Audio Tape Interface Revives Microcassettes As Storage Medium (hackaday.com) 78
Zack Nelson decided to go back in time and add a suitably classic storage medium to a retrocomputing project, in the form of a cassette interface. Hackaday reports: The cassette player he had available was a Pearlcorder L400, which uses the smaller microcassette instead of the familiar audio tapes used in your Walkman or boombox. [Zack] designed the entire thing from the ground up: first he decided to use differential Manchester encoding, which provides immunity against common disturbances like speed variations (which cause wow and flutter). The data is encoded in the frequency range from 1 kHz to 2 kHz, which suits the bandwidth of the cassette player. Next, he designed the interface between the computer and the tape recorder; built from an op-amp and a comparator with a handful of discrete components, it filters the incoming signal and clips it to provide a clean digital signal to be read out directly by the computer.
The system is demonstrated by hooking it up to an Arduino Nano, which reads out the data stream at about 3000 baud. The noise it makes should bring back memories to anyone brought up with the "PRESS PLAY ON TAPE" message.
The system is demonstrated by hooking it up to an Arduino Nano, which reads out the data stream at about 3000 baud. The noise it makes should bring back memories to anyone brought up with the "PRESS PLAY ON TAPE" message.
Re: (Score:2)
Ah the memories... (Score:5, Informative)
The noise it makes should bring back memories to anyone brought up with the "PRESS PLAY ON TAPE" message.
Yeah it does: it reminds me how great things are today because we don't have to deal with this shit no more. Or vinyls. Or diskettes. Or VHS.
Retro tech is what people who haven't had to actually use it think is cool. For those of us old enough to remember them in actual use, it's fucking dreadful.
My sister summed it up perfectly: "vintage" is all the shit young people spend ungodly amounts of money on today that we couldn't wait to get rid of.
Re: (Score:2)
I use cassettes and reel to reel tapes and records. And VHS. Never really stopped. Well, I bought a VCR in 2008, before then I only had players.
However, I do not use the old devices exclusively. Some things are simpler, more reliable or more convenient with analog (or just using separate devices, digital could be OK as well), but not everything.
I have a collection of both analog and digital media. I also am happy that right now I have options and am not limited to just using the od devices, like I was in th
Re:Ah the memories... (Score:4, Funny)
https://www.youtube.com/watch?... [youtube.com]
Re: (Score:2)
Indeed. I had an Exatron "Stringy Floppy" for my TRS-80 Model I that actually did use microcasettes! Oddly, the company still exists:
https://en.wikipedia.org/wiki/... [wikipedia.org]
Most of the other home computer interfaces used regular "Compact Cassette" full-sized regular audio 1/4" tapes, and at the time all I could do is save up the $400.00 I needed to buy a SINGLE 5-1/4" FLOPPY DISK DRIVE! I was so happy when I got it, and I was so frustrated from the many, many hours and programs lost from data errors and unrelia
Re: (Score:2)
Microcassette as the same tape as a regular cassette (1/8"), just that the tape is thinner and it runs at half the speed. 1/4" would be reel to reel tape or elcaset.
Re: (Score:1)
I loved my Stringy Floppy and I still have it, although I have not tried to use it since the early 80's. Great little thing.
For those of you who do not understand what it looked like, the cassettes were about the same size as the cards used by Mr. Spock in the original Star Trek series. They operated like an 8-track tape - not reel to reel but a loop. Serial interface sounded like a mini siren when spooled up.
I wrote a couple of routines to do I/O on them. Interesting little toy.
Re: (Score:2, Insightful)
My sister summed it up perfectly: "vintage" is all the shit young people spend ungodly amounts of money on today that we couldn't wait to get rid of.
Yes, but in some cases you were wrong, a lot of vintage stuff is actually superior like toasters, tools, furniture... but not computing hardware. That stuff was primitive, which is the real differentiator, not merely whether it's old or not. And in computing, as you know, development is still moving very quickly.
Re: (Score:2)
Don't confuse "built better before" and "better tech before".
Toasters of yesteryear were built like tanks, because one of the selling points was durability. But clearly digital toasters offer more toasting options and are better. Or rather, they could be better: toasters today are built to fail, require an internet connection and won't work without connecting to the mothership.
Build a digital toaster for durability without stupid IoT cruft and it'll be vastly superior to vintage stuff.
Re: (Score:3)
Build a digital toaster for durability without stupid IoT cruft and it'll be vastly superior to vintage stuff.
Right, but here's the thing, they don't do that. The simple toasters without the bullshit are uniformly built like garbage. And so are the expensive ones, AFAICT. Nobody is building them worth a fuck for any amount of money.
Re: (Score:2)
Interesting, do they make 120V models? It would be inconvenient for me to add a 240V outlet.
Re: (Score:2)
Since getting a big Breville - it will take a quarter-sheet (13"x9") pan, though not a glass 13x9 casserole that has handles - I almost never turn my big ovens on except for holiday meals. I can have Brussels sprouts or broccolini fully cooked in about the same time that the big ovens take to get up to temp.
Re: (Score:2)
You do realize Japan has the oddball 100V system, right? Which is generally "close enough" to work on a North American 120V system, especially for simple things like toasters. The only irritating thing would be that the plug isn't polarized (or grounded) typically.
The problem with toasters is well, the digital circuit is just one chip nowadays that is little more than a timer. That's why they're like $10 nowadays. An
Re: (Score:2)
Japanese brands make good toasters, but they cost 5-10x what everyone else's do.
Re: (Score:2)
Japanese brands, or made-in-Japan Japanese brands?
A lot of Japanese products are made in China, Taiwan and Singapore.
Re: (Score:2)
Nah, you're running into survivorship bias here. The stuff that still exists from yesteryear (such as toasters, mixers, etc... ) is good because it was built well. They also built a lot of crap in yesteryear. It's now mostly in your back cabinet, garage, or the landfill.
Re: (Score:2)
I'd be curious how much better old furniture was. I guess I can see where structurally it might be better (more high quality hardwood), but is the cushion filling in a couch built in 1925 really better than a quality modern one with advanced synthetic foams?
I'd also call into question some tools. I would not trade a modern 24v Lithium-battery variable speed, reversible electric drill for a singe-speed electric drill from the 1950s.
A lot of the older electrical appliances were long-lasting because they wer
Re: (Score:2)
I'd be curious how much better old furniture was. I guess I can see where structurally it might be better (more high quality hardwood), but is the cushion filling in a couch built in 1925 really better than a quality modern one with advanced synthetic foams?
Let me answer the last question first, IME they both wear out and need replacement, but the synthetics are cheaper. That said, you have it on the nose, the structural materials were higher quality in the past. Not only was quality wood cheaper, but back in the day they didn't make couches out of plywood and cardboard, which is literally how most of them are built now. There's arguments to be made for plywood in some furniture contexts, but none for cardboard.
Re: (Score:2)
Actual plywood isn't a bad material, really. Cross-grain plies insure excellent dimensional stability and resistance to warping.and splitting. But please, no ridiculously heavy MDF or particle board, and chipboard is not a good substitute for actual plywood.
The bigger problem is the least-possible-material engineering. IMHO, other than aiding manufacturing profits, there's no reason to frame furniture like it was a mass reducing exercise for a NASA voyage to the edge of the solar system.
Cardboard is a re
Re: Ah the memories... (Score:2)
Re: (Score:2)
Adhesives have a role, but often the joints lack enough mating surface for glue to be effective on its own.
One one piece my wife bought and I immediately regretted her buying when I started assembling I literally used two-part marine epoxy and some small sheets of fiberglass.
Re: (Score:2)
Old stuff were supposed to last a lifetime. That is why people still are happy to find a good worn dress shoe, and use it as new.
Yet, that model is already gone. Furniture? We have IKEA, if you are lucky it will survive your next move. But the thing is, people do actually move, and every time we moved the old furniture became a burden. And the modern mindset is just replacing most of them.
Toasters? Sure the $10 one will give way. But a simple solid one will last years.
Tools? Look for Wera on Amazon. They ar
Re: (Score:3)
>"Retro tech is what people who haven't had to actually use it think is cool. For those of us old enough to remember them in actual use, it's fucking dreadful."
Indeed. I remember many hours on my 4K Tandy CoCo 1, loading from cassette tape, connecting to a BBS at 300 baud, and waiting 20 minutes for a couple of pages to print. It was certainly neat at the time. But I would never want to go through THAT again. I have no need to relive it; besides, it would probably ruin what good memories I still have
Re: (Score:2)
I remember that loading Popeye on a CoCo 2 tape system was a building block in learning what patience really means.
Re: (Score:2)
>"I remember that loading Popeye on a CoCo 2 tape system was a building block in learning what patience really means."
It is amazing what we put up with.
I was hot stuff once I could afford a floppy drive. What a difference. Then, to finally have an actual hard drive and OS-9, just wow. Full windowing, true multitasking, multiuser, realtime, in the mid 80's on a CoCo3. From there to SunOs, Solaris, Altos Xenix, Interactive Unix, SCO-Unixware, then Linux!
Re: (Score:2)
No Amiga in your timeline?
Re: (Score:2)
>"No Amiga in your timeline?"
Nope, never did anything Commodore, Atari, or Apple.
Re: (Score:2)
OS-9 was all text when I was porting it to transputers in the early 90s. What is this windowing system you speak of?
Re: (Score:2)
>"OS-9 was all text when I was porting it to transputers in the early 90s. What is this windowing system you speak of?"
OS-9 level 2 Multi-Vue.
https://www.youtube.com/watch?... [youtube.com]
Wow, watching that video brings back memories. I was running it from a hard drive by then, though.
Re: (Score:2)
Very nice. The VME based platforms I was porting to didn't have a screen. I was using RS232 to talk to it and before I got that far I had to write the serial driver because it didn't grok the one on the board (the board was the product we designed, so it didn't grok any of the hardware). Fun times. I was pretty fresh out of college back then and I think the boss was throwing things at me to see if I could do them (I couldn't, but I faked it till I maked it, which I how most things work in tech). The code we
Re: Ah the memories... (Score:1)
Re: (Score:2)
From the same master, nope.
The only case where some vinyl can sound better than digital is when shitty mastering [wikipedia.org] is done for digital and that horrible over-compressed mess just cannot be transposed to vinyl so engineers either create a vinyl-dedicated alternative master that's not as bad, or in the case of older records, just use the original master that predates that loudness madness.
Re: (Score:2)
Exactly this. If you try to cut a record from the hot garbage that passes for a master these days you'll burn out some expensive equipment and the record won't play.
Re: (Score:2)
It's different to experience something when you know you can just go can to modern tech any time.
Also great for showing kids some history.
Re: Ah the memories... (Score:3)
Re: (Score:2)
Yeah it does: it reminds me how great things are today because we don't have to deal with this shit no more
Oh but here's the kicker. 20 years from now people will be saying the same thing about the state of the art crap you think is so wonderful today.
After all, cassettes were a thousand times better than punch cards.
Re: (Score:2)
After all, cassettes were a thousand times better than punch cards.
And really really cheap, too.
Re: (Score:2)
...and you could buy cassettes just about anywhere.
Re: (Score:2)
But organic vinyl sounds the best.
Re: (Score:2)
Still not good enough. I only listen to organic, free-range vinyls.
Re: (Score:2)
I don't know if you're trolling or not, but "SuperVinyl" was the best material they made records out of.
The company that used it mostly was MFSL, who produced a famous half-speed master series of the most famous albums around the 1960-1980 period including some gems like Dark Side of The Moon, I Robot, Crime of The Century, and Fly Like an Eagle to name but a few.
Those special releases were cut at half-speed so the mastering cutter would track more accurately as the master was made, and then the final consu
Re: (Score:1)
Re: (Score:1)
Why Manchester encoding? (Score:2)
The chosen encoding is very inefficient. I couldn't figure out from a quick read what the actual bit rate is; there are various numbers in kHz, but this encoding needs a lot of bandwidth relative to the bit rate.
I wonder how much bit rate you could squeeze out of a cassette player using a more advanced encoding such as QPSK [wikipedia.org] , QAM [wikipedia.org], or OFDM [wikipedia.org]. Of course, an interpreted Python script on an Arduino would not be able to do the signal processing in real time. Is there no encoding that can be done efficiently withou
Re: (Score:2)
Let's not forget another limitation: old computers did not have a proper ADC on their line input. They only had a comparator, which means that the incoming tape signal only had 1 bit, and the maximum achievable sample rate was something like 50-100 kHz at best. This alone already rules out QAM.
Re: (Score:2)
Well yes, but if you choose to use an Arduino to do the signal processing, then you could just as well try to squeeze out all performance that you can from the Arduino. :)
Re: (Score:1)
Well yes, but if you choose to use an Arduino to do the signal processing, then you could just as well try to squeeze out all performance that you can from the Arduino. :)
I'm currently doing a project where we do signal processing on an Arduino. The hardware counter/timers do nearly all the work. The only downside is it is clocked by a shitty 16 Mhz resonator. But you can clock the timers from an external clock and you can use an external voltage reference if you need more accuracy. We didn't. The Arduino Mega 2560 was a prefect match for this project. Low-level real time in C with decent libraries and no OS to mess things up. For dev you can just run the board of USB
Re: (Score:1)
It would still be two very different projects to (1) squeeze as much bit rate as possible from the analog tape using Arduino and a proper ADC, and (2) squeeze as much bit rate as possible from the analog tape using Arduino and a fixed-voltage comparator. And my bet is that the resulting modulation schemes will be different.
Re: (Score:2)
>"Is there no encoding that can be done efficiently without FFT methods and still work with the given bandwidth?"
I was wondering that myself. He probably just did what he could do. Certainly, if we used modern technology, we would try to use ALL the bandwidth of the tape, with parallel data streams in many frequencies not just 1Khz to 2Khz! Plus, the tape could be sped up, too. Throw in ECC and even encryption.
And you end up with something much, much better- faster, more reliable, and more secure. A
Re: (Score:3)
I'm reading on how the 14.4 kbps V.32 bis modems [itu.int] of the 1990s worked, which had a similar bandwidth limitation as this cassette player. Seems that it is QAM with some smart preprocessing (Trellis encoding) of the bit stream. For QAM you don't need FFTs; just a sin and cos oscillator and some mechanism to keep he oscillator in the decoder synchronized with the input signal.
Re: (Score:2)
And then the tape gets a small bad spot on it that the old tech wouldn't even notice that totally screws up the readability now that there's no redundancy left in the encoding.
Re: (Score:2)
That's what error correction is for.
Re: (Score:2)
So doubling the density and then cutting it in half again gains you what?
Re: (Score:2)
So doubling the density and then cutting it in half again gains you what?
Well, now it's much more difficult to debug, understand why it doesn't work and write a fix - which puts it at par with most modern technology!
Re: (Score:2)
Usually error correction performs better. That is, you double the density and then give 25% of the new capacity to error correction with a net gain of 50% density..
Hard drive manufacturers seem to do this now, record the data as densely as possible, as long as error correction is able to cope with it. As do digital broadcasts (DVB and such).
Re: (Score:2)
It works well enough when the underlying tech is good enough to support it. Analog audio tape is not. I don't jkust mean the tape itself, I mean the drive technology (audio recorder), the tape transport, and the cartridge it's stored in.
You can make tape work, but by the time you're done, it is no longer inexpensive audio tape cheaply re-purposed.
By the time you do enough engineering to only lose 25% of the capacity to error correction in order to double the density, the underlying tape technology bears mor
Re: (Score:2)
Still, a dial-up modem can fit 33.6kbps over a rather noisy phone line with something like 3.4kHz bandwidth. Audio tape has at least double the bandwidth (100Hz-10kHz would be OK to use) and has two channels.
As I understand though, doing all of this at the time would have been really expensive because of the processing power required and, as such, audio tape for data storage would lose its main advantage - cost.
Re: (Score:2)
I do agree that audio tape itself could have been pushed a little harder and that newer tech should have been able to achieve at least 33.6K, but by the time that tech became affordable, everything had gone over to Floppy and RLL HDs.
Re: (Score:2)
Still, a dial-up modem can fit 33.6kbps over a rather noisy phone line with something like 3.4kHz bandwidth. Audio tape has at least double the bandwidth (100Hz-10kHz would be OK to use)
These micro tapes were for voice memos. The article has a response spectrum; it's just 1 kHz - 4 kHz (at the -3 dB points). The signal/noise ratio might also be worse than a telephone line.
Very cool, retro (Score:2)
Very cool, retro...when I first started to learn to program, my ZX-81 I connected to a Sears & Roebuck cassette recorder, and then later the VIC-20 and Commodore-64 had their own tape drives. The only problem was the cassette recorder could "eat" the tape, or partially erase it if your sister decided to make her own mix tape. :)
Josh K.
Re: (Score:3)
My sister once left a magnetic toy on top of one of my tapes.
Sisters, eh?
Re: (Score:2)
Yes, sisters indeed.
When I got a credit card as a college student, I remember my mom or sister put it on the refrigerator with a cute butterfly magnet. :)
Of course, I still remember bubble memory, I always thought I'd have a "bubble memory" pack attached to my ZX-81 or VIC-20.
Josh K.
I did better (Score:2)
Re: (Score:1)
Re: (Score:1)
"They say he carved it from a bigger spoon!"
Re: (Score:2)
Soooo, that's what you did with that nice piece of Ash?
Reminds me... (Score:2)
This reminds me at some point I need to find my old cassette tapes from my TI-99/4A and digitize them all. Apparently you can decode the data in a PC from the digitized audio. I'm very curious to see what silly BASIC a 10-year-old-me was writing back in the day. I remember making some kinda movie-prop like robot interface (with a simple little sine-wave animation and other odds and ends) inspired from Disney's Black Hole movie (anyone else remember V.I.N.CENT [fandom.com]??). I think it would also turn the screen all w
Re: (Score:2)
I did not know the TI99/4A had a cassette reader. I heard about it thanks to retro people doing a demo on it and learnt it used ROM cartridges.
Re: (Score:2)
TI actually had their own brand cassette player, which I happened to have.
This exact model (and color). [preview.redd.it]
It connected to the joystick port. Which meant swapping devices constantly if you were loading games that used the joysticks.
Re: (Score:2)
No.
The 99/4A's cassette port was dedicated, but it looked exactly the same as the joystick port. (that being a 9 pin DIN)
It was monaural, with a rem port line. Any cassette recorder that does monaural recording and has a rem port will work.
From first hand experience, you can use a modern walmart store brand (ONN) shoebox recorder. The only downside is the lack of a minute counter, meaning storing multiple programs per tape requires some alternative solution.
There is a small basic program for the 4A that c [atariage.com]
Re: (Score:2)
Sounds different from my TRS-80 CoCo (Score:2)
Sounds different from my TRS-80 Color Computer. Not sure what encoding it used but it definitely didn't sound like that and probably delivered far lower bandwidth.
Re: (Score:2)
https://youtu.be/I3lhtcGzhkc?t... [youtu.be]