[CoCo] 720kb vs 1.4mb 3.5" disks

[Gene Heskett]

On Saturday 13 December 2003 00:06, KnudsenMJ at aol.com wrote:
>In a message dated 12/12/03 10:01:25 PM Eastern Standard Time,
> wb8tyw at qsl.net
>
>writes:
>> Most of the issues come from operating systems deferring updates
>> to the disk from having large file cache memory.
>>  The COCO does not have lots of memory to buffer up I/O, so it is
>> less likely to have a power interruption at a critical time that
>> would leave orphaned space.
>
>Funny -- Coco's small RAM became a virtue here.  But ISTR that OS-9
> and OS-K have a deliberate design philosophy of no such write
> deferrals, so power losses and other crashes cannot trash the file
> system or even one file (unless it was being written at the
> moment).  The price, of course, is the slow disk writing we all
> suffer thru under OS-9 -- I think it writes one sector, updates the
> FAT, writes another, hits another FAT bit, etc.  But very safe.

Not quite Mike, it allocates one "sas" sectors, updates the fd sector, 
writes those, and if not done, allocates another "sas" sectors, 
updates the fd sector,  writes those, and continues till done.  When 
its done, it updates the allocation map and the fd sector in two 
final writes, trimming the last "sas" allocated down to just using 
the last sector needed and no more.  All that seeking on a floppy 
takes time.  And its the major reason I often set sas to FF before 
writing a big file, and revert it to 20 for normal use.  At 8, you 
soon run out of fd.seg space in the descriptor.

>>  If there is not a utility to check the integrity of the OS-9 file
>>  system, then there should be.
>
>There's a stock one under OS-K, but ask me its name and I'll flash
> my AARP card.  It collects orphaned sectors, catches sectors
> allocated to more than one file (bad!), etc.  Maybe it's called
> DCHECK, same as in DOS.  And ISTR there's a watered-down version
> under OS-9 too.
>
>>  The only issue I ever heard about that is long term storage of
>> the floppies with the heads clamped down was bad.
>
>It's physically the same issue.  Early Tandy drives had no head-load
>solenoid, so the head was clamped against the disk all the time that
> the lever was clamped over.  That would put a "dimple" in your
> diskette.
>
>  Later drives only load the head against the surface when the
> solenoid is energized by the driver software, and your drive can be
> set to unload the head when the disk is not doing I/O.  Power
> failure causes the head to retract -- hopefully *before* any
> glitches get to the heads.
>
>Newer hardware seems to be more forgiving.  --Mike K.

-- 
Cheers, Gene
AMD K6-III at 500mhz 320M
Athlon1600XP at 1400mhz  512M
99.22% setiathome rank, not too shabby for a WV hillbilly
Yahoo.com attornies please note, additions to this message
by Gene Heskett are:
Copyright 2003 by Maurice Eugene Heskett, all rights reserved.