home *** CD-ROM | disk | FTP | other *** search
- Path: inasa.shizuoka.org!not-for-mail
- From: wileyc@moscow.com (Christopher Wiles)
- Newsgroups: alt.folklore.computers,comp.sys.cbm
- Subject: Re: Vic-20/C-64 serial ports
- Followup-To: alt.folklore.computers,comp.sys.cbm
- Date: 8 Jan 1996 21:05:27 GMT
- Organization: Advanced Genki Studies Institute
- Message-ID: <4cs0un$r5e@whale.moscow.com>
- References: <4cn3ba$ifj@celebrian.otago.ac.nz>
- NNTP-Posting-Host: ppp108.moscow.com
- X-Newsreader: TIN [UNIX 1.3 950824BETA PL0]
-
- The Arch-Deviant (simon.brady@stonebow.otago.ac.nz) wrote:
-
- > My question is, why did Commodore implement the serial protocol in software?
- > Both the 6522 and 6526 had hardware shift registers and associated interrupt
- > capability, but until the 128's fast serial mode these were never used. Does
- > anyone know why not?
-
- 'Cause they were insane. Remember, these were the folks that opted to
- not fully decode the $D000-$DFFF region, so that all of the memory-mapped
- devices (VIC-II, SID, both CIAs) were shadowed every 4k ...
-
- Seriously, hooking a UART to the user port would probably obviated using
- the port for non-RS232 devices. Things were *much* more flexible with
- the CIA (CIA2? It's been a while).
-
- Anyone know why the RS232-emulation kernel code was so inefficient?
- Slipshod design, or what?
-
- -- Chris (wileyc@moscow.com) -- PGP public key available from keyservers
- http://www.eecs.wsu.edu/~cwiles/
-