3 Apr
2024
3 Apr
'24
8:01 a.m.
Vintage computer enthusiasts might want to keep track of where to find
CRT-based analog oscilloscopes, for use as output devices.
The early MIT and Lincoln Labs computers used D/A converters to steer
and activate the beam on analog scopes to draw vector images.
Working on Whirlwind simulation, we've been able to get this technique
to work with "real" oscilloscopes, e.g., Tek 475, but we have not yet
found a single DSO that has X/Y _and_ Z inputs (let alone the required
phosphor fade).
Myself, I have a couple scopes with backups, so I'm not in the market
for another one. But others might consider the option...
/guy fedorkow
Date: Mon, 01 Apr 2024 22:33:38 +0000
From: Just Kant<kantexplain(a)protonmail.com>
Subject: [cctalk] oscilloscopes
To: "General Discussion: On-Topic and Off-Topic Posts"
<cctalk(a)classiccmp.org>
Message-ID: <xVKPUUPAeHv0_Gdsaq0al3Lji6a5VLAX2wRSUo5YhZwiADr5zn171nRBM
NxX-XE1ru7jPj1GFbu0aE3uBwUsVKxlU_XAYKX9B0x42zJZADc=(a)protonmail.com>
Content-Type: text/plain; charset=utf-8
I have more then I need. All the working ones are HP w/color crts, and as far as older,
verifiably vintage tools (right down to the 680x0 processor in either) I have to admit I
favor them as a brand. Call we an oddball, weird egg, badges I wear with pride.
But who could resist the allure of the newer ultra portable, even handheld units (some
with bandwidth or sampling rates to 50mhz). I'm a big cheapo. But there's no real
reason to agonize over a 65 - 200$ or thereabouts acquisition. It's a bit tiring to
wade through the piles of availability. I favor a desktop unit, larger screen (but not
always, careful). But most of those need wall current I think? The convenience of a
handheld battery powered unit obviously has it's benefits.
I will always love and dote upon my color crt based HPs. But the damned things are so
heavy, so unwieldy. Judy-Jude knocked my 54111d over, hit the paved floor, shook the
house. And still works! Built to withstand an atomic bombardment.
3 Apr
3 Apr
8:16 a.m.
New subject: oscilloscopes
On Apr 3, 2024, at 11:01 AM, Guy Fedorkow via cctalk
<cctalk(a)classiccmp.org> wrote:
Vintage computer enthusiasts might want to keep track of where to find CRT-based analog
oscilloscopes, for use as output devices.
The early MIT and Lincoln Labs computers used D/A converters to steer and activate the
beam on analog scopes to draw vector images.
Working on Whirlwind simulation, we've been able to get this technique to work with
"real" oscilloscopes, e.g., Tek 475, but we have not yet found a single DSO that
has X/Y _and_ Z inputs (let alone the required phosphor fade).
So did a whole range of DEC computers, of course. And the famous CDC mainframe console
(DD60) though it did vectors only for text (graphics was dot-mode only since it wasn't
a major use case for that device).
I once built a graphics display setup for an 11/20 lab machine (in college) using DEC D/A
modules (AA-01?) with an RC-11 disk serving as the refresh memory, DMA direct to the D/A
data register.
paul
8:17 a.m.
New subject: oscilloscopes
DEC used Tektronix R503 scopes for a display on many of their early
machines.
On Wed, Apr 3, 2024 at 11:16 AM Paul Koning via cctalk <
cctalk(a)classiccmp.org> wrote:
--
Michael Thompson
On Apr 3, 2024, at 11:01 AM, Guy Fedorkow via
cctalk <
cctalk(a)classiccmp.org> wrote:
Vintage computer enthusiasts might want to keep track of where to find
CRT-based analog oscilloscopes, for use as output devices.
The early MIT and Lincoln Labs computers used D/A
converters to steer
and activate the beam on analog scopes to draw vector images.
Working on Whirlwind simulation, we've been
able to get this technique
to work with "real" oscilloscopes, e.g., Tek 475, but we have not yet
found
a single DSO that has X/Y _and_ Z inputs (let alone the required phosphor
fade).
So did a whole range of DEC computers, of course. And the famous CDC
mainframe console (DD60) though it did vectors only for text (graphics was
dot-mode only since it wasn't a major use case for that device).
I once built a graphics display setup for an 11/20 lab machine (in
college) using DEC D/A modules (AA-01?) with an RC-11 disk serving as the
refresh memory, DMA direct to the D/A data register.
paul
8:26 a.m.
New subject: oscilloscopes
On Apr 3, 2024, at 11:16 AM, Paul Koning via cctalk
<cctalk(a)classiccmp.org> wrote:
The DD60 and its associated controller in the mainframe (6612 or 6602) was an interesting
beast. The interface between controller and display is a hybrid, with the positioning
information delivered as 9 bits each of X and Y, but the character vectors are generated
in the controller and sent to the display as analog waveforms, X and Y on differential
pairs.
Another oddity is the character waveform generation: that uses two pairs of A/D
converters, and the converters are essentially base one -- 6 equally weighted inputs to
produce output values 0..6. And since ROMs were hard to come by in 1964, at least ones
with 100 ns cycle time, the digital inputs for the waveform generators are an amazingly
large pile of gates.
paul
On Apr 3, 2024, at 11:01 AM, Guy Fedorkow via
cctalk <cctalk(a)classiccmp.org> wrote:
Vintage computer enthusiasts might want to keep track of where to find CRT-based analog
oscilloscopes, for use as output devices.
The early MIT and Lincoln Labs computers used D/A converters to steer and activate the
beam on analog scopes to draw vector images.
Working on Whirlwind simulation, we've been able to get this technique to work with
"real" oscilloscopes, e.g., Tek 475, but we have not yet found a single DSO that
has X/Y _and_ Z inputs (let alone the required phosphor fade).
So did a whole range of DEC computers, of course. And the famous CDC mainframe console
(DD60) though it did vectors only for text (graphics was dot-mode only since it wasn't
a major use case for that device). 
10:49 a.m.
New subject: oscilloscopes
Paul wrote:
The DD60 and its associated controller in the
mainframe (6612 or 6602) was an > interesting beast. The interface between controller
and display is a hybrid, > with the positioning information delivered as 9 bits each of
X and Y, but the > character vectors are generated in the controller and sent to the
display as
analog waveforms, X and Y on differential pairs.
Another oddity is the character waveform generation:
that uses two pairs of
A/D converters, and the converters are essentially base one -- 6 equally
weighted inputs to produce output values 0..6. And since ROMs were hard to
come by in 1964, at least ones with 100 ns cycle time, the digital inputs for > the
waveform generators are an amazingly large pile of gates.
I was a systems operator (as it was called back in the day) on a Control Data Cyber 73 at
Tektronix for a number of years. The round console displays on the machine were simply
beautiful. The characters were very clear and easy to read, even though they packed a lot
of them on the screen. One thing that I thought was particularly cool about it is that
you could pull up a page of any section of memory and watch the live data flicker around
as the machine was running. Pretty amazing.
On a similar display topic, I have a very unusual old electronic calculator made by a US
company called Wyle Laboratories. The machine was designed in the 1962-1063 timeframe.
It is all Germanium-Transistor and Resistor logic(RTL). The Model WS-01/WS-02 (two
versions, both using the same display subsystem, but varied in terms of the main register
storage) calculators used a CRT display for showing the content of the registers to the
user. The display
consists of six lines of 24 digits each. The six lines represent the content of three
memory registers, the accumulator, a multiplier-quotient register, and the numeric entry
register.
Even with only having to render the digits zero through nine and a decimal point (the
calculator didn't support negative numbers; they were represented using tens
complement form), the display generator also used a batch of diode-transistor gates to
generate the digits. The interesting thing about it is that instead of generating
strokes to create the digits, the machine uses sine/cosine waveforms that are gated by the
character generation logic to draw the digits on the screen. The position of the digits,
like the CDC scopes, is derived by precision resistor DACs, and then a mixer takes over as
the character is drawn using gated segments of the sine and cosine waveforms mixed
together with the position voltage. The result is really beautifully rendered digits
that look almost like they are drawn by a draftsperson who is extremely consistent in the
drawing of each digit. The CRT has yellow-orange phosphor with a moderate persistence, so
when the digits change, they look like they quickly morph from one digit to the next.
The digits are among the nicest looking digits that I've ever seen on a CRT display,
including those on the CDC scopes as well as IBM console displays.
They are far more aesthetically (to my taste, anyway) pleasing than the segmented digits
drawn on other CRT-display calculators such as the HP 9100A/B, Friden EC-130/132 and 116x,
Victor 3900 and 14-321/14-322, SCM Cogito 240/240SR, Busicom 202/207/2017, and some
calculators (usually clones of US or Japanese-designed machines) made in the former Soviet
Union.
-Rick
11:20 a.m.
New subject: oscilloscopes
On Apr 3, 2024, at 1:49 PM, Rick Bensene via cctalk
<cctalk(a)classiccmp.org> wrote:
...
Even with only having to render the digits zero through nine and a decimal point (the
calculator didn't support negative numbers; they were represented using tens
complement form), the display generator also used a batch of diode-transistor gates to
generate the digits. The interesting thing about it is that instead of generating
strokes to create the digits, the machine uses sine/cosine waveforms that are gated by the
character generation logic to draw the digits on the screen. The position of the digits,
like the CDC scopes, is derived by precision resistor DACs, and then a mixer takes over as
the character is drawn using gated segments of the sine and cosine waveforms mixed
together with the position voltage. The result is really beautifully rendered digits
that look almost like they are drawn by a draftsperson who is extremely consistent in the
drawing of each digit. The CRT has yellow-orange phosphor with a moderate persistence, so
when the digits change, they look like they quickly morph from one digit to the next.
The digits are among the nicest looking digits that I've ever seen on a CRT display,
including those on the CDC scopes as well as IBM console displays.
I have, somewhere, a copy of a paper that describes analog circuits for generating
waveforms for digits along the lines you describe. Might have been from MIT, in the
1950s, but right now I can't find it.
The CDC console waveforms start out as step function waveforms, with delta x and/or y of
+/- 1 or 2 units, at 100 ns intervals. Given the bandwidths of the circuits involved they
get rounded off in the generation, and a whole lot more in the DD60 deflection amplifier
signal chain. I've tried to create a SPICE model of that signal path to try to
reproduce what we know actually showed up on the display screen, but haven't had much
luck. Too much of the circuit involves parts with unknown properties, starting with the
transistors, on to the wirewound resistors that apparently show up in various places, and
ending with the deflection plates of the CRTs themselves. Still, a crude IIR filter
mimicking some of the more obvious contributions do produce acceptable character shapes in
my DD60 emulation software.
Speaking of nice looking numeric displays: probably the best ever are the projection
displays made by IEE, in the 1960s I think.
https://www.antiqueradios.com/forums//viewtopic.php?f=12&t=341355 shows a sample. A
few computers from that era used them for the console, the CDC 1604 seems to be an
example.
paul
11:44 a.m.
New subject: oscilloscopes
On Apr 3, 2024, at 2:20 PM, Paul Koning via cctalk
<cctalk(a)classiccmp.org> wrote:
Found it (on paper): "Generating characters" by Kenneth Perry and Everett Aho,
Electronics, Jan 3, 1958, pp. 72-75.
Bitsavers has it in the MIT/LincolnLaboratory section:
https://bitsavers.org/pdf/mit/lincolnLaboratory/Perry_and_Aho_-_Generating_…
paul
On Apr 3, 2024, at 1:49 PM, Rick Bensene via
cctalk <cctalk(a)classiccmp.org> wrote:
...
Even with only having to render the digits zero through nine and a decimal point (the
calculator didn't support negative numbers; they were represented using tens
complement form), the display generator also used a batch of diode-transistor gates to
generate the digits. The interesting thing about it is that instead of generating
strokes to create the digits, the machine uses sine/cosine waveforms that are gated by the
character generation logic to draw the digits on the screen. The position of the digits,
like the CDC scopes, is derived by precision resistor DACs, and then a mixer takes over as
the character is drawn using gated segments of the sine and cosine waveforms mixed
together with the position voltage. The result is really beautifully rendered digits
that look almost like they are drawn by a draftsperson who is extremely consistent in the
drawing of each digit. The CRT has yellow-orange phosphor with a moderate persistence, so
when the digits change, they look like they quickly morph from one digit to the next.
The digits are among the nicest looking digits that I've ever seen on a CRT display,
including those on the CDC scopes as well as IBM console displays.
I have, somewhere, a copy of a paper that describes analog circuits for generating
waveforms for digits along the lines you describe. Might have been from MIT, in the
1950s, but right now I can't find it.
3:32 p.m.
New subject: oscilloscopes
I wrote:
> The digits are among the nicest looking digits
that I've ever seen
> on a CRT display, including those on the CDC scopes as well as IBM >> console
displays.
To which Paul responded:
I have, somewhere, a copy of a paper that describes
analog circuits > for generating waveforms for digits along the lines you describe.
Might have been from MIT, in the 1950s, but right now I can't find > it.
Found it (on paper): "Generating characters"
by Kenneth Perry and
Everett Aho, > Electronics, Jan 3, 1958, pp. 72-75.
Bitsavers has it in the MIT/LincolnLaboratory section:
https://bitsavers.org/pdf/mit/lincolnLaboratory/Perry_and_Aho__Generating_C…
Very interesting. Here's a link to the patent for the display system on the Wyle
Labs calculator:
https://patentimages.storage.googleapis.com/17/51/58/89c19cee6c60e2/US33058…
The concepts are very similar to the paper written up in ELECTRONICS magazine in early
1958 that you found. Your memory is incredible to have been able to have this pop into
your mind when you read my description of the way the calculator generates its display.
Thank you for looking up this article! It'll provide some nice background for the
concepts of generating characters this way when I finally get to documenting the Wyle
WS-01/WS-02 calculators in an Old Calculator Museum exhibit.
I wonder if the inventor of the display system for the calculator (in fact, the inventor
of the entire Wyle Labs calculator architecture) had read this article at some point
prior?
I scanned through the patent for the calculator display system looking for any reference
to the article or any document from MIT relating, and I couldn't find anything.
The inventor is still alive, and I have talked to him on the telephone a couple of times.
For his advanced age, he is still quite sharp, and remembers a lot of the challenges
involved with trying to make a solid-state electronic calculator that would fit on a
(large) desktop using early 1960's technology.
Here's a link to a little information about the calculator:
https://oldcalculatormuseum.com/w-wyle.html
It's hard to tell from the photo (from advertising material of the day) how large the
machine is, but the dimensions and weight (50 pounds!) are included in the specifications
at the end of the page.
It's quite a monster.
The two models of the machine differed in that the WS-01 utilized a home-brewed(e.g., made
entirely within Wyle Labs) rotating magnetic memory as the storage for the working
registers (as well as for timing tracks that provided clocking signals). It proved to be
very temperamental with many failures in the hands of customers that gave the machine
somewhat of a tarnished reputation in the market. A magnetostrictive delay line that was
much more reliable replaced the rotating magnetic memory in the re-worked WS-02 model of
the calculator.
-Rick
4 Apr
4 Apr
6:12 a.m.
New subject: oscilloscopes
On Apr 3, 2024, at 6:32 PM, Rick Bensene
<rickb(a)bensene.com> wrote:
I wrote:
I didn't see any either, and the patent examiners didn't cite any. Then again,
it's amazing how often patent examiners miss relevant prior art. One example I like
to mention is Edwin Armstrong's patent for FM radio, which doesn't cite an actual
earlier US patent, 1,648,402 from 1927, actually filed 12 years before Armstrong's.
Or the prior art centuries preceding US 6469...
On the other hand, while the concept is similar the details are rather different, and the
Wyle design is clearly a whole lot simpler.
> The digits are among the nicest looking
digits that I've ever seen
> on a CRT display, including those on the CDC scopes as well as IBM >> console
displays.
To which Paul responded:
I have, somewhere, a copy of a paper that
describes analog circuits > for generating waveforms for digits along the lines you
describe.
Might have been from MIT, in the 1950s, but right now I can't find > it.
Found it (on paper): "Generating
characters" by Kenneth Perry and
Everett Aho, > Electronics, Jan 3, 1958, pp. 72-75.
Bitsavers has it in the MIT/LincolnLaboratory
section:
https://bitsavers.org/pdf/mit/lincolnLaboratory/Perry_and_Aho__Generating_C…
Very interesting. Here's a link to the patent for the display system on the Wyle
Labs calculator:
https://patentimages.storage.googleapis.com/17/51/58/89c19cee6c60e2/US33058…
The concepts are very similar to the paper written up in ELECTRONICS magazine in early
1958 that you found. Your memory is incredible to have been able to have this pop into
your mind when you read my description of the way the calculator generates its display.
Thank you for looking up this article! It'll provide some nice background for the
concepts of generating characters this way when I finally get to documenting the Wyle
WS-01/WS-02 calculators in an Old Calculator Museum exhibit.
I wonder if the inventor of the display system for the calculator (in fact, the inventor
of the entire Wyle Labs calculator architecture) had read this article at some point
prior?
I scanned through the patent for the calculator display system looking for any reference
to the article or any document from MIT relating, and I couldn't find anything.
The inventor is still alive, and I have talked to him
on the telephone a couple of times. For his advanced age, he is still quite sharp, and
remembers a lot of the challenges involved with trying to make a solid-state electronic
calculator that would fit on a (large) desktop using early 1960's technology.
It would be neat to ask him about that MIT article.
paul
7:22 a.m.
New subject: oscilloscopes
This 'scope clock also uses circle generators rather than vectors to
produce well-formed characters. It mentions a Teensy controller so I don't
think it's the original made in this way - the first I heard of was too
long ago for that. But I don't know if it's an update or a separate design.
https://scopeclock.com/
On Thu, Apr 4, 2024 at 2:59 PM Paul Koning via cctalk <cctalk(a)classiccmp.org>
wrote:
https://bitsavers.org/pdf/mit/lincolnLaboratory/Perry_and_Aho__Generating_C…
https://patentimages.storage.googleapis.com/17/51/58/89c19cee6c60e2/US33058…
On Apr 3, 2024, at 6:32 PM, Rick Bensene
<rickb(a)bensene.com> wrote:
I wrote:
>> The digits are among the nicest looking digits that I've ever seen
>> on a CRT display, including those on the CDC scopes as well as IBM >>
console displays.
To which Paul responded:
> I have, somewhere, a copy of a paper that describes analog circuits >
for generating waveforms for digits along the lines you describe.
Might
have been from MIT, in the 1950s, but right now I can't find > it.
Found it (on paper): "Generating
characters" by Kenneth Perry and
Everett Aho, > Electronics, Jan 3, 1958, pp. 72-75.
> Bitsavers has it in the MIT/LincolnLaboratory section:
Very interesting. Here's a link to the patent for the display system
on the Wyle Labs calculator:
The concepts are very similar to the paper written up in ELECTRONICS
magazine in early 1958 that you found. Your memory is incredible to have
been able to have this pop into your mind when you read my description of
the way the calculator generates its display.
Thank you for looking up this article! It'll provide some nice
background for the concepts of generating characters this way when I
finally get to documenting the Wyle WS-01/WS-02 calculators in an Old
Calculator Museum exhibit.
I wonder if the inventor of the display system for the calculator (in
fact, the inventor of the entire Wyle Labs calculator architecture) had
read this article at some point prior?
I scanned through the patent for the calculator display system looking
for any reference to the article or any document from MIT relating, and I
couldn't find anything.
I didn't see any either, and the patent examiners didn't cite any. Then
again, it's amazing how often patent examiners miss relevant prior art.
One example I like to mention is Edwin Armstrong's patent for FM radio,
which doesn't cite an actual earlier US patent, 1,648,402 from 1927,
actually filed 12 years before Armstrong's. Or the prior art centuries
preceding US 6469...
On the other hand, while the concept is similar the details are rather
different, and the Wyle design is clearly a whole lot simpler.
The inventor is still alive, and I have talked to
him on the telephone a
couple of times. For his advanced age, he is still quite sharp, and
remembers a lot of the challenges involved with trying to make a
solid-state electronic calculator that would fit on a (large) desktop using
early 1960's technology.
It would be neat to ask him about that MIT article.
paul
12:12 p.m.
New subject: oscilloscopes
On 2024Apr 4,, at 7:22 AM, Adrian Godwin via cctalk
<cctalk(a)classiccmp.org> wrote:
This 'scope clock also uses circle generators rather than vectors to
produce well-formed characters. It mentions a Teensy controller so I don't
think it's the original made in this way - the first I heard of was too
long ago for that. But I don't know if it's an update or a separate design.
https://scopeclock.com/
<https://scopeclock.com/>
Technically, the scopeClock is generating neither curves nor vectors, it's
generating pixels in an XY display - it's just that they’re of fine enough resolution
and fast enough that they’re seen as a smooth-enough curve on the CRT.
The MIT/Electronics-magazine and Wyle techniques are using analog electronics to generate
portions of sine waves for selected phase periods and phase relation such that when
applied to the XY cartesian display you get continuous portions (chords) of circles. Some
digital logic gates the analog sine generators appropriately to produce the chords and
line segments, with offsets, in a sequence to form characters.
The scopeClock, in contrast, is using DACs in the microcontroller to generate (discrete
approximations of) sine wave segments - which is to say it’s relying on the abilities of
inexpensive current-day high-speed digital electronics.
Thanks Paul, for pulling up the article, I recalled it as well when Rick mentioned the
technique of the Wyle. I think I ran across it many years ago when sorting through stacks
of Electronics magazines at our radio museum.
12:35 p.m.
New subject: oscilloscopes
On Apr 4, 2024, at 3:12 PM, Brent Hilpert via cctalk
<cctalk(a)classiccmp.org> wrote:
That is similar to what the CDC 6612 controller for the DD60 console display does. So
given that you're sending the resulting step waveform through a deflection circuit
with finite bandwidth, you do in fact end up with a continuous vector with rounded
features. How nicely rounded depends on the bandwidth and the number and size of the
steps. For example, the DD60 display doesn't look all that elegant, but it is
definitely well rounded, simply because the step clock is 10 MHz and the deflection chain
bandwidth isn't a whole lot more than that. So the fact that you're dealing with
what originally was a step waveform with just 7 positions for X and Y isn't at all
obvious in the final image.
paul
On 2024Apr 4,, at 7:22 AM, Adrian Godwin via
cctalk <cctalk(a)classiccmp.org> wrote:
This 'scope clock also uses circle generators rather than vectors to
produce well-formed characters. It mentions a Teensy controller so I don't
think it's the original made in this way - the first I heard of was too
long ago for that. But I don't know if it's an update or a separate design.
https://scopeclock.com/
<https://scopeclock.com/>
Technically, the scopeClock is generating neither
curves nor vectors, it's generating pixels in an XY display - it's just that
they’re of fine enough resolution and fast enough that they’re seen as a smooth-enough
curve on the CRT.
The MIT/Electronics-magazine and Wyle techniques are using analog electronics to generate
portions of sine waves for selected phase periods and phase relation such that when
applied to the XY cartesian display you get continuous portions (chords) of circles. Some
digital logic gates the analog sine generators appropriately to produce the chords and
line segments, with offsets, in a sequence to form characters.
The scopeClock, in contrast, is using DACs in the microcontroller to generate (discrete
approximations of) sine wave segments - which is to say it’s relying on the abilities of
inexpensive current-day high-speed digital electronics.
1:21 p.m.
New subject: oscilloscopes
On Thu, Apr 4, 2024 at 8:36 PM Paul Koning via cctalk <cctalk(a)classiccmp.org>
wrote:
it's generating pixels in an XY display - it's just that they’re of fine
enough resolution and fast enough that they’re seen as a smooth-enough
curve on the CRT.
That's disappointing. There was such a clock, and I thought I'd found it
since it described using circle generators - but was surprised to find it
used a teensy.
Does anyone know which clock used analog circuits to form the characters ?
https://scopeclock.com/
<https://scopeclock.com/>
Technically, the scopeClock is generating neither curves nor vectors,
1:33 p.m.
New subject: oscilloscopes
This one: https://scopeclock.com/
You can see the connected arcs in the pic on the main page.
I built an older model years ago and it is still running...
--tom
On 4/4/24 15:21, Adrian Godwin via cctalk wrote:
On Thu, Apr 4, 2024 at 8:36 PM Paul Koning via cctalk
<cctalk(a)classiccmp.org>
wrote:
it's generating pixels in an XY display - it's just
that they’re of fine
enough resolution and fast enough that they’re seen as a smooth-enough
curve on the CRT.
That's disappointing. There was such a clock, and I thought I'd found it
since it described using circle generators - but was surprised to find it
used a teensy.
Does anyone know which clock used analog circuits to form the characters ?
https://scopeclock.com/
<https://scopeclock.com/>
Technically, the scopeClock is generating neither
curves nor vectors, 
5 Apr
5 Apr
12:23 a.m.
New subject: oscilloscopes
Found it (on paper): "Generating
characters" by Kenneth Perry and
Everett Aho, > Electronics, Jan 3, 1958, pp. 72-75.
> Bitsavers has it in the MIT/LincolnLaboratory section:
Very interesting. Here's a link to the patent for the display system
on the Wyle Labs calculator:
The concepts are very similar to the paper written up in ELECTRONICS
magazine in early 1958 that you found. Your memory is incredible to have
been able to have this pop into your mind when you read my description of
the way the calculator generates its display.
3 Apr
3 Apr
8:21 a.m.
New subject: oscilloscopes
I'm surprised some digital scope manufacturer hasn't implemented X-Y-Z
control as an option. Driving X-Y was fairly common for certain types
of signals. And many also used the Z input.
Back in the day there were many companies that made X-Y or X-Y-Z displays.
On 4/3/2024 10:01 AM, Guy Fedorkow via cctalk wrote:
Vintage computer enthusiasts might want to keep track
of where to find
CRT-based analog oscilloscopes, for use as output devices.
The early MIT and Lincoln Labs computers used D/A converters to steer
and activate the beam on analog scopes to draw vector images.
Working on Whirlwind simulation, we've been able to get this technique
to work with "real" oscilloscopes, e.g., Tek 475, but we have not yet
found a single DSO that has X/Y _and_ Z inputs (let alone the required
phosphor fade).
Myself, I have a couple scopes with backups, so I'm not in the
market for another one. But others might consider the option...
/guy fedorkow
Date: Mon, 01 Apr 2024 22:33:38 +0000
From: Just Kant<kantexplain(a)protonmail.com>
Subject: [cctalk] oscilloscopes
To: "General Discussion: On-Topic and Off-Topic Posts"
<cctalk(a)classiccmp.org>
Message-ID: <xVKPUUPAeHv0_Gdsaq0al3Lji6a5VLAX2wRSUo5YhZwiADr5zn171nRBM
NxX-XE1ru7jPj1GFbu0aE3uBwUsVKxlU_XAYKX9B0x42zJZADc=(a)protonmail.com>
Content-Type: text/plain; charset=utf-8
I have more then I need. All the working ones are HP w/color crts, and
as far as older, verifiably vintage tools (right down to the 680x0
processor in either) I have to admit I favor them as a brand. Call we
an oddball, weird egg, badges I wear with pride.
But who could resist the allure of the newer ultra portable, even
handheld units (some with bandwidth or sampling rates to 50mhz). I'm a
big cheapo. But there's no real reason to agonize over a 65 - 200$ or
thereabouts acquisition. It's a bit tiring to wade through the piles
of availability. I favor a desktop unit, larger screen (but not
always, careful). But most of those need wall current I think? The
convenience of a handheld battery powered unit obviously has it's
benefits.
I will always love and dote upon my color crt based HPs. But the
damned things are so heavy, so unwieldy. Judy-Jude knocked my 54111d
over, hit the paved floor, shook the house. And still works! Built to
withstand an atomic bombardment.
8:27 a.m.
New subject: oscilloscopes
On Apr 3, 2024, at 11:21 AM, Mike Katz via cctalk
<cctalk(a)classiccmp.org> wrote:
I'm surprised some digital scope manufacturer hasn't implemented X-Y-Z control as
an option. Driving X-Y was fairly common for certain types of signals. And many also
used the Z input.
Oh, they offer X/Y display, but sampled just as the normal operation is. Some of the
applications we're talking about here don't appreciate the digitization
artefacts.
paul
8:47 a.m.
New subject: oscilloscopes
I have a pair of CDC6600 console CRTs (~12" diameter electrostatic deflection
vector), I've been
working on "restoring" a console which I salvaged from Purdue surplus many years
ago, but have kind
of stalled on at present and it takes up a significant amount of space. If someone would
like to
take over the project, let's talk...
For those who are not familiar, it looks like this:
https://en.wikipedia.org/wiki/CDC_6600#/media/File:Control_Data_6600_mainfr…
I have the remains of most of the drive electronics, but the design runs on 400Hz power,
so I am
planning on using a solid state power converter to run it. The ultimate plan of course is
to run a
Pi (or FPGA) CDC6600 simulation to drive the console.
--tom
On 4/3/24 10:01, Guy Fedorkow via cctalk wrote:
Vintage computer enthusiasts might want to keep track
of where to find CRT-based analog
oscilloscopes, for use as output devices.
The early MIT and Lincoln Labs computers used D/A converters to steer and activate the
beam on
analog scopes to draw vector images.
Working on Whirlwind simulation, we've been able to get this technique to work with
"real"
oscilloscopes, e.g., Tek 475, but we have not yet found a single DSO that has X/Y _and_ Z
inputs
(let alone the required phosphor fade).
Myself, I have a couple scopes with backups, so I'm not in the market for another
one. But
others might consider the option...
/guy fedorkow
Date: Mon, 01 Apr 2024 22:33:38 +0000
From: Just Kant<kantexplain(a)protonmail.com>
Subject: [cctalk] oscilloscopes
To: "General Discussion: On-Topic and Off-Topic Posts"
<cctalk(a)classiccmp.org>
Message-ID: <xVKPUUPAeHv0_Gdsaq0al3Lji6a5VLAX2wRSUo5YhZwiADr5zn171nRBM
NxX-XE1ru7jPj1GFbu0aE3uBwUsVKxlU_XAYKX9B0x42zJZADc=(a)protonmail.com>
Content-Type: text/plain; charset=utf-8
I have more then I need. All the working ones are HP w/color crts, and as far as older,
verifiably
vintage tools (right down to the 680x0 processor in either) I have to admit I favor them
as a
brand. Call we an oddball, weird egg, badges I wear with pride.
But who could resist the allure of the newer ultra portable, even handheld units (some
with
bandwidth or sampling rates to 50mhz). I'm a big cheapo. But there's no real
reason to agonize
over a 65 - 200$ or thereabouts acquisition. It's a bit tiring to wade through the
piles of
availability. I favor a desktop unit, larger screen (but not always, careful). But most
of those
need wall current I think? The convenience of a handheld battery powered unit obviously
has it's
benefits.
I will always love and dote upon my color crt based HPs. But the damned things are so
heavy, so
unwieldy. Judy-Jude knocked my 54111d over, hit the paved floor, shook the house. And
still works!
Built to withstand an atomic bombardment.
9:13 a.m.
New subject: oscilloscopes
rfgh
-----Original Message-----
From: Guy Fedorkow via cctalk [mailto:cctalk@classiccmp.org]
Sent: 03 April 2024 16:02
To: cctalk(a)classiccmp.org
Cc: Guy Fedorkow <fedorkow(a)mit.edu>
Subject: [cctalk] Re: oscilloscopes
Vintage computer enthusiasts might want to keep track of where to find CRT-based analog
oscilloscopes, for use as output devices.
The early MIT and Lincoln Labs computers used D/A converters to steer and activate the
beam on analog scopes to draw vector images.
Working on Whirlwind simulation, we've been able to get this technique to work with
"real" oscilloscopes, e.g., Tek 475, but we have not yet found a single DSO that
has X/Y _and_ Z inputs (let alone the required phosphor fade).
Myself, I have a couple scopes with backups, so I'm not in the market for another
one. But others might consider the option...
/guy fedorkow
Date: Mon, 01 Apr 2024 22:33:38 +0000
From: Just Kant<kantexplain(a)protonmail.com>
Subject: [cctalk] oscilloscopes
To: "General Discussion: On-Topic and Off-Topic Posts"
<cctalk(a)classiccmp.org>
Message-ID: <xVKPUUPAeHv0_Gdsaq0al3Lji6a5VLAX2wRSUo5YhZwiADr5zn171nRBM
NxX-XE1ru7jPj1GFbu0aE3uBwUsVKxlU_XAYKX9B0x42zJZADc=(a)protonmail.com>
Content-Type: text/plain; charset=utf-8
I have more then I need. All the working ones are HP w/color crts, and as far as older,
verifiably vintage tools (right down to the 680x0 processor in either) I have to admit I
favor them as a brand. Call we an oddball, weird egg, badges I wear with pride.
But who could resist the allure of the newer ultra portable, even handheld units (some
with bandwidth or sampling rates to 50mhz). I'm a big cheapo. But there's no real
reason to agonize over a 65 - 200$ or thereabouts acquisition. It's a bit tiring to
wade through the piles of availability. I favor a desktop unit, larger screen (but not
always, careful). But most of those need wall current I think? The convenience of a
handheld battery powered unit obviously has it's benefits.
I will always love and dote upon my color crt based HPs. But the damned things are so
heavy, so unwieldy. Judy-Jude knocked my 54111d over, hit the paved floor, shook the
house. And still works! Built to withstand an atomic bombardment.
9:28 a.m.
New subject: oscilloscopes
Ignore my last - incontinence or is it incompetence
A fairly ordinary GPU, in a PC, could almost certainly provide an XY display with Z fade
(long persistance phosphor). I use them for waterfall displays and they keep up - the
data does of course arrive by E'net.
Equally, FPGAs / SOCs can implement frame buffers; eg to output waterfall displays. The
fading memory would have to be in DRAM, FPGA memory is fast but small 3 ns access time but
only 240 ki by .. 2.18 Mi by (Zynq 10 .. 45, the '45 is a corporate purchase). A ping
pong buffer arrangement could implement fading - computed in either processor (vector
instructions) or logic (raw muscle). The DAC input lines could supply the data.
Martin
-----Original Message-----
From: Guy Fedorkow via cctalk [mailto:cctalk@classiccmp.org]
Sent: 03 April 2024 16:02
To: cctalk(a)classiccmp.org
Cc: Guy Fedorkow <fedorkow(a)mit.edu>
Subject: [cctalk] Re: oscilloscopes
Vintage computer enthusiasts might want to keep track of where to find CRT-based analog
oscilloscopes, for use as output devices.
The early MIT and Lincoln Labs computers used D/A converters to steer and activate the
beam on analog scopes to draw vector images.
Working on Whirlwind simulation, we've been able to get this technique to work with
"real" oscilloscopes, e.g., Tek 475, but we have not yet found a single DSO that
has X/Y _and_ Z inputs (let alone the required phosphor fade).
Myself, I have a couple scopes with backups, so I'm not in the market for another
one. But others might consider the option...
/guy fedorkow
Date: Mon, 01 Apr 2024 22:33:38 +0000
From: Just Kant<kantexplain(a)protonmail.com>
Subject: [cctalk] oscilloscopes
To: "General Discussion: On-Topic and Off-Topic Posts"
<cctalk(a)classiccmp.org>
Message-ID: <xVKPUUPAeHv0_Gdsaq0al3Lji6a5VLAX2wRSUo5YhZwiADr5zn171nRBM
NxX-XE1ru7jPj1GFbu0aE3uBwUsVKxlU_XAYKX9B0x42zJZADc=(a)protonmail.com>
Content-Type: text/plain; charset=utf-8
I have more then I need. All the working ones are HP w/color crts, and as far as older,
verifiably vintage tools (right down to the 680x0 processor in either) I have to admit I
favor them as a brand. Call we an oddball, weird egg, badges I wear with pride.
But who could resist the allure of the newer ultra portable, even handheld units (some
with bandwidth or sampling rates to 50mhz). I'm a big cheapo. But there's no real
reason to agonize over a 65 - 200$ or thereabouts acquisition. It's a bit tiring to
wade through the piles of availability. I favor a desktop unit, larger screen (but not
always, careful). But most of those need wall current I think? The convenience of a
handheld battery powered unit obviously has it's benefits.
I will always love and dote upon my color crt based HPs. But the damned things are so
heavy, so unwieldy. Judy-Jude knocked my 54111d over, hit the paved floor, shook the
house. And still works! Built to withstand an atomic bombardment.
10:38 a.m.
New subject: oscilloscopes
On Apr 3, 2024, at 12:28 PM, Martin Bishop via cctalk
<cctalk(a)classiccmp.org> wrote:
Ignore my last - incontinence or is it incompetence
A fairly ordinary GPU, in a PC, could almost certainly provide an XY display with Z fade
(long persistance phosphor). I use them for waterfall displays and they keep up - the
data does of course arrive by E'net.
Yes, and some emulations have done this, such as Phil Budne's famous work in SIMH. I
adopted some of those ideas for DtCyber (CDC 6000 emulation) and it works well, though the
timing is marginal given the graphics subsystem I use. That is WxWidgets -- changes are
SDL would do better and some day I will try that.
Equally, FPGAs / SOCs can implement frame buffers; eg
to output waterfall displays. The fading memory would have to be in DRAM, FPGA memory is
fast but small 3 ns access time but only 240 ki by .. 2.18 Mi by (Zynq 10 .. 45, the
'45 is a corporate purchase). A ping pong buffer arrangement could implement fading -
computed in either processor (vector instructions) or logic (raw muscle). The DAC input
lines could supply the data.
Agreed, and that would be an elegant way to emulate a CDC DD60. Or a GT40. You'd
presumably want to use at least an HD level display signal (1920 by 1080), if not double
that, to make it look right; less than that would give pixel artefacts that make it not
look as vector-like as you would want.
paul
6 Apr
6 Apr
8:40 a.m.
New subject: oscilloscopes
Paul Koning wrote:
Yes, and some emulations have done this, such as Phil
Budne's famous work in SIMH.
Famous?? I'm famous???!!!
To be fair, I started with Douglas W. Jones' PDP8 Emulator.
Which reminds me of:
If I have seen farther than others, it is because I was standing on the shoulders of
giants. -- Isaac Newton
In the sciences, we are now uniquely privileged to sit side by side with the giants on
whose shoulders we stand. -- Gerald Holton
If I have not seen as far as others, it is because giants were standing on my
shoulders. -- Hal Abelson
In computer science, we stand on each other's feet. -- Brian K. Reid
It was certainly an awakening to the inherent parallelism of "analog"
natural processes... I wrote and tuned the code twenty years ago, but
haven't looked at whether better results might be possible by wasting
the capabilities of current systems (SIMD libaries and/or multiple
cores). I felt like I only was able to give a slim impression, and
not an immersive experience. I've also wondered what could be done
with 4K HDR displays: making points round(!) and simulating the
"bloom" and intensity of repeatedly or highly intensified spots.
phil
7 Apr
7 Apr
1:18 a.m.
New subject: oscilloscopes
Phil Budne wrote:
I wrote and tuned the code twenty years ago, but
haven't looked at
whether better results might be possible by wasting the capabilities
of current systems (SIMD libaries and/or multiple cores). I felt like
I only was able to give a slim impression, and I've also wondered what
could be done with 4K HDR displays: making points round(!) and
simulating the "bloom" and intensity of repeatedly or highly
intensified spots.
I have done some experiments in this direction. There are two sample
pictures here which are supposed to look similar to a P7 CRT.
https://github.com/larsbrinkhoff/crt-simulation
1:47 a.m.
New subject: oscilloscopes
I too have experimented with OpenGL to simulate phosphor-decay.
I never got to a satisfactory solution. The learning curve for OpenGL is
steep.
On Sun, Apr 7, 2024 at 4:43 PM Lars Brinkhoff via cctalk <
cctalk(a)classiccmp.org> wrote:
Phil Budne wrote:
I wrote and tuned the code twenty years ago, but
haven't looked at
whether better results might be possible by wasting the capabilities
of current systems (SIMD libaries and/or multiple cores). I felt like
I only was able to give a slim impression, and I've also wondered what
could be done with 4K HDR displays: making points round(!) and
simulating the "bloom" and intensity of repeatedly or highly
intensified spots.
I have done some experiments in this direction. There are two sample
pictures here which are supposed to look similar to a P7 CRT.
https://github.com/larsbrinkhoff/crt-simulation
10:51 a.m.
New subject: oscilloscopes
On Apr 6, 2024, at 11:40 AM, Phil Budne via cctalk
<cctalk(a)classiccmp.org> wrote:
Paul Koning wrote:
Well said, indeed!
Yes, and some emulations have done this, such as
Phil Budne's famous work in SIMH.
Famous?? I'm famous???!!!
To be fair, I started with Douglas W. Jones' PDP8 Emulator.
Which reminds me of:
If I have seen farther than others, it is because I was standing on the shoulders of
giants. -- Isaac Newton
In the sciences, we are now uniquely privileged to sit side by side with the giants on
whose shoulders we stand. -- Gerald Holton
If I have not seen as far as others, it is because giants were standing on my
shoulders. -- Hal Abelson
In computer science, we stand on each other's feet. -- Brian K. Reid It was certainly an awakening to the inherent
parallelism of "analog"
natural processes... I wrote and tuned the code twenty years ago, but
haven't looked at whether better results might be possible by wasting
the capabilities of current systems (SIMD libaries and/or multiple
cores). I felt like I only was able to give a slim impression, and
not an immersive experience. I've also wondered what could be done
with 4K HDR displays: making points round(!) and simulating the
"bloom" and intensity of repeatedly or highly intensified spots.
I did these things, in an emulation of the CDC 6600 console (DD60). It paints the
"dot" on the screen using a 2D Gaussian distribution around the nominal center,
with its parameters adjusted by the "focus" and "intensity" controls
just like on the original. And each visit of the spot is summed into the current screen
data using saturated arithmetic. So you get intensification at no extra charge, and if a
spot is drawn many times it also gets a bit blurrier due to the skirts of the Gaussian
distribution becoming more visible. At one point I had the spot as an RGB value with a
touch of blue in it, so the "bloom" would be bluer than normal lines. I took
that out because I don't remember what the real screen actually does. But clearly a
color shift like that could be simulated.
paul
11:01 a.m.
New subject: oscilloscopes
On 2024-04-06 9:40 a.m., Phil Budne via cctalk wrote:
Paul Koning wrote:
I need to write a emulator for a new cpu design I have. The 1086 cpu,
nice 20 bit cpu design from about 1968 using more modern parts. The
problem is the modern parts are just too fast and I have timing issues.
I can read/ write from memory using the front panel but not run code
for some reason. Jumps seem to work but HALT does not. This requires
a whole new bunch of PCB's with test points and other bug fixes, and I
have few weeks waiting for more parts.
I figure 90% of the code will be planning simple IO for this beast and
10% the emulator itself. I like real hardware that you can put a scope
too. All I can say modern programming is a about using a GUI and very
little about console IO. This is a bit change from using a PDP-8 with
TTY, and 4K focal.
Yes, and some emulations have done this, such as
Phil Budne's famous work in SIMH.
Famous?? I'm famous???!!!
To be fair, I started with Douglas W. Jones' PDP8 Emulator.
Which reminds me of:
If I have seen farther than others, it is because I was standing on the shoulders of
giants. -- Isaac Newton
In the sciences, we are now uniquely privileged to sit side by side with the giants
on whose shoulders we stand. -- Gerald Holton
If I have not seen as far as others, it is because giants were standing on my
shoulders. -- Hal Abelson
In computer science, we stand on each other's feet. -- Brian K. Reid
It was certainly an awakening to the inherent parallelism of "analog"
natural processes... I wrote and tuned the code twenty years ago, but
haven't looked at whether better results might be possible by wasting
the capabilities of current systems (SIMD libaries and/or multiple
cores). I felt like I only was able to give a slim impression, and
not an immersive experience. I've also wondered what could be done
with 4K HDR displays: making points round(!) and simulating the
"bloom" and intensity of repeatedly or highly intensified spots.
phil
2:58 a.m.
New subject: oscilloscopes
A little digging later ... I implemented the waterfall display eight years ago, outputing
to a 1280 x 1024 monitor. 1920 x 1080 was supported, but I was outputing 1 Ki pt FFTs.
The hardware platform was a Xilinx Zynq.
An indication of 4k video capabilities is
https://xilinx-wiki.atlassian.net/wiki/spaces/A/pages/2611216385/Zynq+Ultra…
Note the higher octane hardware used.
Martin
-----Original Message-----
From: Paul Koning [mailto:paulkoning@comcast.net]
Sent: 03 April 2024 18:38
<< snip >>
Equally, FPGAs / SOCs can implement frame buffers; eg
to output waterfall displays. The fading memory would have to be in DRAM, FPGA memory is
fast but small 3 ns access time but only 240 ki by .. 2.18 Mi by (Zynq 10 .. 45, the
'45 is a corporate purchase). A ping pong buffer arrangement could implement fading -
computed in either processor (vector instructions) or logic (raw muscle). The DAC input
lines could supply the data.
Agreed, and that would be an elegant way to emulate a CDC DD60. Or a GT40. You'd
presumably want to use at least an HD level display signal (1920 by 1080), if not double
that, to make it look right; less than that would give pixel artefacts that make it not
look as vector-like as you would want.
paul
29
days inactive
33
days old
26 comments
14 participants
participants (14)
-
Adrian Godwin -
ben -
Brent Hilpert -
Guy Fedorkow -
Lars Brinkhoff -
Martin Bishop -
Mattis Lind -
Michael Thompson -
Mike Katz -
Paul Koning -
Phil Budne -
Rick Bensene -
Tom Hunter -
Tom Uban