From: Marko Mäkelä (marko.makela_at_hut.fi)
Date: 2003-03-14 13:20:15
On Fri, Mar 14, 2003 at 10:44:59AM +0100, Nicolas Welte wrote:
> I have a Turbo232, which is compatible with the SL. I didn't even know
> cbmlink supports the 6551, I might give it a try sometime.
It does. The 6502 side is configured for maximum baud rate, which is
19,200 bps for the plus/4 and the CBM II (the SuperPET is not yet
supported). As far as I've understood, the maximum baud rate on the
SwiftLink is 38,400 bps. So, it won't be any speed improvement over
the C2N232.
> If this gives me a speed boost over the PC64 cable I might be very
> much interested in getting this working, but I doubt it. C2N232 is
> slower than PC64 (20 minutes instead of 15 for a .d82 image), and
> it already runs at 38kbps.
If I remember correctly, I measured some 16 kB/s transfer rates with
the PC64 cable already in 1995 or so. The cbmlink server communicates
with one interface at a time: it'll load a sector, send it to the
PC, load another sector, and so on. With an IRQ loader on the serial
or IEEE-488 bus, it could do a little better, but I don't think the
time difference between parallel and serial connection would become
any smaller in that way.
> Now I wonder what is the fastest speed the 6551 can run in the Plus/4 and
> CBM-II?
The SwiftLink overclocks the 6551 to double speed by using a higher-frequency
crystal. I don't know, maybe it could be overclocked further by changing
the crystal, but there must be some limitation. 76,800 bps (with quadruple
crystal frequency) doesn't sound familiar, and neither does 153,600. You'd
have to configure the other end of the RS-232 line to use the non-standard
speed as well, or you'd need an odd-frequency crystal.
> And since the CBM-II runs at 2MHz it should give some speed boost over the
> C64, and it has an IEEE-488 port built right in. Maybe I found a new transfer
> machine :-) Now I only need my PLAs back to fire up the 610 again.
I find it somewhat annoying that the CBM II has an on-board cassette
connector but no ROM support for it. Thus, the machine cannot be
bootstrapped via the C2N232; the software has to be transferred to disk
first.
Marko
Message was sent through the cbm-hackers mailing list
Archive generated by hypermail 2.1.6.