@alum.dartmouth.org
http://www-personal.umich.edu/~sarr/
Pat Farrell
> Ok, It's been over a week since I've seen anything here? Are
> things just really quiet, or has my nntp source lost its feed
> for this group?
The group goes quiet periodically, there hasn't been all that much
development in the KL and KS lines in a while.
--
Pat Farrell
http://www.pfarrell.com/
jmfbahciv
> Ok, It's been over a week since I've seen anything here? Are
> things just really quiet, or has my nntp source lost its feed
> for this group?
>
/BAH
Morten Reistad
Pat Farrell <pfar...@pfarrell.com> wrote:
>Sarr J. Blumson wrote:
>> Ok, It's been over a week since I've seen anything here? Are
>> things just really quiet, or has my nntp source lost its feed
>> for this group?
>
>The group goes quiet periodically, there hasn't been all that much
>development in the KL and KS lines in a while.
Yes, and Tops20 is seriously lagging in its readiness for ipv6.
-- mrr
Mark Crispin
> Yes, and Tops20 is seriously lagging in its readiness for ipv6.
I believe that by the time IPv4 ceases to be viable, everyone who has even
heard of TOPS-20 will be long deceased.
-- Mark --
http://panda.com/mrc
Democracy is two wolves and a sheep deciding what to eat for lunch.
Liberty is a well-armed sheep contesting the vote.
Stan Barr
Well, I've been slowly working through the TOPS-10 Operator's Guide and
trying stuff out, but that'll all be old hat for youse guys :-)
I've no doubt when I have a better idea of what I'm doing I'll be
bothering you with silly questions!
--
Cheers,
Stan Barr plan.b .at. dsl .dot. pipex .dot. com
The future was never like this!
Pat Farrell
> On Sat, 21 Nov 2009, Morten Reistad posted:
>> Yes, and Tops20 is seriously lagging in its readiness for ipv6.
>
> I believe that by the time IPv4 ceases to be viable, everyone who has
> even heard of TOPS-20 will be long deceased.
Most of us have been out of it for 25 to 30 years, which means we are at
least 40, and more likely pushing 60.
Since IPv6 was claimed to be in critical need to solve all known
problems, it sure would seem to be needed before we all kick the bucket.
But at current adoption rates, Mark may be correct.
My local ISP has no plans to deploy IPv6 even to business accounts.
Closer to home, there has been a fair amount of press about the 40th
anniversary of Multics, which while different in design and goals, was
on weird 36 bit hardware and designed for timesharing.
And press about one or the other 40th of the Apranet, which talks about
wierd machines like Sigmas.= I came late to the 'net party, (since DEC
would not talk to commercial customers about TCP/IP) but I thought that
Tops-10 and Tops-20 were there from the start. Guess its more "very
early days" rather than start.
Mark Crispin
> Most of us have been out of it for 25 to 30 years, which means we are at
> least 40, and more likely pushing 60.
I'd say that the youngest of the young'uns are 45.
I'm 53, and at the trailing edge of the PDP-10 generation. The leading
edge is much older; 70s or even 80s.
> Since IPv6 was claimed to be in critical need to solve all known
> problems, it sure would seem to be needed before we all kick the bucket.
The need for IPv6 is an invented crisis. As with other invented crises
(e.g., global warming and health care), the proferred solution is quite
expensive, doesn't solve the problems that it claims to solve (NAT will
very much still be around in an IPv6-only world!), and the best result we
can hope is to end up more or less the same as we were before.
There is, however, a lot of money to be made in whispering "the sky is
falling" into the ears of the Chicken Littles. Speaking of which, most
people never learned the full story. Chicken Little convinced the other
chickens to break out of the fenced in yard and flee into the nearby
cave...where the fox had a delightful chicken dinner.
> And press about one or the other 40th of the Apranet, which talks about
> wierd machines like Sigmas.= I came late to the 'net party, (since DEC
> would not talk to commercial customers about TCP/IP) but I thought that
> Tops-10 and Tops-20 were there from the start. Guess its more "very
> early days" rather than start.
Tenex was very much part of ARPAnet from the start. Tenex eventually
involved into TOPS-20.
John Francis
Mark Crispin <m...@panda.com> wrote:
>
>I'm 53, and at the trailing edge of the PDP-10 generation. The leading
>edge is much older; 70s or even 80s.
Yep. I was somewhat late to the party; I started on a KA-10 in 1972.
But that was my first job out of college, so I'm only 60 - a youngster.
Charles Richmond
> On Sat, 21 Nov 2009, Morten Reistad posted:
>> Yes, and Tops20 is seriously lagging in its readiness for ipv6.
>
> I believe that by the time IPv4 ceases to be viable, everyone who has
> even heard of TOPS-20 will be long deceased.
>
ISTM that one of our goals here is to *prevent* TOPS-20 from being
forgotten. That's why some here work for the Computer Museum and
some for PDP Planet.
--
+----------------------------------------+
| Charles and Francis Richmond |
| |
| plano dot net at aquaporin4 dot com |
+----------------------------------------+
Pat Farrell
>> would not talk to commercial customers about TCP/IP) but I thought that
>> Tops-10 and Tops-20 were there from the start. Guess its more "very
>> early days" rather than start.
>
> Tenex was very much part of ARPAnet from the start. Tenex eventually
> involved into TOPS-20.
Perhaps not from the first days in 69? It would have been a fine time
for a BBN hacked KA. When did they start in academic use?
glen herrmannsfeldt
(snip)
> I'd say that the youngest of the young'uns are 45.
> I'm 53, and at the trailing edge of the PDP-10 generation. The leading
> edge is much older; 70s or even 80s.
Only 51 for me. I first used TOPS-10 when I started college in 1976.
By senior year it was replaced by a VAX 11/780. I think there was
some overlap of the two, though.
In grad school we had TOPS-20 which, with one week of notice, was
replaced by three VAX 11/750's. That would have been about 1983.
It seems that someone else wanted it, but only if they could have
it right away. There was enough time to get the files off onto
tape and transfer them to the VAX.
-- glen
Ken McM.
http://www.faqs.org/rfcs/rfc308.html
On page 3 of RFC-308 is:
"6528 Surveys from 09:55:18 on 12/13/71 to 08:50:41 on 03/20/72
--HOST-- -#- -%-UP- -RESP- *
ucla-nmc 001 064% 00.93
sri-arc 002 064% 02.96
ucsb-75 003 048% 00.53
utah-10 004 057% 03.23
bbn-ncc 005 000% 00.00
multics 006 047% 05.60
rand-rcc 007 000% 00.13
sdcadept 010 000% 02.60
harv-10 011 022% 00.23
ll-67 012 021% 03.70
su-sail 013 000% 00.00
ill-11 014 000% 00.06
case-10 015 000% 00.00
..."
Case-10 (KA-10) started on the Arpanet in January,1971 running Tenex.
So the question is - were Utah-10 or Harv(ard)-10 running Tenex?
Ken McM.
Digging further:
http://www.livinginternet.com/i/ii_arpanet.htm
"The fourth ARPANET site was added in December 1969 at the University of
Utah Graphics Department, running on a DEC PDP-10 computer using the
Tenex operating system."
http://en.wikipedia.org/wiki/Talk:TOPS-20
"Harvard ran TOPS-10 (1 machine)"
John Otken
> Mark Crispin <m...@panda.com> wrote:
>
> (snip)
>
> > I'd say that the youngest of the young'uns are 45.
> > I'm 53, and at the trailing edge of the PDP-10 generation. The leading
> > edge is much older; 70s or even 80s.
>
> Only 51 for me. I first used TOPS-10 when I started college in 1976.
> By senior year it was replaced by a VAX 11/780. I think there was
> some overlap of the two, though.
My first job at a KA site was the summer of 79 at Pan Am's in
Edinburgh, Texas.
In late 79, I got a job at UT Austin. They had a KI. I just turned
48. Do I win?
John
Bob, K1BC
> I'm 53, and at the trailing edge of the PDP-10 generation. The leading
> edge is much older; 70s or even 80s.
Well, depends on how you define leading edge, I guess. Alan Kotok and
I were the logic designers, and the first to toggle any code into a
KA-10. That's kinda leading edge, but of course there was the PDP-6
before that. We also worked on those. I turned 67 last month and
Alan was two years older. He and Gordon designed the PDP-6. I didn't
get to DEC until June of '64 by which time a couple of 6's were running.
As to TENEX and the ARPAnet, I moved to BBN in January of 1972. At that
time the BBN Pager had been built and was starting to be deployed beyond
BBN. And the ARPAnet was spreading, too. I had hung out at BBN in my
off hours for some time before 1972 watching all that happen, and
decided to go play there rather than at DEC.
/Rcc
Morten Reistad
A glimpse into another world where 65% availability was considered
normal, because someone had to run the machine offline during some
godforsaken shift.
And connection times in minutes. When current projects work to
shave milliseconds off web response times, and uptime is measured
in "nines".
>Digging further:
>http://www.livinginternet.com/i/ii_arpanet.htm
They even mention EUnet, but the technical part is a little
inaccurate. X.25 transport was mainly a matter of cost. They
network ran uucp and ip from the start.
>"The fourth ARPANET site was added in December 1969 at the University of
>Utah Graphics Department, running on a DEC PDP-10 computer using the
>Tenex operating system."
>
>http://en.wikipedia.org/wiki/Talk:TOPS-20
>"Harvard ran TOPS-10 (1 machine)"
-- mrr
leec
Why would Pan Am have a KA in an out of the way place like Edinburgh -
for the wurst?
Lee C.
Pat Farrell
> My first job at a KA site was the summer of 79 at Pan Am's in
> Edinburgh, Texas.
wow, they were old by then. Some of us argue that the KL was too old by
79. Let alone the KA.
Was that a Tops-10 site? I assume not Tenex since it was commercial.
You might be a winner being under 50.
Morten Reistad
Thorsten Lockert may have a claim. He is several year younger
than me, and I was 50 this year. He worked on the NHH Tops20
installation in the eighties.
-- mrr
jmfbahciv
places. I don't remember the year, but Edinburgh hosted a
European DECUS one year. JMF went and even ate some haggis at
the big dinner.
/BAH
Jonathan Knight
> I'd say that the youngest of the young'uns are 45.
I'm 44 and the DEC-KL10 at Middlesex Poly was the first computer I ever
used. I did my O and A-level course work on it between 1979 and 1984
I should think there are a few younger than me that used the DEC-10
systems at school. Our education authority paid for a 110 baud serial
line to the DEC-KL10 for all the schools in the area, although we were
limited to 30mins of online time per week at the school and 20 hours
total for each month.
Offline debugging, testing and dry running skills were essential.
Jon.
David Powell
Can't remember an EU one in Edinburgh, but there was a UK one there,
at Herriot Watt Uni, 1980ish. That was because the then chairman of
DECUS UK was a Scot (Arrick Wilkinson -- a PDP 10 guy), so the rest
of us had to travel to that out-of-the-way place in Terra Incognita
Borealis.
Most of us were old enough and wise enough to give the haggis a miss,
but the scotch was worth the journey.
Regards,
David P.
John Otken
> John Otken wrote:
> > My first job at a KA site was the summer of 79 at Pan Am's in
> > Edinburgh, Texas.
>
> wow, they were old by then. Some of us argue that the KL was too old by
> 79. Let alone the KA.
I recall having two jobs. Remapping the core memory when a box failed
and making a paper tape bootstrap since they had only one.
> Was that a Tops-10 site? I assume not Tenex since it was commercial.
Yes, TOPS-10. That's why I still have a soft spot for TOPS-10. UT
Austin also ran TOPS-10 on their KI and it was on the ARPAnet. UT got
a 2060 sometime in the early 80s, but I had moved on by then.
>Why would Pan Am have a KA in an out of the way place like
>Edinburgh - for the wurst?
More like tacos. This Edinburgh is in South Texas. The KA was the
University's mainframe.
Mark Crispin
> Mark Crispin wrote:
>> I'd say that the youngest of the young'uns are 45.
> I'm 44 and the DEC-KL10 at Middlesex Poly was the first computer I ever
> used. I did my O and A-level course work on it between 1979 and 1984
Wow. You were 18 when the PDP-10 was cancelled. The kidlets at that time
were all about "I don't want to use this old crap" and thus never learned
many things.
Then again, it's much like the kidlets today who don't want to have
anything to do with any UNIX shells and email because it is "old crap".
> I should think there are a few younger than me that used the DEC-10
> systems at school. Our education authority paid for a 110 baud serial
> line to the DEC-KL10 for all the schools in the area, although we were
> limited to 30mins of online time per week at the school and 20 hours
> total for each month.
Not exactly a pleasant experience I'm sure.
Mark Crispin
> Mark Crispin wrote:
>> I'm 53, and at the trailing edge of the PDP-10 generation. The leading
>> edge is much older; 70s or even 80s.
> Well, depends on how you define leading edge, I guess.
I was thinking of people like John McCarthy, who is in his 80s now (and,
sadly, in failing health; I saw him yesterday). Gordon Bell must be up
there too.
Morten Reistad
>> Mark Crispin wrote:
>Wow. You were 18 when the PDP-10 was cancelled. The kidlets at that time
>were all about "I don't want to use this old crap" and thus never learned
>many things.
But then there were some who saw the light. Not many, but they
were treated generously by the old crowd. Lots of the tops20 stuff
found itself migrated to smaller systems; often as rewrites.
>Then again, it's much like the kidlets today who don't want to have
>anything to do with any UNIX shells and email because it is "old crap".
I still set a lot of people along the unix path. And I find lots
of Linux kids around.
And the Mac is all unix inside.
I even show some of them klh. And show them that there are applications
from the early 1970s that run just fine today too.
-- mrr
Mike Ross
wrote:
I pulled the last working -10 out of Scotland back in... '94 or so? Had been
running until then at Robert Gordons college in Aberdeen. It's slowly coming
back to life, in my copious free time, in my workshop.
And I *still* want to know what happened to the last known KI in the UK (which
was running in Edinburgh) which was carted off to an alleged computer museum
with DEC assistance, and never seen again!
I still have the DECUS list of all UK -10 installations, circa 1990 - I'll scan
it and post it at some point.
Mike
--
http://www.corestore.org
'As I walk along these shores
I am the history within'
Mike Ross
>I even show some of them klh. And show them that there are applications
>from the early 1970s that run just fine today too.
I've been known to do the same with Hercules, to show them 1960s applications
running just fine :-)
jmfbahciv
/BAH
jmfbahciv
> On Tue, 24 Nov 2009 02:41:28 +0100, Morten Reistad <fi...@last.name> wrote:
>
>> I even show some of them klh. And show them that there are applications
>>from the early 1970s that run just fine today too.
>
> I've been known to do the same with Hercules, to show them 1960s applications
> running just fine :-)
>
with DDT.
/BAH
Morten Reistad
I show them DDT and emacs, and the original adventures. And the
comnd% interface, and show them it can be useful even at
300 baud.
But Paul Allen has beat me to the lowest power PDP10 ever made. (I
had it in a 600Mhz via Eden with flash disk, could be run permanently
on a car battery and a small photoviltatic device.
-- mrr
David Evans
Morten Reistad <fi...@last.name> wrote:
>But Paul Allen has beat me to the lowest power PDP10 ever made. (I
>had it in a 600Mhz via Eden with flash disk, could be run permanently
>on a car battery and a small photoviltatic device.
>
Is there a contest? :P I've gone as far as a Bifferboard, something
like 1W (maybe 1.5W with Ethernet).
--
David Evans dfe...@bbcr.uwaterloo.ca
Research Associate http://bbcr.uwaterloo.ca/~dfevans/
Computer Laboratory, University of Cambridge
jmfbahciv
> In article <o2eut6-...@laptop.reistad.name>,
> Morten Reistad <fi...@last.name> wrote:
>> But Paul Allen has beat me to the lowest power PDP10 ever made. (I
>> had it in a 600Mhz via Eden with flash disk, could be run permanently
>> on a car battery and a small photoviltatic device.
>>
>
> Is there a contest? :P I've gone as far as a Bifferboard, something
> like 1W (maybe 1.5W with Ethernet).
>
the SUBTTL of pre-7.03 TOPS-10 sources) would have dreamed of this.
We all probably would have had comments about pigs flying or dreams
if somebody had predicted this. :-)
/BAH
David Evans
Oh yeah, it's insane. But the fun/effort ratio is pretty attractive.
Re-reading Morten's post, he might have meant computational power. I
haven't even bothered to see how my tiny -10 fares in that regard.
Certainly typing stuff at TOPS-20 on it is fine. (I don't know any
TOPS-10, Barb; sorry!) This certainly shows the effort that went into
performance in KLH10; my PDP-11/73 crushes one in SIMH on the
Bifferboard for feel.
jmfbahciv
> In article <hej9k...@news5.newsguy.com>, jmfbahciv <jmfbahciv@aol> wrote:
>> David Evans wrote:
>>> In article <o2eut6-...@laptop.reistad.name>,
>>> Morten Reistad <fi...@last.name> wrote:
>>>> But Paul Allen has beat me to the lowest power PDP10 ever made. (I
>>>> had it in a 600Mhz via Eden with flash disk, could be run permanently
>>>> on a car battery and a small photoviltatic device.
>>>>
>>> Is there a contest? :P I've gone as far as a Bifferboard, something
>>> like 1W (maybe 1.5W with Ethernet).
>>>
>> None of us (JMF,TW, nor any of the others who have their initials in
>> the SUBTTL of pre-7.03 TOPS-10 sources) would have dreamed of this.
>> We all probably would have had comments about pigs flying or dreams
>> if somebody had predicted this. :-)
>>
>
> Oh yeah, it's insane. But the fun/effort ratio is pretty attractive.
>
> Re-reading Morten's post, he might have meant computational power. I
> haven't even bothered to see how my tiny -10 fares in that regard.
> Certainly typing stuff at TOPS-20 on it is fine.
However, you can do all the typing you want _and_ have printing,
networking, assembling, and all other kinds of stuff running without
handholding under other jobs. You can't do that with this "modern"
stuff.
> (I don't know any
> TOPS-10, Barb; sorry!)
that's OK. :-)
>This certainly shows the effort that went into
> performance in KLH10; my PDP-11/73 crushes one in SIMH on the
> Bifferboard for feel.
>
It's the feel that matters. Having to wait for an echo is
one of the most frustrating things. Type ahead, done
properly, is also very important. Keeps the human critters
happy.
/BAH
/BAH
Chris Brannon
> I'd say that the youngest of the young'uns are 45.
>
> I'm 53, and at the trailing edge of the PDP-10 generation. The
> leading edge is much older; 70s or even 80s.
Do the people who use emulators count? I'm 30. I had KLH-10 + Panda up and
running on my local network, and I learned the *basics* of MACRO. I
know a handful of people in their late teens who are at least familiar
with TOPS-20. All of us are totally blind,
and we use Linux with spoken output.
> Democracy is two wolves and a sheep deciding what to eat for lunch.
> Liberty is a well-armed sheep contesting the vote.
That's a great quote, and I borrow it regularly.
-- Chris
Morten Reistad
>Mark Crispin <m...@panda.com> writes:
>
>> I'd say that the youngest of the young'uns are 45.
>>
>> I'm 53, and at the trailing edge of the PDP-10 generation. The
>> leading edge is much older; 70s or even 80s.
>
>Do the people who use emulators count? I'm 30. I had KLH-10 + Panda up and
>running on my local network, and I learned the *basics* of MACRO. I
>know a handful of people in their late teens who are at least familiar
>with TOPS-20. All of us are totally blind,
>and we use Linux with spoken output.
I would like to hear more about how you handle linux and other systems.
And, how does the comnd% interface (the tops20 one with esc and "?" completion)
work for you?
>> Democracy is two wolves and a sheep deciding what to eat for lunch.
>> Liberty is a well-armed sheep contesting the vote.
>
>That's a great quote, and I borrow it regularly.
Indeed.
-- mrr
Morten Reistad
>David Evans wrote:
>> In article <hej9k...@news5.newsguy.com>, jmfbahciv <jmfbahciv@aol> wrote:
>>> David Evans wrote:
>>>> In article <o2eut6-...@laptop.reistad.name>,
>>
>>
>> Re-reading Morten's post, he might have meant computational power. I
>> haven't even bothered to see how my tiny -10 fares in that regard.
>> Certainly typing stuff at TOPS-20 on it is fine.
No, I mean electrical power. My little Eden (the microcode engine,
running linux, is ping6able on eden.reistad.name, 2001:16d8:ee61:2::8)
uses 6 watts all inclusive when not writing heavily to it's little
ssd disk. It runs about twice as fast as a 2065. I can do a monitor
build in less than 10k joules.
>However, you can do all the typing you want _and_ have printing,
>networking, assembling, and all other kinds of stuff running without
>handholding under other jobs. You can't do that with this "modern"
>stuff.
I never have a problem with this on Linux, or BSDs. We put Linux to
the test of writing three DVDs, running 50 RTP-based telephone calls,
(50k packets per second, just there) and doing two large software builds.
The machine was slow, all right, but didn't miss a beat.
>> (I don't know any
>> TOPS-10, Barb; sorry!)
>
>that's OK. :-)
>
>>This certainly shows the effort that went into
>> performance in KLH10; my PDP-11/73 crushes one in SIMH on the
>> Bifferboard for feel.
>>
>It's the feel that matters. Having to wait for an echo is
>one of the most frustrating things. Type ahead, done
>properly, is also very important. Keeps the human critters
>happy.
Have you got klh on the mac yet?
-- mrr
David Evans
Morten Reistad <fi...@last.name> wrote:
>
>No, I mean electrical power. My little Eden (the microcode engine,
>running linux, is ping6able on eden.reistad.name, 2001:16d8:ee61:2::8)
>uses 6 watts all inclusive when not writing heavily to it's little
>ssd disk. It runs about twice as fast as a 2065. I can do a monitor
>build in less than 10k joules.
>
I should measure how well my thing does.
>Have you got klh on the mac yet?
>
FWIW, I did this a few years ago under 10.4 (I think I was rotting
at the airport). No drama.
jmfbahciv
> In article <hem0d...@news1.newsguy.com>, jmfbahciv <jmfbahciv@aol> wrote:
>> David Evans wrote:
>>> In article <hej9k...@news5.newsguy.com>, jmfbahciv <jmfbahciv@aol> wrote:
>>>> David Evans wrote:
>>>>> In article <o2eut6-...@laptop.reistad.name>,
>>> Oh yeah, it's insane. But the fun/effort ratio is pretty attractive.
>>>
>>> Re-reading Morten's post, he might have meant computational power. I
>>> haven't even bothered to see how my tiny -10 fares in that regard.
>>> Certainly typing stuff at TOPS-20 on it is fine.
>
> No, I mean electrical power. My little Eden (the microcode engine,
> running linux, is ping6able on eden.reistad.name, 2001:16d8:ee61:2::8)
> uses 6 watts all inclusive when not writing heavily to it's little
> ssd disk. It runs about twice as fast as a 2065. I can do a monitor
> build in less than 10k joules.
>
>> However, you can do all the typing you want _and_ have printing,
>> networking, assembling, and all other kinds of stuff running without
>> handholding under other jobs. You can't do that with this "modern"
>> stuff.
>
> I never have a problem with this on Linux, or BSDs. We put Linux to
> the test of writing three DVDs, running 50 RTP-based telephone calls,
> (50k packets per second, just there) and doing two large software builds.
> The machine was slow, all right, but didn't miss a beat.
then you are running software that doesn't trip over its bits. I keep
having to wait until the disk clattering dies down and does (what sounds
like a) checkpoint. Then I know I won't have problems.
>
>>> (I don't know any
>>> TOPS-10, Barb; sorry!)
>> that's OK. :-)
>>
>>> This certainly shows the effort that went into
>>> performance in KLH10; my PDP-11/73 crushes one in SIMH on the
>>> Bifferboard for feel.
>>>
>> It's the feel that matters. Having to wait for an echo is
>> one of the most frustrating things. Type ahead, done
>> properly, is also very important. Keeps the human critters
>> happy.
>
> Have you got klh on the mac yet?
>
Not yet.
/BAH
jmfbahciv
> In article <qrh4u6-...@laptop.reistad.name>,
> Morten Reistad <fi...@last.name> wrote:
>> No, I mean electrical power. My little Eden (the microcode engine,
>> running linux, is ping6able on eden.reistad.name, 2001:16d8:ee61:2::8)
>> uses 6 watts all inclusive when not writing heavily to it's little
>> ssd disk. It runs about twice as fast as a 2065. I can do a monitor
>> build in less than 10k joules.
>>
>
> I should measure how well my thing does.
>
>> Have you got klh on the mac yet?
>>
>
> FWIW, I did this a few years ago under 10.4 (I think I was rotting
> at the airport). No drama.
>
/BAH
Morten Reistad
>Morten Reistad wrote:
>> In article <hem0d...@news1.newsguy.com>, jmfbahciv <jmfbahciv@aol> wrote:
>>> David Evans wrote:
>>>> In article <hej9k...@news5.newsguy.com>, jmfbahciv <jmfbahciv@aol> wrote:
>> I never have a problem with this on Linux, or BSDs. We put Linux to
>> the test of writing three DVDs, running 50 RTP-based telephone calls,
>> (50k packets per second, just there) and doing two large software builds.
>> The machine was slow, all right, but didn't miss a beat.
>
>then you are running software that doesn't trip over its bits. I keep
>having to wait until the disk clattering dies down and does (what sounds
>like a) checkpoint. Then I know I won't have problems.
That is how a BSD system that is severely memory-starved will feel.
That is the "purge" that happens when the "rear arm" catches up
with the "front arm" in the clock algorihm for paging/disk cache.
More memory.
-- mrr
Stan Barr
<dfe...@bcr10.uwaterloo.ca> wrote:
> In article <qrh4u6-...@laptop.reistad.name>,
> Morten Reistad <fi...@last.name> wrote:
>>
>>No, I mean electrical power. My little Eden (the microcode engine,
>>running linux, is ping6able on eden.reistad.name, 2001:16d8:ee61:2::8)
>>uses 6 watts all inclusive when not writing heavily to it's little
>>ssd disk. It runs about twice as fast as a 2065. I can do a monitor
>>build in less than 10k joules.
>>
>
> I should measure how well my thing does.
I was looking at the idea of one if these:
http://beagleboard.org
fitted inside a compact keyboard as a PDP-11 emulator, of course with
simh there are a number of possibilities :-)
--
Cheers,
Stan Barr plan.b .at. dsl .dot. pipex .dot. com
The future was never like this!
David Evans
Stan Barr <pla...@dsl.pipex.com> wrote:
>
>I was looking at the idea of one if these:
>http://beagleboard.org
>fitted inside a compact keyboard as a PDP-11 emulator, of course with
>simh there are a number of possibilities :-)
>
Yeah, the Beagleboard is pretty nice. Pretty much the same
price/performance as the Bifferboard, probably, though the Bifferboard
is only £29.
jmfbahciv
/BAH
David Evans
Morten Reistad <fi...@last.name> wrote:
>No, I mean electrical power. My little Eden (the microcode engine,
>running linux, is ping6able on eden.reistad.name, 2001:16d8:ee61:2::8)
>uses 6 watts all inclusive when not writing heavily to it's little
>ssd disk. It runs about twice as fast as a 2065. I can do a monitor
>build in less than 10k joules.
>
You have me beat: I come in at something like 22k. Mine is lower
power but slower by far more!
Morten Reistad
There are limits to what software can do in a real time environment
when the hardware in inadequate. One of the signatures of being fatally
short of memory in BSD systems is that the disk goes into turbo mode
with only a tiny provocation; like a mouse movement over another window,
a resize or a cut&paste. (This is just like a checkpoint; which is
actually what it is - it checkpoints everything so it can discard
pages from memory). Apple OSX is also a BSD system.
We did wonder how you managed with only 128Mb ram and OSX. Now we
know the answer.
When in this mode, all bets are off for realtime performance. Devices
which are realtime sensitive, like CD/DVD writing, tape, and UDP based
network protocols will go into thrashing and probably die after a while.
TCP based network stuff, USB&firewire and most of the connected stuff,
disks and printers will work, but very slowly.
-- mrr
jmfbahciv
of the comm software. Haven't figured out how to reproduce the most
annoying one but I have figured out how to work around it.
/BAH
Chris Brannon
> In article <87zl69r...@cox.net>, Chris Brannon <cmbr...@cox.net> wrote:
>>Do the people who use emulators count? I'm 30. I had KLH-10 + Panda up and
>>running on my local network, and I learned the *basics* of MACRO. I
>>know a handful of people in their late teens who are at least familiar
>>with TOPS-20. All of us are totally blind,
>>and we use Linux with spoken output.
>
> I would like to hear more about how you handle linux and other systems.
I use a screenreader, a program that speaks the text that is displayed
on the screen. Command-line interfaces are pretty straightforward. I
type, and responses are spoken. Graphical user interfaces are a little
more difficult. The mouse isn't an option. We end up using lots
of keyboard shortcuts. Objects are rendered verbally.
Under Linux, the only accessible GUI is via the gnome desktop, so I
don't use it. 'tis too bloated for my taste. I stick to the text console.
> And, how does the comnd% interface (the tops20 one with esc and "?"
> completion) work for you?
Extremely well. I especially like abbreviated input and the ubiquitous
help facility. All in all, its a very human-friendly and discoverable
interface.
-- Chris
Mark Crispin
>> And, how does the comnd% interface (the tops20 one with esc and "?"
>> completion) work for you?
> Extremely well. I especially like abbreviated input and the ubiquitous
> help facility. All in all, its a very human-friendly and discoverable
> interface.
I've always wondered about that. I remember on multiple occasions hearing
testimony about how well TOPS-20 and its applications worked for the
visually-impaired; and suspected that the modern replacements would work
less well.
It'd be ironic if governments started mandating, as part of accessibility
legislation, that software be text-based once again...
-- Mark --
Chris Brannon
> It'd be ironic if governments started mandating, as part of
> accessibility legislation, that software be text-based once again...
That probably won't happen, because most blind folks
bought in to graphical user interfaces. To be honest, they're more or
less usable under most operating systems, especially the Microsoft
ones. For years, the big problem was cost.
Commercially-produced screenreading programs for Windows have a price
tag of around $1000. Then, there's a yearly maintenance fee of between
$100 and $250. Most of this software is purchased by governments,
corporations, and other institutions that can and will afford it.
As a rule, government-imposed accessibility standards are met by
insuring that content is accessible under one of those commercial Windows
programs.
-- Chris
Morten Reistad
Mark Crispin <m...@panda.com> wrote:
>On Sun, 29 Nov 2009, Chris Brannon posted:
>>> And, how does the comnd% interface (the tops20 one with esc and "?"
>>> completion) work for you?
>> Extremely well. I especially like abbreviated input and the ubiquitous
>> help facility. All in all, its a very human-friendly and discoverable
>> interface.
Have you tried any of the modern replacements, like kermit or the
cisco router consoles?
>I've always wondered about that. I remember on multiple occasions hearing
>testimony about how well TOPS-20 and its applications worked for the
>visually-impaired; and suspected that the modern replacements would work
>less well.
Which modern replacements are there? The Columbia/Kermit one is pretty
complicated in terms of licenses. Are there others?
>It'd be ironic if governments started mandating, as part of accessibility
>legislation, that software be text-based once again...
I am pondering if it would be possible to make a 3-way driver for
comnd%, web and window interfaces from the same API, so you could choose
freely how to approach the application.
After all, the comnd% parsing is pretty similar to how you parse input
from gnome, motif etc, and with some stored state it could work for
web-pages as well.
-- mrr
Chris Brannon
> Have you tried any of the modern replacements, like kermit or the
> cisco router consoles?
I don't have access to a Cisco router, but that sounds like fun. I've
used C-Kermit for years. I've also used the C port of the mm mail
client. Its command-line parser is the ccmd library, a C implementation
of COMND% functionality.
I started writing my own exec using ccmd some time ago, but that got
side-tracked by other things.
-- Chris
Rich Alderson
> On Sat, 21 Nov 2009, Pat Farrell posted:
>> Most of us have been out of it for 25 to 30 years, which means we are at
>> least 40, and more likely pushing 60.
> I'd say that the youngest of the young'uns are 45.
> I'm 53, and at the trailing edge of the PDP-10 generation. The leading
> edge is much older; 70s or even 80s.
Well, let's see. Freshmen were still being introduced to TOPS-20 at LOTS in
1991, although far fewer than only 3 years earlier. Even they would be 36 to
38 by now--and probably members of the "why should I use *this* old crap?"
group, for the most part.
--
Rich Alderson "You get what anybody gets. You get a lifetime."
ne...@alderson.users.panix.com --Death, of the Endless
Rich Alderson
> In article <alpine.OSX.2.00.0...@hsinghsing.panda.com>,
> Mark Crispin <m...@panda.com> wrote:
>> I'm 53, and at the trailing edge of the PDP-10 generation. The leading
>> edge is much older; 70s or even 80s.
> Yep. I was somewhat late to the party; I started on a KA-10 in 1972.
> But that was my first job out of college, so I'm only 60 - a youngster.
I'm 58.
I started with a 1401 my last semester of high school, 40 years ago, and
didn't get started with the KL till 1977 when I was in grad school. On the
other hand, I *had* heard of a DECsystem-10 in 1970, thanks to a Datamation
advert that one of the System/360 sysprogs showed me at the CAI Lab at the
University of Texas.
Weird to think of myself as one of the last guardians of the knowledge!
Rich Alderson
> On Nov 21, 7:58=A0pm, glen herrmannsfeldt <g...@ugcs.caltech.edu> wrote:
>> Mark Crispin <m...@panda.com> wrote:
>> (snip)
>>> I'd say that the youngest of the young'uns are 45.
>>> I'm 53, and at the trailing edge of the PDP-10 generation. The leading
>>> edge is much older; 70s or even 80s.
>> Only 51 for me. =A0I first used TOPS-10 when I started college in 1976.
>> By senior year it was replaced by a VAX 11/780. =A0I think there was
>> some overlap of the two, though.
> My first job at a KA site was the summer of 79 at Pan Am's in
> Edinburgh, Texas.
> In late 79, I got a job at UT Austin. They had a KI. I just turned
> 48. Do I win?
So, you worked with Clive?
Rich Alderson
> On Sun, 22 Nov 2009, Bob, K1BC posted:
>> Mark Crispin wrote:
>>> I'm 53, and at the trailing edge of the PDP-10 generation. The leading
>>> edge is much older; 70s or even 80s.
>> Well, depends on how you define leading edge, I guess.
> I was thinking of people like John McCarthy, who is in his 80s now (and,
> sadly, in failing health; I saw him yesterday). Gordon Bell must be up
> there too.
I *think* Gordon is about 70, but not (much) older.
Rich Alderson
> But Paul Allen has beat me to the lowest power PDP10 ever made. (I
> had it in a 600Mhz via Eden with flash disk, could be run permanently
> on a car battery and a small photoviltatic device.
????
I think there's some confusion here. What system(s) did you have in mind?
Rich Alderson
> In article <hem0d...@news1.newsguy.com>, jmfbahciv <jmfbahciv@aol> wrote:
[ things that led Morten to ask ]
> Have you got klh on the mac yet?
Umm, no, she doesn't. I've been tied up for a while getting the museum into
shape while being supportive of my wife while she dealt with her mother's
Alzheimer's (and as of last week, with her mother's passing).
I'm working on it.
Rich Alderson
> Morten Reistad wrote:
>>> Morten Reistad wrote:
>> More memory.
Sometimes the correct answer *is* to throw more resources at a problem. I know
it goes against the Tops-10 grain, but sometimes it was the only resolution of
TOPS-20 performance issues.
Get that extra memory installed! You're blaming software for not working well
in a hardware resource-starved environment. (Hey, imagine running 6.03A on a
16KW KA10.)
Rich Alderson
> On Sun, 29 Nov 2009, Chris Brannon posted:
>>> And, how does the comnd% interface (the tops20 one with esc and "?"
>>> completion) work for you?
>> Extremely well. I especially like abbreviated input and the ubiquitous
>> help facility. All in all, its a very human-friendly and discoverable
>> interface.
> I've always wondered about that. I remember on multiple occasions hearing
> testimony about how well TOPS-20 and its applications worked for the
> visually-impaired; and suspected that the modern replacements would work
> less well.
There were a number of very happy sight-impaired LOTS users during my tenure
there, back when text-to-speech was very primitive and ASCII-to-Braille readers
were common.
John Francis
Rich Alderson <ne...@alderson.users.panix.com> wrote:
>
>Sometimes the correct answer *is* to throw more resources at a problem. I know
>it goes against the Tops-10 grain, but sometimes it was the only resolution of
>TOPS-20 performance issues.
And TOPS-10 issues, too.
The first KA-10 machine I worked on had 48K (words) of memory. Theoretically
the 32K it was bought with was enough for it to work well, but in practice that
wasn't the case. DEC gave UMIST an extra 16K memory cabinet to fix the issues.
>Get that extra memory installed! You're blaming software for not working well
>in a hardware resource-starved environment. (Hey, imagine running 6.03A on a
>16KW KA10.)
I suspect you probably needed more than 16K to run 6.03 - I belive we upped the
memory to 80K somewhere between 4.whatever that we were running when I joined
and 6.03 coming along.
Morten Reistad
Rich Alderson <ne...@alderson.users.panix.com> wrote:
>Morten Reistad <fi...@last.name> writes:
>
>> But Paul Allen has beat me to the lowest power PDP10 ever made. (I
>> had it in a 600Mhz via Eden with flash disk, could be run permanently
>> on a car battery and a small photoviltatic device.
>
>????
>
>I think there's some confusion here. What system(s) did you have in mind?
I just read a posting that klh was running on some even smaller device,
cannot remember the details. It was from a third party.
Anyway; what I have has been documented earlier.
Via C3 Eden, 633 Mhz, 133Mhz bus, 512 megabytes memory, 2 flash cards
in IDE busses, one 1G and one 4G, rhine ethernet chip. Three power
saving states, stepping down 133Mhz at a time. Bought from www.mini-itx.com
nearly 5 years ago. The exact model is discontinued, but similar ones are
still on sale.
A stripped Fedora core 3 and KLH; running panda. Peak power consumption
when writing intensly to both flash chips is 13W, just idling is 3.5W;
normal tops20 emacsing around 5W, monitor build at 8W. I had my boat
power monitor plugged in to check, running it from a 105Ah marine battery
for a little more than three days (it requires 12.2V voltage, so it will
not drain the battery further.). You can ping6 the outside linux.[1]
It measures 3860 dhrystones; where a geniune 2065 measured 1835.
I have made myself a number of "RP06" memory sticks. They fit just
exactly on 512M USB sticks. No file system except the RP06 one.
BTW: Are there any genuine 9-track tape devices available that
can be connected without too much of an investment?
-- mrr
[1] Are we the only people in the world to actually USE ipv6 in a
production setting [2] ? We deployed ipv6 as internal network, and it just
works out of the box, mac, bsds, linux, windows. No NAT.
[2] At least the spammers haven't caught on. Having one IPv6-only host
as primary MX (with ipv4 forced to use secondaries) loses around 80%
of the spam, right there.
David Evans
Morten Reistad <fi...@last.name> wrote:
>
>It measures 3860 dhrystones; where a geniune 2065 measured 1835.
>
Yeah, my Bifferboard-based thing gets about 1400. (I got about
4000ish...can't remember...from a TOAD-1.)
jmfbahciv
> jmfbahciv <jmfbahciv@aol> writes:
>
>> Morten Reistad wrote:
>
>>> In article <heolb...@news5.newsguy.com>, jmfbahciv <jmfbahciv@aol> wrote:
>
>>>> Morten Reistad wrote:
>
>>>>> I never have a problem with this on Linux, or BSDs. We put Linux to
>>>>> the test of writing three DVDs, running 50 RTP-based telephone calls,
>>>>> (50k packets per second, just there) and doing two large software builds.
>>>>> The machine was slow, all right, but didn't miss a beat.
>
>>>> then you are running software that doesn't trip over its bits. I keep
>>>> having to wait until the disk clattering dies down and does (what sounds
>>>> like a) checkpoint. Then I know I won't have problems.
>
>>> That is how a BSD system that is severely memory-starved will feel.
>>> That is the "purge" that happens when the "rear arm" catches up
>>> with the "front arm" in the clock algorihm for paging/disk cache.
>
>>> More memory.
>
>> More memory won't fix the software bugs.
>
> Sometimes the correct answer *is* to throw more resources at a problem.
But that doesn't solve the underlying bugs that are annoying.
> I know
> it goes against the Tops-10 grain, but sometimes it was the only resolution of
> TOPS-20 performance issues.
There are performance issues and then there are bugs. I know how to
wait w.r.t. performance problems.
>
> Get that extra memory installed!
Coincidentally, I unpacked the memory card the day before Morten asked
the question.
You're blaming software for not working well
> in a hardware resource-starved environment.
Errrmm....no. I'm not. I know the difference.
> (Hey, imagine running 6.03A on a
> 16KW KA10.)
>
I don't have to imagine it. :-) although 16K wasn't the COREMAX.
/BAH
jmfbahciv
> In article <mddr5rc...@panix5.panix.com>,
> Rich Alderson <ne...@alderson.users.panix.com> wrote:
>> Sometimes the correct answer *is* to throw more resources at a problem. I know
>> it goes against the Tops-10 grain, but sometimes it was the only resolution of
>> TOPS-20 performance issues.
>
> And TOPS-10 issues, too.
>
> The first KA-10 machine I worked on had 48K (words) of memory. Theoretically
> the 32K it was bought with was enough for it to work well, but in practice that
> wasn't the case. DEC gave UMIST an extra 16K memory cabinet to fix the issues.
The sceduler was improved by the 6-series monitor. Shuffling didn't
work well.
>
>> Get that extra memory installed! You're blaming software for not working well
>> in a hardware resource-starved environment. (Hey, imagine running 6.03A on a
>> 16KW KA10.)
>
> I suspect you probably needed more than 16K to run 6.03 - I belive we upped the
> memory to 80K somewhere between 4.whatever that we were running when I joined
> and 6.03 coming along.
>
>
48K ran 4S72. the 5-series monitors needed a lot of work for the first
major releases. I don't recall the version number of the disaster
ship which cuased the monitor development group to change its work
procedures to always run the last edit of the monitor (done every
Tuesday).
/BAH
jmfbahciv
> jo...@panix.com (John Francis) writes:
>
>> In article <alpine.OSX.2.00.0...@hsinghsing.panda.com>,
>> Mark Crispin <m...@panda.com> wrote:
>
>>> I'm 53, and at the trailing edge of the PDP-10 generation. The leading
>>> edge is much older; 70s or even 80s.
>
>> Yep. I was somewhat late to the party; I started on a KA-10 in 1972.
>> But that was my first job out of college, so I'm only 60 - a youngster.
>
> I'm 58.
>
> I started with a 1401 my last semester of high school, 40 years ago, and
> didn't get started with the KL till 1977 when I was in grad school. On the
> other hand, I *had* heard of a DECsystem-10 in 1970, thanks to a Datamation
> advert that one of the System/360 sysprogs showed me at the CAI Lab at the
> University of Texas.
>
> Weird to think of myself as one of the last guardians of the knowledge!
>
The feel of the machine working cannot be described in ASCII English.
/BAH
jmfbahciv
> Morten Reistad <fi...@last.name> writes:
>
>> In article <hem0d...@news1.newsguy.com>, jmfbahciv <jmfbahciv@aol> wrote:
>
> [ things that led Morten to ask ]
>
>> Have you got klh on the mac yet?
>
> Umm, no, she doesn't.
I don't even have the printer I bought last year on-line yet.
> I've been tied up for a while getting the museum into
> shape while being supportive of my wife while she dealt with her mother's
> Alzheimer's (and as of last week, with her mother's passing).
>
Sorry to hear that. It's sadness combined with a huge relief. Dad
had something similar but died in August before he got really bad
mentally.
> I'm working on it.
>
I am not in any hurry. :-) There are other things which have been
keeping me distracted.
/BAH
John Otken
wrote:
> So, you worked with Clive?
Yes I did. He caught me in his office late one Friday or Saturday
night drinking beer and playing DecWars at 9600 baud.
I haven't spoken with Clive in years, but I still have lunch with Rick
Watson. He wrote a bunch of front-end code for the KI.
John Otken
Morten Reistad
Did anyone else notice that the KL10 has a distinct smell? Still
electronic-ish, but very distinct. PDP11s and peripherals smell
differently. I have only encountered that smell from the KL10
cabinets.
-- mrr
John Francis
>John Francis wrote:
>>
>> I suspect you probably needed more than 16K to run 6.03 - I belive we upped the
>> memory to 80K somewhere between 4.whatever that we were running when I joined
>> and 6.03 coming along.
>
>48K ran 4S72. the 5-series monitors needed a lot of work for the first
>major releases. I don't recall the version number of the disaster
>ship which cuased the monitor development group to change its work
>procedures to always run the last edit of the monitor (done every
>Tuesday).
Thanks for filling in the details.
I should have gone with my instincts - I almost typed "4.72" rather than
"4.whatever", based purely on almost-40-year-old memories.
I don't think UMIST ever ran a 5-series monitor; it didn't offer anything
that we needed, and by then it was my call as to whether we upgraded :-)
Rich Alderson
> In article <mddws14...@panix5.panix.com>,
> Rich Alderson <ne...@alderson.users.panix.com> wrote:
>> Morten Reistad <fi...@last.name> writes:
>>> But Paul Allen has beat me to the lowest power PDP10 ever made. (I
>>> had it in a 600Mhz via Eden with flash disk, could be run permanently
>>> on a car battery and a small photoviltatic device.
>> ????
>> I think there's some confusion here. What system(s) did you have in mind?
> I just read a posting that klh was running on some even smaller device,
> cannot remember the details. It was from a third party.
I was just trying to figure out where Paul Allen fits in to this.
I think you may be thinking of Jake Nelson's Zaurus port, back when he and I
both worked at XKL.
(I occasionally run klh10 on my desktop to check out some Tops-10 behaviour,
but nothing simulated runs out of the museum. Unless you're a microcode
snob^Wpurist. :-)
jmfbahciv
I had no idea that KLs had a smell until I visited the last one
in Marlboro after Jim died. It invoked all kinds of "coming
home" reactions.
/BAH
Rhialto
> On Sat, 21 Nov 2009, Pat Farrell posted:
>> Most of us have been out of it for 25 to 30 years, which means we are at
>> least 40, and more likely pushing 60.
>
I lurk here sometimes because I like reading about old real computers,
and I'm 42.
> I'm 53, and at the trailing edge of the PDP-10 generation. The leading
> edge is much older; 70s or even 80s.
>
>> Since IPv6 was claimed to be in critical need to solve all known
>> problems, it sure would seem to be needed before we all kick the bucket.
>
> The need for IPv6 is an invented crisis. As with other invented crises
> (e.g., global warming and health care),
Let's just say we probably disagree on most political topics, but I
enjoy your knowledge of DEC things.
> -- Mark --
--
___ Olaf 'Rhialto' Seibert -- You author it, and I'll reader it.
\X/ rhialto/at/xs4all.nl -- Cetero censeo "authored" delendum esse.
Rhialto
> In article <qrh4u6-...@laptop.reistad.name>,> FWIW, I did this a few years ago under 10.4 (I think I was rotting
> at the airport). No drama.
One could add all sorts of GUI tools to do debugging from the outside,
similar to something I was pointed to recently for another (much later
and less powerful) computer,
http://icu64.blogspot.com/2009/09/first-public-release-of-icu64frodo.html
-Olaf