Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
Before the PDP-8
126 views
Skip to first unread message
Quadibloc
May 18, 2013, 7:32:37 AM5/18/13
to
I see that the SPC-12 manual is on Bitsavers; I had looked for it
there before long ago, and had not found it then.
I looked now because an issue of Computers and Automation from 1969
mentioned two 12-bit computers - the ELBIT 100 and the Motorola
MDP-1000.
Those aren't described on Al Kossow's site, but I see the Computer
History Museum has a copy of the Elbit 100 technical spec, so I
suppose it will come soon.
In my Google for Elbit 100 information, I saw this paper
http://www.math.tau.ac.il/~matias/99-SPRING-TALKS/even.html
and the brief CV of the author at the end claims that the Elbit 100
has the historical significance commonly attributed to the PDP-8!!!
John Savard
there before long ago, and had not found it then.
I looked now because an issue of Computers and Automation from 1969
mentioned two 12-bit computers - the ELBIT 100 and the Motorola
MDP-1000.
Those aren't described on Al Kossow's site, but I see the Computer
History Museum has a copy of the Elbit 100 technical spec, so I
suppose it will come soon.
In my Google for Elbit 100 information, I saw this paper
http://www.math.tau.ac.il/~matias/99-SPRING-TALKS/even.html
and the brief CV of the author at the end claims that the Elbit 100
has the historical significance commonly attributed to the PDP-8!!!
John Savard
Rod Speed
May 18, 2013, 11:02:30 AM5/18/13
to
"Quadibloc" <jsa...@ecn.ab.ca> wrote in message
news:3ceb5fc4-5472-4283...@a8g2000yqp.googlegroups.com...
Quadibloc
May 19, 2013, 2:40:22 AM5/19/13
to
On May 18, 5:32 am, I wrote:
> Those aren't described on Al Kossow's site, but I see the Computer
> History Museum has a copy of the Elbit 100 technical spec, so I
> suppose it will come soon.
I see I must thank Al Kossow for reading my posts, and being guided by
> Those aren't described on Al Kossow's site, but I see the Computer
> History Museum has a copy of the Elbit 100 technical spec, so I
> suppose it will come soon.
them in his scanning!
I modified my page on 12-bit systems
http://www.quadibloc.com/comp/cp0305.htm
and other miscellaneous systems to include the Elbit 100, as well as
to include a more extensive discussion of the Honeywell H-112.
I recently corrected the page on 24-bit systems
http://www.quadibloc.com/comp/cp0303.htm
as I had somehow managed to get the instruction format of the SDS-940
and related computers wrong.
Incidentally, some things weren't added.
The information on the H-300, unfortunately, didn't include the
instruction formats. They did include the floating-point format, which
is the same as that used on some other computers already listed on my
page on that subject; I haven't noted the additional computer there
yet.
The Elbit computer uses a different notation for hexadecimal numbers.
It appears to be a capitalized version of Honeywell notation, but I
can't yet verify that:
0123456789BCDEFG
John Savard
John Byrns
May 19, 2013, 10:12:02 AM5/19/13
to
In article <3ceb5fc4-5472-4283...@a8g2000yqp.googlegroups.com>,
The Motorola MDP-1000 is a rebadged and repackaged General Automation SPC-12.
It is essentially an SPC-12 stood on its nose with the switches and blinken
lights rotated around to the new front side. This gave it a svelte profile for
mounting in 19 inch racks without requiring the excessive depth of other
computers, although it was much taller as a result.
Back in the 1970-71 time frame I wrote a hyperbolic navigation package for the
MDP-1000, I threw out the manuals years ago, but the SPC-12 programming manuals
will work.
--
Regards,
John Byrns
Surf my web pages at, http://fmamradios.com/
It is essentially an SPC-12 stood on its nose with the switches and blinken
lights rotated around to the new front side. This gave it a svelte profile for
mounting in 19 inch racks without requiring the excessive depth of other
computers, although it was much taller as a result.
Back in the 1970-71 time frame I wrote a hyperbolic navigation package for the
MDP-1000, I threw out the manuals years ago, but the SPC-12 programming manuals
will work.
--
Regards,
John Byrns
Surf my web pages at, http://fmamradios.com/
Quadibloc
May 19, 2013, 10:58:09 AM5/19/13
to
On May 19, 8:12 am, John Byrns <byr...@sbcglobal.net> wrote:
> The Motorola MDP-1000 is a rebadged and repackaged General Automation SPC-12.
Ah. And that's an 8-bit computer with a 12-bit address, which is why
> The Motorola MDP-1000 is a rebadged and repackaged General Automation SPC-12.
it isn't included in that section.
John Savard
Quadibloc
May 19, 2013, 11:02:13 AM5/19/13
to
On May 19, 12:40 am, Quadibloc <jsav...@ecn.ab.ca> wrote:
> They did include the floating-point format, which
> is the same as that used on some other computers already listed on my
> page on that subject; I haven't noted the additional computer there
> yet.
I've not only added in the H-300, but I've also finally included that
> They did include the floating-point format, which
> is the same as that used on some other computers already listed on my
> page on that subject; I haven't noted the additional computer there
> yet.
of the Datamatic D-1000, also used on the H-400, the H-800, and so on.
It turns out that this floating-point format has an interesting story.
Basically, it's the original System/360 format, but 48 bits long, used
years before the 360. However, Honeywell had a better excuse than IBM
for using a base-16 exponent: the same format was also used for packed
decimal floating-point!
I had somewhat disdained this computer, because also for decimal/
binary compatibility, its binary integers used the first four bits
rather than the first bit for the sign, wasting three bits.
John Savard
Al Kossow
May 19, 2013, 12:59:48 PM5/19/13
to
On 5/18/13 11:40 PM, Quadibloc wrote:
> The information on the H-300, unfortunately, didn't include the
> instruction formats.
I didn't have time to get to the H300 programming manual. I'll try to get to it
> The information on the H-300, unfortunately, didn't include the
> instruction formats.
in the next week or two.
Quadibloc
May 19, 2013, 2:50:58 PM5/19/13
to
On May 19, 10:59 am, Al Kossow <a...@bitsavers.org> wrote:
> I didn't have time to get to the H300 programming manual. I'll try to get to it
> in the next week or two.
That's wonderful! Thank you very much. I thought that you simply
> I didn't have time to get to the H300 programming manual. I'll try to get to it
> in the next week or two.
didn't have one available.
The issues of Computers and Automation provided clues and suggestions
for further inquiry. But I've also now come across what seems like a
sad story. The editor was the author of one of the first books about
computers, "Giant Brains". But I see that in the late 1960s, he
published some controversial articles in the magazine, and I suspect
that this may have driven many advertisers away, leading to the
publication dwindling.
John Savard
Quadibloc
May 19, 2013, 5:21:04 PM5/19/13
to
On May 19, 12:40 am, I wrote:
> The Elbit computer uses a different notation for hexadecimal numbers.
> It appears to be a capitalized version of Honeywell notation, but I
> can't yet verify that:
I missed a footnote (in color) on the instruction opcode page; it is
> The Elbit computer uses a different notation for hexadecimal numbers.
> It appears to be a capitalized version of Honeywell notation, but I
> can't yet verify that:
indeed Honeywell notation, so I have now updated the page.
John Savard
Shmuel Metz
May 20, 2013, 3:55:55 AM5/20/13
to
In <byrnsj-4175BC....@news.giganews.com>, on 05/19/2013
PDP-5.
BTW, do they mention the CDC 160?
--
Shmuel (Seymour J.) Metz, SysProg and JOAT <http://patriot.net/~shmuel>
Unsolicited bulk E-mail subject to legal action. I reserve the
right to publicly post or ridicule any abusive E-mail. Reply to
domain Patriot dot net user shmuel+news to contact me. Do not
reply to spam...@library.lspace.org
at 09:12 AM, John Byrns <byr...@sbcglobal.net> said:
>In article
><3ceb5fc4-5472-4283...@a8g2000yqp.googlegroups.com>,
> Quadibloc <jsa...@ecn.ab.ca> wrote:
>> http://www.math.tau.ac.il/~matias/99-SPRING-TALKS/even.html
>>
>> and the brief CV of the author at the end claims that the Elbit 100
>> has the historical significance commonly attributed to the PDP-8!!!
What signifance is that? The PDP-8 is just a new iteration of the
>In article
><3ceb5fc4-5472-4283...@a8g2000yqp.googlegroups.com>,
> Quadibloc <jsa...@ecn.ab.ca> wrote:
>> http://www.math.tau.ac.il/~matias/99-SPRING-TALKS/even.html
>>
>> and the brief CV of the author at the end claims that the Elbit 100
>> has the historical significance commonly attributed to the PDP-8!!!
PDP-5.
BTW, do they mention the CDC 160?
--
Shmuel (Seymour J.) Metz, SysProg and JOAT <http://patriot.net/~shmuel>
Unsolicited bulk E-mail subject to legal action. I reserve the
right to publicly post or ridicule any abusive E-mail. Reply to
domain Patriot dot net user shmuel+news to contact me. Do not
reply to spam...@library.lspace.org
Quadibloc
May 20, 2013, 9:02:53 AM5/20/13
to
On May 20, 1:55 am, Shmuel (Seymour J.) Metz
commercially successful minicomputer, an honor usually bestowed on the
PDP-8.
The thing the PDP-8 had that the PDP-5 didn't was that it was cheaper.
(Well, it had a program counter too.)
John Savard
<spamt...@library.lspace.org.invalid> wrote:
> What signifance is that? The PDP-8 is just a new iteration of the
> PDP-5.
The web site linked actually stated that the Elbit 100 was the first
> What signifance is that? The PDP-8 is just a new iteration of the
> PDP-5.
commercially successful minicomputer, an honor usually bestowed on the
PDP-8.
The thing the PDP-8 had that the PDP-5 didn't was that it was cheaper.
(Well, it had a program counter too.)
John Savard
Quadibloc
May 20, 2013, 9:07:41 AM5/20/13
to
Speaking of the Honeywell 300, I noted in the material that was
available that it was touted as not requiring air conditioning because
of an amazing new feature - lithium cores, which aren't temperature-
sensitive like ordinary cores.
As there don't seem to have been later compatible successors of the
Honeywell 300, I thought that perhaps it was a horrible failure, and
maybe the lithium cores didn't work - because I would have expected to
have heard of something so revolutionary.
It turns out that was not the case.
The HP 9100 calculator used lithium-ferrite cores; but it still needed
to have drive circuits designed to compensate for the remaining
changes in the magnetic characteristics of the cores with temperature.
And lithium-ferrite cores were also used with the Apollo Guidance
Computer.
John Savard
available that it was touted as not requiring air conditioning because
of an amazing new feature - lithium cores, which aren't temperature-
sensitive like ordinary cores.
As there don't seem to have been later compatible successors of the
Honeywell 300, I thought that perhaps it was a horrible failure, and
maybe the lithium cores didn't work - because I would have expected to
have heard of something so revolutionary.
It turns out that was not the case.
The HP 9100 calculator used lithium-ferrite cores; but it still needed
to have drive circuits designed to compensate for the remaining
changes in the magnetic characteristics of the cores with temperature.
And lithium-ferrite cores were also used with the Apollo Guidance
Computer.
John Savard
John Levine
May 20, 2013, 11:06:47 AM5/20/13
to
>The web site linked actually stated that the Elbit 100 was the first
>commercially successful minicomputer, an honor usually bestowed on the
>PDP-8.
I'd say the PDP-1 was adequately successful.
>commercially successful minicomputer, an honor usually bestowed on the
>PDP-8.
The PDP-8 was the first computer priced under $20K, so it certainly
brought computing to a lot of places where it wouldn't have been
affordable before.
--
Regards,
John Levine, jo...@iecc.com, Primary Perpetrator of "The Internet for Dummies",
Please consider the environment before reading this e-mail. http://jl.ly
Quadibloc
May 20, 2013, 11:32:09 AM5/20/13
to
On May 20, 9:06 am, John Levine <jo...@iecc.com> wrote:
> I'd say the PDP-1 was adequately successful.
True enough, but the PDP-4 was considerably more successful. But by
> I'd say the PDP-1 was adequately successful.
some definitions, it was not quite a minicomputer.
John Savard
hanc...@bbs.cpcn.com
May 20, 2013, 12:17:15 PM5/20/13
to
On May 20, 11:06 am, John Levine <jo...@iecc.com> wrote:
> The PDP-8 was the first computer priced under $20K, so it certainly
> brought computing to a lot of places where it wouldn't have been
> affordable before.
Would anyone know how it compared in price to an IBM 1130 configured
> The PDP-8 was the first computer priced under $20K, so it certainly
> brought computing to a lot of places where it wouldn't have been
> affordable before.
to the same level as the PDP-8? The minimum price of an 1130 with
bare bones config?
John Levine
May 20, 2013, 2:31:51 PM5/20/13
to
>> The PDP-8 was the first computer priced under $20K, so it certainly
>> brought computing to a lot of places where it wouldn't have been
>> affordable before.
>
>Would anyone know how it compared in price to an IBM 1130 configured
>to the same level as the PDP-8? The minimum price of an 1130 with
>bare bones config?
It's hard to make a meaningful comparison. A bare bones PDP-8 had 4K
>> brought computing to a lot of places where it wouldn't have been
>> affordable before.
>
>Would anyone know how it compared in price to an IBM 1130 configured
>to the same level as the PDP-8? The minimum price of an 1130 with
>bare bones config?
of core, an ASR-33 Teletype which was both the console and the program
load device for paper tape, and whatever other peripherals you wanted
to add to make it useful.
A bare bones 1130 had the keyboard and console typewriter, but it
wasn't usable without something for program loading beyond the console
switches. IBM sold a paper tape reader, but I never saw one, and all
of the program development was card oriented.
A little googlage says a basic system was $32K, one with a disk was
$41K. A basic PDP-8 with the TTY was $20K and I can report from
experience that it was possible to do program development and other
useful work if you were patient enough to wait for the paper tapes to
chug through the reader.
http://www.ibm1130.net/1130Release.html
hanc...@bbs.cpcn.com
May 20, 2013, 2:48:15 PM5/20/13
to
On May 20, 2:31 pm, John Levine <jo...@iecc.com> wrote:
> A bare bones 1130 had the keyboard and console typewriter, but it
> wasn't usable without something for program loading beyond the console
> switches. IBM sold a paper tape reader, but I never saw one, and all
> of the program development was card oriented.
The console typewriter of an 1130 could be used for program loading
> A bare bones 1130 had the keyboard and console typewriter, but it
> wasn't usable without something for program loading beyond the console
> switches. IBM sold a paper tape reader, but I never saw one, and all
> of the program development was card oriented.
and as an I/O device. However, there needed to be some way to store a
program after it was keyed in.
In a production environment, I would expect the Bell System DEC
machines to have a tape or disk to store the application programs,
instead of loading them up from paper tape every day (too slow, plus
risk of tape breakage from frequent handling.)
Simon Brown
May 20, 2013, 3:23:27 PM5/20/13
to
<hanc...@bbs.cpcn.com> wrote in message
news:314391ff-fa65-4bda...@z10g2000yqd.googlegroups.com...
claims it was priced from $32K.
Presumably the 8S, the serial one, was significantly cheaper than the 1130.
I don't recall that IBM ever did go the serial route with the 1130.
Simon Brown
May 20, 2013, 3:38:16 PM5/20/13
to
<hanc...@bbs.cpcn.com> wrote in message
news:658cc1c5-0dd6-4e12...@o2g2000yqb.googlegroups.com...
> (too slow, plus risk of tape breakage from frequent handling.)
Jon Elson
May 20, 2013, 4:31:46 PM5/20/13
to
Shmuel (Seymour J.) Metz wrote:
> What signifance is that? The PDP-8 is just a new iteration of the
> PDP-5.
Yes, indeed I worked on a PDP-5. It used an older version of DEC
> What signifance is that? The PDP-8 is just a new iteration of the
> PDP-5.
logic modules, about 4 x 6" with an aluminum frame, and Elco Varilock
connectors. Discrete Germanium transistors, I'm pretty sure.
It was also slower than the straight-8, but maybe a bit faster
than an 8S.
Jon
Jon Elson
May 20, 2013, 4:33:25 PM5/20/13
to
Quadibloc wrote:
> The thing the PDP-8 had that the PDP-5 didn't was that it was cheaper.
> (Well, it had a program counter too.)
Umm, how did it work without a program counter. Are you saying
> The thing the PDP-8 had that the PDP-5 didn't was that it was cheaper.
> (Well, it had a program counter too.)
the PC was in core memory? I suppose they could have done that
and saved 12 FFs.
Jon
Quadibloc
May 20, 2013, 5:23:08 PM5/20/13
to
On May 20, 2:33 pm, Jon Elson <jmel...@wustl.edu> wrote:
> Umm, how did it work without a program counter. Are you saying
> the PC was in core memory? I suppose they could have done that
> and saved 12 FFs.
Yes, they did exactly that on the PDP-5 (it was location zero), and of
> Umm, how did it work without a program counter. Are you saying
> the PC was in core memory? I suppose they could have done that
> and saved 12 FFs.
course that meant two core accesses for every instruction.
John Savard
Rod Speed
May 20, 2013, 6:08:41 PM5/20/13
to
"Jon Elson" <jme...@wustl.edu> wrote in message
news:h9-dnWbRcOXeGwfM...@giganews.com...
> Shmuel (Seymour J.) Metz wrote:
>
>
>> What signifance is that? The PDP-8 is just a new iteration of the
>> PDP-5.
> Yes, indeed I worked on a PDP-5. It used an older version of DEC
> logic modules, about 4 x 6" with an aluminum frame, and Elco Varilock
> connectors.
> Discrete Germanium transistors, I'm pretty sure.
Yes, that's what the 8 and 9 used.
>
>
>> What signifance is that? The PDP-8 is just a new iteration of the
>> PDP-5.
> Yes, indeed I worked on a PDP-5. It used an older version of DEC
> logic modules, about 4 x 6" with an aluminum frame, and Elco Varilock
> connectors.
> Discrete Germanium transistors, I'm pretty sure.
Quadibloc
May 20, 2013, 6:10:34 PM5/20/13
to
http://www.computerhistory.org/collections/accession/102725154
in searching for information after seeing an announcement, in the
December 1963 issue of Computers and Automation, of the DSI-1000
computer by Data Systems Incorporated - a 12-bit computer in a compact
box - along with the much larger 12-bit PDP-5 also noted as new that
month!
Somehow, they "had designs for a 16-bit computer" with the same name,
when they were purchased by HP, according to
http://en.wikipedia.org/wiki/HP_2100
which eventually became the HP 2100!!!
Today, a DSI 1000 is a stereo amplifier by Crown, as far as Google is
concerned.
John Savard
Peter Flass
May 20, 2013, 6:57:47 PM5/20/13
to
On 5/19/2013 2:50 PM, Quadibloc wrote:
> On May 19, 10:59 am, Al Kossow <a...@bitsavers.org> wrote:
>
>> I didn't have time to get to the H300 programming manual. I'll try to get to it
>> in the next week or two.
>
> That's wonderful! Thank you very much. I thought that you simply
> didn't have one available.
I think Al once mentioned that he had a whole warehouse of unscanned
> On May 19, 10:59 am, Al Kossow <a...@bitsavers.org> wrote:
>
>> I didn't have time to get to the H300 programming manual. I'll try to get to it
>> in the next week or two.
>
> That's wonderful! Thank you very much. I thought that you simply
> didn't have one available.
manuals.
--
Pete
Peter Flass
May 20, 2013, 7:07:05 PM5/20/13
to
for around $100/mo, IIRC). That includes 1 10MG removable disk, 80lpm
printer, 300cpm card reader, 80 col/sec punch, and a console typewriter.
It had a maximum 32K (16-bit) words of memory, I think 8K was most
common. Wonderful machine.
--
Pete
Peter Flass
May 20, 2013, 7:09:18 PM5/20/13
to
On 5/20/2013 2:48 PM, hanc...@bbs.cpcn.com wrote:
> On May 20, 2:31 pm, John Levine <jo...@iecc.com> wrote:
>
>> A bare bones 1130 had the keyboard and console typewriter, but it
>> wasn't usable without something for program loading beyond the console
>> switches. IBM sold a paper tape reader, but I never saw one, and all
>> of the program development was card oriented.
>
> The console typewriter of an 1130 could be used for program loading
> and as an I/O device. However, there needed to be some way to store a
> program after it was keyed in.
This is not my recollection, but that may be parochial. I know you
> On May 20, 2:31 pm, John Levine <jo...@iecc.com> wrote:
>
>> A bare bones 1130 had the keyboard and console typewriter, but it
>> wasn't usable without something for program loading beyond the console
>> switches. IBM sold a paper tape reader, but I never saw one, and all
>> of the program development was card oriented.
>
> The console typewriter of an 1130 could be used for program loading
> and as an I/O device. However, there needed to be some way to store a
> program after it was keyed in.
could toggle in a program from the console switches.
--
Pete
Quadibloc
May 20, 2013, 9:13:39 PM5/20/13
to
On May 20, 5:09 pm, Peter Flass <Peter_Fl...@Yahoo.com> wrote:
The 1130 usually came with a disk pack with a one-megabyte capacity.
John Savard
John Savard
Quadibloc
May 20, 2013, 9:14:51 PM5/20/13
to
On May 20, 4:57 pm, Peter Flass <Peter_Fl...@Yahoo.com> wrote:
> I think Al once mentioned that he had a whole warehouse of unscanned
> manuals.
Which is why he has something new to scan in every week... ah, well,
> I think Al once mentioned that he had a whole warehouse of unscanned
> manuals.
it gives us something to look forward to.
John Savard
John Levine
May 21, 2013, 12:17:57 AM5/21/13
to
>Presumably the 8S, the serial one, was significantly cheaper than the 1130.
It was about $10K. The S stood for Slow, or perhaps Stupendously Slow.
>I don't recall that IBM ever did go the serial route with the 1130.
memory. They were all pretty slow. The usual model had 3.6 us core
and an add from memory took 8us. The PDP-8 had 1.5us core and an add
was 3us.
Rod Speed
May 21, 2013, 3:02:11 AM5/21/13
to
John Levine <jo...@iecc.com> wrote
>> Presumably the 8S, the serial one, was significantly cheaper than the
>> 1130.
> It was about $10K.
That�s certainly what
http://www.ricomputermuseum.org/Home/equipment/pdp-8s
says. And they have an ad to prove that too.
> The S stood for Slow, or perhaps Stupendously Slow.
I measured fluorescent decay at the sub nS level with it.
>>I don't recall that IBM ever did go the serial route with the 1130.
> The various models of 1130 differed primarily in the speed of the core
> memory.
Yeah, that�s what I meant.
> They were all pretty slow.
Faster than the 8S tho.
> The usual model had 3.6 us core and an add from memory
> took 8us. The PDP-8 had 1.5us core and an add was 3us.
But the 8S was ten times slower than that.
>> Presumably the 8S, the serial one, was significantly cheaper than the
>> 1130.
> It was about $10K.
http://www.ricomputermuseum.org/Home/equipment/pdp-8s
says. And they have an ad to prove that too.
> The S stood for Slow, or perhaps Stupendously Slow.
>>I don't recall that IBM ever did go the serial route with the 1130.
> The various models of 1130 differed primarily in the speed of the core
> memory.
> They were all pretty slow.
> The usual model had 3.6 us core and an add from memory
> took 8us. The PDP-8 had 1.5us core and an add was 3us.
Peter Flass
May 21, 2013, 8:17:02 AM5/21/13
to
--
Pete
Quadibloc
May 21, 2013, 10:51:08 AM5/21/13
to
Charlie Gibbs
May 21, 2013, 12:44:54 PM5/21/13
to
In article
<3f69434b-f14c-47fd...@n11g2000yqe.googlegroups.com>,
The Univac 9300's program counter, instruction register, general
purpose registers, I/O buffer control words, and printer buffer
were all in memory. (Yup, 40% memory utilization while the printer
was figuring out which hammers to fire at which time.) But why not
save the money - typical applications were I/O-bound anyway...
--
/~\ cgi...@kltpzyxm.invalid (Charlie Gibbs)
\ / I'm really at ac.dekanfrus if you read it the right way.
X Top-posted messages will probably be ignored. See RFC1855.
/ \ HTML will DEFINITELY be ignored. Join the ASCII ribbon campaign!
<3f69434b-f14c-47fd...@n11g2000yqe.googlegroups.com>,
purpose registers, I/O buffer control words, and printer buffer
were all in memory. (Yup, 40% memory utilization while the printer
was figuring out which hammers to fire at which time.) But why not
save the money - typical applications were I/O-bound anyway...
--
/~\ cgi...@kltpzyxm.invalid (Charlie Gibbs)
\ / I'm really at ac.dekanfrus if you read it the right way.
X Top-posted messages will probably be ignored. See RFC1855.
/ \ HTML will DEFINITELY be ignored. Join the ASCII ribbon campaign!
Quadibloc
May 21, 2013, 1:06:59 PM5/21/13
to
On May 21, 9:58 am, "Charlie Gibbs" <cgi...@kltpzyxm.invalid> wrote:
> In article
> <3f69434b-f14c-47fd-86b5-a99fe6ff3...@n11g2000yqe.googlegroups.com>,
microcode was in memory as well.
John Savard
> In article
> <3f69434b-f14c-47fd-86b5-a99fe6ff3...@n11g2000yqe.googlegroups.com>,
>
> jsav...@ecn.ab.ca (Quadibloc) writes:
> > On May 20, 2:33 pm, Jon Elson <jmel...@wustl.edu> wrote:
>
> >> Umm, how did it work without a program counter. Are you saying
> >> the PC was in core memory? I suppose they could have done that
> >> and saved 12 FFs.
>
> > Yes, they did exactly that on the PDP-5 (it was location zero),
> > and of course that meant two core accesses for every instruction.
>
> The Univac 9300's program counter, instruction register, general
> purpose registers, I/O buffer control words, and printer buffer
> were all in memory. (Yup, 40% memory utilization while the printer
> was figuring out which hammers to fire at which time.) But why not
> save the money - typical applications were I/O-bound anyway...
Some of the low-end 360s had that characteristic too, and the
> jsav...@ecn.ab.ca (Quadibloc) writes:
> > On May 20, 2:33 pm, Jon Elson <jmel...@wustl.edu> wrote:
>
> >> Umm, how did it work without a program counter. Are you saying
> >> the PC was in core memory? I suppose they could have done that
> >> and saved 12 FFs.
>
> > Yes, they did exactly that on the PDP-5 (it was location zero),
> > and of course that meant two core accesses for every instruction.
>
> The Univac 9300's program counter, instruction register, general
> purpose registers, I/O buffer control words, and printer buffer
> were all in memory. (Yup, 40% memory utilization while the printer
> was figuring out which hammers to fire at which time.) But why not
> save the money - typical applications were I/O-bound anyway...
microcode was in memory as well.
John Savard
Jon Elson
May 21, 2013, 2:52:43 PM5/21/13
to
The rest would be mostly core speed.
Having the PC in memory could lead to some really hard to debug
errors! I sure don't recall this "never write to loc 0" dictum,
but it must have been written somewhere. (It has been a while since
I dealt with the PDP-5, like 40 years.)
Jon
Shmuel Metz
May 21, 2013, 2:51:48 PM5/21/13
to
In <kne97u$75c$3...@dont-email.me>, on 05/20/2013
scans.
--
Shmuel (Seymour J.) Metz, SysProg and JOAT <http://patriot.net/~shmuel>
Unsolicited bulk E-mail subject to legal action. I reserve the
right to publicly post or ridicule any abusive E-mail. Reply to
domain Patriot dot net user shmuel+news to contact me. Do not
reply to spam...@library.lspace.org
at 06:57 PM, Peter Flass <Peter...@Yahoo.com> said:
>I think Al once mentioned that he had a whole warehouse of unscanned
>manuals.
Don't we all? I've got 20 shelves worth that I'd happily trade for OCR
>I think Al once mentioned that he had a whole warehouse of unscanned
>manuals.
scans.
--
Shmuel (Seymour J.) Metz, SysProg and JOAT <http://patriot.net/~shmuel>
Unsolicited bulk E-mail subject to legal action. I reserve the
right to publicly post or ridicule any abusive E-mail. Reply to
domain Patriot dot net user shmuel+news to contact me. Do not
reply to spam...@library.lspace.org
Shmuel Metz
May 21, 2013, 8:42:32 AM5/21/13
to
In <h9-dnWbRcOXeGwfM...@giganews.com>, on 05/20/2013
Wow! I thought that they were dead by the time the PDP-5 came out.
BTW, wasn't the PDP-4 fairly popular a few years earlier?
Wow! I thought that they were dead by the time the PDP-5 came out.
BTW, wasn't the PDP-4 fairly popular a few years earlier?
Shmuel Metz
May 21, 2013, 8:39:50 AM5/21/13
to
In <avvce4...@mid.individual.net>, on 05/21/2013
type of memory. There are applications where a nonvolatile memory is
important, but in normal usage the machine was not shut down very
often.
>You don't have frequent handling with the code for core machines.
What gives you that idea? You need to read in new jobs periodically.
at 05:38 AM, "Simon Brown" <s...@kigfr.com> said:
>> In a production environment, I would expect the Bell System DEC
>> machines to have a tape or disk to store the application programs,
>> instead of loading them up from paper tape every day
>You don't have to do that with core machines.
You have to reload them when the program changes, regardless of the
>> In a production environment, I would expect the Bell System DEC
>> machines to have a tape or disk to store the application programs,
>> instead of loading them up from paper tape every day
>You don't have to do that with core machines.
type of memory. There are applications where a nonvolatile memory is
important, but in normal usage the machine was not shut down very
often.
>You don't have frequent handling with the code for core machines.
Jon Elson
May 21, 2013, 3:47:52 PM5/21/13
to
Shmuel (Seymour J.) Metz wrote:
> In <h9-dnWbRcOXeGwfM...@giganews.com>, on 05/20/2013
> at 03:31 PM, Jon Elson <jme...@wustl.edu> said:
>
>>Discrete Germanium transistors, I'm pretty sure.
>
> Wow! I thought that they were dead by the time the PDP-5 came out.
The PDP-5 was developed in 1963, maybe not sold until 1964.
> at 03:31 PM, Jon Elson <jme...@wustl.edu> said:
>
>>Discrete Germanium transistors, I'm pretty sure.
>
> Wow! I thought that they were dead by the time the PDP-5 came out.
It was built with DEC "system modules" which I think dated back
to the PDP-1 and were used for everything DEC made at the time.
So, until they moved forward to the flip-chip modules, it may
have been partly for historical reasons the -5 used the same
old technology.
I've worked on a number of other pieces of gear from the mid-60's
and they often used all Ge transistors, but maybe has Si diodes.
Jon
Quadibloc
May 21, 2013, 4:29:41 PM5/21/13
to
> > In <h9-dnWbRcOXeGwfMnZ2dnUVZ_rSdn...@giganews.com>, on 05/20/2013
DEC System Modules, and E. Gordon Bell's book Computer Engineering
noted that most logic operations were performed with germanium
transistors at the time of those modules.
Germanium transistors and diodes have a lower threshold voltage than
silicon ones, which is one reason they were preferred in the early
days. Another is simply that the first transistors were made only from
germanium; it wasn't until 1954 that the first silicon transistors
were made, so there was less experience with them.
John Savard
> > at 03:31 PM, Jon Elson <jmel...@wustl.edu> said:
>
> >>Discrete Germanium transistors, I'm pretty sure.
>
> > Wow! I thought that they were dead by the time the PDP-5 came out.
>
> The PDP-5 was developed in 1963, maybe not sold until 1964.
> It was built with DEC "system modules" which I think dated back
> to the PDP-1 and were used for everything DEC made at the time.
> So, until they moved forward to the flip-chip modules, it may
> have been partly for historical reasons the -5 used the same
> old technology.
>
> I've worked on a number of other pieces of gear from the mid-60's
> and they often used all Ge transistors, but maybe has Si diodes.
I checked this out; a web site noted that the PDP-5 was made with the
>
> >>Discrete Germanium transistors, I'm pretty sure.
>
> > Wow! I thought that they were dead by the time the PDP-5 came out.
>
> The PDP-5 was developed in 1963, maybe not sold until 1964.
> It was built with DEC "system modules" which I think dated back
> to the PDP-1 and were used for everything DEC made at the time.
> So, until they moved forward to the flip-chip modules, it may
> have been partly for historical reasons the -5 used the same
> old technology.
>
> I've worked on a number of other pieces of gear from the mid-60's
> and they often used all Ge transistors, but maybe has Si diodes.
DEC System Modules, and E. Gordon Bell's book Computer Engineering
noted that most logic operations were performed with germanium
transistors at the time of those modules.
Germanium transistors and diodes have a lower threshold voltage than
silicon ones, which is one reason they were preferred in the early
days. Another is simply that the first transistors were made only from
germanium; it wasn't until 1954 that the first silicon transistors
were made, so there was less experience with them.
John Savard
Scott Lurndal
May 21, 2013, 4:39:11 PM5/21/13
to
Note: so know I know what "BBN" stood for.
======
THE FIRST "PROGRAMMED DATA PROCESSOR"
DELIVERED AND IN USE
Digital Equipment Corp.
Maynard, Mass.
A new low-to-medium-priced, solid-state,
digital computer designed and manufactured by
this company, which is capable of operating
at twice the speed of most large computers in
use today, has been formally accepted by Bolt,
Beranek & Newman, Inc., Cambridge, Mass.
The computer, called the DEC Programmed
Data Processor (PDP), will be used by BBN on
many advanced,research projects it conducts
for government and for private industry.
Several of these projects involve the use of
the computer as an automatic teaching machine
for instruction in such different subjects as
foreign language vocabulary and Sonar sound
recognition. The computer's high speed will
enable it to teach different subjects simul-
taneously to different people, making use of
time-sharing.
...elided...
Standard equipment on the DEC computer
consists of a paper tape reader, a typewriter
for online input and output operations, and
a paper tape punch. Optional input-output
equipment includes sequence break, cathode
ray tube visual display, a "light pen" (which
can write with light on the cathode ray tube
visual display), card punch and card reader
controls, tape units, and tape control units.
Built-in multiply and divide orders are
available as a central processor option.
The random access core memory is expandable
in modules of 4096 words of 18 binary digits.
The price of a PDP with standard equipment
and one memory bank is approximately $110,000.
It operates in binary; programs are used to
convert from decimal to binary and vice versa.
The computer has a word length of 18
binary digits. Instructions are carried out
in multiples of the memory cycle time of
five microseconds. Add, subtract, deposit,
and load, for example, are two-cycle instruc-
tions requiring 10 microseconds. Multipli-
cation by programmed subroutine requires 350
microseconds on the average, and division
takes about 600.
Program features include single address
instructions, multiple step indirect address-
ing and logical and arithmetical commands.
Among the console features are flip-flop in-
dicators grouped for convenient octal reading,
six program flags for automatic setting and
computer sensing, and six sense switches for
manual setting and computer sensing.
_Computers and Automation_, April 1961
======
THE FIRST "PROGRAMMED DATA PROCESSOR"
DELIVERED AND IN USE
Digital Equipment Corp.
Maynard, Mass.
A new low-to-medium-priced, solid-state,
digital computer designed and manufactured by
this company, which is capable of operating
at twice the speed of most large computers in
use today, has been formally accepted by Bolt,
Beranek & Newman, Inc., Cambridge, Mass.
The computer, called the DEC Programmed
Data Processor (PDP), will be used by BBN on
many advanced,research projects it conducts
for government and for private industry.
Several of these projects involve the use of
the computer as an automatic teaching machine
for instruction in such different subjects as
foreign language vocabulary and Sonar sound
recognition. The computer's high speed will
enable it to teach different subjects simul-
taneously to different people, making use of
time-sharing.
...elided...
Standard equipment on the DEC computer
consists of a paper tape reader, a typewriter
for online input and output operations, and
a paper tape punch. Optional input-output
equipment includes sequence break, cathode
ray tube visual display, a "light pen" (which
can write with light on the cathode ray tube
visual display), card punch and card reader
controls, tape units, and tape control units.
Built-in multiply and divide orders are
available as a central processor option.
The random access core memory is expandable
in modules of 4096 words of 18 binary digits.
The price of a PDP with standard equipment
and one memory bank is approximately $110,000.
It operates in binary; programs are used to
convert from decimal to binary and vice versa.
The computer has a word length of 18
binary digits. Instructions are carried out
in multiples of the memory cycle time of
five microseconds. Add, subtract, deposit,
and load, for example, are two-cycle instruc-
tions requiring 10 microseconds. Multipli-
cation by programmed subroutine requires 350
microseconds on the average, and division
takes about 600.
Program features include single address
instructions, multiple step indirect address-
ing and logical and arithmetical commands.
Among the console features are flip-flop in-
dicators grouped for convenient octal reading,
six program flags for automatic setting and
computer sensing, and six sense switches for
manual setting and computer sensing.
_Computers and Automation_, April 1961
hanc...@bbs.cpcn.com
May 21, 2013, 5:09:59 PM5/21/13
to
On May 20, 7:09 pm, Peter Flass <Peter_Fl...@Yahoo.com> wrote:
> > The console typewriter of an 1130 could be used for program loading
> > and as an I/O device. However, there needed to be some way to store a
> > program after it was keyed in.
>
> This is not my recollection, but that may be parochial. I know you
> could toggle in a program from the console switches.
I asked the 1130 systems programmer about using the console to enter
> > The console typewriter of an 1130 could be used for program loading
> > and as an I/O device. However, there needed to be some way to store a
> > program after it was keyed in.
>
> This is not my recollection, but that may be parochial. I know you
> could toggle in a program from the console switches.
programs, and after a few minutes he was able to set it up (I have no
idea what he did). However, I don't think this was a standard
practice, just something that was possible by reassigning some fields.
If memory serves, the 1130 console keyboard was in the same format as
an 029 keypunch, meaning the punctuation and other special characters
were located in odd places as a shift. Also, it was not an
"interactive terminal", but merely a raw input pipe. In essence, it
was easier to keypunch a program in advance than key it in that way.
I wish I could go back and experiment with a Fortran program that read
data into an array, did some calculations on it, then printed the data
out from the array. This was to see if separating reading and
printing would result in faster performance. The slow I/O used to
drive us crazy.
hanc...@bbs.cpcn.com
May 21, 2013, 5:18:38 PM5/21/13
to
On May 21, 4:39 pm, sc...@slp53.sl.home (Scott Lurndal) wrote:
> Note: so know I know what "BBN" stood for.
> Note: so know I know what "BBN" stood for.
> use today, has been formally accepted by Bolt,
> Beranek & Newman, Inc., Cambridge, Mass.
> Beranek & Newman, Inc., Cambridge, Mass.
> _Computers and Automation_, April 1961
They're still around:
http://en.wikipedia.org/wiki/BBN_Technologies
A variety of electronic and industrial firms are mentioned in ads in
C&A. Some are used as examples of satisfied customers of a computer
vendor; others are vendors or component suppliers.
In 1961 vacuum tube manufacture was still going strong--Corning
developed a tube testing turntable to expedite manufacturing.
Checking the address of a company from back then on google-maps is
interesting. So far, it seems most companies, if they're still in
existence, are no longer at the cited location. One company's 1961 HQ
was an ugly small brick building. Many locations have been bulldozed
and redeveloped with new buildings or highways.
One company, an instrument lamp maker, is still in business, though in
suburban New Jersey instead of Brooklyn.
Lots of help wanted ads.
hanc...@bbs.cpcn.com
May 21, 2013, 5:19:06 PM5/21/13
to
On May 21, 4:39 pm, sc...@slp53.sl.home (Scott Lurndal) wrote:
> Note: so know I know what "BBN" stood for.
> use today, has been formally accepted by Bolt,
> Beranek & Newman, Inc., Cambridge, Mass.
> Beranek & Newman, Inc., Cambridge, Mass.
> _Computers and Automation_, April 1961
hanc...@bbs.cpcn.com
May 21, 2013, 5:19:31 PM5/21/13
to
On May 21, 4:39 pm, sc...@slp53.sl.home (Scott Lurndal) wrote:
> Note: so know I know what "BBN" stood for.
> use today, has been formally accepted by Bolt,
> Beranek & Newman, Inc., Cambridge, Mass.
> Beranek & Newman, Inc., Cambridge, Mass.
> _Computers and Automation_, April 1961
Rod Speed
May 21, 2013, 5:30:11 PM5/21/13
to
"Quadibloc" <jsa...@ecn.ab.ca> wrote in message
news:122b19f9-d6d1-4c94...@wg15g2000pbb.googlegroups.com...
> On May 21, 1:47 pm, Jon Elson <jmel...@wustl.edu> wrote:
>> Shmuel (Seymour J.) Metz wrote:
>>
>> > In <h9-dnWbRcOXeGwfMnZ2dnUVZ_rSdn...@giganews.com>, on 05/20/2013
>> > at 03:31 PM, Jon Elson <jmel...@wustl.edu> said:
>>
>> >>Discrete Germanium transistors, I'm pretty sure.
>>
>> > Wow! I thought that they were dead by the time the PDP-5 came out.
>>
>> The PDP-5 was developed in 1963, maybe not sold until 1964.
>> It was built with DEC "system modules" which I think dated back
>> to the PDP-1 and were used for everything DEC made at the time.
>> So, until they moved forward to the flip-chip modules, it may
>> have been partly for historical reasons the -5 used the same
>> old technology.
>>
>> I've worked on a number of other pieces of gear from the mid-60's
>> and they often used all Ge transistors, but maybe has Si diodes.
>
> I checked this out; a web site noted that the PDP-5 was made with the
> DEC System Modules, and E. Gordon Bell's book Computer Engineering
> noted that most logic operations were performed with germanium
> transistors at the time of those modules.
>
> Germanium transistors and diodes have a lower threshold voltage than
> silicon ones, which is one reason they were preferred in the early days.
That wasn't the reason.
>> Shmuel (Seymour J.) Metz wrote:
>>
>> > In <h9-dnWbRcOXeGwfMnZ2dnUVZ_rSdn...@giganews.com>, on 05/20/2013
>> > at 03:31 PM, Jon Elson <jmel...@wustl.edu> said:
>>
>> >>Discrete Germanium transistors, I'm pretty sure.
>>
>> > Wow! I thought that they were dead by the time the PDP-5 came out.
>>
>> The PDP-5 was developed in 1963, maybe not sold until 1964.
>> It was built with DEC "system modules" which I think dated back
>> to the PDP-1 and were used for everything DEC made at the time.
>> So, until they moved forward to the flip-chip modules, it may
>> have been partly for historical reasons the -5 used the same
>> old technology.
>>
>> I've worked on a number of other pieces of gear from the mid-60's
>> and they often used all Ge transistors, but maybe has Si diodes.
>
> I checked this out; a web site noted that the PDP-5 was made with the
> DEC System Modules, and E. Gordon Bell's book Computer Engineering
> noted that most logic operations were performed with germanium
> transistors at the time of those modules.
>
> Germanium transistors and diodes have a lower threshold voltage than
> silicon ones, which is one reason they were preferred in the early days.
> Another is simply that the first transistors were made only from
> germanium; it wasn't until 1954 that the first silicon transistors
> were made, so there was less experience with them.
Peter Flass
May 21, 2013, 8:02:42 PM5/21/13
to
--
Pete
Peter Flass
May 21, 2013, 8:04:25 PM5/21/13
to
though, couldn't they?
--
Pete
Rich Alderson
May 21, 2013, 8:04:22 PM5/21/13
to nobody
Shmuel (Seymour J.) Metz <spam...@library.lspace.org.invalid> writes:
> In <h9-dnWbRcOXeGwfM...@giganews.com>, on 05/20/2013
> at 03:31 PM, Jon Elson <jme...@wustl.edu> said:
>> Discrete Germanium transistors, I'm pretty sure.
> Wow! I thought that they were dead by the time the PDP-5 came out.
> BTW, wasn't the PDP-4 fairly popular a few years earlier?
The PDP-5 was developed almost simultaneously with the PDP-4 (NB: There was
> In <h9-dnWbRcOXeGwfM...@giganews.com>, on 05/20/2013
> at 03:31 PM, Jon Elson <jme...@wustl.edu> said:
>> Discrete Germanium transistors, I'm pretty sure.
> Wow! I thought that they were dead by the time the PDP-5 came out.
> BTW, wasn't the PDP-4 fairly popular a few years earlier?
no PDP-2 or PDP-3, though designs for both were created), for a particular
customer who needed a small system to act as data collection points for the
central PDP-4. (Reported in Bell/Mudge/McNamara, _Computer Engineering_)
The PDP-4 was actually *not* particularly popular. It was built as a follow-
on to the PDP-1 for people looking for a cheaper machine. DEC learned from it
that 2/3 the price for 1/2 the speed was not the way to build share in the
computer marketplace. (Or it might have been 1/2 the price at 2/3 the speed.
I don't have my copy of the cited work handy.) People want cheaper *and*
faster, not *or*.
--
Rich Alderson ne...@alderson.users.panix.com
the russet leaves of an autumn oak/inspire once again the failed poet/
to take up his pen/and essay to place his meagre words upon the page...
Joe Pfeiffer
May 21, 2013, 8:48:34 PM5/21/13
to
use silicon transistors. While Thornton's book points out the
advantages of silicon in several places, I'm not able to find an actual
confirmation at the moment. I do note that
http://www.computerhistory.org/semiconductor/timeline.html says in 1961
"Computer architect Seymour Cray funds development of the first silicon
device to meet the performance demands of the world’s fastest machine."
Walter Bushell
May 21, 2013, 9:21:01 PM5/21/13
to
In article <udadnVxxgOERXQbM...@giganews.com>,
Hey, you didn't need a separate unconditional jump with PC at location
0.
--
Gambling with Other People's Money is the meth of the fiscal industry.
me -- in the spirit of Karl and Groucho Marx
0.
--
Gambling with Other People's Money is the meth of the fiscal industry.
me -- in the spirit of Karl and Groucho Marx
Charlie Gibbs
May 21, 2013, 11:20:49 PM5/21/13
to
In article <519bc244$2$fuzhry+tra$mr2...@news.patriot.net>,
Me too.
</aol>
It's required for "gold" membership in a.f.c, isn't it?
spam...@library.lspace.org.invalid (Seymour J.) writes:
> In <kne97u$75c$3...@dont-email.me>, on 05/20/2013
> at 06:57 PM, Peter Flass <Peter...@Yahoo.com> said:
>
>> I think Al once mentioned that he had a whole warehouse of unscanned
>> manuals.
>
> Don't we all? I've got 20 shelves worth that I'd happily trade for OCR
> scans.
<aol>
> In <kne97u$75c$3...@dont-email.me>, on 05/20/2013
> at 06:57 PM, Peter Flass <Peter...@Yahoo.com> said:
>
>> I think Al once mentioned that he had a whole warehouse of unscanned
>> manuals.
>
> Don't we all? I've got 20 shelves worth that I'd happily trade for OCR
> scans.
Me too.
</aol>
It's required for "gold" membership in a.f.c, isn't it?
Charlie Gibbs
May 21, 2013, 11:22:23 PM5/21/13
to
In article <edCdndkbGarnUAbM...@giganews.com>,
You know when you're working when old gear when you come across
selenium rectifiers.
selenium rectifiers.
Charlie Gibbs
May 21, 2013, 11:25:26 PM5/21/13
to
In article <mdd61yb...@panix5.panix.com>,
crippled hardware being sold for less, and the various hacks
people did to make said hardware run at full speed.
ne...@alderson.users.panix.com (Rich Alderson) writes:
> The PDP-4 was actually *not* particularly popular. It was built as a
> follow-on to the PDP-1 for people looking for a cheaper machine. DEC
> The PDP-4 was actually *not* particularly popular. It was built as a
> learned from it that 2/3 the price for 1/2 the speed was not the way
> to build share in the computer marketplace. (Or it might have been
> 1/2 the price at 2/3 the speed. I don't have my copy of the cited
> work handy.) People want cheaper *and* faster, not *or*.
Not necessarily. There have been several threads here in about
> to build share in the computer marketplace. (Or it might have been
> 1/2 the price at 2/3 the speed. I don't have my copy of the cited
> work handy.) People want cheaper *and* faster, not *or*.
crippled hardware being sold for less, and the various hacks
people did to make said hardware run at full speed.
Quadibloc
May 21, 2013, 10:53:08 PM5/21/13
to
On May 20, 9:06 am, John Levine <jo...@iecc.com> wrote:
> I'd say the PDP-1 was adequately successful.
DEC sold _fifty_ PDP-1 computers, and _40,000_ PDP-8 computers, as I
found out while perusing Computer Engineering once again.
The PDP-1 was profitable for DEC, and thus allowed the company to
continue making computers, so a term like "commercially successful"
indeed might apply to it too; the PDP-8, though, was the first
minicomputer to have a commercial success of significance.
And the PDP-1 was not all that "mini", as I've already noted.
John Savard
> I'd say the PDP-1 was adequately successful.
DEC sold _fifty_ PDP-1 computers, and _40,000_ PDP-8 computers, as I
found out while perusing Computer Engineering once again.
The PDP-1 was profitable for DEC, and thus allowed the company to
continue making computers, so a term like "commercially successful"
indeed might apply to it too; the PDP-8, though, was the first
minicomputer to have a commercial success of significance.
And the PDP-1 was not all that "mini", as I've already noted.
John Savard
Quadibloc
May 22, 2013, 12:51:46 AM5/22/13
to
On May 21, 8:25 pm, "Charlie Gibbs" <cgi...@kltpzyxm.invalid> wrote:
> In article <mdd61ybkhh5....@panix5.panix.com>,
sense of buying a small computer instead of a bigger one. But his
point is valid enough too when it comes to successor systems - thus,
in Computer Engineering, Bell gives precisely this as the reason for
the lukewarm reception of the PDP-7.
John Savard
> In article <mdd61ybkhh5....@panix5.panix.com>,
> n...@alderson.users.panix.com (Rich Alderson) writes:
>
> > People want cheaper *and* faster, not *or*.
>
> Not necessarily. There have been several threads here in about
> crippled hardware being sold for less, and the various hacks
> people did to make said hardware run at full speed.
Of course people will buy cheaper even if it isn't faster, in the
>
> > People want cheaper *and* faster, not *or*.
>
> Not necessarily. There have been several threads here in about
> crippled hardware being sold for less, and the various hacks
> people did to make said hardware run at full speed.
sense of buying a small computer instead of a bigger one. But his
point is valid enough too when it comes to successor systems - thus,
in Computer Engineering, Bell gives precisely this as the reason for
the lukewarm reception of the PDP-7.
John Savard
Howard S Shubs
May 22, 2013, 12:58:44 AM5/22/13
to
In article <721.924T24...@kltpzyxm.invalid>,
"Charlie Gibbs" <cgi...@kltpzyxm.invalid> wrote:
> In article <519bc244$2$fuzhry+tra$mr2...@news.patriot.net>,
> spam...@library.lspace.org.invalid (Seymour J.) writes:
>
> > Don't we all? I've got 20 shelves worth that I'd happily trade for OCR
> > scans.
>
> <aol>
> Me too.
> </aol>
>
> It's required for "gold" membership in a.f.c, isn't it?
Some of us did it quite a while ago. :-)
> In article <519bc244$2$fuzhry+tra$mr2...@news.patriot.net>,
> spam...@library.lspace.org.invalid (Seymour J.) writes:
>
> > Don't we all? I've got 20 shelves worth that I'd happily trade for OCR
> > scans.
>
> <aol>
> Me too.
> </aol>
>
> It's required for "gold" membership in a.f.c, isn't it?
Quadibloc
May 22, 2013, 9:04:11 AM5/22/13
to
On May 21, 10:51 pm, Quadibloc <jsav...@ecn.ab.ca> wrote:
> Of course people will buy cheaper even if it isn't faster, in the
> sense of buying a small computer instead of a bigger one. But his
> point is valid enough too when it comes to successor systems - thus,
> in Computer Engineering, Bell gives precisely this as the reason for
> the lukewarm reception of the PDP-7.
Actually, the PDP-4, on page 147.
> Of course people will buy cheaper even if it isn't faster, in the
> sense of buying a small computer instead of a bigger one. But his
> point is valid enough too when it comes to successor systems - thus,
> in Computer Engineering, Bell gives precisely this as the reason for
> the lukewarm reception of the PDP-7.
John Savard
hanc...@bbs.cpcn.com
May 22, 2013, 12:45:31 PM5/22/13
to
On May 21, 10:53 pm, Quadibloc <jsav...@ecn.ab.ca> wrote:
> DEC sold _fifty_ PDP-1 computers, and _40,000_ PDP-8 computers, as I
> found out while perusing Computer Engineering once again.
>
> The PDP-1 was profitable for DEC, and thus allowed the company to
> continue making computers, so a term like "commercially successful"
> indeed might apply to it too; the PDP-8, though, was the first
> minicomputer to have a commercial success of significance.
In 1959, computers were still new enough that fifty is a pretty good
> DEC sold _fifty_ PDP-1 computers, and _40,000_ PDP-8 computers, as I
> found out while perusing Computer Engineering once again.
>
> The PDP-1 was profitable for DEC, and thus allowed the company to
> continue making computers, so a term like "commercially successful"
> indeed might apply to it too; the PDP-8, though, was the first
> minicomputer to have a commercial success of significance.
order, especially for a very young company.
John Levine
May 22, 2013, 1:07:02 PM5/22/13
to
>> Don't we all? I've got 20 shelves worth that I'd happily trade for OCR
>> scans. ...
>It's required for "gold" membership in a.f.c, isn't it?
I did find my MUMPS manuals from 1976. It was originally written for
a PDP-9. Does that count?
And does anyone else remember the Lockheed MAC-16? It was an adequate 16 bit
mini from 1969.
--
Regards,
John Levine, jo...@iecc.com, Primary Perpetrator of "The Internet for Dummies",
Please consider the environment before reading this e-mail. http://jl.ly
Shmuel Metz
May 22, 2013, 1:04:07 PM5/22/13
to
In <721.924T24...@kltpzyxm.invalid>, on 05/21/2013
off our hands in exchange for OCR scans?
at 07:20 PM, "Charlie Gibbs" <cgi...@kltpzyxm.invalid> said:
>In article <519bc244$2$fuzhry+tra$mr2...@news.patriot.net>,
>spam...@library.lspace.org.invalid (Seymour J.) writes:
>> In <kne97u$75c$3...@dont-email.me>, on 05/20/2013
>> at 06:57 PM, Peter Flass <Peter...@Yahoo.com> said:
>>
>>> I think Al once mentioned that he had a whole warehouse of unscanned
>>> manuals.
>>
>> Don't we all? I've got 20 shelves worth that I'd happily trade for OCR
>> scans.
><aol>
>Me too.
></aol>
>It's required for "gold" membership in a.f.c, isn't it?
Do we lose our gold memberships if someone offers to take the manuals
>In article <519bc244$2$fuzhry+tra$mr2...@news.patriot.net>,
>spam...@library.lspace.org.invalid (Seymour J.) writes:
>> In <kne97u$75c$3...@dont-email.me>, on 05/20/2013
>> at 06:57 PM, Peter Flass <Peter...@Yahoo.com> said:
>>
>>> I think Al once mentioned that he had a whole warehouse of unscanned
>>> manuals.
>>
>> Don't we all? I've got 20 shelves worth that I'd happily trade for OCR
>> scans.
><aol>
>Me too.
></aol>
>It's required for "gold" membership in a.f.c, isn't it?
off our hands in exchange for OCR scans?
Jon Elson
May 22, 2013, 3:07:35 PM5/22/13
to
Peter Flass wrote:
> They could have gotten by without an unconditional branch instruction
> though, couldn't they?
>
>
Seems like it. Except this would require the branch target to
> They could have gotten by without an unconditional branch instruction
> though, couldn't they?
>
>
be in the accumulator.
Jon
Jon Elson
May 22, 2013, 3:13:28 PM5/22/13
to
Quadibloc wrote:
> I checked this out; a web site noted that the PDP-5 was made with the
> DEC System Modules, and E. Gordon Bell's book Computer Engineering
> noted that most logic operations were performed with germanium
> transistors at the time of those modules.
>
> Germanium transistors and diodes have a lower threshold voltage than
> silicon ones, which is one reason they were preferred in the early
> days. Another is simply that the first transistors were made only from
> germanium; it wasn't until 1954 that the first silicon transistors
> were made, so there was less experience with them.
Ge transistors were also faster than Si, but were VERY temperature
> I checked this out; a web site noted that the PDP-5 was made with the
> DEC System Modules, and E. Gordon Bell's book Computer Engineering
> noted that most logic operations were performed with germanium
> transistors at the time of those modules.
>
> Germanium transistors and diodes have a lower threshold voltage than
> silicon ones, which is one reason they were preferred in the early
> days. Another is simply that the first transistors were made only from
> germanium; it wasn't until 1954 that the first silicon transistors
> were made, so there was less experience with them.
sensitive, it was REAL hard to make circuits that would run at both room
temp and at even 50C or so. I saw another doc that said DEC
shipped the first -5 in 63, and another place that said the first
epitaxial Si transistors were made especially for CDC and were delivered
in 1964. Now, those may have been special high-speed devices, and
more modest performance units may have been available commercially
before then. I'm guessing IBM had Si transistors for the 360
before then, at least in development.
The earlier point-contact transistors were great to study in the lab,
but were not something that could be commercially manufactured in the
quantities needed for a computer.
Jon
hanc...@bbs.cpcn.com
May 22, 2013, 3:09:02 PM5/22/13
to
On May 22, 3:13 pm, Jon Elson <jmel...@wustl.edu> wrote:
> The earlier point-contact transistors were great to study in the lab,
> but were not something that could be commercially manufactured in the
> quantities needed for a computer.
"IBM's Early Computers" goes into a great detail about their research
> The earlier point-contact transistors were great to study in the lab,
> but were not something that could be commercially manufactured in the
> quantities needed for a computer.
into the properties and utilization of various transistors through the
1950s.
The S/360 book also has detail on the early component chips.
Nick Spalding
May 22, 2013, 4:23:59 PM5/22/13
to
Jon Elson wrote, in <guydne6UZIN1iwDM...@giganews.com>
on Wed, 22 May 2013 14:13:28 -0500:
>
> The earlier point-contact transistors were great to study in the lab,
> but were not something that could be commercially manufactured in the
> quantities needed for a computer.
Back in the late '50s I went to some evening lectures (at Acton Tech
IIRC) on these new-fangled transistors. One of the lecturers had
worked at Phillips on point contact transistors. He described how a
semi-automatic machine switched between discharging capacitors
through the junctions and testing the resultant device.
"You go on doing this until you find you have a transistor, then you
sell it - quick"
--
Nick Spalding
on Wed, 22 May 2013 14:13:28 -0500:
>
> The earlier point-contact transistors were great to study in the lab,
> but were not something that could be commercially manufactured in the
> quantities needed for a computer.
IIRC) on these new-fangled transistors. One of the lecturers had
worked at Phillips on point contact transistors. He described how a
semi-automatic machine switched between discharging capacitors
through the junctions and testing the resultant device.
"You go on doing this until you find you have a transistor, then you
sell it - quick"
--
Nick Spalding
Charles Richmond
May 22, 2013, 4:45:35 PM5/22/13
to
"Charlie Gibbs" <cgi...@kltpzyxm.invalid> wrote in message
news:1196.924T11...@kltpzyxm.invalid...
conditioning cooling towers...
--
numerist at aquaporin4 dot com
news:1196.924T11...@kltpzyxm.invalid...
> In article <edCdndkbGarnUAbM...@giganews.com>,
> jme...@wustl.edu (Jon Elson) writes:
>
>> Shmuel (Seymour J.) Metz wrote:
>>
>>> In <h9-dnWbRcOXeGwfM...@giganews.com>, on 05/20/2013
>>> at 03:31 PM, Jon Elson <jme...@wustl.edu> said:
>>>
>>>>Discrete Germanium transistors, I'm pretty sure.
>>>
>>> Wow! I thought that they were dead by the time the PDP-5 came out.
>>
>> The PDP-5 was developed in 1963, maybe not sold until 1964.
>> It was built with DEC "system modules" which I think dated back
>> to the PDP-1 and were used for everything DEC made at the time.
>> So, until they moved forward to the flip-chip modules, it may
>> have been partly for historical reasons the -5 used the same
>> old technology.
>>
>> I've worked on a number of other pieces of gear from the mid-60's
>> and they often used all Ge transistors, but maybe has Si diodes.
>
> You know when you're working when old gear when you come across
> selenium rectifiers.
>
I remember those selenium rectifiers!!!! They look like little air
> jme...@wustl.edu (Jon Elson) writes:
>
>> Shmuel (Seymour J.) Metz wrote:
>>
>>> In <h9-dnWbRcOXeGwfM...@giganews.com>, on 05/20/2013
>>> at 03:31 PM, Jon Elson <jme...@wustl.edu> said:
>>>
>>>>Discrete Germanium transistors, I'm pretty sure.
>>>
>>> Wow! I thought that they were dead by the time the PDP-5 came out.
>>
>> The PDP-5 was developed in 1963, maybe not sold until 1964.
>> It was built with DEC "system modules" which I think dated back
>> to the PDP-1 and were used for everything DEC made at the time.
>> So, until they moved forward to the flip-chip modules, it may
>> have been partly for historical reasons the -5 used the same
>> old technology.
>>
>> I've worked on a number of other pieces of gear from the mid-60's
>> and they often used all Ge transistors, but maybe has Si diodes.
>
> You know when you're working when old gear when you come across
> selenium rectifiers.
>
conditioning cooling towers...
--
numerist at aquaporin4 dot com
Nick Spalding
May 22, 2013, 4:56:15 PM5/22/13
to
Charles Richmond wrote, in <knjahs$ldo$1...@dont-email.me>
on Wed, 22 May 2013 15:45:35 -0500:
> "Charlie Gibbs" <cgi...@kltpzyxm.invalid> wrote in message
> news:1196.924T11...@kltpzyxm.invalid...
--
Nick Spalding
on Wed, 22 May 2013 15:45:35 -0500:
> "Charlie Gibbs" <cgi...@kltpzyxm.invalid> wrote in message
> news:1196.924T11...@kltpzyxm.invalid...
> > You know when you're working when old gear when you come across
> > selenium rectifiers.
> >
>
> I remember those selenium rectifiers!!!! They look like little air
> conditioning cooling towers...
And had a very characteristic smell when they failed.
> > selenium rectifiers.
> >
>
> I remember those selenium rectifiers!!!! They look like little air
> conditioning cooling towers...
--
Nick Spalding
Rich Alderson
May 22, 2013, 8:01:09 PM5/22/13
to
Quadibloc <jsa...@ecn.ab.ca> writes:
I'm glad you made that correction. The PDP-7 was successful (120 sold); IIRC
only about 40 PDP-4's went out the door.
only about 40 PDP-4's went out the door.
Rich Alderson
May 22, 2013, 8:05:23 PM5/22/13
to
"Charlie Gibbs" <cgi...@kltpzyxm.invalid> writes:
> In article <mdd61yb...@panix5.panix.com>,
> ne...@alderson.users.panix.com (Rich Alderson) writes:
>> The PDP-4 was actually *not* particularly popular. It was built as a
>> follow-on to the PDP-1 for people looking for a cheaper machine. DEC
>> learned from it that 2/3 the price for 1/2 the speed was not the way
>> to build share in the computer marketplace. (Or it might have been
>> 1/2 the price at 2/3 the speed. I don't have my copy of the cited
>> work handy.) People want cheaper *and* faster, not *or*.
> Not necessarily. There have been several threads here in about
> crippled hardware being sold for less, and the various hacks
> people did to make said hardware run at full speed.
Yes, but the PDP-4 was not code-compatible with the PDP-1. It was not simply
> In article <mdd61yb...@panix5.panix.com>,
> ne...@alderson.users.panix.com (Rich Alderson) writes:
>> The PDP-4 was actually *not* particularly popular. It was built as a
>> follow-on to the PDP-1 for people looking for a cheaper machine. DEC
>> learned from it that 2/3 the price for 1/2 the speed was not the way
>> to build share in the computer marketplace. (Or it might have been
>> 1/2 the price at 2/3 the speed. I don't have my copy of the cited
>> work handy.) People want cheaper *and* faster, not *or*.
> Not necessarily. There have been several threads here in about
> crippled hardware being sold for less, and the various hacks
> people did to make said hardware run at full speed.
a crippleware -1.
(12 bits of address, 6 bits of instruction in the PDP-1 vs. 13 bits of address
and 5 bits of instruction in the PDP-4/7/9/15, with bank switching in the 9/15
and paging in the 15, but the same basic instruction set.)
Joe Morris
May 22, 2013, 8:23:52 PM5/22/13
to
"Quadibloc" <jsa...@ecn.ab.ca> wrote:
>On May 20, 9:06 am, John Levine <jo...@iecc.com> wrote:
>> I'd say the PDP-1 was adequately successful.
>DEC sold _fifty_ PDP-1 computers, and _40,000_ PDP-8 computers, as I
>found out while perusing Computer Engineering once again.
>The PDP-1 was profitable for DEC,
H'mmm...the usual figure I've heard for the estimated price of a PDP-1 was
>On May 20, 9:06 am, John Levine <jo...@iecc.com> wrote:
>> I'd say the PDP-1 was adequately successful.
>DEC sold _fifty_ PDP-1 computers, and _40,000_ PDP-8 computers, as I
>found out while perusing Computer Engineering once again.
>The PDP-1 was profitable for DEC,
~$100K (in 1962 dollars), so fifty sold would be "only" $5 million gross
revenue. Serious question: does your source say that the return on sales of
the PDP1 by iself fully recovered the development costs?
Of course, it's not unreasonable to say that the acceptance by customers of
the PDP-1 convinced the Powers That Be in DEC to continue developing new
systems, and that the revenue from selling those later designs would not
have existed had the PDP-1 not been developed.
> and thus allowed the company to
>continue making computers, so a term like "commercially successful"
>indeed might apply to it too; the PDP-8, though, was the first
>minicomputer to have a commercial success of significance.
>And the PDP-1 was not all that "mini", as I've already noted.
unlike the big mainframes such as the 7090 programmers actually operated it.
And it was lots of fun to use, especially when we got the first drum from
DEC: IIRC it had 32 tracks of 4K words. (First drum-aware application I
recall seeing: Spacewar, of course, with a hook for DDT so you could change
the so-called "constants" on the fly.)
Joe
Walter Bushell
May 22, 2013, 10:58:09 PM5/22/13
to
In article <knjni...@news4.newsguy.com>,
"Joe Morris" <j.c.m...@verizon.net> wrote:
> Well...it was "mini" by comparison to many other systems of its day. And
> unlike the big mainframes such as the 7090 programmers actually operated it.
> And it was lots of fun to use, especially when we got the first drum from
> DEC: IIRC it had 32 tracks of 4K words.
Ah, as much memory on drum as the first Macintosh had in RAM.
> Well...it was "mini" by comparison to many other systems of its day. And
> unlike the big mainframes such as the 7090 programmers actually operated it.
> And it was lots of fun to use, especially when we got the first drum from
> DEC: IIRC it had 32 tracks of 4K words.
Charlie Gibbs
May 23, 2013, 1:51:16 AM5/23/13
to
In article <519cfa87$38$fuzhry+tra$mr2...@news.patriot.net>,
spam...@library.lspace.org.invalid (Seymour J.) writes:
> In <721.924T24...@kltpzyxm.invalid>, on 05/21/2013
> at 07:20 PM, "Charlie Gibbs" <cgi...@kltpzyxm.invalid> said:
>
>> In article <519bc244$2$fuzhry+tra$mr2...@news.patriot.net>,
>> spam...@library.lspace.org.invalid (Seymour J.) writes:
>
>>> In <kne97u$75c$3...@dont-email.me>, on 05/20/2013
>>> at 06:57 PM, Peter Flass <Peter...@Yahoo.com> said:
>>>
>>>> I think Al once mentioned that he had a whole warehouse of unscanned
>>>> manuals.
>>>
>>> Don't we all? I've got 20 shelves worth that I'd happily trade for
>>> OCR scans.
>>
>> <aol>
>> Me too.
>> </aol>
>>
>> It's required for "gold" membership in a.f.c, isn't it?
>
> Do we lose our gold memberships if someone offers to take the manuals
> off our hands in exchange for OCR scans?
Put the scans on Bitsavers and you qualify for "emeritus" status.
> In <721.924T24...@kltpzyxm.invalid>, on 05/21/2013
> at 07:20 PM, "Charlie Gibbs" <cgi...@kltpzyxm.invalid> said:
>
>> In article <519bc244$2$fuzhry+tra$mr2...@news.patriot.net>,
>> spam...@library.lspace.org.invalid (Seymour J.) writes:
>
>>> In <kne97u$75c$3...@dont-email.me>, on 05/20/2013
>>> at 06:57 PM, Peter Flass <Peter...@Yahoo.com> said:
>>>
>>>> I think Al once mentioned that he had a whole warehouse of unscanned
>>>> manuals.
>>>
>>> Don't we all? I've got 20 shelves worth that I'd happily trade for
>>> OCR scans.
>>
>> <aol>
>> Me too.
>> </aol>
>>
>> It's required for "gold" membership in a.f.c, isn't it?
>
> Do we lose our gold memberships if someone offers to take the manuals
> off our hands in exchange for OCR scans?
Al Kossow
May 23, 2013, 9:32:52 AM5/23/13
to
On 5/22/13 10:51 PM, Charlie Gibbs wrote:
> Put the scans on Bitsavers and you qualify for "emeritus" status.
>
That would be helpful. Space is still tight and there is still a big backlog of paper to scan.
> Put the scans on Bitsavers and you qualify for "emeritus" status.
>
I can batch process TIFF->PDF/OCR and have just brought a faster machine on line to do that so
sending just the 400dpi bitonal G4 TIFFs is fine.
hanc...@bbs.cpcn.com
May 23, 2013, 10:28:22 AM5/23/13
to
Shmuel Metz
May 23, 2013, 12:18:25 PM5/23/13
to
In <1189.925T8...@kltpzyxm.invalid>, on 05/22/2013
go to bitsavers as well as to me. Unfortunately, bitsavers has a long
queue and inadequate resources.
at 09:51 PM, "Charlie Gibbs" <cgi...@kltpzyxm.invalid> said:
>Put the scans on Bitsavers and you qualify for "emeritus" status.
One of my conditions for giving away my manuals is that the OCR scans
>Put the scans on Bitsavers and you qualify for "emeritus" status.
go to bitsavers as well as to me. Unfortunately, bitsavers has a long
queue and inadequate resources.
Charlie Gibbs
May 23, 2013, 1:31:35 PM5/23/13
to
In article
<bc0a6e91-7baf-4d0f...@i4g2000vbl.googlegroups.com>,
Is there some sort of howto for people considering doing this?
<bc0a6e91-7baf-4d0f...@i4g2000vbl.googlegroups.com>,
Scott Lurndal
May 23, 2013, 1:45:33 PM5/23/13
to
Shmuel (Seymour J.) Metz <spam...@library.lspace.org.invalid> writes:
>In <1189.925T8...@kltpzyxm.invalid>, on 05/22/2013
> at 09:51 PM, "Charlie Gibbs" <cgi...@kltpzyxm.invalid> said:
>
>>Put the scans on Bitsavers and you qualify for "emeritus" status.
>
>One of my conditions for giving away my manuals is that the OCR scans
>go to bitsavers as well as to me. Unfortunately, bitsavers has a long
>queue and inadequate resources.
>
Maybe we could start a kickstarter for Al to hire some interns or something
>In <1189.925T8...@kltpzyxm.invalid>, on 05/22/2013
> at 09:51 PM, "Charlie Gibbs" <cgi...@kltpzyxm.invalid> said:
>
>>Put the scans on Bitsavers and you qualify for "emeritus" status.
>
>One of my conditions for giving away my manuals is that the OCR scans
>go to bitsavers as well as to me. Unfortunately, bitsavers has a long
>queue and inadequate resources.
>
to help speed things along?
scott
Al Kossow
May 23, 2013, 1:55:56 PM5/23/13
to
On 5/23/13 10:45 AM, Scott Lurndal wrote:
> Maybe we could start a kickstarter for Al to hire some interns or something
> to help speed things along?
>
It would probably be better to get something going on the east coast to keep from
> Maybe we could start a kickstarter for Al to hire some interns or something
> to help speed things along?
>
having to ship paper across the country. I saw Jason Scott at a conference we both
attended, and he's out here at the IA for a couple weeks, maybe I should see if he's
got anyone that wants to take on scanning. I've got a second Panasonic scanner that
I could send out there.
It isn't really practical for me to commit to anyone locally. My days are 7AM to 9PM
at the Museum, and we are really tight on work space.
Dan Espen
May 23, 2013, 2:47:27 PM5/23/13
to
they'd be interested. No coding involved.
--
Dan Espen
Michael Black
May 23, 2013, 6:41:24 PM5/23/13
to
On Thu, 23 May 2013, Dan Espen wrote:
>> It isn't really practical for me to commit to anyone locally. My days are 7AM to 9PM
>> at the Museum, and we are really tight on work space.
>
> Google does run those Google Summer of Code things, but I'm not sure
> they'd be interested. No coding involved.
>
How will the interns be able to program the PDP-8 if they don't have
>> It isn't really practical for me to commit to anyone locally. My days are 7AM to 9PM
>> at the Museum, and we are really tight on work space.
>
> Google does run those Google Summer of Code things, but I'm not sure
> they'd be interested. No coding involved.
>
documentation?
Michael
Quadibloc
May 23, 2013, 11:50:47 PM5/23/13
to
On May 23, 4:41 pm, Michael Black <et...@ncf.ca> wrote:
> How will the interns be able to program the PDP-8 if they don't have
> documentation?
There's already quite a bit of PDP-8 documentation on the site.
> How will the interns be able to program the PDP-8 if they don't have
> documentation?
John Savard
John Levine
May 24, 2013, 12:18:23 AM5/24/13
to
>How will the interns be able to program the PDP-8 if they don't have
>documentation?
If they can't figure out how to program a PDP-8, they're not very good
>documentation?
interns.
I haven't programmed a PDP-8 for 40 years, but I think I still
remember enough to toggle in programs from the switches.
Ahem A Rivet's Shot
May 24, 2013, 3:26:32 AM5/24/13
to
On Tue, 21 May 2013 13:52:43 -0500
Jon Elson <jme...@wustl.edu> wrote:
> Having the PC in memory could lead to some really hard to debug
> errors! I sure don't recall this "never write to loc 0" dictum,
It's just as well C wasn't around then.
--
Steve O'Hara-Smith | Directable Mirror Arrays
C:>WIN | A better way to focus the sun
The computer obeys and wins. | licences available see
You lose and Bill collects. | http://www.sohara.org/
Jon Elson <jme...@wustl.edu> wrote:
> Having the PC in memory could lead to some really hard to debug
> errors! I sure don't recall this "never write to loc 0" dictum,
It's just as well C wasn't around then.
--
Steve O'Hara-Smith | Directable Mirror Arrays
C:>WIN | A better way to focus the sun
The computer obeys and wins. | licences available see
You lose and Bill collects. | http://www.sohara.org/
Shmuel Metz
May 24, 2013, 10:27:24 AM5/24/13
to
In <knmpmf$4ci$1...@leila.iecc.com>, on 05/24/2013
not very good. OTOH, if they couldn't find the manuals in bitsavers[1]
then I might be a bit disappointed.
>I haven't programmed a PDP-8 for 40 years, but I think I still
>remember enough to toggle in programs from the switches.
That's not figuring it out. Try figuring out how to program a machine
whose instruction set you've never encountered, without recourse to
documentation.
[1] Yes, they are there.
at 04:18 AM, John Levine <jo...@iecc.com> said:
>If they can't figure out how to program a PDP-8, they're not very
>good interns.
ITYM that if they can't figure it out from the manuals then they're
>If they can't figure out how to program a PDP-8, they're not very
>good interns.
not very good. OTOH, if they couldn't find the manuals in bitsavers[1]
then I might be a bit disappointed.
>I haven't programmed a PDP-8 for 40 years, but I think I still
>remember enough to toggle in programs from the switches.
whose instruction set you've never encountered, without recourse to
documentation.
[1] Yes, they are there.
Jon Elson
May 24, 2013, 3:14:09 PM5/24/13
to
Ahem A Rivet's Shot wrote:
> On Tue, 21 May 2013 13:52:43 -0500
> Jon Elson <jme...@wustl.edu> wrote:
>
>> Having the PC in memory could lead to some really hard to debug
>> errors! I sure don't recall this "never write to loc 0" dictum,
>
> It's just as well C wasn't around then.
>
Oh, my! The PDP-8 architecture might not be real efficient at
> On Tue, 21 May 2013 13:52:43 -0500
> Jon Elson <jme...@wustl.edu> wrote:
>
>> Having the PC in memory could lead to some really hard to debug
>> errors! I sure don't recall this "never write to loc 0" dictum,
>
> It's just as well C wasn't around then.
>
running C code. No index registers, and that annoying page
scheme. a = *(p+27)++; would take quite a few instructions.
Jon
hanc...@bbs.cpcn.com
May 29, 2013, 4:59:25 PM5/29/13
to
On May 18, 7:32 am, Quadibloc <jsav...@ecn.ab.ca> wrote:
> and the brief CV of the author at the end claims that the Elbit 100
> has the historical significance commonly attributed to the PDP-8!!!
The $10,000 Question:
Can a small computer
find happiness
as part of your instrument?
[picture of a cute little stuffed bear hugging a small CPU].
Time was, general purpose computers were expensive
to buy, expensive to use - unapproachable. No more.
If you build instruments or systems that analyze, or
measure, or compute, or adjust, or coni.r,-I :)rocesses,
consider this:
For less than $10,000 (much less, if you order several at
once) the PDP-SIS - a full, real-time, on-line
4096 word, FORTRAN speaking, general purpose
tal computer can be part of your system. To analyze,
measure, compute, adjust. or control processes.
If you make more than one kind of system, you still may
need only one kind of computer. It's gene; al purpose,
you see. And if your requirements are big, we have big
fast machines too, upwards compatible. And a com·
plete line of modules for interfacing.
One advantage for your product is clear: if your customer
needs more or different capability after he buys,
he adapts by plugging in options, or writing new programs,
or changing them, or expanding them. And
your product just might be easier to sell if there's a
computer inside.
The PDP-8/S offers security. Security in
change. It is priced lower than many speCial
purpose machines. More than 300 have
been sold in the past three months. And
chances are still good that your competition
hasn't even looked into it. Why don't you.
d|i|g|i|t|a|l
DIGITAL EQUIPMENT CORPORATION. Maynard, Massachusetts 01754.
http://bitsavers.informatik.uni-stuttgart.de/pdf/computersAndAutomation/196701.pdf
page 68
> and the brief CV of the author at the end claims that the Elbit 100
> has the historical significance commonly attributed to the PDP-8!!!
The $10,000 Question:
Can a small computer
find happiness
as part of your instrument?
[picture of a cute little stuffed bear hugging a small CPU].
Time was, general purpose computers were expensive
to buy, expensive to use - unapproachable. No more.
If you build instruments or systems that analyze, or
measure, or compute, or adjust, or coni.r,-I :)rocesses,
consider this:
For less than $10,000 (much less, if you order several at
once) the PDP-SIS - a full, real-time, on-line
4096 word, FORTRAN speaking, general purpose
tal computer can be part of your system. To analyze,
measure, compute, adjust. or control processes.
If you make more than one kind of system, you still may
need only one kind of computer. It's gene; al purpose,
you see. And if your requirements are big, we have big
fast machines too, upwards compatible. And a com·
plete line of modules for interfacing.
One advantage for your product is clear: if your customer
needs more or different capability after he buys,
he adapts by plugging in options, or writing new programs,
or changing them, or expanding them. And
your product just might be easier to sell if there's a
computer inside.
The PDP-8/S offers security. Security in
change. It is priced lower than many speCial
purpose machines. More than 300 have
been sold in the past three months. And
chances are still good that your competition
hasn't even looked into it. Why don't you.
d|i|g|i|t|a|l
DIGITAL EQUIPMENT CORPORATION. Maynard, Massachusetts 01754.
http://bitsavers.informatik.uni-stuttgart.de/pdf/computersAndAutomation/196701.pdf
page 68
Quadibloc
Jun 6, 2013, 11:54:24 AM6/6/13
to
On May 19, 12:50 pm, Quadibloc <jsav...@ecn.ab.ca> wrote:
> On May 19, 10:59 am, Al Kossow <a...@bitsavers.org> wrote:
>
> > I didn't have time to get to the H300 programming manual. I'll try to get to it
> > in the next week or two.
>
> That's wonderful! Thank you very much. I thought that you simply
> didn't have one available.
I am still interested, but I see that you've been very busy of late
scanning all sorts of other very interesting documents.
John Savard
> On May 19, 10:59 am, Al Kossow <a...@bitsavers.org> wrote:
>
> > I didn't have time to get to the H300 programming manual. I'll try to get to it
> > in the next week or two.
>
> That's wonderful! Thank you very much. I thought that you simply
> didn't have one available.
I am still interested, but I see that you've been very busy of late
scanning all sorts of other very interesting documents.
John Savard
yigal...@gmail.com
Sep 11, 2013, 3:32:44 AM9/11/13
to
As a 12 years old kid, back in 1978, I had an access to an 'old' Elbit 100 and wrote it some programs in assembly language for my own joy (as strange as it may sound).
The Elbit 100 was declared to be:
1. The first that has a CPU 100% built by Integrated Circuits (wo/single transistors)
2. The first mini computer with a cost/benefit ratio that suits commercial uses.
(starting at 5000$, and in ~10K$: 4086 x 12bit memory, punched strip reader, assembler).
If that makes it a pioneer in the same line as the PDP 8? Probably not. But on the other hand it worth mentioning much more than it is now.
Yigal
The Elbit 100 was declared to be:
1. The first that has a CPU 100% built by Integrated Circuits (wo/single transistors)
2. The first mini computer with a cost/benefit ratio that suits commercial uses.
(starting at 5000$, and in ~10K$: 4086 x 12bit memory, punched strip reader, assembler).
If that makes it a pioneer in the same line as the PDP 8? Probably not. But on the other hand it worth mentioning much more than it is now.
Yigal
0 new messages