[Gllug] Historical question

[Richard Jones]

On Sat, Aug 14, 2004 at 01:30:35PM +0100, SteveC wrote:
> sat they were 2400 bit/sec and an hour for both sides
> 
> 2400 bit/sec * 3600 sec / 8 bit/byte / 1024 byte/k
> 
> == 1054 k
> 
> hmmmmmm

Hmmmm also.

The Speccy by default got 1500 bps (incorrectly called 'baud', but
IIRC they could only encode one bit per cycle anyway).  The ZX81 was
much worse - 300 bps.

There were various fast loading schemes.  On the Amstrad CPCs you
could get around 3000 bps before you started to get errors.

The capacity of audio tape though is surely much greater than this.
This stands to reason if you think about the information content of
music which is routinely recorded on audio tapes.

The problems with old 8 bit computers seemed to be threefold:

(a) the tape machines themselves didn't usually run at a very constant
speed;

(b) the CPUs could only execute 100-200 instructions per bit, so
you're limited by what sort of signal processing you can do; and

(c) the hardware was limited to simple on/off signal levels - eg.  no
automatic sinusoid generation/detection, or even amplitude control.

I'm quite sure that with modern hardware and signal processing
techniques, and using both stereo tracks on each side, you could get
much much higher data rates from audio tape.  (Anyone fancy
implementing QAM for audio tape ?!?)

Rich.

-- 
Richard Jones. http://www.annexia.org/ http://www.j-london.com/
Merjis Ltd. http://www.merjis.com/ - improving website return on investment
http://www.YouUnlimited.co.uk/ - management courses
-- 
Gllug mailing list  -  Gllug at gllug.org.uk
http://lists.gllug.org.uk/mailman/listinfo/gllug