DSP programming in Atari days

Hello! 

I'm new to DSP and programming, so I wonder how you guys did DSP programming in Atari days back in 80's. How did you figure out to do it? I guess education flow is something like that: 
1) Learn Assembly language; 
2) Get and learn Atari or IBM PC technical reference manuals; 
3) Read some DSP books; 
4) Start programming simple batch programs; 
Am I right? 


Can you please suggest some learning resources that were valuable back in a day? I was advised to start writing code for different platforms, so I inspired and wonder how dsp programming was done for Atari, IBM PC, etc.

Thanks! 

Michael 

<versa20@gmail.com> wrote:

>I'm new to DSP and programming, so I wonder how you guys did DSP
>programming in Atari days back in 80's. How did you figure out to do it?
>I guess education flow is something like that: 
>1) Learn Assembly language; 
>2) Get and learn Atari or IBM PC technical reference manuals; 
>3) Read some DSP books; 
>4) Start programming simple batch programs; 
>Am I right? 

That wasn't what it was like for me.

The go-to component .. if your project could afford it, was the
TRW 16-bit multiplier (later, multiplier-accumulator).  Another popular
component was the 2901 bit-slice.  Most requirements could be met
by surrounding one or both of these with appropriate circuitry.

Not much DSP was done on microprocessors, but projects did include
devices such as 1802's or 6809's.  Intel processors were avoided due 
to their jittery execution times.

Steve

On Wednesday, August 29, 2018 at 10:02:13 AM UTC+6, Steve Pope wrote:
> <versa20@gmail.com> wrote:
> 
> >I'm new to DSP and programming, so I wonder how you guys did DSP
> >programming in Atari days back in 80's. How did you figure out to do it?
> >I guess education flow is something like that: 
> >1) Learn Assembly language; 
> >2) Get and learn Atari or IBM PC technical reference manuals; 
> >3) Read some DSP books; 
> >4) Start programming simple batch programs; 
> >Am I right? 
> 
> That wasn't what it was like for me.
> 
> The go-to component .. if your project could afford it, was the
> TRW 16-bit multiplier (later, multiplier-accumulator).  Another popular
> component was the 2901 bit-slice.  Most requirements could be met
> by surrounding one or both of these with appropriate circuitry.
> 
> Not much DSP was done on microprocessors, but projects did include
> devices such as 1802's or 6809's.  Intel processors were avoided due 
> to their jittery execution times.
> 
> Steve

I mean software side of Digital Signal Processing (e.g. how to make FFT, audio effect, etc), not hardware. And I would like to know knowlegde flow.

How did people get knowledge at that time? How did they understand how to correctly write programs that made digital signal processing?

On Wed, 29 Aug 2018 04:02:10 +0000 (UTC), spope384@gmail.com (Steve
Pope) wrote:

><versa20@gmail.com> wrote:
>
>>I'm new to DSP and programming, so I wonder how you guys did DSP
>>programming in Atari days back in 80's. How did you figure out to do it?
>>I guess education flow is something like that: 
>>1) Learn Assembly language; 
>>2) Get and learn Atari or IBM PC technical reference manuals; 
>>3) Read some DSP books; 
>>4) Start programming simple batch programs; 
>>Am I right? 
>
>That wasn't what it was like for me.
>
>The go-to component .. if your project could afford it, was the
>TRW 16-bit multiplier (later, multiplier-accumulator).  Another popular
>component was the 2901 bit-slice.  Most requirements could be met
>by surrounding one or both of these with appropriate circuitry.
>
>Not much DSP was done on microprocessors, but projects did include
>devices such as 1802's or 6809's.  Intel processors were avoided due 
>to their jittery execution times.
>
>Steve

I remember when TI came in to show our engineering department their
brand new TMS320 (32010 ?) DSP chips (40 pin DIP ?)  which we used for
evaluating ECGs in our stress test systems at Quinton Instruments in
the early-ish 1980s.  That was a pretty cool time !

boB

versa20@gmail.com writes:

> On Wednesday, August 29, 2018 at 10:02:13 AM UTC+6, Steve Pope wrote:
>> <versa20@gmail.com> wrote:
>> 
>> >I'm new to DSP and programming, so I wonder how you guys did DSP
>> >programming in Atari days back in 80's. How did you figure out to do it?
>> >I guess education flow is something like that: 
>> >1) Learn Assembly language; 
>> >2) Get and learn Atari or IBM PC technical reference manuals; 
>> >3) Read some DSP books; 
>> >4) Start programming simple batch programs; 
>> >Am I right? 
>> 
>> That wasn't what it was like for me.
>> 
>> The go-to component .. if your project could afford it, was the
>> TRW 16-bit multiplier (later, multiplier-accumulator).  Another popular
>> component was the 2901 bit-slice.  Most requirements could be met
>> by surrounding one or both of these with appropriate circuitry.
>> 
>> Not much DSP was done on microprocessors, but projects did include
>> devices such as 1802's or 6809's.  Intel processors were avoided due 
>> to their jittery execution times.
>> 
>> Steve
>
> I mean software side of Digital Signal Processing (e.g. how to make
> FFT, audio effect, etc), not hardware. And I would like to know
> knowlegde flow.
>
> How did people get knowledge at that time? How did they understand how
> to correctly write programs that made digital signal processing?

For me, the first and foremost component was study. I took classes in
mathematics, e.g., linear algebra - an FIR filter can be viewed as an
inner product, probability theory, complex variables [complexvariables],
applied math, stochastic processes, and engineering. One foundational
class was based on [signalsandsystems], and that is where I learned
about all the transforms, including Fourier, the DFT, the DTFT, and
z-transforms. I also had courses in estimation [schwartz], adaptive
filtering [widrow], Optimal Filters (Anderson and Moore), digital signal
processing (Stanley), modern control systems [brogan], etc.

Then I bought a PC (286) computer in the late 80s and wrote a few of my
own routines, e.g., a signal plotter, noise generator, filter, FFT,
using C. That was instructive. I remember modeling

At least that's how it started for me. IOW, a helluva lot of work..

@BOOK{complexvariables,
  title = "{Complex Variables and Applications}",
  author = "{Ruel~V.~Churchill, James~Ward~Brown}",
  publisher = "McGraw-Hill",
  edition = "fifth",
  year = "1990"}

@BOOK{signalsandsystems,
  title = "{Signals and Systems}",
  author = "{Alan~V.~Oppenheim, Alan~S.~Willsky, with Ian~T.~Young}",
  publisher = "Prentice Hall",
  year = "1983"}

@book{schwartz,
  title = "Signal Processing: Discrete Spectral Analysis, Detection, and Estimation",
  author = "{Mischa~Schwartz and Leonard~Shaw}",
  publisher = "McGraw-Hill",
  year = "1975"}

@book{widrow,
  title = "Adaptive Signal Processing",
  author = "Bernard Widrow and Samuel D. Stearns",
  publisher = "Prentice-Hall",
  edition = "first",
  year = "1985"}

@book{brogan,
  title = "Modern Control Theory",
  author = "William L. Brogan",
  publisher = "Prentice-Hall",
  edition = "second",
  year = "1985"}

-- 
Randy Yates, DSP/Embedded Firmware Developer
Digital Signal Labs
http://www.digitalsignallabs.com

Randy Yates <yates@digitalsignallabs.com> writes:

> I remember modeling

... the Gibbs phenomenom by adding more and more sinusoidal
terms for a square wave.

But even in the late 80s the 286 was slow - it took over a minute
to do a 1024-point FFT. Adding a 287 coprocessor took the time
down to 10 seconds (which I was elated about!).

--Randy
>
> At least that's how it started for me. IOW, a helluva lot of work..
>
> @BOOK{complexvariables,
>   title = "{Complex Variables and Applications}",
>   author = "{Ruel~V.~Churchill, James~Ward~Brown}",
>   publisher = "McGraw-Hill",
>   edition = "fifth",
>   year = "1990"}
>
> @BOOK{signalsandsystems,
>   title = "{Signals and Systems}",
>   author = "{Alan~V.~Oppenheim, Alan~S.~Willsky, with Ian~T.~Young}",
>   publisher = "Prentice Hall",
>   year = "1983"}
>
> @book{schwartz,
>   title = "Signal Processing: Discrete Spectral Analysis, Detection, and Estimation",
>   author = "{Mischa~Schwartz and Leonard~Shaw}",
>   publisher = "McGraw-Hill",
>   year = "1975"}
>
> @book{widrow,
>   title = "Adaptive Signal Processing",
>   author = "Bernard Widrow and Samuel D. Stearns",
>   publisher = "Prentice-Hall",
>   edition = "first",
>   year = "1985"}
>
> @book{brogan,
>   title = "Modern Control Theory",
>   author = "William L. Brogan",
>   publisher = "Prentice-Hall",
>   edition = "second",
>   year = "1985"}

-- 
Randy Yates, DSP/Embedded Firmware Developer
Digital Signal Labs
http://www.digitalsignallabs.com

Am 29.08.2018 um 06:02 schrieb Steve Pope:

> 
> That wasn't what it was like for me.
> 
> The go-to component .. if your project could afford it, was the
> TRW 16-bit multiplier (later, multiplier-accumulator).  Another popular
> component was the 2901 bit-slice.  Most requirements could be met
> by surrounding one or both of these with appropriate circuitry.
> 
> Not much DSP was done on microprocessors, but projects did include
> devices such as 1802's or 6809's.  Intel processors were avoided due
> to their jittery execution times.
> 
> Steve


That's how I did a FIR filter in 1984:

< 
http://www.hoffmann-hochfrequenz.de/project_gallery/project_gallery.html 
               >

It was a Co-processor to a 8085; my customer used Intel/Siemens/AMD
8080 / 8085 / 8086, the Interface was Siemens SMP Bus.

In contrary to that said  above, Intel was definitely the No. 1
because they had 2nd sources. I was even unable to lure my customers
to Z80 that still sorta belonged to the family.

regards, Gerhard

(I admit, my web site needs an update desperately.)

On Tue, 28 Aug 2018 22:25:59 -0700 (PDT), versa20@gmail.com wrote:

>On Wednesday, August 29, 2018 at 10:02:13 AM UTC+6, Steve Pope wrote:
>> <versa20@gmail.com> wrote:
>> 
>> >I'm new to DSP and programming, so I wonder how you guys did DSP
>> >programming in Atari days back in 80's. How did you figure out to do it?
>> >I guess education flow is something like that: 
>> >1) Learn Assembly language; 
>> >2) Get and learn Atari or IBM PC technical reference manuals; 
>> >3) Read some DSP books; 
>> >4) Start programming simple batch programs; 
>> >Am I right? 
>> 
>> That wasn't what it was like for me.
>> 
>> The go-to component .. if your project could afford it, was the
>> TRW 16-bit multiplier (later, multiplier-accumulator).  Another popular
>> component was the 2901 bit-slice.  Most requirements could be met
>> by surrounding one or both of these with appropriate circuitry.
>> 
>> Not much DSP was done on microprocessors, but projects did include
>> devices such as 1802's or 6809's.  Intel processors were avoided due 
>> to their jittery execution times.
>> 
>> Steve
>
>I mean software side of Digital Signal Processing (e.g. how to make FFT, audio effect, etc), not hardware. And I would like to know knowlegde flow.
>
>How did people get knowledge at that time? How did they understand how to correctly write programs that made digital signal processing?

I wrote a pretty subtantial library of DSP functions including FFTs,
Hartley transforms, filters, etc., etc., in both C and 68000 assembly
language on the Amiga back in those days.

Meanwhile at work we were making SAR radar processors with discrete
Fairchild FAST TTL parts, 22V10 PALs, and whatever else we could make
work.   The FFT butterflies used the TRW multipliers that Steve
mentioned, but most everything else was discrete processing, ROM LUTs,
and PALs.

Then TI came out with a DSP, that was pretty clunky.   Motorola had
the 56000 and ADI had the 2101/2105 series, which I used a lot.

Getting knowledge was the same as anything.   You could take graduate
classes in DSP, there were a lot of good books.   Oppenheim and Shafer
was already out at that time.

On Wednesday, August 29, 2018 at 3:37:55 PM UTC+12, ver...@gmail.com wrote:
> Hello!=20
>=20
> I'm new to DSP and programming, so I wonder how you guys did DSP programm=
ing in Atari days back in 80's. How did you figure out to do it? I guess ed=
ucation flow is something like that:=20
> 1) Learn Assembly language;=20
> 2) Get and learn Atari or IBM PC technical reference manuals;=20
> 3) Read some DSP books;=20
> 4) Start programming simple batch programs;=20
> Am I right?=20
>=20
>=20
> Can you please suggest some learning resources that were valuable back in=
 a day? I was advised to start writing code for different platforms, so I i=
nspired and wonder how dsp programming was done for Atari, IBM PC, etc.
>=20
> Thanks!=20
>=20
> Michael

The theory has been known for a very long time. We only lacked the computin=
g power. TMS320 was the first half decent one and yes we used assembler fix=
ed point. A nightmare. Eventually c cross compilers came along. Man many ye=
ars later. In fact I cannot recall when. Maybe 90s. We simulated stuff a lo=
t and developed theory and could use mini-computers. I had  Masscomp for in=
stance in the late 80s.

=D1=81=D1=80=D0=B5=D0=B4=D0=B0, 29 =D0=B0=D0=B2=D0=B3=D1=83=D1=81=D1=82=D0=
=B0 2018 =D0=B3., 19:45:19 UTC+6 =D0=BF=D0=BE=D0=BB=D1=8C=D0=B7=D0=BE=D0=B2=
=D0=B0=D1=82=D0=B5=D0=BB=D1=8C Eric Jacobsen =D0=BD=D0=B0=D0=BF=D0=B8=D1=81=
=D0=B0=D0=BB:
> On Tue, 28 Aug 2018 22:25:59 -0700 (PDT), versa20@gmail.com wrote:
>=20
> >On Wednesday, August 29, 2018 at 10:02:13 AM UTC+6, Steve Pope wrote:
> >> <versa20@gmail.com> wrote:
> >>=20
> >> >I'm new to DSP and programming, so I wonder how you guys did DSP
> >> >programming in Atari days back in 80's. How did you figure out to do =
it?
> >> >I guess education flow is something like that:=20
> >> >1) Learn Assembly language;=20
> >> >2) Get and learn Atari or IBM PC technical reference manuals;=20
> >> >3) Read some DSP books;=20
> >> >4) Start programming simple batch programs;=20
> >> >Am I right?=20
> >>=20
> >> That wasn't what it was like for me.
> >>=20
> >> The go-to component .. if your project could afford it, was the
> >> TRW 16-bit multiplier (later, multiplier-accumulator).  Another popula=
r
> >> component was the 2901 bit-slice.  Most requirements could be met
> >> by surrounding one or both of these with appropriate circuitry.
> >>=20
> >> Not much DSP was done on microprocessors, but projects did include
> >> devices such as 1802's or 6809's.  Intel processors were avoided due=
=20
> >> to their jittery execution times.
> >>=20
> >> Steve
> >
> >I mean software side of Digital Signal Processing (e.g. how to make FFT,=
 audio effect, etc), not hardware. And I would like to know knowlegde flow.
> >
> >How did people get knowledge at that time? How did they understand how t=
o correctly write programs that made digital signal processing?
>=20
> I wrote a pretty subtantial library of DSP functions including FFTs,
> Hartley transforms, filters, etc., etc., in both C and 68000 assembly
> language on the Amiga back in those days.
>=20
> Meanwhile at work we were making SAR radar processors with discrete
> Fairchild FAST TTL parts, 22V10 PALs, and whatever else we could make
> work.   The FFT butterflies used the TRW multipliers that Steve
> mentioned, but most everything else was discrete processing, ROM LUTs,
> and PALs.
>=20
> Then TI came out with a DSP, that was pretty clunky.   Motorola had
> the 56000 and ADI had the 2101/2105 series, which I used a lot.
>=20
> Getting knowledge was the same as anything.   You could take graduate
> classes in DSP, there were a lot of good books.   Oppenheim and Shafer
> was already out at that time.

Thank you! I see these books you mentioned is for understanding of DSP. But=
 what resources did you use to learn programming? Do manual were distribute=
d with computers? How these manuals were useful for learning programming?