4 ms·

I wonder what the maximum bitrate of an audio cassette is, using modern encoding techniques like OFDM with error correction? I searched once, and saw estimates

by p1mrx 5y ago

I wonder what the maximum bitrate of an audio cassette is, using modern encoding techniques like OFDM with error correction? I searched once, and saw estimates in the 25 kbps range, but no usable implementation.

LeoPanthera 5y ago
The sound quality of an audio cassette is at least as good, and almost certainly better, than the quality of an analogue telephone line. So I bet you could set a lower bound of at least 33.6kbit/s, since that was as fast as telephone modems went. 56k modems relied on your phone line being digital after the switchboard, but I would imagine that 56k should be doable as well.
- p1mrx 5y ago
  I wouldn't automatically assume that tape is better than a phone line. For instance, a phone line will always clock at 1 second per second, because everything runs in real time with no storage.
  - LeoPanthera 5y ago
    I can’t imagine that clock recovery would make a huge amount of difference.
  - TedDoesntTalk 5y ago
    Analog phone lines were limited to 4 KHz frequency range in the US. Compare that to tapes close to 20 KHz.
- Const-me 5y ago
  Not sure it's directly comparable. When modem flips one bit, the CRC checksum in IP packet becomes incorrect, someone drops the IP packet, the condition is detected and the packet is re-transmitted. This slows down things substantially but user still gets their data. I think tapes need more reliability because there's only 1 try.
  - LeoPanthera 5y ago
    I remember back in the 80s when my BBC Micro would miss a bit it would just beep and tell you to "Rewind tape", which meant, go back about 10 seconds. And then it would figure it out from there.
  - tenebrisalietum 5y ago
    The Commodore 64 (and possibly earlier Commodore systems such as the PET and VIC20) actually recorded each "block" twice on the tape, and could correct up to 31 errors per block as long as an error didn't affect the same exact byte in each block (it didn't do anything like ECC). This also made loading/saving from tape really slow - 50bps.
    - krallja 5y ago
      SiriusXM uses a similar method, plus buffering, to ensure your satellite radio keeps playing, even if you dare to drive under such rare things as traffic lights and tree branches.
  - adrianmonk 5y ago
    You could choose to do something similar with cassette. It's common for tape drives to verify the data that was written. Rewinding is one way, but it's slow. A faster way is with two separate heads: as the tape moves, it passes over a write head first and then a read head. You could maintain a buffer in RAM of data that has been written but not yet verified (as readable without errors). If you can verify it, you discard it from the buffer. If you can't, then you queue it up to be rewritten. This wouldn't even require a special cassette deck. There were plenty of three-head audio cassette decks that allowed you to listen to the read head while making a recording. That made it much easier to set the right recording level. (Basically, there were two common types of cassette deck: two-head and three-head. A two-head cassette deck has an erase head and a dual-purpose read/write head. A three-head deck has erase, read, and write heads. At least, that's the basic idea. Things get more complicated with auto-reverse cassette decks.)
sprash 5y ago
I was able to fit about 60 MB per channel on a 90 min audio cassette [1]. The bias frequency of that particular deck was about 60 kHz. This means bypassing the recording circuitry could lead to much higher data rates. 1.: https://news.ycombinator.com/item?id=28193079 https://news.ycombinator.com/item?id=28193079
tyingq 5y ago
They were getting 33.6kbps out of a fully analog phone line with 300Hz-4Khz usable frequency range. Depends on the equipment and tape, but I imagine even cheapish equipment would have at least twice that amount of bandwidth to work with, maybe even up to 10-12kHz instead of a 4kHz top end. High end equipment and high end tape shows up to ~20kHz or so: https://audiochrome.blogspot.com/2019/12/cassette-tape-comparative-measurements.html https://audiochrome.blogspot.com/2019/12/cassette-tape-compa...
femto 5y ago
S/N for a cassette is variously given as 50dB, 58dB or 67dB [1,2]. Bandwidth is given as 12kHz (standard), 16kH (medium Fi), 19kHz (HiFi) for different quality cassette systems [3]. Let's simplify by assuming the frequency cutoff is a brick wall and the S/N is constant across the band. The maximum channel capacity for a cassette can be calculated using Shannon's Theorem [4]: C = B*log2(S/N+1) Cheap and cheerful (S/N=50dB, B=12kHz): C= 199kbit/s Mid range (S/N=58dB, B=16kHz): C= 297kbit/s Hi Fi (S/N=67dB, B=19kHz): C= 422kbit/s Modern coding (LPDC) can achieve within a fraction of a dB of the Shannon Bound, so those numbers will be close to what a state-of-the-art encoding could do with a cassette. (A new hobby for someone?) [1] http://www.tapeheads.net/showthread.php?t=35390 http://www.tapeheads.net/showthread.php?t=35390 [2] http://www.muzines.co.uk/articles/choosing-a-cassette-tape/1249 http://www.muzines.co.uk/articles/choosing-a-cassette-tape/1... [3] https://arstechnica.com/civis/viewtopic.php?t=848978 https://arstechnica.com/civis/viewtopic.php?t=848978 [3] https://en.wikipedia.org/wiki/Shannon%E2%80%93Hartley_theorem https://en.wikipedia.org/wiki/Shannon%E2%80%93Hartley_theore... Edit: Here's a dose of reality from another forum: https://dsp.stackexchange.com/questions/70700/why-can-so-little-digital-information-be-stored-on-a-cassette-tape https://dsp.stackexchange.com/questions/70700/why-can-so-lit... My guess is the the Shannon based predictions are realistic, as they are lower than the same rates as achieved by specialised data recording systems. They are also a bit better than what a narrow band radio system can achieve, and a cassette will be a better channel than a typical radio channel.
- adrianmonk 5y ago
  There's also stereo to consider, right? So you could get up to 2X bandwidth. (Channel separation issues might mean less than 2X in practice, though.) If you want to stretch the definition of "cassette", tapes are two-sided, and there were 4-track recorders that used all 4 tracks in one direction. They weren't anything too exotic. Musicians used them for cheap recordings for demos and such.
  - hulitu 5y ago
    You don't want stereo. Only as an error correction maybe.

by p1mrx 5y ago

LeoPanthera 5y ago

The sound quality of an audio cassette is at least as good, and almost certainly better, than the quality of an analogue telephone line. So I bet you could set a lower bound of at least 33.6kbit/s, since that was as fast as telephone modems went. 56k modems relied on your phone line being digital after the switchboard, but I would imagine that 56k should be doable as well.

p1mrx 5y ago
I wouldn't automatically assume that tape is better than a phone line. For instance, a phone line will always clock at 1 second per second, because everything runs in real time with no storage.
- LeoPanthera 5y ago
  I can’t imagine that clock recovery would make a huge amount of difference.
- TedDoesntTalk 5y ago
  Analog phone lines were limited to 4 KHz frequency range in the US. Compare that to tapes close to 20 KHz.
Const-me 5y ago
Not sure it's directly comparable. When modem flips one bit, the CRC checksum in IP packet becomes incorrect, someone drops the IP packet, the condition is detected and the packet is re-transmitted. This slows down things substantially but user still gets their data. I think tapes need more reliability because there's only 1 try.
- LeoPanthera 5y ago
  I remember back in the 80s when my BBC Micro would miss a bit it would just beep and tell you to "Rewind tape", which meant, go back about 10 seconds. And then it would figure it out from there.
- tenebrisalietum 5y ago
  The Commodore 64 (and possibly earlier Commodore systems such as the PET and VIC20) actually recorded each "block" twice on the tape, and could correct up to 31 errors per block as long as an error didn't affect the same exact byte in each block (it didn't do anything like ECC). This also made loading/saving from tape really slow - 50bps.
  - krallja 5y ago
    SiriusXM uses a similar method, plus buffering, to ensure your satellite radio keeps playing, even if you dare to drive under such rare things as traffic lights and tree branches.
- adrianmonk 5y ago
  You could choose to do something similar with cassette. It's common for tape drives to verify the data that was written. Rewinding is one way, but it's slow. A faster way is with two separate heads: as the tape moves, it passes over a write head first and then a read head. You could maintain a buffer in RAM of data that has been written but not yet verified (as readable without errors). If you can verify it, you discard it from the buffer. If you can't, then you queue it up to be rewritten. This wouldn't even require a special cassette deck. There were plenty of three-head audio cassette decks that allowed you to listen to the read head while making a recording. That made it much easier to set the right recording level. (Basically, there were two common types of cassette deck: two-head and three-head. A two-head cassette deck has an erase head and a dual-purpose read/write head. A three-head deck has erase, read, and write heads. At least, that's the basic idea. Things get more complicated with auto-reverse cassette decks.)

sprash 5y ago

I was able to fit about 60 MB per channel on a 90 min audio cassette [1]. The bias frequency of that particular deck was about 60 kHz. This means bypassing the recording circuitry could lead to much higher data rates. 1.: https://news.ycombinator.com/item?id=28193079 https://news.ycombinator.com/item?id=28193079

tyingq 5y ago

They were getting 33.6kbps out of a fully analog phone line with 300Hz-4Khz usable frequency range. Depends on the equipment and tape, but I imagine even cheapish equipment would have at least twice that amount of bandwidth to work with, maybe even up to 10-12kHz instead of a 4kHz top end. High end equipment and high end tape shows up to ~20kHz or so: https://audiochrome.blogspot.com/2019/12/cassette-tape-comparative-measurements.html https://audiochrome.blogspot.com/2019/12/cassette-tape-compa...

femto 5y ago

S/N for a cassette is variously given as 50dB, 58dB or 67dB [1,2]. Bandwidth is given as 12kHz (standard), 16kH (medium Fi), 19kHz (HiFi) for different quality cassette systems [3]. Let's simplify by assuming the frequency cutoff is a brick wall and the S/N is constant across the band. The maximum channel capacity for a cassette can be calculated using Shannon's Theorem [4]: C = B*log2(S/N+1) Cheap and cheerful (S/N=50dB, B=12kHz): C= 199kbit/s Mid range (S/N=58dB, B=16kHz): C= 297kbit/s Hi Fi (S/N=67dB, B=19kHz): C= 422kbit/s Modern coding (LPDC) can achieve within a fraction of a dB of the Shannon Bound, so those numbers will be close to what a state-of-the-art encoding could do with a cassette. (A new hobby for someone?) [1] http://www.tapeheads.net/showthread.php?t=35390 http://www.tapeheads.net/showthread.php?t=35390 [2] http://www.muzines.co.uk/articles/choosing-a-cassette-tape/1249 http://www.muzines.co.uk/articles/choosing-a-cassette-tape/1... [3] https://arstechnica.com/civis/viewtopic.php?t=848978 https://arstechnica.com/civis/viewtopic.php?t=848978 [3] https://en.wikipedia.org/wiki/Shannon%E2%80%93Hartley_theorem https://en.wikipedia.org/wiki/Shannon%E2%80%93Hartley_theore... Edit: Here's a dose of reality from another forum: https://dsp.stackexchange.com/questions/70700/why-can-so-little-digital-information-be-stored-on-a-cassette-tape https://dsp.stackexchange.com/questions/70700/why-can-so-lit... My guess is the the Shannon based predictions are realistic, as they are lower than the same rates as achieved by specialised data recording systems. They are also a bit better than what a narrow band radio system can achieve, and a cassette will be a better channel than a typical radio channel.

adrianmonk 5y ago
There's also stereo to consider, right? So you could get up to 2X bandwidth. (Channel separation issues might mean less than 2X in practice, though.) If you want to stretch the definition of "cassette", tapes are two-sided, and there were 4-track recorders that used all 4 tracks in one direction. They weren't anything too exotic. Musicians used them for cheap recordings for demos and such.
- hulitu 5y ago
  You don't want stereo. Only as an error correction maybe.