Hello everyone,
Back in the early 1980s, I had the unique opportunity to work at Capitol Records in the Toronto, Ontario area. While the company produced both tapes and records, my role was centred in the tape manufacturing plant.
At that time, quality control was a completely manual process—each product was reviewed by ear, making the results highly subjective. To improve consistency and reliability, three of us collaborated to design and implement a new quality control system. This system consisted of two custom cards: one plugged into an Apple ][+ computer running custom firmware, and the other was an analog board equipped with band-pass filters targeting the specific audio frequencies we wanted to monitor. The analog board interfaced directly with the digital board inside the Apple computer.
Remarkably, our system remained in use for over a decade, right up until pre-recorded tapes were eventually phased out.
Now that I’m retired, I’ve put together a video sharing more about these experiences. If you’re interested, you can watch it on YouTube here:
Thanks for reading!
That's pretty neat.
Tell us more about the supporting hardware.
You correctly indicated that the Apple II is incapable of analyzing the audio spectrum in real time and displaying it on the screen, so the supporting hardware must have done most of the heavy lifting.
Also, from the bootup sequence, this appears to be a DOS 3.3 based system, and at least the introductory screen seems to be written in AppleSoft, or am I wrong?
I'm curious about this
This project was originally based on the Eventide Clockworks card, which was a single-channel, real-time 1/3-octave spectrum analyzer. We redesigned it and laid out new boards using vellum and colored tape—both the digital board (which plugged into the Apple II) and the analog band-pass filter board.
Obviously, we couldn’t fit a full 1/3-octave spectrum for both channels, so we selected 15 band-pass frequencies that were most relevant to high-speed duplication—though in practice, we really only needed about 4 or 5.
The digital board was then modified to accept two analog channels from the filter board. Each channel went through a pair of AD7523s that controlled the signal gain before feeding it to the analog board.
The analog board used four CD4051 multiplexers, one for each frequency plus level of each channel, and the apple had to pick one at a time.
Both outputs were then routed back to the digital board to a pair of ADC0800 8-bit ADCs for the apple to read. Aside from various support chips, that was the core of the system.
You're exactly right—it was running DOS 3.3 and Applesoft. While the original Eventide design suggested using Integer BASIC, I didn’t want to reload Integer BASIC every time on a ][+, so we went with Applesoft instead. That decision didn’t affect the test tone readings, since those routines were written in assembly anyway.
On a personal note, I graduated from school in 1979 with very little computer experience, so this project was completely new to me. Fortunately, my company brought in someone to help get me up to speed with programming, which gave me the kickstart I needed.
Once I had the system running in the QC room, I convinced my Boss that I needed another Apple II for development. But my boss couldn’t just buy one outright—it wasn’t in the budget. So, thinking creatively, he said I could buy the parts and build one myself.
The original machine I bought from ComputerLand in Mississauga: a 48K Apple ][+ with a single floppy drive. It cost nearly $5,000 CAD at the time.
Back then, you couldn’t buy genuine Apple parts easily, so I tracked down a real Apple ][ Rev. 0 main board at a computer store in Toronto, along with photocopied diagrams showing what traces and jumpers were needed to work with North American monitors. I also bought a keyboard that claimed to be Apple-compatible—but wasn’t. I found someone to burn the ROMs for me, and after many late nights in my basement assembling everything, I powered it up—and it worked on the first try!
I still have that system today, mounted in a Hammond metal case, along with a pair of Quenton floppy drives I added later. Unfortunately I upgraded the rev 0 board to a clone board so that I sould have 4 colors available. Too bad, I wish I still had that board.
Oh and I replaced the keyboard with one from Keytronics which was desicned to look like an IBM PC keyboard complete with coily cord which I thought was cool. Recently as I was restoring the computer I really learned how crappy that keytronic board was with disintegrating foam keys. And I seem to recall it was around $400.00 usd.
Those really were fun times—even back then!
Very nice historical perspective video on YouTube. I appreciate you sharing this uncommon back-story!