Friday Hack Chat: Reverse Engineering the Digital Compact Cassette

For this week’s Hack Chat, we’re talking about reverse engineering the Digital Compact Cassette. Why should we care about an obsolete format that was only on the market for four years?  Because if a copy of the Spin Doctor’s Pocket Full of Kryptonite costs $50 USD on the used market, it has to be good.

In the early 1990s, several different digital magnetic tape formats came onto the scene. The MiniDisc was magneto-optical, yes, but back in the day it was amazing for recording bootlegs. DAT also appeared in the early 90s, and it was a godsend for recording studios. There was another format introduced in 1992, the Digital Compact Cassette. It was backward compatible with standard audio cassettes, an important feature, because no one would want to replace their entire cassette-based music collection with a new-fangled digital format. That would be just lunacy.

Our guest for this week’s Hack Chat will be [Jac Goudsmit], prolific creator on Hackaday.io, with projects ranging from the L-Star Software Defined 6502 Computer to a GPS Controlled FischerTechnik Clock. [Jac] grew up on a PET 2001, and in the years since he’s worked on projects ranging from motion control systems for lithography equipment, pick and place machines, and even at a Radio Shack. In this Hack Chat, he’ll be discussing the history of the Digital Compact Cassette, the behind the scenes on how stereo PCM is recorded to tape, and other topics like the difference between CS/EE careers in the Netherlands and the USA.

join-hack-chatOur Hack Chats are live community events on the Hackaday.io Hack Chat group messaging. Our Hack Chats usually happen on Fridays at Noon, so buckle up because this is going down Friday, December 1, at 12:00 PST. What time is that where you live? Who cares! Here’s a time zone converter!

Click that speech bubble to the left, and you’ll be taken directly to the Hack Chat group on Hackaday.io.

You don’t have to wait until Friday; join whenever you want and you can see what the community is talking about.

31 thoughts on “Friday Hack Chat: Reverse Engineering the Digital Compact Cassette

  1. IIRC, this format was killed by Big Music because they feared lossless copying. Which was to emerge later anyway with MP3 (although the format isn’t lossless, duplicating the files is)

    1. I also remember something about “for Professional Use Only”…cassettes and recording equipment were not available retail, you had to buy them from some special outlet. Playback only equipment was generally available, though priced so as not to sell.

    2. Lossless copying came about with CD-R before that.

      Which is why the music industry collectively clapped their hands when Apple came up with FairPlay.

      The reason why MiniDisc or DAT never became a hit with regular consumers is because of Sony’s suicidal licensing policies – they’ve killed more media formats than anyone else has invented.

      1. Minidisc was popular outside the USA, but since most of the tech press is in the USA, when they write those “Top 10 formats that failed” type things they always throw Minidisc in there because they are too lazy to do research.

        1. Every country has that problem to a certain extent but the US do seem to be the ones that seem to care the least about bothering to find out how things work outside of their own state let alone in other countries.

          1. The “It’s a small world after all” syndrome, where everyone else in the world is basically just an American with a different hat. They really seem to be like that, not realising that American culture actually is American culture, and not just the way all human beings have always done everything.

        2. No, the minidisc died elsewhere as well because sony refused to let anyone else make the players or the discs until it was too late and the MP3 players started taking over the market.

          Sony could have made the MD into a generic data device as well as a recording format at the time when everyone was trying to come up with a good alternative to the diskette – it was vastly better than the zip/jaz drives etc. but they simply had their heads so far up their asses and dollar signs in their eyes that they couldn’t see they were killing it.

          1. @Bitblade – In the US or not, the fact that people see a mere 10 years as being long enough of a survival time to consider a media format as being successful says a lot about what is wrong with the world. This is media, it’s the stuff that our culture is stored on. Wake me when someone once again comes up with something that might last 50 years at least.

      2. “Lossless copying came about with CD-R before that.”

        CD-R recording was developed by Philips around the same time as DCC. The first CD recorder (as far as I know) was a Yamaha; it was a rackmounted device, 19″ wide and probably about 10″ high. It recorded at single speed and was very unreliable.

        The second CD-recorder was the Philips CD-521 which I worked with in 1994 and 1995 but around that time it was already old. It was also unreliable and it tended to turn your discs into coasters if your computer was even THINKING of not keeping it fed with data.

        In the late 1990s CD-recorders finally started to come down in price and started getting more reliable. By that time, recorders also started to appear that didn’t have the buffer underrun problem. I was a beta tester for Plextor for a while; they featured “Burn-PROOF” which made it possible to continue after a buffer underrun.

        Audio CD-R recorders were never very popular because unlike computer recorders, they needed the more expensive “audio only” discs.

        DAT predates DCC and was the reason why Philips developed DCC: They wanted to have a format that would be less fragile than DAT and would be interesting to people who already had a large analog tape collection. They failed to think ahead in the way Sony did with MiniDisc.

        I don’t think the music industry killed DCC. It simply wasn’t a very user friendly format and consumers (outside the Netherlands) weren’t buying it. There were no recorders that could record analog cassettes so you still had to keep your old cassette deck handy to record analog cassettes for your portable player or for the car. And if you wanted to record more than 45 minutes of continuous music, you couldn’t do it (I know many people who just recorded long continuous music such as radio shows by using hifi VHS).

    3. DCC was a Philips invention. At that time, Philips / Phonogram was one of the biggest music company conglomorates. But Minidisc was Sony’s and they were expanding their own music & entertainment companies and that kinda sealed the fate for both formats. Lesson learned: avoid format wars if at all possible. They both made billions on the compact cassette and discs.

    4. no, sorry, you’re wrong. both DCC and MD (minidisc) used LOSSY compression, so even spdif back to back copies (if you diabled scms) would still be NOT bit-perfect.

      DAT was the only digital format at the time (and in non-video tape format) that was lossless.

      but let me tell you a story about buzzsaws and tape tension. see, DAT was for audio and DDS was for data. tapes were nearly identical. audio used 60meter (2 hour) tapes but later DDS drives could take 90m and 120m length tapes. thinner tapes. tapes that needed custom tension settings. guess what did NOT have a variable tension system? yes, consumer and even pro dat decks. what you got when the tape ‘slipped’ was a digital buzzsaw. very ruining for the music on it; usually live music.

      add to this that dat decks needed annual alignment and cleaning, and that most consumers never did this (even studios often skipped it for long periods of time) – and you have a mixture for all kinds of failure.

      oh, and dat tape decks sold to consumers would not allow recording of analog to 44.1, only 32 and 48 (32lp was a sony mode, most didn’t implement that, btw). ‘pro decks’ could record to 44.1 Fs and that’s the same value used to master redbook audio cd’s. coincidence? hmmm. consumer pany decks (like sv3200) was about half the price of the pro version (sv3700) and the only real diff was that lock-out stuff and the front panel color ;)

      I’m so glad dat is dead. then again, I still have my sv3700 which is badly in need of a CLA and no one is left in the US that I know of who still CAN or WILL do that kind of work.

      1. “even spdif back to back copies (if you diabled scms) would still be NOT bit-perfect.”

        On a side note: on the DCC-L discussion list in the late 1990s, the question often came up how much loss there would be if you just kept copying one DCC cassette to the next, generation after generation. It is now clear to me that starting with the second generation, the audio quality should not get any worse.

        This is why: The PASC compression basically first converts the PCM data stream to a frequency based datastream (like a frequency analyzer or a waterfall diagram on a software defined radio, pretty much). Then it figures out what frequencies you can’t hear anyway, and it leaves those frequencies out of the recording, while recording the other frequencies with greater precision. When it reads this signal back from tape, there is no “decompression” step as such; it’s basically just synthesizing the PCM signal from the frequency-based signal, and all the things you can’t hear, aren’t synthesized. If you send that signal back to a PASC encoder in another recorder, it basically finds out that there is nothing to hide and it records the signal with full precision. I don’t have proof for this but this is implied by the way the PASC (or MPEG-1 layer 1) encoding works.

        “oh, and dat tape decks sold to consumers would not allow recording of analog to 44.1, only 32 and 48”

        That’s the first time I heard that. I’ve worked with several different DAT decks and they could all record at 44,1kHz from analog as well as digital sources. Professional decks would have EBU connectors for the digital audio (and would ignore SCMS) and consumer decks would use S/PDIF (with SCMS).

        By the way DCC could ONLY record at 44.1 from analog sources, and SCMS allowed unrestricted further copying of tapes recorded in analog mode.

  2. i never messed with DCC, but I was a SONY fanboy. DAT was introduced as a high-end linear digital system, but the main adopters were professionals, who wanted something smaller and cheaper than PCM adaptors and U-Matic VCRs. FOSTEX managed to fit SMPTE timecode into the DAT format, which finally made it more useable for location and post -production sound.

    I still have some MiniDisc gear (nostalgia) . It was a great hobby field format (eg bootlegging as previously mentioned) or just some fast street recording.

  3. Two things helped kill DCC.

    Serial copy protection. You could make only 1st generation digital copies of a DCC. None of the consumer/prosumer recorders would allow 2nd generation copies.

    No analog recording capability. DCC decks could only *play* Compact Cassettes. Audiophiles wanted to *replace* their analog decks, not add another big lump of gear to their systems. So if they still wanted to record to analog tape, they had to hand onto their old deck.

    Were there ever any in-dash DCC players for vehicles? There were some walkman style machines so a car player would have been possible. If they’d gotten Ford, GM, Chrysler, VW etc on board with DCC…

    1. there were car dat decks, though! a friend and I both had the sony dat car decks (pullout style, that’s how old it was). fragile and such, but it actually worked well even though the car is harsh for vcr-like spinning heads.

      I didn’t know anyone who was into dcc or md; we all were into dat and ran ‘tape trees’ to send digital copies of live shows to our tree ‘children’. it was quite the thing back in the early 90’s.

    2. I think SCMS was widely regarded as a necessary evil and didn’t keep people from buying digital audio equipment. So you couldn’t make a digital copy of another digital copy; it was easy to just copy the second copy via analog, and it would sound just as good if you were careful not to let the signal clip. And it was well documented and easy to circumvent. In the Netherlands, Elektor magazine published several articles with SCMS defeaters.

      The inability to record analog cassettes is related to the heads they used in DCC. The bias currents needed to record analog tapes would have destroyed the fragile magneto-resistive heads, and a separate erase head would be needed for which there was no space (DCC doesn’t need an erase head). I agree, if they would have managed to make DCC recorders record analog cassettes (in analog format), it would have given a major boost to sales because ironically, even though DCC was compatible with analog cassettes, the people who would be most interested in them (i.e. the people who owned a lot of cassettes) would also feel forced to keep the analog recorder in addition to the digital one!

      Another factor in the demise of DCC was that the longest contiguous recording was 45 minutes. Then the tape would have to reverse to the B side and that would always takes some time. If they would have made stationary recorders with stationary (i.e. non-pivoting) heads, and then add some smart electronics to buffer the audio while the mechanism changed direction, so that continuous audio was possible, it would have made a lot more people happy. But this is technically hard (you can’t change the laws of physics and at recording time you would have to overlap some data from side A with side B) so they probably never would have done that. Instead they could have made a DCC-LongPlay format where they record the tape at half speed with a 192kbps MP3 stream. There are hints in the Technics RS-DC8 service manual that they may have been working on that, but there are no datasheets of the latest Matsushita chipset online.

      There were two different DCC car stereos: the DCC-811 and the DCC-822 (I own the DCC-824 which is basically a DCC-822 with a remote control added in the box; there was also a DCC-850 which was also the same as the 822 but with a CD changer and the remote control). I’m donating it to the DCC museum this weekend and we may post a video about it soon.

      1. Hi Jac.
        What also not helped DCC going was the fact that the way the pick-up head was made. at least not at the lodewijkstraat plant. the machine which made the radius on the head didnt function as promised. let alone the way the head was welded into it’s housing. the worst engineering I ever saw.

    1. Is there an Australian tape manufacturer or do they farm production out to National Audio Company in the USA? NAC is the only company in the USA still producing Compact Cassettes. They produced the movie tie-in Awesome Mix tapes for the Guardians of the Galaxy movies, and they do all the tapes for every company having tapes made for USA sales.

      As the other companies quit tapes, NAC sales *have never gone down*. They didn’t have a slump followed by a resurgence as tapes regained popularity, they just bought their equipment and took over whatever market the other guys still had. Last I heard, they were looking for some new people able to keep their old equipment running. Most of their employees are getting up there in years.

      HaD should see if they can get a factory tour at NAC. https://www.youtube.com/watch?v=CMTpvr9HXeI

  4. More info on DCC : https://www.youtube.com/watch?v=CkGMJBqZawA&t=6m20s

    And the last release on DCC… and then a new release in 2017, thanks to “Doctor DCC” and “The DCC Museum” : https://www.youtube.com/watch?v=unj79pzEHNQ&t=9m30s

    How they produced it was interesting, they had to use a portable DCC deck which had a computer hook-up option that let you transfer the audio tracks digitally and even encode title data just like a proper DCC release.

    Interestingly, that same album then later go released on ELCASET because why not? (It was previously released on MiniDISC, vinyl, cassette, CD, … ) : https://www.youtube.com/watch?v=lVyanU3GgG8

    Repair of the portable DCC deck mentioned in above video : https://www.youtube.com/watch?v=tCpXx181QsQ

    Techmoan is a great channel if you want to learn about esoteric or old audio or video systems.

  5. Correct. PASC is pretty much the same as MPEG-1 layer 1 (I should look up which came first but I have a funny feeling that MPEG-1 L1 was based on PASC) and SCMS is easy to defeat.

    The things I would like to do are: (1) reverse engineer the ITTS format to make it possible to see the text and pictures that are stored on prerecorded DCC’s, and (2) build a device that can format tapes the same way as prerecorded tapes, so when you want to fast forward or rewind to a different track, you can choose the track title to play, instead of just choose how many tracks backward or forward to go.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

w

Connecting to %s