catching up from a mmonth

[jmk3@crux3.cit.cornell.edu (Jay Krell)]

>A bunch of replies: ROM 1/3, ORCA/C 2, error codes
>non-existant memory, A2U Compression, Winchester/HD?, Apple ///
 
>re: ROM 1 vs. ROM 3
 
re: speed
 
Apparently, a number of major optimizations have been made to Quickdraw II.
They are all in ROM in ROM 3, and _some_ of them have been added to ROM 1 via
RAM patches. Quickdraw is a very significant bottleneck.
 
re: Memory Peeker and Visit Monitor CDA's
 
>And yet another... One version of the ROM 3 had memory peeker etc locked in..
>I would prefer this because they are in ROM and therefore not taking up any
>RAM (aside from whatever they might use when executed)..
 
There are not 2 versions of ROM 3. To install Memory Peeker and Visit Monitor,
type # at the system monitor prompt. Also, a bit in battery RAM allows them to
be permenently installed. I wrote 2 small apps to set and clear this bit. GS+
has a control panel that twiddles these bits. If the CDA's are permenenant on
your ROM 3 and you don't have any program to disable them, press
option-command-control-reset and set the battery RAM settings to the defaults
(option 2, unless you have 50 Hz monitor).
 
> do CDAs in ROM get transfered into RAM to run?
 
Abolutely not.
 
>That means that if you are using an accelerator, ROM 1 will be faster since
>cache RAM is now faster than ROM.
 
Once code is run from ROM, it will be placed in the cache. The cache is not a
cache for RAM but a cache of everything the processor accesses.
 
>Andy McFadden's A2U Compress course on GEnie
 
Can Andy's courses be distributed on internet? After the course is over?
I would think not, but I didn't see any notices on GEnie to the contrary.
 
>Subject: GS/OS error $0046 FF0F
 
I strongly recommend Nifty List 3.4 to everyone who ever wondered what an error
code meant. It is a fairly advanced programmers' tool, but answering the above
question is as easy as typing
"46\err"
Typing "0\err" gives you a complete list of error codes. Granted, many are
cryptic to users, but OS ones are most likely to be seen by users and many are
not too technical. NL _is_ overkill for looking up errors, but it is the quick
and easy. For self-test codes, there is an article floating around that is
actually a technote missing its header.
 
>From: ci167@cleveland.Freenet.Edu (David Hoppe)
>Subject: Apple ///(not a misprint)
 
>  Ok, I just recieved a couple of Apple /// computers, along with
>a couple Apple HD's, and disk drves. Wht I need issome software,
 
Dave,
  CompuServe has an Apple /// area in the Apple II area that might be worth
checking out.
 
 
>ORCA/C 2.0
 
I believe ORCA/C 2.0 has been shipping for about a week.
 
 
>From: sadasiva@utkvx3.utk.edu (Sadasivam, Krishna Madur)
>Subject: Winchester drive question
 
>Can someone tell me if a Winchester drive is a HD? If it's different than
 
I _think_ Winchester was an early manufacturer of hard drives.
 
 
>From: gdesroch@slate.mines.colorado.edu (DESROCHERS GARY FREDERIC)
>Subject: O.K. So the abort pin isn't connected. Now...
 
>    Then what actually happends when I access an address that does not
>exist.  Again this is on a IIgs.  I would like as in depth as you can give
 
Hardware Ref 2nd p. 52 under Ghost addresses: I don't fully understand the
preceding context, but it says "whenever accessing addresses beyond RAM
expansion-card limit, locations in a corresponding low bank are accessed."
On my ROM 3, accesses up to bank $7F show information that is obviously in
lower banks, as I only have 2 MB. Above bank $7F, access in a bank just return
the bank. ie- a dump of bank $80 reveals all $80's
 
 
  --Jay, jay.krell@cornell.edu