informis
/
textfiles
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
textfiles/messages/ALANWESTON/1990/os905_17.txt

2217 lines
72 KiB
Plaintext
Raw Blame History

This file contains invisible Unicode characters

This file contains invisible Unicode characters that are indistinguishable to humans but may be processed differently by a computer. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

29417 13-MAY 12:24 General Information
RE: Config (Re: Msg 28143)
From: RTHOM To: RADICAL
No, the documentation writers of course are not intentionally muddying the
waters just to sell more after-market books; but if they WOULD write more
legibly--understandably, I mean-- the aftermarket how-to books would have no
place to sell--they wouldnt be needed.
I thimk it's as you say--they're incapable of clear concise writing, and it's a
shame they aren't...
-*-
29429 13-MAY 16:33 General Information
RE: Config (Re: Msg 29417)
From: GREGL To: RTHOM (NR)
Ray,
While it's true that many manuals that come bundled with the software is not
very well written, the same goes for a lot of the after-market books. Come to
think of it, I've seen a lot of after-market books that were worse than the
original manuals. What I think is far worse is that a good many of the
after-market books are written for the absolute idiot that can't find the ANY
KEY on the keyboard and never go beyond simple addition in a spreadsheet or very
simplistic formatting in a word processor.
Perhaps I'm in a different level of sophistication than some users that have
never turned on a computer. I don't like manuals that tell you that you can do
"many wonderful and wonderous things" without telling you how to do it. But the
after-market books that tell you how to enter your name by telling you how to
press each and every key I dislike even more. Those make me feel like I'm being
belittled, a child that hasn't learned how to read and write. But it's a
tradeoff. It's difficult to write manuals as you have to calculate and write for
the widest market. If the company is going after the beginner then you'll get a
manual that tells you how to turn on the computer and get the program running by
telling you to type each and every keystroke. Yuck.
But if the company is targeting the technical market then you'll get manuals
that contain complex mathematical formulas and overviews of the commands.
Unfortunately, it seems that very few manuals hit the right combination to make
it usable for both the beginner and the advanced user.
-- Greg
-*-
End of Thread.
-*-
29418 13-MAY 12:44 General Information
RE: EMAIL (Re: Msg 29389)
From: MRGOOD To: ZACKSESSIONS
So, I guess a nice improvement in Delphi's modus operandi would be to make
binary file mail easier to receive? Just stating the obvious!
Hugo
-*-
29424 13-MAY 14:48 General Information
RE: EMAIL (Re: Msg 29418)
From: ZACKSESSIONS To: MRGOOD
Unfortunately that would be a major undertaking at Delphi. You see, "Delphi" is
actually a DEC VAXCluster system. They at Delphi decided to utilize the provided
mail interface, VAXMail, for Delphi users. VAXMail is owned by Digital, and I
doubt that they are going to modify VAXMail to suit Delphi's needs. Ya takes
what ya gets. Delphi could, however, custom design their own mail interface
which is more user friendly in the way of uploading and downloading, but that
was that "major undertaking" I alluded to in my first sentance.
Zack
-*-
29436 13-MAY 22:42 General Information
RE: EMAIL (Re: Msg 29424)
From: EDDIEKUNS To: ZACKSESSIONS
Not quite true, Zack. If Delphi added "send" (as in bitnet send) then it would
be trivial to send and recieve files. You would just:
send filename username
(perhaps from your workspace) to send a file to someone and you would:
recieve *
to recieve a file sent to you. I agree tho that to do this within MAIL would be
extremely unlikely.
Of course, it would be nice if Delphi did the above and then actually connected
to the bitnet. <Grin> Then I could be eddiekuns@delphi as well as ekuns@zodiac
and ekuns@fnald. <big grin> Gosh that would be nice.
Eddie
-*-
29440 13-MAY 23:16 General Information
RE: EMAIL (Re: Msg 29436)
From: ZACKSESSIONS To: EDDIEKUNS
Everything I said was true, but could be analyzed in different ways, which you
did.
There indeed could be a "send" command. Actually, no receive command would be
necessary, once the send command copied the file to your workspace, it's there.
Mail would not be used at all, unless not notify you that someone had "sent" you
a file, and what it's name/size is.
We have a similar facility in the General Electric DECnet network, the command
is called TRANSFER.
Zack
-*-
29450 14-MAY 01:22 General Information
RE: EMAIL (Re: Msg 29440)
From: EDDIEKUNS To: ZACKSESSIONS
Right -- everything you said was true. I guess I worded what I did too
strongly. :)
Anyway -- a 'recieve' command is a good thing when you get charged for disk
space! That way, if you happen to be gone for a month for whatever reason but
someone has sent you a 500k file, you won't lose an arm & leg in disk charges.
Also, it's how the VAX/VMS bitnet file transfer stuff works. :) So it already
exists and works. Of course, Delphi isn't on the bitnet, but I believe that the
software works anyway. (Well, I *know* it works to send files to folks on the
machine you're on. I just don't know if it required any bitnet stuff to be
there.)
Yearh -- stream of consciousness here. I should stop now (and goto bed?).
Anyway, it would be nice if there were an easier way of sending files to people
which was available to all users. Sending files via MAIL is tried-n-true, but
kinda klunky esp for binary files.
Eddie
-*-
End of Thread.
-*-
29419 13-MAY 13:14 General Information
RE: In the beginning...... (Re: Msg 29398)
From: CBJ To: DBEARISTO
Darryn,
If you don't already have it STARTING OS-9 by Paul K. Ward is an excellent
book for beginners. I have found it easier to use than the RAINBOW guides and
it is well paced for a beginner. I'm not saying that the other books aren't any
good but if you only buy one book this should be it. dPaul also joins us on
DELPHI from time to time and can probably answer questions on the book. It is
the only reason I have gotten my OS-9 package off the shelf again. I wish you
lots of luck with OS-9. It is fascinating.
---Carl---
-*-
29435 13-MAY 21:41 General Information
RE: In the beginning...... (Re: Msg 29419)
From: DBEARISTO To: CBJ
Would you happen to know Paul Ward's username so that I could ask him where I
may be able to get a copy of STARTING OS-9?
Darryn.
-*-
29439 13-MAY 22:48 General Information
RE: In the beginning...... (Re: Msg 29435)
From: EDDIEKUNS To: DBEARISTO (NR)
Paul Ward's username is PKW.
Eddie
-*-
29454 14-MAY 14:20 General Information
RE: In the beginning...... (Re: Msg 29398)
From: KEITHMARCH To: DBEARISTO (NR)
LOOK IN THE OS-9 MANUAL:
Page 5-9 of the section called Technical Reference.
*******************************
NOTE All numbers are in HEX
Info. in the () you do not have
to type in.
In the following file (after the c)
the numbers are as follows:
First Number = Relative Address (place in memory) (Ex: 14)
Second Number = Old Value (Ex: 00)
Third Number = New Value (Ex: 03)
If you need help with the BUILD and COBBLER commands.
See page 6-10 and 6-16 of the section called OS-9 Commands.
*******************************
Also do [TYPE] this at the OS-9 Prompt:
build patchmydisk <enter>
l XX (link in memory to XX = Drive number to patch Ex: /DD)
c 14 00 03 ($00=30ms $01=20ms $02=12ms $03=6ms Step Rate)
c 17 00 00 (You really don't need this line) (No changes)
c 18 23 28 ($23=35 Tks $28=40 Tks $50=80 Tks)
c 19 01 02 (Change from SS to DS)
v (Update the Header and CRC Values)
<enter>
Now TYPE this at the OS-9 Prompt:
modpatch patchmydisk <enter>
(Take off the Write Protect tab of your BOOT Disk)
cobbler /dd <enter> (/dd = Your BOOT Disk - You may have to change this)
If /D0 = /DD then (edit patchmydisk) for the other one also
-eof-
Now when you boot up, you'll have a 6 ms 40 Trk DS Disk Drive
Keith
-*-
End of Thread.
-*-
29423 13-MAY 13:36 Utilities
3 1/2" drive
From: BANCROFT To: ALL
I am going to reenter my previous message, as it was partially lost. I have a
mitsubishi 355c high density drive. I have put it on the coco 3. I have switch
the jumper from high density mode to switchable density mode. It is set up as
drive 1. My drive 0 is a TEAC 40 track single sided. My controler is the R.S.
fd-501. I used DMODE to set the device descriptor as folows: DRV=1, STP=0,
TYP=20, DNS=3, CYL=50, SID=2, VFY=0, SCT=12, TOS=12, ILV=3, SAS=8. I format it
and I get it formated with no errors. I can do a free and it gives me the same
number of sectors as the format does. But when ever I write to it, it starts
writing a LS 0, and overwrites the disk info. I have re-cleaned all connectors,
re-configed my system so it comes up with the new descriptor, but still have no
luck. Any ideas? By the way, I am going through a R.S. multi-pak. The newer
model, that is supposed to not need the pal chip upgrade. I also am using a
Burke & Burke 40 meg hard drive in slot 4 I mean 3, my disk is in slot 4. I
have never had problems with the hard drive, drive 0, or my previous drive 1,
which was also a TEAC 40 track single sided drive. Thanks for any help you can
give me.
-*-
29425 13-MAY 15:49 Utilities
RE: 3 1/2" drive (Re: Msg 29423)
From: BANCROFT To: BANCROFT (NR)
I have done some checking and found out that if the number of sectors I format
is less then hex 800, I am ok, and have no problems. But when the number of
sectors is hex 800 or more, than I start getting errors. My guess is that I
cannot have the high order bit set on the number of sectors? Aany takers on
this one?
-*-
29426 13-MAY 16:13 Utilities
RE: 3 1/2" drive (Re: Msg 29423)
From: OS9UGVP To: BANCROFT (NR)
I'm not familiar with that exact drive, but if its similar to my 3.5" Toshiba
high capacity drive, you should make sure that pin 2 of the floppy connector is
tied to the appropriate level. To a pull up resistor if the drive is set for
pin 2 high equals normal capacity, or to ground if the drive is set for pin 2
high equals high capacity. Perhaps your best bet is to set it back in
auto-sense mode, and then use only 720K disks (without the hole that indicates
high capacity media).
Bruce
-*-
29508 17-MAY 04:02 Utilities
RE: 3 1/2" drive (Re: Msg 29423)
From: OS9UGPRES To: BANCROFT (NR)
Are you using normal diskettes? You shouldn't use ones marked as being hi-
density (with two notches).
Oh. And what cable are you using? Do you have one with all pins intact, so that
the other side of the disk is actually being used? - kev
-*-
End of Thread.
-*-
29427 13-MAY 16:18 Utilities
RE: 3 1/2" (Re: Msg 29410)
From: GREGL To: BANCROFT (NR)
Brian,
Try changing DNS=03 to DNS=01. You aren't really using a 96 TPI disk or
drive, even though it is 80 tracks. It could be that OS-9 might be getting
confused and is trying to double step the heads. If I remember correctly, I had
similar problems when I used DNS=03 on my 3.5 inch drives and changing it
worked. Can't hurt to try it.
-- Greg
-*-
29428 13-MAY 16:21 General Information
RE: OS9 via RS-DOS? (Re: Msg 29413)
From: GREGL To: BILLBEISSERT (NR)
Bill,
There's a couple of utilities in the databases here that will make an RS-DOS
disk readable under OS-9. One is called RS-OS9.BAS and the other, I think, is
called COCOOS9.BAS. With these utilities you can download a terminal program
with your existing terminal, copy the file onto a blank (freshly formatted) disk
and then run RS-OS9.BAS on it. Once it's finished you can boot OS-9 and copy the
file onto a regular OS-9 disk.
-- Greg
-*-
29430 13-MAY 16:46 General Information
Multisync pinout
From: KEITHMARCH To: OS9UGPRES
This is the pinout for the hi-res DB-15 and DB-9 connectors:
PIN Descrition
1 Red
2 Green
3 Blue
4 Ground
5 Ground
6 Red Ground
7 Green Ground
8 Blue Ground
9 N.C.
10 Ground
11 Ground
12 N.C.
13 H. Sync
14 V. Sync
15 N.C.
PIN Descrition
1 Red
2 Green
3 Blue
4 H. Sync
5 V. Sync
6 Ground
7 Ground
8 Ground
9 Ground
-*-
29431 13-MAY 18:14 General Information
RE: smartwatch help! (Re: Msg 29372)
From: RICKLT To: PAGAN
Thanks for your input Stephen, but it sounds like you didn`t have any sockets
in your controller. My controller already had the 24 pin socket installed.
Every thing goes well as long as the smartwatch is not installed. But as soon
as I install the smart watch all I get is extended color basic the proces- or
refuses to see disk basic. Sorry I took so long to answer your help message,
but I have been very busy with other matters. Thanks again
Ricklt (Rick Tackett)
-*-
29472 15-MAY 20:18 General Information
RE: smartwatch help! (Re: Msg 29431)
From: ARTFLEXSER To: RICKLT (NR)
This is probably an unnecessary suggestion, but....are you sure you put the
SmartWatch in the correct socket?? Although the SmartWatch has 28 pins, it does
NOT go in the 28-pin socket of the controller, which is reserved for the disk
controller chip. It goes in the 24-pin socket, underneath the DOS chip, with
pins 1, 2, 27, and 28 hanging off the end. In addition, you must jumper the
hanging-off pin 28 to pin 26 to supply the SmartWatch with the needed +5 volts.
Another thing to watch out for is making sure the SmartWatch's notch faces the
same direction as that of the DOS chip.
-*-
End of Thread.
-*-
29432 13-MAY 18:59 Graphics & Music
RE: Umuse3 update (Re: Msg 26929)
From: JANG To: RAGTIMER
Yes, I got it a while back, and it's another great-one... How's the
"Full-score-print-out" coming..??
JANG
-*-
29446 13-MAY 23:55 Graphics & Music
RE: Umuse3 update (Re: Msg 29432)
From: RAGTIMER To: JANG (NR)
Glad you got the upgrade, finally. Well, I have some work started on the
printout, but there's artuiculation, and TV play, and the Lyra ocnverter, and
zillions of little things. Like the 4.6.0 version won't save your score upon
exit -- amazed nobody's hollered about that one yet, grtin.
-*-
End of Thread.
-*-
29433 13-MAY 19:53 General Information
ViewPerfect Multisync Monitor
From: KEITHMARCH To: OS9UGPRES
Hi Kevin: I got that Multisync Monitor to work with my 512K CoCo 3 I know it
will work with the MM/1
What is the latest info on OSK for the MM/1 and the Shell, (C Compiler if any) ?
:-)
Keith
-*-
29434 13-MAY 20:39 General Information
SCI
From: TEDJAEGER To: DISTO (NR)
I have an old SCI that works fine with a CoCo I but not my CoCo III. Do I
remember correctly that there is a fix (free?) for this? --Thanks, TedJaeger
-*-
29437 13-MAY 22:45 General Information
OS9 vs. UniFlex
From: MRGOOD To: PKW
I saw over on CIS that another operating system is being considered for the MM1
(as an additional choice or as the only choice, I'm not sure). If UniFlex, which
I know nothing about, is the chosen system, then I may think twice about buying
a new machine. Of course, I'm willing to hear arguments that would possibly sway
me in the other direction...
Seriously, are you in a position to elaborate, or should I wait til August to
find out?
Hugo
-*-
29467 15-MAY 12:18 General Information
RE: OS9 vs. UniFlex (Re: Msg 29437)
From: PKW To: MRGOOD
Hugo,
If you want to know the full story, you should probably give me a phone call at
the KLE offices. 202/232-4246 goes right to my diesk.
The choice of operating system for the MM/1 is flexible until we get the deals
that you people require.
Also, it makes sense that if an OS vendor is our "partner" then we'll make each
other successful. So we are arranging "partnerships" that will make us
successful, the OS vendor successful, and will also serve your needs.
Don't panic.
Paul
-*-
29484 16-MAY 06:52 General Information
RE: OS9 vs. UniFlex (Re: Msg 29467)
From: PAGAN To: PKW (NR)
I'd like to put in my vote for OS9 an the MM/1. I've used MS-DOS, UNIX, CPM and
the Mac-OS at one time or another over the years and still prefer OS9 on my
little COCO III over any of them. I havn't yet recieved any specs on the MM/1
yet but if it uses a 68X00 processor OS9 would be my "drug-of-choice" for it's
flexibility and speed.
Stephen(PAGAN)
-*-
29486 16-MAY 19:14 General Information
RE: OS9 vs. UniFlex (Re: Msg 29467)
From: MRGOOD To: PKW (NR)
Thanks for the reply concerning operating systems. I fully understand the 'deal'
aspect. It's just that I've invested several years in OS9. Not sure if I want to
start from scratch again, know what I mean?
I won't bother you at your office. I'm willing to wait and see. Of course, my
vote is for OS9.
Hugo
-*-
End of Thread.
-*-
29438 13-MAY 22:46 General Information
Shell+
From: EDDIEKUNS To: ALL
I've noticed that when I run make with:
make ^100
that make's priority is not altered. Also, if I do:
maze <>>>/w&<blank>
(ie: a blank after the '&') it runs maze but does weird things and loses it's
prompt! Well, I'll be specific:
OS9[Term]: maze <>>>/w&<blank>
&003
Error #215 - Bad Path Name
B
And thereafter there is no prompt. If I type 'p=' on a line, I get back the
default prompt. (Of course, if I don't put that extra blank there, it works
fine.)
I'm assuming that the above are bugs in shell+ 2.1?
Eddie
-*-
29506 17-MAY 04:02 General Information
RE: Shell+ (Re: Msg 29438)
From: OS9UGPRES To: EDDIEKUNS (NR)
Eddie -
I'm still using shell+ ver 1.3a, but even it had the "&[space]" error 215. Tho
at least I kept my prompt <grin>.
If nothing has changed from my original methods, then shell+ sets the child
priority after forking it. If the command is coming off disk at that moment (or
using the disk), then SetPriority will fail. Why?
Whenever a process is using the disk, RBF jacks up his priority by 3. Afterwards
he sets it back down again by 3. When shell sets his priority inbetween RBF's
shenanigans, it's lost.
kev
-*-
End of Thread.
-*-
29441 13-MAY 23:38 General Information
RE: Help! My computer hates me! <Grin> (Re: Msg 29352)
From: RAGTIMER To: EDDIEKUNS
Eddie, my whole setup still runs pretty cool (slightly warm) with my little fan
under the keyboard. But you are right, that Commode-door supply out way too
much juice. Glad I moved the jack off the Disto board and put in in the bottom
half shell of the case. Heroes on the half shell? Uh, not quite.
-*-
29442 13-MAY 23:40 General Information
RE: Help! My computer hates me! <Grin> (Re: Msg 29353)
From: RAGTIMER To: EDDIEKUNS
Right, the unpatched VDGInt screws up high-performance grafix programs (the ones
with sound, ya know :-).
No, I had upgraded the GIME long before I got the 1 Meg. I just plain don't see
that flash any more. It may have been due to some really heavy I/O going on at
the time -- that can freezxe the machine halfway between windows even.
-*-
29451 14-MAY 01:24 General Information
RE: Help! My computer hates me! <Grin> (Re: Msg 29442)
From: EDDIEKUNS To: RAGTIMER (NR)
Freeze the machine 1/2way between windows? Yikes! That I'd like to see (on
someone elses machine <grin>)! You mean you hit the <Clear> key and it doesn't
quite make it to the next window?
Eddie
-*-
End of Thread.
-*-
29443 13-MAY 23:47 Programmers Den
RE: pointers (Re: Msg 29406)
From: RBSMITH To: GREGL
thanks greg i will study this code and try to get a handle on it.
i appreciate your time and trouble it is often very difficult to find answers .
rbsmith
-*-
29444 13-MAY 23:48 Graphics & Music
RE: UMusE3 (Re: Msg 29399)
From: RAGTIMER To: CTL56
Hi, thanks for getting the program, and the kind words. Yes it was designed so
you could enter music without understanding everything, and I'm glad that works
for you.
I also have an Mt240 (and a few others), and a friend has a 540 or 640, a big
brother. Turns out the instruction manual is a "generic" for all models, and
the 240 CANNOT receive MIDI clock on Channel 4 or any other. A shame, and sort
of false advertising, but I can't get mine to do it either. The rhythm clock
works great on my Yamaha PSS-480, which is about 5 times the synth for maybe 1.5
times the bucks.
You might try getting a 540 or 640 -- the rhythm supposedly works, and you get
Channel 3 back for another instrument voice too. Must admit the 240 has good
sounds, tho!
PS: I hope someone with a 240 breaks in here and says I'm all wet -- I'll take
the disgrace for learning how to drive the rhythm, grin! Enjoy the program, and
UPLOAD, pretty please...? mike k
-*-
29461 14-MAY 23:16 Graphics & Music
RE: UMusE3 (Re: Msg 29444)
From: CTL56 To: RAGTIMER (NR)
Thanks for the information. I do have another question. Can you drive another
keyboard thru the MT240, as to drive another MT240?
It was just a question, by the way, I got a good deal on mine. I got for only
$80.00. It was on sale on the old CVN (shopping channel).
By the way, did you say the 4.6.0 Version is out now? If it is, do I need to
contact Second City for the upgrade?
-*-
End of Thread.
-*-
29445 13-MAY 23:53 General Information
RE: Music Formats (Re: Msg 29415)
From: RAGTIMER To: BRIANWHITE (NR)
Well lets see. Somewhere I have the chip maker's specs on the AY-1013 chip or
whatever it is, of which there are four in the SYm-12. Also I once worked out a
state diagram for an assembly program to drive those chips -- it isn't as simple
as it should be. odds of finding that today are not very good (by today I mean
these days, sorry).
I also have the Speech Systems software for the thing, whose docs include some
chip info. Also their program (this is RSDOS of course) has a very low- level
mode where it tickles the chips pretty directly.
Any info would have to be by US Paper Mail (lick it & stick it). Keep bugging me
and maybe it will happen. If I hadn't figured out a way to afford some
synthesizxers, I probably would have done some neat things with that
Symph-12....mike k
-*-
29447 14-MAY 00:01 General Information
RE: ShellPlus blowing Umuse3 memory (Re: Msg 29347)
From: RAGTIMER To: GREGL
OK, thanks for the deatails. I did experiment a bit and I'm convinced that the
Kreider malloc() is doing the whole thing -- it's making the F$Mem calls and
getting more pages and 8K blocks from OS9.
When I tried running Umuse3 #31, it ran perfect untill I went to do a direcdtory
of the .ume files -- then it #207'ed out. (New verb -- to 207). That's where it
malloc()'s.
I'd expect that OS9 keeps track of 256-byte pages, just for program modules that
are merged together. --mike k
-*-
29448 14-MAY 00:05 General Information
RE: ShellPlus blowing Umuse3 memory (Re: Msg 29347)
From: RAGTIMER To: GREGL
Some more details on my test -- Seemed that malloc() would NOT use memory
already allocated by a # in the command line. But actually it might have,
except the block size my program requested was probably too big to fit betwwen
current memory end and the stack. Malloc() does have to return a solid block of
RAM, so in my case had to go outside to F$MEM.
-*-
29503 17-MAY 04:01 General Information
RE: ShellPlus blowing Umuse3 memory (Re: Msg 29347)
From: OS9UGPRES To: GREGL (NR)
Greg, that's correct. P$PagCnt in the process descriptor keeps track of how many
contiguous pages are allocated within the RAM block(s) assigned to your data
space.
-*-
End of Thread.
-*-
29449 14-MAY 00:11 Graphics & Music
Thanks for Chinese music
From: RAGTIMER To: BAMBOO
John -- thanks a lot for your uploads of Cinese gospel and folk music. I really
like that a lot, having heard some CHinese traditional music fests at work.
Fortunatley we in the West used pretty much the same scale as your people, grin.
I like your orchestration and percussion work.
Of course your previous massive uploads have been top quality too, but the
CHinese is really unique -- our Coco Club was very impressed wehn I gave an
Ultimuse demo featuring "Mountain", and re-orchestrated it for PSS-480 while
they watched and listened.
You put a lot of expression, feeling, and subtlety into those pieces. Have you
got your version 4.6.0 in the mail yet? It has the new Percussion Clef and
separate save/loads for MIDI channels and percussions as well as instruments.
Thanks a lot -- I don't knwow when you sleep or work, but keep the coming! --
mike k
-*-
29452 14-MAY 01:34 Applications
GSHELL +
From: COCOKIWI To: ALL
I HAVE a problem with the IPC patch files that came with it! I cannot patch the
scf file.I get a "size is incorrect" when I try to patch" The SCFFIX pgm worked
fine! But BOTH ver of The Ipc file will not patch. Ver 2.00.01 Os9 Lev II.Org
copy! What gives ! any Ideas! Gshell + works fine,no problems patching! H-E-L-P.
regards Dennis
-*-
29453 14-MAY 02:36 68K-OS9
OSK System For Sale
From: JAYTRUESDALE To: ALL
For Sale: HCS 68008 OS-9/68000 system. This is a big message, a bit over 5000
characters, do you wish to continue reading?
I have a Hazelwood Computers UniQuad I computer which was sold by Frank Hogg
Labs as the QT (this is NOT to be confused with the QT+ which is a Hazelwood
Computers UniQuad II). This is a single board system with the same dimensions
as a 5.25 inch disk drive and includes mounting holes for mounting on such a
drive. The Microprocessor is an 8 Mhz 68008, there is 512K of RAM memory which
is the maximum. The boot EPROM can be up to 32K in size (27256) and can contain
OS9 modules in addition to the boot code, I have not gotten around to doing this
yet (which would save some RAM). The system has four serial ports with jumper
selectable baud rates from 300 to 19.2 BPS. There are two parallel ports, one
of which is used for a battery backed up clock.
Up to four 80 track 96 tpi DS/QD (NOT IBM "HD") floppy drives can be attached, I
only have one floppy drive (a Teac).
The hard disk controller is an Adaptec ACB-4070 RLL controller and is attached
to a Tulin TL240 RLL hard disk drive. This combination gives 44.75 megabytes of
storage according to DCHECK using 256 byte sectors. If storage gets tight, 512
or 1024 byte physical sectors could be used (the HD driver takes care of the
mapping from/to the 256 byte OS9 logical sector) although this can result in
slightly slower performace depending on your applications.
The system is mounted in a Baby AT style flip-top case which has provisions for
mounting three drives, with one internal to the case. Currently I have the hard
disk and the floppy drive mounted in the two externally visible slots, the hard
disk could be mounted in the non-visible location to free up another external
drive slot. The drive LED on the PC panel is not currently used and could be
wired up to the drive to indicate disk activity.
The power supply is a 40 watt switching supply. The case is cooled by a fan.
The computer and hard disk controller are mounted above each other on standoffs
over in the area that would normally be occupied by the PC motherboard. One of
the parallel ports is wired to a male DB-25 connector on the back panel for use
with a printer. All of the serial ports are wired to female DB-25 connectors
located in the area normally used by PC expansion cards, one of the ports has
been wired as a DTE for use with a modem. Using a standard 6 pin dip socket
connector I ran the low (300, 1200, 2400) baud rate selection lines to a switch
located on the bottom of the front panel, this allows changing the modem port
baud rate without having to open up the case and changing the baud rate jumpers.
I have all of the documentation for the hardware which includes pinouts, jumper
settings, parts lists, parts placement diagrams, and schematics (sad to say but
this is MUCH more documentation than is provided with either the 20x or the
K-Bus systems). I have the manual for the hard disk controller and the hard
drive. The controller manual is pretty detailed, the drive manual isn't but it
gives you all the information you really need. I doubt that I have the manual
for the Teac Floppy drive but I might be able to obtain a copy of one that is
pretty close.
I have version 2.2 of OS-9/68000 which is one behind the current version of 2.3;
the upgrade to the Frank Hogg Labs QT20x and QT+ systems at work was $75.00, so
I suspect this would be the same. This includes Microware BASIC and the
assembler/linker/debugger but *NO* C compiler. The board came with the
Stylograph word processor (includes mail merge and spell checker) which I don't
like or use, I did upgrade it to a more recent version. The board also came
with the Dynacalc spread sheet program which I do occasionally use, I'll miss it
if I sell this system! I have the documentation for both programs. Also
included would be the programs "Qcom" and "fbu" from FHL. Qcom is a
communications program that supports Xmodem (note that with the program Zmodem
the x, y, and Zmodem protocols can be used from a sub-shell of Qcom) and fbu
(<F>loppy <B>ackup <U>tility is a hard disk backup program. I use FBU because
it has the ability to go back and verify the integrity of a backup, something
that Microware's fsave and frestore backup programs can't do. The board came
minus OSK manuals so I ended up purcahsing a complete set of Professional
OS-9/68000 manuals from Microware which included the C compiler and uMacs
manuals eventhough I don't have that software. I have all of the receipts for
the hardware and software.
I have some selected software from Compuserve and elsewhere that I'd include
with the system (CKermit, Zoo, Zmodem, etc.). I'll also include a copy of the
OS-9 Users Group ten disk library set and a copy of the TOPS public domain Unix
like software disk set.
I have a copy of the origional brochure for the UniQuad board if anyone would
like more details. If you have any questions feel free to ask me here via mail
or during the day at work at (314) 426-7781 8:30 am - 4:00 PM CST. Considering
all that I've invested in this system, the asking price is $800.00 but I'll at
least consider other offers.
-Jay Truesdale 5/13/90 

-*-
29455 14-MAY 19:55 General Information
RE: Auto-Follow Mouse (Re: Msg 29343)
From: DENNYSKALA To: ZACKSESSIONS
Right you are, Zack. 'Control -e' does indeed do the trick, but it really
trashes an ordinary shell window --- no keyboard echo. I guess I could try
sending the output to a junk window, or perhaps open up a sacrificial window
just for that purpose. Thanks for the suggestion.
***** Dennis *****
-*-
29459 14-MAY 21:18 General Information
RE: Auto-Follow Mouse (Re: Msg 29455)
From: EDDIEKUNS To: DENNYSKALA
I put the control -e in my startup file as the last thing. It doesn't matter
that it trashes the tmode parameters of the startup shell. :)
Eddie
-*-
29462 15-MAY 00:17 General Information
RE: Auto-Follow Mouse (Re: Msg 29455)
From: OS9UGVP To: DENNYSKALA
Dennis,
Yes, you can probably do something like "control -e <>>>/w", which should not
permanently mess up keyboard echo, assuming you have a spare device window
available. An alternative "from the other place" is SetM (SetMouse) by Fred
Sawtelle, I believe. Look for "setm.ar" in their utilities, or maybe CoCo,
library.
Bruce
-*-
29502 17-MAY 04:00 General Information
RE: Auto-Follow Mouse (Re: Msg 29333)
From: OS9UGPRES To: DENNYSKALA
PROCEDURE Automouse
(* Denny, hey so write a little program <grin>!
(* That's what basic09 and syscall are for.
(* Here's an example of writing a "correct" command:
(* Note that it checks for a proper parameter by
(* assigning the param to a variable... failure causes
(* it to goto line 100 (help). - Kevin
PARAM switch$
TYPE stack=CC,A,B,DP:BYTE; X,Y,U:INTEGER
DIM reg:stack
DIM I_SetStt,SS_Mouse:BYTE
I_SetStt=$8E
SS_Mouse=$89
ON ERROR GOTO 100
onoff$=switch$
IF onoff$="on" THEN
reg.X=$0301
reg.Y=$01
ELSE IF onoff$="off" THEN
reg.X=$00
reg.Y=$00
ELSE GOTO 100
ENDIF
ENDIF
reg.A=1
reg.B=SS_Mouse
RUN syscall(I_SetStt,reg)
IF onoff$="on" THEN
RUN gfx2("GCSet",$CA,1)
ELSE
RUN gfx2("GCSet",0,0)
ENDIF
END
100 (* Give Help If Bad Params:
PRINT #2,"Automouse: enable/disable auto mouse and cursor"
PRINT #2,"Usage: automouse on "
PRINT #2," automouse off"
END
-*-
29522 17-MAY 20:17 General Information
RE: Auto-Follow Mouse (Re: Msg 29459)
From: DENNYSKALA To: EDDIEKUNS (NR)
OK thanks, Eddie. That sounds good. Is there any price to pay for leaving the
auto-follow mouse on all the time? And if so, how would one turn it off?
***** Dennis *****
-*-
29523 17-MAY 20:20 General Information
RE: Auto-Follow Mouse (Re: Msg 29502)
From: DENNYSKALA To: OS9UGPRES (NR)
Thank for the input. But its *SOOO* inelegant to write a program if one can
cajole the OS into doing the job. <grin>. I'll give the program a try thoug.
***** Dennis *****
-*-
29524 17-MAY 20:21 General Information
RE: Auto-Follow Mouse (Re: Msg 29462)
From: DENNYSKALA To: OS9UGVP
Thanks for the tip, Bruce. I'll look up that utility.
***** Dennis *****
-*-
End of Thread.
-*-
29456 14-MAY 21:13 General Information
RE: new user (Re: Msg 29355)
From: CAROLYNA To: EDDIEKUNS
Thanks for reply. I'll keep trying. --Carolyn
-*-
29457 14-MAY 21:15 General Information
RE: new user (Re: Msg 29391)
From: CAROLYNA To: CIZZIJR (NR)
Okay! Thanks, I'm going to try it now. Carolyn Agenjo
-*-
29458 14-MAY 21:18 General Information
RE: new user (Re: Msg 29400)
From: CAROLYNA To: NES
Thanks for all that help. Maybe I'll get somewhere now.
--Carolyn Agenjo
-*-
End of Thread.
-*-
29460 14-MAY 22:58 General Information
"Diode Hack"
From: CTL56 To: ALL
I have just learned the reason for my Keyboard locking-up on Comm Program is
that it needs a "Diode Hack", Can someone advise me of where I can get the
information to do this operation? I have some basic information in electronics
so that I could do the changes without any problems, I just need to know what
and where to do it.
Thanks for ths info.
-*-
29464 15-MAY 01:51 General Information
RE: "Diode Hack" (Re: Msg 29460)
From: CTL56 To: CTL56
Well, I found the answer to my own question, right under my own nose. It was on
page 50 of the November issue of Rainbow. I will try the fix and see if it
works.
Thanks for all of the help on other things and I hope this information will
help others.
-*-
29466 15-MAY 03:58 General Information
RE: "Diode Hack" (Re: Msg 29464)
From: KNOT1 To: CTL56
Yes, I just did those hacks on Friday (and these were my FIRST hardware hacks
ever) and they are Fantastic! Marty explained them very well, which was why I
was willing to give them a try. Before I did the fixes I had about a 30% to 40%
chance of successfully downloading a 20K file with YMODEM at 2400 bps.
Afterwards I tested it with a 200K file (30 minutes!) without so much as a bad
block, much less a lock-up! I definitely recommend these fixes to anyone such a
problem! Good luck!
-Jamie Wilmoth (KNOT1)-
-*-
29476 15-MAY 23:10 General Information
RE: "Diode Hack" (Re: Msg 29466)
From: CTL56 To: KNOT1
Thanks for the information. I will try to preform thoses "HACKS" tomorrow. It
does't look that hard and I hope that I don't have to many problems
Thanks again.
-*-
End of Thread.
-*-
29463 15-MAY 01:27 General Information
database
From: RADICAL To: ALL
Current Rainbow os9 report lists several new uploads in the utility section. I
have searched both the Utility and the New Upload areas with no results. Perhaps
it would not hurt for the Rainbow to list some keywords to find these uploads.
In Particular I was looking for th SDIR by ZackSessions and the Wcreate by
KingTrent. Thanks to both for the uploads, but where are they? Len
-*-
29468 15-MAY 18:08 General Information
RE: database (Re: Msg 29463)
From: ZACKSESSIONS To: RADICAL
I can tell you where SDIR is. I decided to have it removed from the library
since I have released it on an OS9 Level 2 Utilitites disk available from me.
Send me mail if interested. (SDIR was in the library for MONTHS, so you must be
looking at an old Rainbow, also, Uploads data in the Rainbow is 3 to 4 month
behind due to publication schedules.)
Zack
-*-
29477 15-MAY 23:21 General Information
RE: database (Re: Msg 29468)
From: RADICAL To: ZACKSESSIONS
Zack, Sorry to disturb you. The Rainbow was June 1990. It is your right to
remove the program, and I guess it is no supprise that they are months behind.
Thanks Len
-*-
29485 16-MAY 19:03 General Information
RE: database (Re: Msg 29477)
From: ZACKSESSIONS To: RADICAL (NR)
Hey, Len, you aren't distubing me at all! As a matter of fact, I just received
my June 1990 Rainbow, and sure enough, SDir was mentioned in the Delphi column,
this was not my original posting of this program, but it WAS a major improvement
over the original posting. I am considering re-uploading it to Delphi ONLY, for
the benefit of Rainbow readers. Will let ya know.
Zack
-*-
End of Thread.
-*-
29465 15-MAY 02:19 Programmers Den
Multi-Vue programming
From: ALPHASOFT To: ALL
Does anyone know if it is possible to have overlay windows on a menu window and
still access the menu bar? I believe this is impossible, but I thought I should
check. GShell appears to be doing it with its dir listing window, but on closer
look, the window is overlaid, drawn, and closed (without redraw).
I also need to know how to have the menu bar transferred to an overlay window.
I OWSet the window, then call _ss_wset to set up the menu, and it half does it,
but the main menu bar is also updated! Any help would be appreciated.
Keith Alphonso
-*-
29469 15-MAY 18:11 Programmers Den
RE: Multi-Vue programming (Re: Msg 29465)
From: ZACKSESSIONS To: ALPHASOFT
I don't think you can get there from here! <grin> Seriously, I don't think you
can access ANYTHING outside of an overlay window as long as the overlay window
is open. I don't think overlay windows support the menu bar as well, umm that is
the same menu bar which is in the window which was overlaid. You can fork a
process in an overlay window which has it's own menu bar, however.
Zack
-*-
29470 15-MAY 18:44 Programmers Den
RE: Multi-Vue programming (Re: Msg 29469)
From: DODGECOLT To: ALPHASOFT
As Zack said, Windint doesn't allow you to select menu items from outside of an
overlay. GShell probably just draws the dir listing on the screen, as all of the
window graphics (close box, title pattern, etc.) are available with the 200,3
font.
As for menu bars on overlay windows, yes and no. Due to bugs in Windint, you
can't have more than two menu bars (main window plus overlay) that will work
properly. This has been fixed for the upgrade (whenever it comes out :(. The
problem stems from Windint's handling of window/overlay coordinates.
Additionally, windint has big problems with a second framed window with scroll
bars! This can be very frustrating when you want to get a click on the up-scroll
icon (it doesn't recognize it.)
God I hope I haven't been too confusing! Yell if you need a more complete
answer...
-Mike
-*-
29509 17-MAY 04:03 Programmers Den
RE: Multi-Vue programming (Re: Msg 29465)
From: OS9UGPRES To: ALPHASOFT (NR)
Keith,
As to transferring the menu bar to an overlay, I suggest (for no really good
reason) having another process handle the overlay. I think windint can get
confused, since it uses the process ID as one method of checking whose menu info
to use. Same ID = weird results.
Overlay windows and keep access to main menubar? Not normally. But there are an
awful lot of tricks you can play... I haven't tried this but it should work.
Here's one example:
1. start with window /WX.
2. turn off window protect on /WX.
3. dwset type 00 /WY on top of /WX.
4. now use /WX for menus, /WY for overlays (output only).
Simple, and I think it should work fine. If you do need keyboard input from the
overlay, just temporarily Select it (1B21). All you're doing is opening two
windows, both sharing the same video memory and screen location.
- kevin
-*-
End of Thread.
-*-
29471 15-MAY 20:18 General Information
my computer hated me!
From: XLIONX To: EDDIEKUNS
Howdy Eddie,
-*-
29479 15-MAY 23:59 General Information
RE: my computer hated me! (Re: Msg 29471)
From: EDDIEKUNS To: XLIONX
Hi! Were you intending to send me a more complete message? 29471 seems to have
been cut a little short.
Eddie
-*-
End of Thread.
-*-
29473 15-MAY 20:34 General Information
Help! My computer hated me!
From: XLIONX To: ALL
Howdy ALL,
Another case of the "1Megger-GIMME-JEEBIES" cured with a $23.00 dose of 1987
GIMME juice! All is well and good in the world again (at least on my computer
bench!).
With the release of the SHELL+ 2.1 patch for memory (shell was too generous) and
the new GIMME, even UltiMuse3, Carmen SanDiego and many mazes work just fine!
Just another update from someone who has been sweating it out, waiting for the
new GIMME to come from TANDY. Just like with cars..."It can't 'just' be a $22.00
part thats gonna fix it...There 'has' to be something wrong $$$!!!
Hope ALL of your fixes go as easy as mine did!
-Mark W. Farrell -XLIONX (DELPHI) -SIGOp ProSIG (Pinball Haven RIBBS
708-428-8445)
(Stop by and visit us sometime!)
-*-
29480 16-MAY 00:01 General Information
RE: Help! My computer hated me! (Re: Msg 29473)
From: EDDIEKUNS To: XLIONX
That's good news, as just a couple of days ago I ordered a new GIME! I hope I
have the same good luck!
Eddie
-*-
29490 16-MAY 22:54 General Information
RE: Help! My computer hated me! (Re: Msg 29480)
From: XLIONX To: EDDIEKUNS
Howdy Eddie, I was at RBFest standing behind you when you were trying to get
Cosmics (I think) no halt going. I think you will have no problems at all as I
had ALL of them before, and none of them after! I can run maze, Carmen and
UltiMuse3 ALL at the same time and no bugs!!!
I am seriously thinking of either getting another COCO3 as a working spare and
/or at least another GIMME as a backup!
It's nice to be able to use the 1Meg upgrade as needed now.
Good luck!
-Mark W. Farrell -SIGOp ProSIG (Pinball Haven RIBBS (708)426-8228) -XLIONX
(DELPHI)
-*-
29492 16-MAY 23:34 General Information
RE: Help! My computer hated me! (Re: Msg 29490)
From: EDDIEKUNS To: XLIONX (NR)
BTW -- have you ever been to a meeting of the Glenside CoCo club? I notice you
live in the general area.
Eddie
-*-
End of Thread.
-*-
29474 15-MAY 20:47 Programmers Den
OSK C Errors
From: JAYTRUESDALE To: ALL
I don't know much about C. I get the following error message from the Microware
OSK C compiler when I compile 'tcmp' (from the UG disks) which was written for
OS-9/6809 C.
"tcmp.c", line 77: **** warning - illegal pointer/integer combination ****
if (s1 != (-1))
^ s1 is defined as a pointer as follows:
char *s1, *s2; /*pointers to current lines being compared*/
The origional source reads:
if (s1 != EOF)
free(s1);
I see the problem but don't know what to do about it, how do I 'legally' fix the
comparison so the compiler is happy?
-J

-*-
29475 15-MAY 20:57 Programmers Den
RE: OSK C Errors (Re: Msg 29474)
From: TRIX To: JAYTRUESDALE (NR)
Seems to me that the source should read:
if(*s1 != EOF)
instead. That way your're comparing what s1 POINTS TO and comparing that to the
EOF character. But that's just my opinion having neither seen the program nor
used OSK.
-John.
-*-
29481 16-MAY 00:32 Programmers Den
RE: OSK C Errors (Re: Msg 29474)
From: EDDIEKUNS To: JAYTRUESDALE (NR)
One way you could fix it would be to:
if (s1 != (char *) EOF)
or something like that. This forces the EOF to be treated like a pointer.
Eddie
-*-
29487 16-MAY 21:13 Programmers Den
RE: OSK C Errors (Re: Msg 29474)
From: GREGL To: JAYTRUESDALE (NR)
Jay,
Apparently the OSk C compiler doesn't like mixing and matching pointers with
integers. In the source code you are using, (s1) contains the address of the
string and (*s1) is the string itself. The routine is probably getting input via
gets(s1) which reads the input and copies it into the string with a null
terminator. The particular method to fix it depends on how EOF is returned by
the specific function.
From the code fragment it appears that the input function used is returning
an "address of" a string or EOF (-1) on end-of-file. Although an address of -1
is illegal it might be treated as 0xFFFFFFFF. In that case you can try to fix it
with:
if(s1 != (char *) EOF)
That will cast EOF (-1) into a character pointer and might bufallo the compiler
enough to work. I have to admit, I'm not too keen on assigning an address of -1
to a pointer - not very portable and probably won't work half the time.
-- Greg
-*-
End of Thread.
-*-
29478 15-MAY 23:52 Utilities
New Dir
From: SCG To: KLINDSAY (NR)
Hello, just wanted to drop you a line and comment on your new dir command. Well
it is just GREAT, something I've been looking for for quite a while. I have one
question tho: when I do a "dir -rn aif.*" I get a display of all the aif files
in my root dir, but it doesn't pick up the aif files in other directories ie:
/H0/APPLICATIONS, /H0/PICS, and so on. Am I doing something wrong?
Just wondering?!
Steve Gilbert (SCG) **OS9 Sysop** ***PC BBS*** 516-795-5974
-*-
29482 16-MAY 06:25 General Information
smartwatch
From: PAGAN To: RICKLT (NR)
Rick,
Did I goof? I thought I made it clear that the 24 pin socket should be replaced
with a 28 pin socket (see message 29472 from ARTFLEXER for an alternative,
though). It's that +5V at pin 28 that gave me the problem.
Stephen (PAGAN)
-*-
29483 16-MAY 06:41 Users Group
RE: MOTD (Re: Msg 29412)
From: PAGAN To: OS9BERT
I can sypmpathize with you about user interfaces. I've been working on a
hypertext program for LII for about a year now and have about 2 dozen modules to
handle the hypertext part but the d**n user interface and text editor parts are
driving me CRAZY.
Stephen (PAGAN)
-*-
29489 16-MAY 22:39 Users Group
RE: MOTD (Re: Msg 29483)
From: OS9BERT To: PAGAN (NR)
This month's issue of Dr. Dobb's Journal has several articles dedicated to
Hypertext! Take care.
Bert Schneider
-*-
End of Thread.
-*-
29488 16-MAY 21:33 Device Drivers
RE: SCII reads, but write hangs (Re: Msg 29324)
From: DISTO To: SANDYT
Sandy, is your multi-Pak modified for the COCO 3? If not, try it without it.
Also, what mods do you have to OS9? Try starting from the original disks and no
other drivers. If this works, continue from there. -Tony.
-*-
29491 16-MAY 23:24 General Information
Problems, problems...
From: KMTHOMPSON To: ALL
I'M EXPERIENCING TECHNICAL DIFFICULTIES
Am having a problem with my CoCo 3. I've had a few minor problems with
my system locking up for no explainable reason, but now I'm having a more
severe problem. Anytime I use a program that accesses the LEFT joystick
port, I get a loud "BZZZZZZZZZZZZZZ" through the speaker. At first I
thought it was the volume up to loud, but that wasn't so. I then unplugged
the joystick thinking perhaps it was fried. No luck. So, I'm lost. As
I'm not a technical person, the only step further I took was to do a Ram
test (OK) and to clean contacts on my GIME and M-Pak. (No luck here,
either.) I have not yet tested it under RS-DOS, I'm unsure whether it will
happen there. Under 512k OS-9 Level II, it is a problem.
Any help would be GREATLY appreciated. When using KQ3 and Koronis Rift,
Rescue on Fractalus, etc. (all use left port) I hear the noise. Under
Microscopic Mission the joystick is even unusable, it just kinda jumps
around. (Left port, again.) Funny--Just tried it with Hordes & Holes with
the Left AND Right Joystick port, and got the sound BOTH places--but I
don't with Hi or Lo-Res mouse in Multi-View. EITHER port. I dunno.
HHHHHHHHHHEEEEEEEEEEEEEEEEEELLLLLLLLLLLLLLLLLLLPPPPPPPPPPPPPPP!!!!!!!!
<Talk about annoying>
Marty, Greg, Kev? Any ideas? Anyone... Anyone... ?
--]<elly
P.S. Kevin, if you're too busy to write up a response to this, just send me
an MM/1, that should cure my problems. <grin>
-*-
29507 17-MAY 04:02 General Information
RE: Problems, problems... (Re: Msg 29491)
From: OS9UGPRES To: KMTHOMPSON (NR)
Kelly -
Well, the PIA at $FF00 controls the multiplexer in IC7, which selects both
joystick inputs and sound inputs at the same time. So either chip could be
wacky, or you could simply have software troubles.
Wish I could remember the poke, but you could try disabling sound at $FF23.
Perhaps I've said enough to get some of the hackers to jump in with ideas and
more info.
(apologies: had rear tooth pulled today, and I'm sleepy ;-)
-*-
29526 17-MAY 21:11 General Information
RE: Problems, problems... (Re: Msg 29491)
From: OS9UGVP To: KMTHOMPSON (NR)
Kelly,
A friend's CoCo 3 had a similar problem, except it "buzzed" all the time, not
just when a particular joystick was accessed. We replaced C63 (220 uF, 25V
electrolytic) which smooths the 8VDC regulator's input voltage. The 8VDC supply
is used in the RF "can" and also in the sound output. We also had to replace
C31 (220 uF, 16V electrolytic) which smooths the -12V (more or less) power to
the "SALT" chip... but I don't think that was contributing to the "buzz".
Bruce
-*-
End of Thread.
-*-
29493 16-MAY 23:34 Programmers Den
RE: RMA/RLink boo-boo. (Re: Msg 29382)
From: TIMKOONCE To: EDDIEKUNS
Declaring arrays as DP vars is definitely evil. Extended addressing is EVIL.
Now, if you mean 16-bit offset indexed addressing <grin>, that's different.
Arrays are usually accessed with indexed addressing, via such statements as:
LEAX ARRAY,Y to get the base address of the array. If ARRAY starts at offset
beyond 127 in the direct page, this _will_ cause a problem. Of course, as
someone mentioned, the C compiler might be smart enough to ignore "direct" for
arrays. Also, anytime you take the address of a variable with the "&" operator,
you risk the same thing. In general, I would strongly recommend not having more
than 127 bytes of DP vars in a C program.
- Tim
-*-
29494 16-MAY 23:37 Programmers Den
RE: RMA/RLink boo-boo. (Re: Msg 29493)
From: EDDIEKUNS To: TIMKOONCE
KBCom now (as of recent versions) has less than 128 bytes of DP vars, but older
versions pretty much filled up the DP! Yes, the C compiler is smart enough to
access DP var arrays (and some pointer references) as indexed addressing (and
thus not DP!). Thus, you get no gain by putting arrays into the DP, as they are
accessed just like they normally would be. I think James Jones pointed this out
to me. :)
Eddie
-*-
29495 16-MAY 23:43 Programmers Den
RE: RMA/RLink boo-boo. (Re: Msg 29376)
From: TIMKOONCE To: GREGL (NR)
First, Greg, I sincerely hope you weren't suggesting someone use the statement
LDA >DP_var under some circumstances, since that generates _extended_
addressing, which is definitely not PIC, and therefore EVIL.
There is a relatively simple way to handle this that would allow RLink to at
least detect this problem. Currently, RLink will report an error if there are
more than 256 bytes of DP vars (I know! :-). If RMA flagged references in the
ROF file as being "signed" or "unsigned" offsets, then it would certainly help.
Clearly, DP addressing is unsigned, and 8-bit offset indexed addressing is
signed. Simply adding this level of discrimination would allow RLink to at
least detect this problem, rather than generating incorrect code!
As far as I can see, the best advice for C programmers is simply to NEVER
have more than 127 bytes of "direct" vars. For RMA programmers, there's a
little more control, but if they do have more than 127 bytes of DP vars, they
should be careful to NOT use 8-bit offset indexed addressing for DP vars, i.e.
they should force 16-bit offsets with P_Var,Y" (or ,U depending on which
convention you use), or should simply avoid indexed offsets altogether.
- Tim
-*-
29497 17-MAY 00:27 Programmers Den
RE: RMA/RLink boo-boo. (Re: Msg 29494)
From: TIMKOONCE To: EDDIEKUNS (NR)
Well, Eddie, the problem is that if you have more than 127 bytes of DP vars,
then _indexed_ accesses to those vars become screwed up! As long as you only use
DP accesses of DP vars, all is fine, but using indexed accesses to DP vars (i.e.
making arrays "direct") can cause problems if there are more than 127 bytes of
DP vars.
In theory, you should be able to access any DP var with indexed addressing.
In practice, RMA will assume an 8-bit offset is sufficient, and RLink is never
told enough to be able to detect when that won't work. Ugh. As I mentioned to
Greg, C programmers should keep their DP size under 128 bytes, since I suspect
many internally-generated calls might try to take the address of a variable,
which can cause problems.
- Tim
-*-
29498 17-MAY 03:18 Programmers Den
RE: RMA/RLink boo-boo. (Re: Msg 29495)
From: KNOT1 To: TIMKOONCE
And just for techicalities, don't forget that "cstart" steals the first gP byte
so that there are no pointers to byte zero. This means you'll need to keep your
DP's to 126 bytes or fewer. Just figured I'd mention that.
-Jamie (KNOT1)-
-*-
End of Thread.
-*-
29496 17-MAY 00:06 General Information
RE: MPI upgrade (Re: Msg 29388)
From: TIMKOONCE To: TEDJAEGER
I've heard mixed stories about the necessity of the upgrade. Perhaps it's worth
a try??
- Tim Koonce
-*-
29525 17-MAY 20:58 General Information
RE: MPI upgrade (Re: Msg 29496)
From: TEDJAEGER To: TIMKOONCE (NR)
Tim, I sent the thing off for the upgrade. Since my old MPI, which is upgraded
by chip, does work in my system, and the new MPI, which is not upgraded, does
not work, the upgrade should be what I need, right? We shall see if logic is any
good!! Keep your fingers crossed, OK? --Bests, TedJaeger
-*-
End of Thread.
-*-
29499 17-MAY 03:19 General Information
RE: Help with Equipment and Downloading (Re: Msg 29336)
From: KNOT1 To: ALL
Of course! My perfect timing again! I just read Tony's May "Turn of the Screw"
(yes, I'm a little behind in my reading! Also have about six or seven books
next to my bed waiting to be read!) and there are all the answers to my
question! How come answers always "pop up" AFTER I dig them out the hard way (
but not _because_ I dug for them)? ;-( I suspect that that is a rhetorical
question, but if anyone DOES have an answer I'd be glad to hear it! <smile>
-Jamie (KNOT1)-
-*-
29500 17-MAY 03:57 Telcom
RE: (Re: Msg 29327)
From: OS9UGPRES To: WB4GCS (NR)
Jim,
Not a Ham, but I think as long as coco's have been around, most hams have had to
place the coco across the room at least.
You are probably right that it's the keyboard scan causing trouble. The only
possibility that i can think of would be to encase the entire coco, and feed
Puppo's IBM keyboard interface cable into it.
I'll ask around... I know hams use the coco, so there must be a way around all
this noise. best - kev
-*-
29504 17-MAY 04:01 General Information
RE: Basic09 Help!!!! (Re: Msg 29351)
From: OS9UGPRES To: OLDGROUCH
Eric - what else is merged in with your RunB and Basic09 commands? Ident them
and check. Also use PMap to study what your maps look like during failure. - kev
-*-
29519 17-MAY 18:01 General Information
RE: Basic09 Help!!!! (Re: Msg 29504)
From: OLDGROUCH To: OS9UGPRES (NR)
Kevin,
Actually, I've found a way to work around the problem. I just call Basic09 via
a "ex CHAIN Basic09 <>>>/1 #4k ProgName" to get it started. Since all the
modules are packed and loaded off disk, I really don't need to assign any more
memory than that. It seems to work well enough for what I'm doing anyway.
Maybe this whole thing was a fluke or maybe a bug. But I really didn't feel
like messing with it any longer so I kept experimenting until I found this
combination that works.
BTW: I had syscall, inkey and gfx2 merged in with my RunB. But now, I merged
those files with my Basic09 because I'm using that new method.
- Eric
-*-
End of Thread.
-*-
29505 17-MAY 04:01 Programmers Den
RE: os9gen trouble (Re: Msg 29403)
From: OS9UGPRES To: NES (NR)
What error do you get when you try to os9gen the disk?
That would help us pin down what's happening. Fragmented disk is what most often
is the trouble. May have to format a new one and gen it before dsaving cmds
over.
-*-
29510 17-MAY 17:03 General Information
RE: BLOB research (Re: Msg 28964)
From: MARTYGOODMAN To: RICKADAMS
(finally reading OS9 forum after an absence of more than two weeks)...
I'll be interested to hear what Kevin Darling has to say about your
observations, for he is to my knowledge THE recognized authority on OS9 / 6809 /
GIME hardware / software ideosyncracies.
One thing you might onsider to further your researches...
Make some kind of TESTER program that folks here can run, to see if THEY get
the same results you do. Upload it, and solicit reports. Ask, too, that those
who report in note whether they have a 1986 or 1987 version GIME chip. (Yes...
I KNOW there were more than just two varients... but those were the only
distinctions one could make from the outside of the chip.)
I LIKE your idea about a module that knows to relocate itself or use a
different piece of itself to get around the BLOB problem. But a "tuner" module
would also be needed for those cases where folks are using commerical or
shareware alterntive disk I/O programs, such as MS DOS to OS9 file converter
packages.
PLEASE keep us informed!
---marty
-*-
29511 17-MAY 17:10 Programmers Den
Development Pack
From: PAULIGHT To: ALL
Is anyone's Development Pack just gathering dust?? Please sell me the one with
the 'cgfx.l'. I want all the original disks and original docs. -will pay $50.
Call me at (817) 773-9332 anytime. Thanks, Paul Light
-*-
29512 17-MAY 17:18 General Information
RE: blob (Re: Msg 28970)
From: MARTYGOODMAN To: OS9UGPRES (NR)
You wouldn't happen to be using a LOGIC ANALYZER, now, would you? <grin>
---marty
-*-
29513 17-MAY 17:24 Programmers Den
SCSI operation
From: MARTYGOODMAN To: OS9UGPRES (NR)
Kevin (or Greg, or Bruce, or anyone who can help...):
I am in need of a simple explantion of how SCSI bus hardware works at the
level of device identification and data exchange.
I mean... I know that there can exist up to eight different devices on a SCSI
bus. And that a host computer requests a device to wake up, and that device
then does respond by making the bus "busy", and data exchange takes place.
Now, I need to know SPECIFICALLY what LINES are involved in this process, and
what the TIMING involved is.
I have the pin out for 50 pin and Mac 25 pin SCSI bus.
I need to know exactly what lines are pulled what way for, say, a host on the
bus to ask device number THREE to wake up and receive data, then for that device
to turn around and SEND data to the host. And in what sequence.
Note that the device in question need NOT be a hard drive. I do not need to
know exact codes for operating a SCSI hard drive controller. What I need is
generic SCSI device identification and data exchange information.
Thanks much in advance.
---marty
-*-
29528 17-MAY 21:17 Programmers Den
RE: SCSI operation (Re: Msg 29513)
From: OS9UGVP To: MARTYGOODMAN (NR)
Marty,
If no one else replies, send me a "wake up" message and I'll attempt to
explain the SCSI arbitration scheme for you. I don't have time right now, but I
don't want to forget about it if no one else can help...
Bruce
-*-
End of Thread.
-*-
29514 17-MAY 17:35 General Information
RE: Essay on Promoting OS9 (Re: Msg 28977)
From: MARTYGOODMAN To: OS9UGPRES (NR)
Addressing the last item first...
It is true that, as I have often admitted, I am not much of an OS9 user. TIME
has been a consideration here more than anything else. RS DOS applications
have until recently met my needs on the CoCo, and most recently I have been
making efforts to become "literate" in the world of MS DOS hardware and
software.
However, I remain a "fan" and "proponent" of Motorola processors, and would be
the first to point out the technical superiority of OS9, which came out years
before Windows or OS2, yet accomplishes much of what those "new innovations" now
do, years before they came out, and in a ten to one hundred times less fancy
hardware environment.
As you apparantly recognized this time, my essay was intended PURELY
constructively, not as any sort of attack or disparagement.
In that essay, I of course did use some hyperbole. Certainly you are
ABSOLUTELY RIGHT that the hardware superiority of the CoCo over the years... for
the price... was a big factor in its unprecedented longevity and popularity in
the market. Naturally, what I MEANT was that hardware superiority alone is NOT
enough. My main point (which which you appear to agree in principle) was that
QUALITY application software for the three or four most common applications for
home use is a crying need in the world of OS9.
Yes... I DO understand the need for powerful development tools (C compiler)
to be available in order for new software to be developed. But, without that
new software, end users will turn elsewhere for their needs.
---marty
-*-
29515 17-MAY 17:41 General Information
RE: Essay on Promoting OS9 (Re: Msg 28994)
From: MARTYGOODMAN To: RAGTIMER (NR)
The CoCo 1, 2, AND the three did indeed suffer from an at best "ho hum" attitude
on the part of Tandy toward marketting. I can speculate on the reasons for this.
It has been alledged repeatedly that throughout the life of the CoCo, from its
very first inception, there were groups within Tandy that wanted it
discontinued, which "warred" internally with its Tandy Corporate proponents.
The Pro CoCo faction succeeded in bringing us the 1, 2, AND the 3 (the last, I
suspect, against great odds!) and deserve great thanks from us users. However,
the "Anti Coco" faction "succeeded" in crippling any concerted, intelligent
effort to promote the machine based around Tandy vending quality application
software for the machine. It is this void that Rainbow Magazine, most
prominently, filled... this was the basis, I would contend, of Rainbow's
success.
I agree with much of what you wrote in response to my observations.
As I noted to Kevin above, my MAIN point was to make a PLEA for more attention
to providing end OS9 users with quality application packages at reasonable cost
for the three or four most common home computer applications.
---marty
-*-
29516 17-MAY 17:47 General Information
RE: Essay on Promoting OS9 (Re: Msg 29019)
From: MARTYGOODMAN To: KMTHOMPSON (NR)
Kelly (and Kevin, too)...
(1) I do like your suggestion that that OS9 application software be provided
in a "ready to run" or "easily prompted for configuring / installing" fashion.
And further suggest that one or more "generic" OS9 system setups be provided in
"spoon feed-able" fashion with any new OS9 machines, so a user can rapidly and
without frustration get up and running... and LATER learn how to better, more
optimally, configure their system to their particular needs.
(2) Both you, Kelly, and you, Keven, may make the mistake that I am equating
Mouse interfaces with "good" interfaces and command line or text/keyboard
interfaces with "bad" interfaces. Nothing could be further from the truth. I
have for YEARS argued that the FORM of the user interface is far less important
than the amount and quality of THOUGHT that has been put into it to make it
"intuitive" and "self prompting". And these last can be EXCELLENTLY acheived
in many cases by EITHER grahical / mouse OR text / menu + hot key interfaces.
---marty
-*-
29529 17-MAY 21:33 General Information
RE: Essay on Promoting OS9 (Re: Msg 28943)
From: THEFERRET To: MARTYGOODMAN (NR)
I can7t remember if I have said this before( The tone seems familiar, but
oh well)
I was enthused by OS9 LLII on the CoCo. I love the OS. But I got put off
writing any serious Apps., 'cause (I hate to say it) the CoCo is too slow (For
what I want to do, anyway).
The new machines fill me with an urge to be creative again. When I get my
hands on one, I will endevour to churn out some stuff myself. Unfortunately,
there is the CoCo gap: I won't have enough money for a while (at least a year)
But on the subject of designing software again:
The stuff that I write now and then is done mainly because _I_ feel a personal
need for it. Unfortunately, there is probably an enormous duplication of effort
occuring in other users. almost none of us "publish", however, because, as
written, the routines are quirky, and very individualised. W they are far from
perfect. In fact, in certain cases, I know that certain things don't work. I
see that quite a few people have downloaded some of them. Yet I have recieved
no mail on them, good or bad. I know some other public programmers experience
this. It's discouraging, to say the least.
Comments?
Philip
P.S. A Decent version of my text formatter should be in the App database
soon: 6 times the speed working embedded commands more memory efficient (no need
to fork shells)
plug, plug, plug....
-*-
End of Thread.
-*-
29517 17-MAY 17:54 General Information
RE: Rainbowfest New 68000 Machine Report (Re: Msg 29254)
From: MARTYGOODMAN To: PKW (NR)
(I've been absent for two weeks or more from here... first due to personal needs
and for the last week due to a VERY nasty viral flu that I am only now almost
recovered from...)
First of all... Paul... I am sorry about the extent to which what should have
been no more than a humorous personal ancedote about this last Rfest (the
oversight in my first report and the "4 AM phone call") got blown up into a
major, multi-network confrontation. As I noted a couple of times, I really
would have preferred it to have originally been relegated to something like the
post "pssst! Marty SNORES!" that appeared on CIS many years ago after one of
the first Rainbowfests.
I am glad that things seem to have settled down considerably.
I do like the CONCEPT behind the MM1... that of making a few compromises in
order to provide maximum power for minimum price in a 68000 based system. This
CLEARLY was the basic idea behind the MM1, no? Certainly, those CoCo 3 OS9
users I spoke to who were excited about the MM1 viewed it in those terms.
I am sure everyone here would welcome periodic progress reports as the MM1
approaches distribution.
---marty
-*-
29518 17-MAY 17:56 General Information
RE: Rainbowfest New 68000 Machine Report (Re: Msg 29255)
From: MARTYGOODMAN To: PKW (NR)
I take it that... a card could be devised for a MM1 to upgrade it to full 32
bit wide 68030 capability?
---marty
-*-
29520 17-MAY 18:03 General Information
RE: Rainbowfest New 68000 Machine Report (Re: Msg 29260)
From: MARTYGOODMAN To: PKW (NR)
question. In my experience on MS DOS systems, 640 by 350 (EGA) or 640 by 480
("old IBM" VGA) resolution is all one needs for text editing... it yeilds an
adequate amount of text on the screen, and provides for a use-able "page view"
mode.
And yes... as you say... "no resolution is enough" at this time for full on
desk top publishing applications. Many folks say 1K by 1K (or there abouts...
perhaps the current 1024 by 768 ultra super VGA) will become a minmum standard
for that environment.
But full on desk top / high speed / high productivity publishing is pretty
specialized... and required by only a small fraction of computer users... and a
still smaller fraction of HOME computer users.
Mundane as they may seem now, I would suspect that simple text editing /
formatting and data base and spread sheet applications and the like will remain
the bread and butter of home computer use.
And the computing and display capabilities of the MM1 are QUITE nicely suited
to those applications... and even to many more fancy graphical applications,
such as low to medium duty CAD and other graphics applications.
Note, too, that even now a decent hi res VGA card for a MS DOS system costs
$150 or more, and a color monitor to go with it costs $450 or more if it is to
support the higher res modes.
---marty
-*-
End of Thread.
-*-
29521 17-MAY 18:28 General Information
RE: hI (Re: Msg 29119)
From: JMURPHY To: KSCALES
Been lurking a bit here, and on the Hayes board. I haven't been able to get out
Thursdays because my girlfriend was going to night school and I got elected to
babysit her little boy. Might be able to make the June 7th meeting. Hope to see
you, but no promises. Later, JohnM.
-*-
29527 17-MAY 21:12 Programmers Den
RE: file size & basic09 (Re: Msg 29407)
From: THEFERRET To: GREGL (NR)
It turns out I was doing THE most incompetant, beggiinner thing to do... I used
$89 instead of $8x for the syscall code. Duh!"! But about your redefining the
EOR: I think I did that once, but can't reemember what I did . There is no
obvious way to change it for disks, as xmode/dmode does not support changing
EOR.
(BTW, my routine works now, anyway)
Philip
-*-
FORUM>Reply, Add, Read, "?" or Exit>
Reference in New Issue View Git Blame Copy Permalink