Re: TCP/IP for 6502 commented source code (Off-topic)

[Linards Ticmanis <ticmanis@coli.uni-sb.de>]

Offtopic warning!

Byron Desnoyers Winmill wrote:

> If everybody was forced to serve webpages off C=64
> maybe the web would be sane.


I doubt it. People found ways to put p*rn on the C64 after all.

> But I was curious: wasn't the C=64 floppy drive handled by the serial
> port?  If this was the case, why would Commodore design a machine
> where the primary source for I/O was crippled?


Actually the drive was sitting on the Serial _bus_, not the serial port. 
That bus was a serial variant of the IEEE-488 parallel standard that's 
sometimes still used for measuring equipment today. However the bus was 
also pretty crippled - limited to about 3000 bps and run by software. 
The hardware was capable of going up to 15 or so times faster though, 
that's why all those "fast loaders" were around.

Why would they do it? When the VIC-20 was designed, the first with the 
"serial IEEE", they found in the last minute that there was a bug in the 
6522 interface chips they were using, in its "serial transfer" function. 
So they tossed the idea and used the normal I/O ports of that chip with 
a piece of firmware quickly and dirtily tossed together. Since the disk 
drive for the 3.5K RAM VIC-20 didn't sell well, as most people went with 
tape, they still had a large supply sitting around on the shelves when 
they made the C64, so although the C64 has unbuggy interface chips, they 
decided to stay with the crippled bus to be able to sell off their 
already manufactured drives.

Oh, and the serial IEEE itself was introduced in the first place because 
there had been a price increase in standard IEEE-488 cables, and the 
goal for the VIC-20 was to sell below $300. For the same reason they ran 
the RS232 through software - you save a 6551 chip.

--

Linards Ticmanis

 The Master said, "The business of laying on the colors follows the
preparation of the plain ground."