Commodore Datasette Does Its Own Calibration

Ah, the beloved Commodore 64. The “best-selling computer system of all time”. And hobbyists are keeping the dream alive, still producing software for it today. Which leads us to a problem with using such old equipment. When you get your copy of Petscii Robots on cassette, and try to fastload it, your machine might just consistently fail to load the program. That’s fine, time to pull out the cue-tips and rubbing alcohol, and give the read heads a good cleaning. But what if that doesn’t do the job? You may just have another problem, like tape speed drift.

There are several different ways to measure the current tape speed, to dial it in properly. The best is probably a reference cassette with a known tone. Just connect your frequency counter or digital oscilloscope, and dial in the adjustment pot until your Datasette is producing the expected tone. Oh, you don’t have a frequency counter? Well good news, [Jan Derogee] has a solution for you. See, you already have your Datasette connected to a perfectly serviceable frequency counter — your Commodore computer. He’s put out a free program that counts the pulses coming from the Datasette in a second. So play a reference cassette, run the program, and dial in your Datasette deck. Simple! Stick around after the break for a very tongue-in-cheek demonstration of the problem and solution.

41 thoughts on “Commodore Datasette Does Its Own Calibration

      1. 1982-09 to 1994-04 Comadore 64 12.5 – 17 million units sold (PAL and NTSC versions)

        I think that “best-selling computer system of all time” still holds, because even though the Raspberry Pi since 2012 has sold more than 40 million units worldwide. But is not a single product that would be Raspberry Pi B, RPi A, RPi B+, RPI A+, RPi 2 B, RPi Zero, Zero W/WH, Zero 2 W, 3 B, 3 A+, 3 B+, 4 B, 4 400, Pico.

      2. Commodore Computers. Long since defunct. For a while there, Commodore was flying high, selling millions of C-64s. It was way ahead of it’s time, but the CEO Jack Tramile ran the company into the ground. Their last PC was the Commodore Amiga, which was an awsome PC, but the company faltered shortly after releasing the Amiga, and a few years later Commodore was no more.

        1. “Ackshually” Jack *Tramiel left Commodore and went to Atari, around the same time that Commodore bought Amiga. The company was run into the ground by the management that lost all sense of direction, marketing teams that had no idea what people wanted (Commodore 65?) and overworked tech teams that lost all motivation.

          Read Brian Bagnall’s book series about Commodore, it’s very interesting.

      1. How do you DIY if you do not have a tape deck that is freshly calibrated to record the tone on?

        Regarding the original question: I found some tapes for sale on test-tape.com but yikes!, they cost €39.90, which is all right if you are going to calibrate more than once but a bit pricey if you just want fix your own datasette.

  1. “Ah, the beloved Commodore 64. The “best-selling computer system of all time”. And hobbyists are keeping the dream alive, still producing software for it today. ”

    Let’s just hope all the dissatisfied C64 users won’t tell their stories here. Otherwise, the comments section would be full. ;) Seriously, the C64 had so many flaws. I don’t know were to start.

    The somewhat limited Basic v2? The noisy VIC II? The noisy RF box? The double broken serial floppy interface? The easily broken PLAs? The easily broken SID (paddle interface)? The missing light-barrier in the 1541 floppy drives? The hard-coded firmware that bumps the drive heads against the wall to make sure it’s at sector 0? The limited RAM inside the 1541s that prevented a complete sector read/write in a single pass (used by GEOS as copy protection)? The missing reset button (had to be installed by users manually)? The elephant feet PSUs that always broke due to overheating? The high amount of RFI that made the C64 hard to use as a RTTY computer among radio enthusiasts? Etc etc.

    “Which leads us to a problem with using such old equipment. When you get your copy of Petscii Robots on cassette, and try to fastload it, your machine might just consistently fail to load the program. ”

    Perhaps also because the copy is stored on a low-end musical cassettes instead of real datasettes with the correct ferro-magnetic surface?

    I remember from back in the day that music cassettes did have the same standards as datasettes/computer tapes. My computer was a Sharp model with an 1200 Baud datasette not too different from the Commodore datasette.

    A new, fresh music cassette from the super market had trouble saving my programs, whereas as a ten year old datasette (labeled as computer cassette) worked flawlessly.

      1. Isn’t that the point of the “/track0” signal on the standard PC floppy drive cable? To be able to know when to stop and not rely on skipped steps when it reaches the end of travel?

        1. I could see a couple of arguments about this. First, it seems that some manufacturers (Apple) wanted to save a few pennies by not implementing this (on either the drive or the controller). Also, it’s not clear if reaching track 0 according to the sensor is the same as having the head physically aligned with track 0. (Although, if you saw the track 0 signal come on, you’d know you wouldn’t have to step much more to hit the physical stop.)

      2. Most computers just relied on the head stepper alignment and used the index hole to locate sector 0 and track 0. IIRC there may also have been an optical sensor on full height 5.25″ drives so they could tell when the heads were all the way towards the outside of the disk.

    1. Ahhh… those were the days. But don’t forget all brands had problems of some kind, it was the dawn of home (affordable computing (though, still expensive).

      Strangely I also have fond memories of the playground, although many times I’ve been pushed of swing, fell of the stairs leading towards the slide. Ran across the field, but overlooked a small hole, stepped into it limped for a week. Tripped over the items that belonged in the sandbox. Don’t ask me about the seesaw, getting on to it was a problem, getting of it too. I never won anything at the yearly bingo and somehow the weather was either too hot or too wet. But I kept coming back. I guess it was because there just wasn’t anything better around at the time.

    2. What design is not without flaws? Bill Herd (I think he might lurk around here…) Et Al. had a lot of constraints when designing these early systems, as well as quite a bit of pressure to get things out the door rather than make them perfect because of how fast paced the industry was at the time.

      Many programmers cut their teeth programming the C64 and others (look at how the Apple II line found sector 0) despite their flaws and I shudder to think how the modern Internet and tech would look had those engineers stopped to make the “perfect” rather than the “great” and never made it to market.

      1. So in short, the designer did a botch job due to time constraints. That’s understandable. I don’t blame them.

        Gratefully, the PRESTEL standard from the 1970s did predate the VC20/C64 times. PRESTEL was what inspired BTX/Videotex/Minitel here in Europe, afaik.
        It was a true forerunner to the internet/www to most people.
        (The businesses used X.25 links like Datex-P.)

        Fun fact: The C64 didn’t meet the official minimum requirements for BTX, by the way. It couldn’t display the complete colour palette, afaik.

        https://en.wikipedia.org/wiki/Videotex

        https://en.wikipedia.org/wiki/Bildschirmtext

        On the other hand, PlayNet/Q-Link from the mid-80s likely never existed without the C64/C128. Or AOL.
        CompuServe and Prodigy on the other hand, were never dependable of the Commodore machines.

  2. A little story by me about the Datasette was published in the April 1998 issue of C=Hacking ( https://codebase64.org/doku.php?id=magazines:chacking15 ). Perhaps you all may find it interesting. (And the software lead/manager for the project had a masters in music and had worked for a company that developed educational software for Commodore 64s. He, of course, did the music and music programming for the applications. He eventually ended up at our company.) Here’s the story:

    ***** From 1998 *****

    Our company is building the ground control system for AT&T’s next Telstar satellite, to be launched in May. (As you might have heard, one of the existing Telstar satellites was killed off by the solar magnetic burst in January.) Although the telemetry rate is fairly slow, the system is pretty high-tech: Solaris workstations, VME PowerPCs running LynxOS, and special hardware for encoding/decoding the command and telemetry data streams.

    We’ve been flowing simulated telemetry data through the system, but a couple of weeks ago, we were provided with actual spacecraft telemetry (from ground tests of the satellite). When we opened the box that had been shipped to us, we found: an ordinary looking cassette tape and an old COMMODORE tape drive with a little adapter box for the type of cable we use! We plugged it in, pushed the PLAY button, and watched as our system locked on to real telemetry data.

    (The tape drive was one of those about the size of a small book with rounded edges. That must have been a newer model than the big, boxy drive I had on my VIC-20.)

  3. Too many comments are off the subject, including data vs music tapes — really? No mention of high bias vs chromium vs basis ferric oxide. And no one mentioned that what you really need to do is “transcribe” the old crappy tapes to digital then you can probably store all the C64 tapes ever made on a 4GB micro SD and never have to worry about that nonsense ever again. Even on a crappy transport, you can digitize and capture the audio and post-process it using “modern” tools. While it’s simple enough to do simple speed shifts, but onerous to fix wow and flutter (W&F) require more serious effort to overcome the (pseudo) random speed variations. BTW, a transport with bad W&F can make it worse to play back in the same machine the audio was recorded on, since the W&F is not correlated with previous recordings, unlike simple speed variations due to speed differences between transports. Science!

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.