Jac Goudsmit says: 4:08 PM The encoded audio signal also contains what they call SYSINFO. The exact format of the SYSINFO is not documented but it contains information such as filter settings and SCMS copy protection information. Jac Goudsmit says: 4:08 PM SCMS is the system that lets you copy a CD to a tape digitally but then denies you from copying the tape to another tape. Jac Goudsmit says: 4:09 PM SYSINFO also (probably) contains ITTS info which I'll get back to in a minute. Jac Goudsmit says: 4:10 PM So a DCC recorder uses 9 tracks: 8 for audio and 1 for auxiliary data (AUX). The AUX track runs much slower, about 12kbits per second. Christian says: 4:10 PM Are these filter settings for audio or for the tape recording process? Jac Goudsmit says: 4:11 PM The way it's encoded on tape makes it possible to write and read markers, and they can be recognized even when the tape is winding or rewinding. Jac Goudsmit says: 4:13 PM @Christian The filter settings that are recorded on tape are used to control the audio filters at playback time. I don't know exactly how they work, I errrr... was sick on the day they explained filters when I studied EE for a year... I guess. ;) Jac Goudsmit says: 4:14 PM The first and second generation recorders can record markers that indicate the start of the track or the end of the recording or a signal to reverse the tape during playback. Jac Goudsmit says: 4:15 PM Some time between the second and third generation recorders, Philips changed the specification and third generation recorders (except portables) can record track titles with the markers. Jac Goudsmit says: 4:15 PM Unfortunately the early recorders can't read the track titles from 3rd generation recordings Jac Goudsmit says: 4:16 PM Prerecorded cassettes have a continuous stream of data in the AUX track, containing a table of contents of the entire cassette, with album title, artist and track titles. So at any time while playing the tape, you can select another track to go to, by title. Jac Goudsmit says: 4:18 PM On self-recorded tapes you can't select tracks by title, you can only go back and forth by a number of tracks, and the recorder of course doesn't know whether it would be more efficient to reverse the tape. James Lewis says: 4:18 PM @Jac Goudsmit regarding the continuous stream on AUX. is that how/when they display information on their LCD? (aka, there is no RAM storing the tape's contents?) Jac Goudsmit says: 4:18 PM There is a service bulletin somewhere that explains why a certain function of the earliest recorders doesn't work as expected and why it says "NO TOC" Jac Goudsmit says: 4:19 PM It explains that they planned to put a feature in the software to let the recorder figure out the timing of each track, and then rewind to the beginning of side A to write the TOC to tape, but they ran out of space in the ROM to finish it. Jac Goudsmit says: 4:20 PM @James Lewis There is RAM where probably the TOC is stored. On prerecorded tapes it's probably refreshed continuously. Jac Goudsmit says: 4:21 PM So a self-recorded tape doesn't have the same information as a prerecorded tape, which is a shame. But understandable. Jac Goudsmit says: 4:22 PM Ralf, @DRDCC, helped a guy called Jeremy Heiden release an album on many different formats this year, including DCC and Elcaset (and that was the first prerecorded Elcaset album ever, as far as we know) Jac Goudsmit says: 4:23 PM He did the recording with a DCC-175 and a computer because the computer software made it possible to automatically record song titles along with the audio, so he didn't have to do a second pass on each tape to record the track titles. Jac Goudsmit says: 4:25 PM So that means that the DCC-175 can probably record a continuous TOC track too, and that's one thing I would like to try and implement. Jac Goudsmit says: 4:26 PM The cable that connects the DCC-175 to the Parallel Port of the PC is not documented and is extremely rare. We know from someone who worked on the software that there were only 1400 cables made. Jac Goudsmit says: 4:27 PM Also because it required bitbanging the parallel port, the software is not compatible with any Windows version after Windows 98. Jac Goudsmit says: 4:28 PM However the service manual is very clear in how it's connected (though it does have a bug where a port is drawn the wrong way around). The PC basically has a direct connection to the subband (PASC) I2S bus, and a bidirectional serial port on the microcontroller in the DCC-175. Jac Goudsmit says: 4:29 PM So it should be possible to use a logic analyzer to analyze what the software sends back and forth to/from the recorder. Jac Goudsmit says: 4:29 PM And it should be possible to emulate that behavior with a microcontroller, even without the cable (which basically only acts to convert everything to a parallel port). Jac Goudsmit says: 4:30 PM Another thing we want to do, and @DRDCC and I are going to get together for some of that tomorrow, is ITTS. DRDCC says: 4:31 PM When recording a pre-recorded tape from the DCC-175 to the pc, you can actually see all the artist, track info etc, but once the recording is done, this info can no longer be retrieved in the software. Jac Goudsmit says: 4:31 PM If you've ever looked at the inlay card of a DCC cassette, you probably noticed something like "This cassette has extra information that you can see if you have a DCC recorder with a TV connection". DRDCC says: 4:32 PM We could look at this from the software point of view as well. If we are allowed to write the TOC info back to the DCC-175? Jac Goudsmit says: 4:33 PM So the specification of the DCC (which we've never seen and don't have access to) mandated that every prerecorded cassette should have ITTS (Interactive Text Service or soemthing) Jac Goudsmit says: 4:35 PM @DRDCC has a box that can be connected to the digital S/PDIF output of a DCC recorder and displays the ITTS information on TV. See demo at https://youtu.be/DcdiJM5yPzs Jac Goudsmit says: 4:36 PM The ITTS format is mentioned but not fully explained in the S/PDIF format sepcification. My project at https://hackaday.io/project/24911-propeller-spdif-receiver is an attempt at decoding it. Jac Goudsmit says: 4:38 PM It's apparently very similar to the format of CD-TEXT by Sony but this format is also not documented. There is a spec for ITTS from the ITU but it costs about $300 just to download the document. Anyone wanna donate that much money? :-) DRDCC says: 4:40 PM If you are sure this is the missing piece, the dccmuseum can help. Jac Goudsmit says: 4:40 PM Anyway, @DRDCC's ITTS box contains a couple of 8031(?) microcontrollers (often used by Philips, they have a full license from Intel to build and use those things anywhere they want), with external EPROMs. It also contains a Teletext decoder. Jac Goudsmit says: 4:41 PM I'm going to read out the EPROMs and disassemble the software to see what it does. Jac Goudsmit says: 4:41 PM But the most exciting possibility for me is that after reading the datasheets for the third generation chipset, I think it's possible to encode continuous TOCs on self-recorded cassettes. Jac Goudsmit says: 4:44 PM I'm thinking of putting a microcontroller in a DCC-951 or DCC-730 to convert regular cassettes into prerecorded cassettes that way. It would have to play the tape once to find out where all the track markers are, and it would have to have some interface (probably serial or audio/video) to enter the track titles. After it has all that information, it would play the tape while re-recording the AUX track. Jac Goudsmit says: 4:45 PM And that's pretty much what I wanted to share today :) Jac Goudsmit says: 4:46 PM For more information go to dccmuseum.com and digitalcompactcassette.github.io. I will also answer further questions at this page here: https://hackaday.io/event/28313-reverse-engineering-the-digital-compact-cassette Christian says: 4:47 PM Thanks for all that information, Jac! Jac Goudsmit says: 4:47 PM Most of all, if you happen to have information about the DCC format that we don't have (or you know someone who does), @DRDCC and I would like to talk to you!