Re: 1541 with an "harddisk"

g.j.p.a.a.baltissen_at_kader.hobby.nl
Date: 2000-09-01 10:29:31

Hallo Martijn,

> This will be a problem, since these lines are bidirectional. 286 and 386
> printerports mostly don't have a bidirectional printerport. Ofcourse,
> you can simulate this by multiplexing the available inputs, but I think
> that you'll be running out of time then. 

You only need two wires to make a bidirectional port of an old unidirectional. 
Works on most older LPT-cards, IBM-monochrome and MGP-cards. Does not work on 
(mostly newer) cards equiped with the 8211 (?) LPT-IC. This IC has an input to 
tristate its databus but the card is missing the unused outputs which the other 
cards have to steer this input.

> (It seems quite time-critical to me as it is right now anyway).

The floppy outputs a byte every 26 to 32 uSec. But the software has a build in 
timeout detector. I don't know the trigger value. I remember something like 64 
uSec. Using ML even IMHO this should be enough  for a 12 MHz AT, see below.

> Remember that the track has to be "GCRed" too. (Or you should use something
> like a G64 image from the start). 

Yes, this is what I have on my mind. If the input file is a D64, it first is 
converted to a G64-file. 

> After that, the only thing needed when the track is being changed is 
> modifying a pointer (which can be as easy as loading a segment register with 
> a new value)

My idea was loading the needed track into memory the moment I detect the 1541 
is changing track. Does anybody know how much time the 1541 reserves to be sure 
that the head is placed correctly above the track?

Groetjes, Ruud


-
This message was sent through the cbm-hackers mailing list.
To unsubscribe: echo unsubscribe | mail cbm-hackers-request@dot.tcm.hut.fi.

Archive generated by hypermail 2.1.1.