Cassettes had one advantage: they weremuch cheaper than floppies.
In the UK this meant that certain budget software houses could churn out games (especially as game production had very low costs in those days and some classics were produced by single-person teams).
I remember for a normal budget game it would cost £2.99 for a normal budget game and £3.99 for the more expensive ones. The full-price releases usually cost £9.99 or more and tbh the budget games were often the best games.
Codemasters actually started out making budget games on the audio tape format.
Also itmade giving away free demos and games with magazines much cheaper.I remember at the end of the Spectrum’s life, you’d could buy Crash magazine and expect to get about 40-50 full games(of varying quality and commericality) spread across 3 tapes.
With respect to the PCM on tape subject, it is possible for the storage format itself to use error correction. Standard audio CDs have roughly 37% extra error-correction information, with an extra 9 bytes for every 12 16-bit samples.
The PCM device mentioned to above doesn’t seem to indicate what, if any, error correction was used on the 16-bit devices. 14-bit devices stored an extra 2 bits, giving only 14% extra information.
In theory either of those formats could have used more or less of that type of eror correction. It’s a purely digital thing; it would merely reduce the amount of space available to record on.
There’s also things that might be called “error correction” that are more format-dependent. On a CD, there’s a bunch more information which prevents device problems on playback, mostly related to the laser technology when it was invented. In total, there’s about 3 times as many bits on the disc as there are playable audio bits.
I’m not sure what was used on the tape devices, but on the other hand it’s not a terribly meaningful metric. Simply having more ‘bits’ that aren’t possible to be used by the format doesn’t mean it is necessarily better or worse - it’s just designed that way.
The early successful personal computers — the Apple II, the TRS-80, the Commodore PET, the Atari 400 & 800 — all had floppy drives available by 1979. The use of cassette tapes to distribute commercial software died quickly after that. (Or it did in the US & Canada anyway.)
I also remember when software titles on floppy disks were enclosed in long ziplock bags.
Hm. I had a Datasette for my Vic-20 and C64 (actually C128) until, oh, 1985 or so. While floppies were more popular, it was not difficult to find cassette versions of software, in my recollection.
I realize this is an old thread, but figured it was worth responding since there was software released on vinyl that you had to record to tape, and then playback on a cassette tape drive. This was more of a gimmic because it was included as a track on an album, and not something you would have purchased at a computer software store. Here is a video explaining the process.
Pretty interesting. To add to earlier parts of the thread I bought software games for my Timex Sinclair 1000 on cassette tape. I think a tape held 16k of data. It took like fifteen minutes to load and had about a 90% chance of loading properly.
I also had an early IBM PC with a cassette drive at first, as the floppy drive was quite expensive. The cassette interface was built in. It had a much better interface and started and stopped the cassette using the remote cable. On the Sinclair you had to use the buttons on the tape deck. I don’t remember any software for the PC on cassette but there might have been some.
Mostly right. The only limit to the amount of data a tape could hold was the length of the tape. The Timex Sinclair itself only had 2K of RAM, but it was very common for people to have expansion modules that boosted this to 16K (as showin in the picture here).
I remember keeping several different programs on the same cassette, and using the counter on the tape recorder to keep track of where on the tape I had to fast-forward to to load a particular program.
Theoretically, there’s no reason why a computer program for something like a Timex Sinclair 1000 couldn’t be distributed on any audio media. I remember, from my childhood, playable records made of plastic or cardboard, distributed in magazines and on cereal boxes. So computer software could have been distributed that way too, though I don’t remember actually seeing any.
Hm. MicroSD cards weigh around a quarter gram. A USPS first class letter in the US is up to one ounce. So you could mail around 100 TB of data for 55 cents.
Distributing programs on vinyl (most likely Flexidisc, for cost reasons) would probably have become a standard if the technological progress had been much slower in the 1970’s.
Many schemes were eclipsed by rapid progress. Floppy drives, a luxury item (I purchased a double-floppy Micropolis drive for $1800 ca. 1976, capacity 360k per drive), soon became affordable and much more reliable than cassette or vinyl.
At the same time, programs were getting bigger, and small storage devices necessarily gave way to larger ones.
Much the same thing happened to a barcode I co-invented, the Bytewrite code. It was capable of encoding all 255 ASCII (or hex) characters in a single barcode of up to 250 characters in length, with checksum. It could be concatenated with other barcode lines to make a larger program. It was designed to be easily and cheaply printed on almost any paper – magazines, newspapers, much cheaper than vinyl or cassettes could be made. It could be read quickly (300 baud?) as long as you had a barcode reader wand (about $250 at the time).
It didn’t catch on, for the same reasons as given above – programs got bigger and other storage mediums got cheaper.
Musicat: fascinating! I vaguely remember the Bytewrite bar codes in some computer magazine I was subscribing to at the time. Those were the days of thick magazines filled with computer code that you had to type in by hand and then save to several different cassette tapes (because there was no real error check so one little flaw in the tape meant retyping everything all over again).
Bytewrite seemed like a great way to, well, not spend weeks typing stuff. I don’t remember many magazines that used 5.25" floppies but there were a bunch of magazines with 3.5" stiffies. Probably more resistant to getting mangled in the mail.
Behold the power of being in the right place at the wrong time.
I think pretty obviously nobody ever did it in any serious form, but if one wanted to design the hardware for it, one could obviously encode digital data onto a vinyl record. Just as interesting and totally useless thought experiment, how much data could you manage to cram onto a standard sized LP record given the tolerances involved in mechanical playback? And how fast could it be read (you don’t have to presume you read it at 33 1/3 RPM). For reference, paper tape encoded about 10 characters per inch and the best optical readers in use while paper tape was still a commonly used medium seem to have been able to manage about 5000 characters/second. the crappy reader on the side of your model 33 teletype was far less than that, of course.
.