Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!utgpu!water!watnot!watmath!clyde!rutgers!mit-eddie!uw-beaver!tektronix!tekgen!tekigm!tekigm2!neals
From: neals@tekigm2.UUCP
Newsgroups: comp.sys.m6809
Subject: Re: CoCo3, Disto Controller, OS-9
Message-ID: <1558@tekigm2.TEK.COM>
Date: Fri, 13-Mar-87 20:42:10 EST
Article-I.D.: tekigm2.1558
Posted: Fri Mar 13 20:42:10 1987
Date-Received: Sat, 14-Mar-87 15:27:37 EST
References: <1608@lsuc.UUCP> <15129@amdcad.UUCP> <1389@loral.UUCP>
Reply-To: neals@tekigm2.UUCP (Neal A. Sedell)
Distribution: na
Organization: Tektronix, Inc., Beaverton, OR.
Lines: 36

In article <1389@loral.UUCP> dml@loral.UUCP (Dave Lewis) writes:
>In article <15129@amdcad.UUCP> bandy@amdcad.UUCP (Andy Beals) writes:
>>In article <1608@lsuc.UUCP> jimomura@lsuc.UUCP (Jim Omura) writes:
>>>problem is that the WD-1773 chip doesn't seem to read/write single density.
>>
>>The 1773 definitely does read/write single density.  Your vendor probably
>>doesn't program it correctly. :-(
>>	andy
>
>  The single/double density select on Radio Shack and J & M disk controllers
>is address $FF40, bit 5 ($20): 0 = single density, 1 = double density. Other
>manufacturers (Disto, etc.) probably do the same thing, but I can't be sure
>as I don't have a Disto controller.
>
>  The Radio Shack BASIC disk routines, and the RS/Microware CCDisk in OS-9,
>are both hard-coded with this bit set to a 1 for double density. I have
>considered writing single-density support into the new disk driver that I'm
>working on, but it would be a lot of trouble (and code, on a memory-limited
>machine) for little return.

Just thought I'd mention, before someone runs out and mod's Newdisk to clear
bit 5 of $FF40 to do single density, that besides being the double density
enable bit, it also enables the HALT circuitry.  This is because a 6809 at 
0.9MHz isn't fast enough to poll the status register and do data transfers
within the 23.5usec worst case (minimum) cycle time for double density, but
it is VERY close.  You'll have to include the polling stuff in the read/write
routines.

I think a nice solution would be for Tandy to include an internal disk
controller that uses DMA.  I wouldn't mind too much if they ripped the
COCO 2 compatibility stuff out of the GIME ;-).


Now coming to you from tekigm2,

Neal Sedell