#: 14399 S12/OS9/68000 (OSK) 22-Feb-92 10:35:47 Sb: #14397-#upload space Fm: Jim Sutemeier 70673,1754 To: Mike Ward 76703,2013 (X) size of uploads--> Mike, the cmds for the C++ Compiler take up 690,000 bytes (compressed). (The main module for this compiler is 849,000 bytes, uncompressed!) I had also intended on uploading a GCC compiler, which is ANSI compatible. It is smaller in size than the C++ Compiler, should only be about 750-800,000 bytes total for that set. When I started the uploading, wasn't aware that there were any space probs here on CI$. I'll try to get the C++ cmds in now, please advise me on the ANSI compiler. Thanks. jim Sutemeier There is 1 Reply. #: 14436 S12/OS9/68000 (OSK) 27-Feb-92 00:13:50 Sb: #14399-upload space Fm: Wayne Day 76703,376 To: Jim Sutemeier 70673,1754 (X) Jim, I've made a little bit more space available, so if you run into problems again, please let us know. There aren't really any space problems here on CompuServe..... just a problem with us not always erasing the files that folks are no longer accessing. Houskeeping chores, ya know. Wayne #: 14401 S7/Telecommunications 22-Feb-92 12:23:41 Sb: #IRQ Hack Query Fm: Ches Looney 73016,1336 To: Kevin Darling 76703,4227 (X) Kevin, I recently started using STERM and began to experience lockups that Mark Griffith thinks may be due to lost IRQs. He urges me to install the IRQ hack so I looked in the files and found the Lyall hack and the diode hack. Apparently both are intended to tie the IRQ from the RS232 pak to the IRQ input on the 6809. What is the problem with the original design? It looks like a multiplexer should be sending the IRQs through, so is the problem that the multiplexer misses IRQs under certain conditions? On the diode hack, what is intended by the direction to tie the #8 ports together? Tie all pin 8s on the pak sockets together? What about tying the combined #8s to the #8 plug to the CoCo? In other words, is the diode hack intended to bypass the multiplexer by the "brutal" method of shorting the four inputs to the single output? In the diode hack, is it necessary that the diode be germanium? Is the input that sensitive that the .4 volt difference between germanium and silicon diodes is important? Thanks for your help. Regards, Ches. There is 1 Reply. #: 14406 S7/Telecommunications 23-Feb-92 07:16:41 Sb: #14401-#IRQ Hack Query Fm: Kevin Darling 76703,4227 To: Ches Looney 73016,1336 (X) Hi Ches! Background: the GIME's external IRQ input is edge-sensitive, instead of level-sensitive... meaning yes, interrupts can be lost (or ignored). It seems that the GIME design was changed to this method towards the end, but the updated info slipped through the cracks... L-II still expected otherwise. The other needed background is that the MPI only allows one slot to pass interrupts at a time (this is left over from the ROMPAK startup days). So, right, the basic idea of all the "IRQ Hacks" is to bypass the GIME and/or the MPI irq-slot selection. (There are also software hacks to bypass the GIME, and many aftermarket drivers do that, but...). MPI: You only need the MPI hack (pins 8 tied together) if you have more than one device out there which uses interrupts. So, for example, if you only have one such device (the RS232 pak), then no problem. If you have more (a couple of RS232 paks, or a pak plus a no-halt controller, or etc), then you clearly have to bypass the MPI selector. GIME: Some people simply bypass it all together, and run all the external IRQs straight to the 6809 (you should see my lazy computer mess, btw: my SC-II irqs come thru my stock MPI, and so I have an external wire running to the cpu from my PBJ dual rs232 pak - ugh). Anyway, the diode hack takes any interrupts that can get out the MPI (modified or not) and bypasses the GIME. The diode part is so that old ROMPAKs can also be used. I believe that it does have to be germanium, from what people have found. Of course, if you do the MPI hack, I'm not sure the diode hack (vs just using a jumper) makes sense. Clear as mud? :-) Basically, the MPI is a SP4T switch... tieing pins 8 bypass it. The GIME is a latching doorbell... so you sometimes can't tell how many people are pushing it. - kevin There are 2 Replies. #: 14407 S7/Telecommunications 23-Feb-92 07:33:40 Sb: #14406-IRQ Hack Query Fm: Ches Looney 73016,1336 To: Kevin Darling 76703,4227 (X) Thanks for info, Kevin. Now off to study. Regards, Ches. #: 14414 S7/Telecommunications 23-Feb-92 14:49:30 Sb: #14406-#IRQ Hack Query Fm: Ches Looney 73016,1336 To: Kevin Darling 76703,4227 (X) OK, Kev. I've put in the diode hack as follows: tied the rom-pak-socket ends of r11, r12, r13, r14 together in the MPI, and put in a 1n34 between the IRQ end of r2 and pin 8 of the rom-pak-plug of the CoCoIII. One attempt of a download using STERM on a 180K file in the non-sliding-window mode of B+ protocol worked without a flaw. Trying the sliding window mode seems not to work - first and second blocks come thru head to tail, then there's a long wait followed by 3 and 4 then a long wait (perhaps as much as a minute) then 5 & 6, etc. Should I have also tied the #8 pin of the MPI to the 4 resistor ends? Tried a silicon diode without success, btw. Comments?? Ches. There are 2 Replies. #: 14417 S7/Telecommunications 23-Feb-92 16:45:20 Sb: #14414-IRQ Hack Query Fm: Kevin Darling 76703,4227 To: Ches Looney 73016,1336 (X) Ches - I'll have to look up the schematics, but it sounds right. The STerm thing may simply be because of the smaller serial input buffers on the coco... they could be overrun and it takes a while for CIS to notice the reported error and resend and slow down because it knows things aren't going so hot, etc. Perhaps someone else can confirm (or deny :) this. - kev #: 14418 S7/Telecommunications 23-Feb-92 17:15:27 Sb: #14414-#IRQ Hack Query Fm: Steve Wegert 76703,4255 To: Ches Looney 73016,1336 (X) Ches, I've lost track of your set up so .... The sterm '-f' option works only if you are downloading to a ram disk or have a fast hard drive. Also ... make sure the buffers are as large as you can make 'em. Steve There is 1 Reply. #: 14419 S7/Telecommunications 23-Feb-92 20:53:17 Sb: #14418-#IRQ Hack Query Fm: Ches Looney 73016,1336 To: Steve Wegert 76703,4255 (X) Steve, I hadn't thought of using ramdisk - I'll try that. Still haven't doped out how to change buffer size (but haven't given up yet) Ches. There is 1 Reply. #: 14426 S7/Telecommunications 24-Feb-92 20:52:27 Sb: #14419-IRQ Hack Query Fm: Steve Wegert 76703,4255 To: Ches Looney 73016,1336 (X) Ches, Take a peek at PAKMOD.TXT and PAKMOD.SRC in LIB 10 if you're still using the stock stuff from Tandy. That will get you a couple of more extra bytes. You might also want to look at Bruces Isted's SACIA stuff. I believe it allows for varriabe size buffers. Should also be in LIB 10. Steve #: 14404 S12/OS9/68000 (OSK) 22-Feb-92 23:24:00 Sb: #14390-#GNU C Compiler Fm: Robert Heller 71450,3432 To: Jim Sutemeier 70673,1754 (X) I already snarfed it from the InterNet. It is great. It would be nice if it generated srcdbg compatable debug info though... While I can survive with printf()s (err cout <<'s), it would be nice to be able to use the debugger... Robert There is 1 Reply. #: 14405 S12/OS9/68000 (OSK) 23-Feb-92 00:02:19 Sb: #14404-#GNU C Compiler Fm: Jim Sutemeier 70673,1754 To: Robert Heller 71450,3432 (X) I haven't gotten much further than to start listing out the defs files, myself. But, 'cause I ALWAYS try something out before reading the dox (grin), I simply HAD to run a test comparison between the MW Compiler and this GNU C++ Compiler, as it will 'try to imitate traditional C'. I compiled a program of 27,375 bytes first with the MW compiler...it took 2:14 minutes, final outfile was 23,428 bytes. Using GNU 1.40, it took 2:21 minutes, and, using the -O (for optimizer) option, the final outfile was 22,806 bytes. That really impressed the heck out of me. Over 600 bytes optimized out, on a compiler that says it will "TRY" to compile it. Now, I gotta go buy some books on C++, so I can start to learn that implementation of C. Have fun! jim Sutemeier There is 1 Reply. #: 14425 S12/OS9/68000 (OSK) 24-Feb-92 20:49:55 Sb: #14405-#GNU C Compiler Fm: Bob van der Poel 76510,2203 To: Jim Sutemeier 70673,1754 (X) I tried to recompile a few of my programs (Ved and Vprint) and found that the code actually increased in size a little. Guess it depends on what's going on. But you're correct--they are very nice offerings. Not sure if I'm going to get into C++ or not. Oh, are the doc files you uploaded for the C++ or C compiler? There is 1 Reply. #: 14429 S12/OS9/68000 (OSK) 25-Feb-92 15:32:15 Sb: #14425-GNU C Compiler Fm: Jim Sutemeier 70673,1754 To: Bob van der Poel 76510,2203 (X) dox files for C Compiler--> Bob, I uploaded two different compilers from GNU, the ANSI C Compiler and the C++ C Compiler. The ANSI C Compiler is complete and intact, just as it sits (all dox and cmds are in the same archive.) The C++ Compiler, on the other hand, are so extensive (and VERY long) that I left those in idividual archives in the database. So, the DOX 1 & 2 belong to the C++ Compiler, as does the lfiles, defs, and gnu cmds. I haven't decided if I want to tackle C++ myself.....yet.....but I LIKE the option of being able to do so if I want to. Take care. jim Sutemeier #: 14408 S12/OS9/68000 (OSK) 23-Feb-92 09:00:34 Sb: #14393-/P ports Fm: Jim Peasley 72726,1153 To: Steve Wegert 76703,4255 (X) Steve; No resolution switch, but it does have a PC/MS switch on the bottom and it is 3 button - model AM25. ...Jim #: 14409 S14/misc/info/Soapbox 23-Feb-92 09:26:59 Sb: CoCo List postings Fm: Steve Wegert 76703,4255 To: All Just a short note to let everyone know that the COCO LIST postings have resumed in Library 14. Their absence over the last few weeks has been due to getting my MM/1 up and in a state where I can be productive. I'm finally there! The wait for the I/O board was certainly worth it. This thing zips along! Now .... the hunt for all those utilities I've grown accustomed to on the CoCo continues! Steve #: 14411 S10/OS9/6809 (CoCo) 23-Feb-92 12:15:13 Sb: CRC stuff Fm: Steve Wegert 76703,4255 To: All Just found this on the CoCo List and thought a few folks might be interested: <<>> From @KSUVM.KSU.EDU:COCO@PUCC.BITNET Sat Feb 22 19:29:10 1992 Received: by wuarchive.wustl.edu (5.65a+/WUSTL-0.3) with SMTP id AA09562; Sat, 22 Feb 92 19:29:08 -0600 Message-Id: <9202230129.AA09562@wuarchive.wustl.edu> Received: from KSUVM.KSU.EDU by KSUVM.KSU.EDU (IBM VM SMTP V2R1) with BSMTP id 6253; Sat, 22 Feb 92 19:29:20 CST Received: from KSUVM.BITNET by KSUVM.KSU.EDU (Mailer R2.07) with BSMTP id 6561; Sat, 22 Feb 92 19:29:20 CST Date: Sat, 22 Feb 1992 19:25:03 CDT Reply-To: "Haywood,Arvin Carl Robert" Sender: COCO - Tandy Color Computer List From: "Haywood,Arvin Carl Robert" Subject: CRC and 1 Meg upgrade To: Multiple recipients of list COCO Status: OR I make a post earlier in the week in which I said that I was looking for the 1 Meg upgrade for the coco 3. I have received numberous messages asking if I had found out a source yet. Well, I got ahold of Tony DiStefano tonight and he informed me that CRC was going to make a limited number of 1 Meg upgrades, Super Controller II's, and 4-N-1 Boards. The Products should be available in a week or so. You can call CRC for details or to order one. This info should make the people that contacted me happy. Carl #: 14412 S3/Languages 23-Feb-92 12:25:15 Sb: #14190-#Help Fm: George Hendrickson 71071,2003 To: Pete Lyall 76703,4230 (X) I think I tried CC but it need RMA and RLINK which I don't have. Where do I get those files? There is 1 Reply. #: 14415 S3/Languages 23-Feb-92 15:41:39 Sb: #14412-Help Fm: Pete Lyall 76703,4230 To: George Hendrickson 71071,2003 George - RMA and RLINK come on the DEVELOPERS PAK disks. Technically, you could use a disk editor and patch the module names of C.asm and C.link to be RMA and RLINK. Then you would of course also have to rename the files to the same names. Optionally, you could get the developers pak. Pete #: 14413 S9/Utilities 23-Feb-92 12:28:48 Sb: #14191-#PAK utility Fm: George Hendrickson 71071,2003 To: Pete Lyall 76703,4230 (X) I like AR except for the fact that when you update a file in the pack, it expands the file and doesn't get rid of the old file. I use AR to backup my BBS every night at maintenance time. I had to delete the old AR files and then archive the directorys I wanted backed up all over again. Its just time consuming thing... BTW, I bought a 386 33mhz system this past saturday and really love the thing! Just blows me away! Now I need to get that cc3disk module that reads IBM disks and start coping files over.... There is 1 Reply. #: 14416 S9/Utilities 23-Feb-92 15:44:03 Sb: #14413-PAK utility Fm: Pete Lyall 76703,4230 To: George Hendrickson 71071,2003 You may want to look at the HDKIT package in DL9... that will allow you to do incremental backups. You could always AR the results later. Pete #: 14420 S14/misc/info/Soapbox 24-Feb-92 03:43:26 Sb: #14387-#OS-9 Community Network Fm: John Wight 76370,2100 To: Robert R MacKay 100036,676 (X) I have noticed there have been a few systems from "down under" to join us on the Fidonet COCO and OS9 echos lately. You may have heard there is soon to be an MM/1 echo as well. One in particular is John Kelly's system (I don't have th name and number with me). He has already expressed interest in receiving files via Fidonet software. I don't know where his system is in relation to yours, but all Australian OS-9 users might want to organize in order to distribute information with those of us in Fidonet Zone 1 more efficiently. To start this information exchange, send me some disks. I can work with 5 1/4 inch 360k floppies, or 3 1/2 inch 720k floppies. I have about 12 or 13 megabyte of information to exchange, which would be about 30 360k floppies, or 15 720k floppies (in COCO OS-9 format). Then I hope either you or John Kelly will (or both of you together) establish a means of distributing this information with the rest of the OS-9 community in Australia. With the number of systems in Australia already picking up the Fido COCO and OS9 echos, those of us are finding OS-9 is more popular in Australia than those of us in North America thought :-). BTW, my Fidonet mail problems seem to be over, so I assume your Fido message will be getting to me anyday now... There is 1 Reply. #: 14434 S14/misc/info/Soapbox 26-Feb-92 18:24:46 Sb: #14420-#OS-9 Community Network Fm: ole hansen 100016,3417 To: John Wight 76370,2100 (X) reply Hello John. I have just been reading the information about the OS-9 community and OCNN I looks great. I live in DENMARK nearby Copenhagen, where I work for a company called 'danelec electronics aps'. We sell VME-boards/systems with MOTOROLA 68K/88K running Unix and of course OS-9/68K. I have been here since 1981 (I believe somebody see this as beeing lazy, but a lot happens on the VME-side as well as OS-9/68K). We have just founded a danish OS-9/68K user-group. It is called : ONUG/DK (Os-Nine User Group/DenmarK). We also run a BBS for MOTOROLA family of micros and OS-9/68K. How do I get onboard this group ?? I know very little about 'fido-net'( I have heard the words ), so do you know of any 'fido-nodes' in DENMARK that I can connect to ?? I have downloaded the documents mentioned as well as the newsletters. Can I post these on our BBS ??? I will give you the Number in a few days, but wright now (today) somebody disconnected both the lines we had. Our bbs is running on a UNIX System V/68 MVME147A 8MB-ram 20MHz 16 seriel lines and 600MB harddisk (300 for BBS (100 for OS-9 alone)). It will contain all the TOP-Munic v2.0 and EFFO software + a lot more. Could you please guide me from here ?? I would not mind to work in the interrest of the OS-9 community. best regards Ole B. Hansen 100016,3417 OZ1JRU Email: ole@danelec.dk There is 1 Reply. #: 14501 S14/misc/info/Soapbox 06-Mar-92 02:54:18 Sb: #14434-#OS-9 Community Network Fm: John Wight 76370,2100 To: ole hansen 100016,3417 (X) Consider yourself a member of the OS-9 Community Network, then, and feel free to spread our files around. I have captured your message and will do some research on the best way to exchange information with those of you in Denmark, including Fidonet. There are Fidonet BBS systems in your area. I will look on my list for one near you and get back to you. Thanks for taking interest. So far we have members in the U.S, Canada, Australia, Belgium, and now Denmark ! BTW, the members in Belgium tell me there is a users group called EUROS9 and EUROCOCO, and have sent alot of information such as their newsletters back to us. It looks like our new network is going to show the rest of the computing world how special our community is ! There is 1 Reply. #: 14533 S14/misc/info/Soapbox 07-Mar-92 18:58:13 Sb: #14501-OS-9 Community Network Fm: ole hansen 100016,3417 To: John Wight 76370,2100 Hello John. I will be looking forward to hear from you. I have been online for 10 years with OS-9/OSK, so it is nice to have someone to share with. best regards ole b. hansen #: 14421 S7/Telecommunications 24-Feb-92 07:01:20 Sb: #STERM etc. Fm: Ches Looney 73016,1336 To: Steve Wegert 76703,4255 (X) Steve, thanks for your help. Using a ramdisk and the IRQ diode hack allows STERM on my system to fly. Downloaded a long (approx 280k) file last night with approx 75% efficiency; it didn't appear to have any break in the received packets at all (watching the LEDs). Many thanks. Ches. There is 1 Reply. #: 14427 S7/Telecommunications 24-Feb-92 20:53:46 Sb: #14421-STERM etc. Fm: Steve Wegert 76703,4255 To: Ches Looney 73016,1336 (X) Glad we got things cooking! Downloading to a RAMDISK practically assures success using the B+ protocol. Steve #: 14422 S7/Telecommunications 24-Feb-92 07:08:45 Sb: IRQ hack etc. Fm: Ches Looney 73016,1336 To: Kevin Darling 76703,4227 (X) Thanks for your guidance, Kevin. After the IRQ diode hack and using a ramdisk, downloading appears to be pause free between packets. BTW, I had to go out to buy some 1N34s (never thought I'd run out of those) and found 1N93s (germanium power diodes), as well as silicon diodes, don't let the IRQ diode hack work. Still uncertain about the original MPI design flaw. Edge detection ? - another way of saying the multiplexing of IRQs is just too slow in the original design? or is it really using the pulse created by differentiating the square wave level change? Or is the multiplexing losing an IRQ in the times when the multiplexor is looking at other IRQ sources? And is there a bad side effect of strapping all the PAK IRQs together so that identification of which IRQ is from which PAK is lost? Anyway, thanks for the encouragement and guidance; STERM is now working at full speed on my system. Regards, Ches. #: 14423 S7/Telecommunications 24-Feb-92 07:11:44 Sb: STERM etc. Fm: Ches Looney 73016,1336 To: Mark Griffith 76070,41 (X) Mark, thanks for your suggestions regarding the IRQ hack and the -f option in STERM. Apparently my hard disk (OWLware) is not fast enough for the -f option, but Steve Wegert urged me to try STERM/-f with a ramdisk and that works like a champ. Downloaded a long file last night with apparently no pauses between packets. Thanks again and regards, Ches. #: 14424 S12/OS9/68000 (OSK) 24-Feb-92 16:14:32 Sb: G-WINDOWS Fm: Ed Gresick 76576,3312 To: ALL To all - G-WINDOWS for the SYSTEM IV computer is released and available for immediate shipment. All orders we have received were shipped today. Owners of SYSTEM IV computers or computers using the PT68K4 board, call for a free demo copy of the G-WINDOWS and G-VIEW. For more information, see upload G-WINDOWS INFO in the library. Ed Gresick - DELMAR CO 302-378-2555 #: 14430 S10/OS9/6809 (CoCo) 25-Feb-92 19:11:08 Sb: #Deskmate Fm: Ian Hodgson 72177,1762 To: All This question should take a few of you back a bit in time. I have recently taken out and dusted off my copy of Deskmake 3, and put it on a real disk with Vdgint, etc. It's not very sophisticated in some ways, but is perfectly suited to the needs of a beginner who uses my CoCo 3 from time to time. Now, for reasons unknown, the mouse/joystick will not move the pointer all the way to the right side, nor to the bottom, of the screen. Normally this is not too much of a problem, but with PAINT it seriously limits the type of drawing you can do. I checked the joystick and it indeed covers the normal 0-63 range horizontally and vertically just fine. Has anyone run into this and found a solution? Secondly, what in Deskmate 3 binds it to /term? It won't run from another VDG screen. Not a serious problem, but I do like to use it concurrently with other tasks and prefer to come up in 80 column mode. Finally, why do most of the modules show up as non-shr in an ident? Does this really mean that they are not reentrant, ie. I can't run two Deskmates at once (provided that they could work from other than /term, of course)? There is 1 Reply. #: 14431 S10/OS9/6809 (CoCo) 26-Feb-92 03:41:23 Sb: #14430-#Deskmate Fm: Kevin Darling 76703,4227 To: Ian Hodgson 72177,1762 (X) Ian, Get DESK3.SCR in Lib 10. It's a modpatch script to make DM3 use the hires mouse, and run from any vdg screen (not just /TERM). For that matter, do a "bro /key:desk*" in Lib 10. There are some interesting files and mods there. I have no idea if more than one DM3 can be run at a time. But heck, change them all to shareable and try it . Luck! kev There is 1 Reply. #: 14437 S10/OS9/6809 (CoCo) 28-Feb-92 22:50:27 Sb: #14431-#Deskmate Fm: Ian Hodgson 72177,1762 To: Kevin Darling 76703,4227 (X) Kevin, I got DESK3.SCR from lib 10 and it worked out just fine. Thanks for the tip. One odd thing, though. The hires mouse works now, but unlike its very smooth movement with MultiVue, with Deskmate the movement is jerky and erratic and difficult to control. Any idea why there shoul be that difference? Come to think of it, just where is the driver for the hires mouse? I have no idea how it works, though I suspect there is a digital ramp generator controlled by software somewhere; 8 bits, I would guess. Given OS9's modularity I would have expected a device driver and descriptor for the mouse, but that isn't the case, so where does it hide? Ian There is 1 Reply. #: 14441 S10/OS9/6809 (CoCo) 29-Feb-92 10:13:13 Sb: #14437-#Deskmate Fm: Kevin Darling 76703,4227 To: Ian Hodgson 72177,1762 (X) Ian, Hmm. Deskmate must sleep a lot between reading mouse/joystick (like CoCo Artist did). Would have to track it down. The hires mouse driver is in CC3IO. You're right: it tells the hires interface to begin charging a capacitor... and the driver shuts off interrupts and counts in software until ramp meets the joystick position. If the mouse is in the lower right corner, this can mean irqs are disabled for a very long time (I think up to 10ms or maybe more). Therefore, when doing nothing else, it's best to move the mouse to the upper-left corner where the delay is negligible. kev There is 1 Reply. #: 14487 S10/OS9/6809 (CoCo) 05-Mar-92 11:48:47 Sb: #14441-Deskmate Fm: Lee Veal 74726,1752 To: Kevin Darling 76703,4227 (X) Or Ian could now install Bruce Isted's serial mouse support (in MOUSE.AR). I'm using a Logitech mouse on one of my CoCo3s and an Impression Mouse (which is a Microsoft-compatible) on another. The problem of loosing IRQs during mouse movements goes away when the serial mouse is used instead of the Hi-Res adaptor. I have no IRQ hacks on either of my systems, but now have the freedom of using GSHELL+ or any other mouse-driven software while doing uploads or downloads from BBSs or other services. I have had NO lock-ups due to IRQs that were lost due to mouse operations since installing Bruce's Serial Mouse Drivers. Incidentally, I use multi-port seral cards on both of my systems, so I'm able to run multiple serial applications simultaneously and the mouse drivers, too. Lee #: 14432 S7/Telecommunications 26-Feb-92 07:20:00 Sb: #STERM Fm: Brother Jeremy, CSJW 76477,142 To: All Dear Friends, I recently started using STERM here on CIS. I seem to be loosing a few characters here and there. Also I am getting "garbage" characters. For example, I have a line on the screen "[JThe OS-9 Forum;70HCompose" Finally, the B protocol worked fine for one dowload, and then every thing locked up. I originally used OSTERM 2.08 for CIS. Do I have to change any settings, patch anything on my system, etc. Thank you for your help. With all best wishes, Br. Jeremy, CSJW There is 1 Reply. #: 14433 S7/Telecommunications 26-Feb-92 07:35:21 Sb: #14432-#STERM Fm: Steve Wegert 76703,4255 To: Brother Jeremy, CSJW 76477,142 (X) Sounds like a couple of problems cooking. First, the weird characters are probabbly VIDTEX control codes. GO DEFAULT and set your terminal to OTHER or DUMBCRT. THat should fix that. Loosing characters is indicative of an IRQ problem. Have you installed the IRQ hack as described in in LIB 10? BROwse IRQ* and see what pops up. There are two ways of doing this ... the diode hack and the original method as told by Pete Lyall. I perfer the later. A third alternative would be to replace your ACIAPAK drivers with Bruce Isted's SACIA package. It fixes in software the problem that causes the missing interupts. Steve There is 1 Reply. #: 14435 S7/Telecommunications 26-Feb-92 20:39:14 Sb: #14433-STERM Fm: Brother Jeremy, CSJW 76477,142 To: Steve Wegert 76703,4255 (X) Dear Steve, I have been thinking of adding the SACIA driver in place of ACIA. I will go ahead and do that and I'll get back to you and let you know what happens. Thank you for your time and response. With all best wishes, Br. Jeremy, CSJW #: 14438 S12/OS9/68000 (OSK) 29-Feb-92 02:04:25 Sb: #MM/1 stereo Fm: Ken Gideon 72270,664 To: all welp, another stupid question bout my MM/1, I've tried Mike K.'s recplays program, works fine, sept when I plug in my stereo to the audio in of the MM/1, the MM/1 is grounding out my stereo, whitch makes the recordings sound like crap... I tested the cable and its not shorted, also all the wires are going to the right place (that tends to help)... so just for kicks I measured the resistance between the audio in signal pin, and the audio in ground... both channels were about 620 ohms... since they were both about the same, I dont think anything is fried or anything like that... but only 620 ohms? shouldn't it be more? thats a real load for my amp, and it cant take it... is that normal though? would puting a op amp betwwen the stereo amp and the MM/1 (with an Av of 1) help? them little suckers have a high Z in...(or hay, even an Av of 100,000... amplify the **** outa it :-) oh well, anyone got any Ideas? Ken or Cosmic@SandV or something, some where on INTERNET (or is it bitnet? I forget...) or you can get me on the Coco listserv... or or or There is 1 Reply. #: 14442 S12/OS9/68000 (OSK) 29-Feb-92 10:17:18 Sb: #14438-MM/1 stereo Fm: Kevin Darling 76703,4227 To: Ken Gideon 72270,664 (X) Ken, If it helps, I think R2 and R6 back by the connector are the resistors to ground on the sound input (600 ohms each). R1 and R5 (100 ohms) are in series, respectively. kev #: 14443 S5/OS9 Users Group 29-Feb-92 11:02:35 Sb: #OS9CN Fm: Jim Sutemeier 70673,1754 To: all I have read carefully the ocnn.0192 file which I obtained from Delphi. (Even printed out a hard copy on my printer so I could read it more carefully). At first read through of the by-laws of the organization, I thought (still do) that the organization appears to be well set up, and most every detail has been well covered. Whoever wrote these by-laws should be congratulated on a job well done. A couple of points stick in my mind, however, after having read this 'edition' of the NetNews. Why were there no other "Portions" mentioned in 17B? Is this because the OS9CN is going to be strictly a domain of FidoNet and RiBBS? I know for a fact that none of the StG SysOp's were contacted about their BBS being part of the OS9CN, and I suspect that none of the AcBBS BBS's were, either. While RiBBS is the 'biggie' of the networking CoCo BBS's at this time, asking AcBBS and StG BBS's to join into the OS9CN would net an additional 30 (or so) BBS's, thus giving the OS9CN a wider potential user base. The point here is that the more people, networks, etc., you contact and set up, the more 'dominant' you become. RiBBS/FidoNet are not the entire world of OS9, there is UseNet, and the other OS9 BBS's that network that should be contacted. If you feel I am being a little 'touchy' on this apparent OS9CN/RiBBS tie-in, might I call you attention to the fact that the NetNews I have just read contains the laws and by-laws of the organization, followed immediately by an apparently complete listing of RiBBS Systems. By attaching that listing to the 'official rules and regulations' ANY user would say that RiBBS and OS9CN are connected to each other. There are 2 Replies. #: 14502 S5/OS9 Users Group 06-Mar-92 03:03:17 Sb: #14443-OS9CN Fm: John Wight 76370,2100 To: Jim Sutemeier 70673,1754 (X) Sorry if we gave you that impression, but actually the opposite is the case. One example is that I came here regularly to exchange information on behalf of the OS-9 Community Network. We are also brainstorming ways to regularly exchange information with those on other networks such as Internet, Usenet, UUCP, Stgnet, Rime, ACBBS network, and there are most likely others. What we need is volunteers to help us brainstorm how to make these gateways come about. For example, MS-DOS systems are capable of handing both Fidonet mail and UUCP mail, as well as other networks. So if we took advantage of those OS-9 fans that run BBS systems on MS-DOS systems, we could make this exchange of information automatic. We do have a few volunteers that are sending information for us to some of the other popular networks. But we do not want The OS-9 Community Network to be based only in Fidonet, as there are members of the OS-9 Community on almost any network. For those of us that use OS-9 on the COCO, using RiBBS and Fidonet is the most feasible and economical way to go (for most of us), and Australian OS-9 users have already joined us on Fidonet. If someone will offer ideas of better ways to do it, or ideas on gatewaying, we are all ears! #: 14503 S5/OS9 Users Group 06-Mar-92 03:06:26 Sb: #14443-OS9CN Fm: John Wight 76370,2100 To: Jim Sutemeier 70673,1754 (X) By the way, StG and ACBBS members have both been contacted long before you left this message. Alan Sheltra was the one I contacted for StG, and he said the author would attempt a port to Fidonet. I have yet to get a response from Carmen Izzi, but have gotten responses from users of ACBBS echo areas. It is slow going but we are keeping these things in mind. #: 14444 S7/Telecommunications 29-Feb-92 11:16:32 Sb: OSK C Kermit 5A beta Fm: James Jones 76257,562 To: All YES!!!!!! Just compiled C Kermit 5A on my MM/1. (I hasten to add that ALL I did was tweak a makefile a couple of times--Chris Hemsing in Germany and Bob Larson, with whom folks who have been here for a while should be familiar, did all the work.) I'm using it now. I am having NONE of the problems I had with version 4F talking to the modem. I've yet to try to transfer any files, but I'll let you know how it comes out. C Kermit 5A has LOTS of bells and whistles: sliding windows (say, is CIS going to do sliding window Kermit for up/downloads?), long packets, attribute packets (a truly Great Thing, because it lets Kermit pre-allocate space for incoming files!!!), an extensive scripting language, the whole nine yards. possible to twist knobs at make time to just get the pieces you want. More news as it happens. This is the beta version, so I don't quite feel comfortable about uploading it. Instead, I'll send off the makefile tweaks to Kermit Central (at Columbia U), and I expect the official 5A version should be uploaded here. (I'll do it if nothing else.) #: 14445 S10/OS9/6809 (CoCo) 29-Feb-92 23:20:21 Sb: #14083-#Rogue under Multivue Fm: Mike Guzzi 76576,2715 To: Bert Schneider 70244,427 (X) Ill try it and see if it works. I always get my MV fonts clobbered after running rogue There is 1 Reply. #: 14460 S10/OS9/6809 (CoCo) 02-Mar-92 18:26:25 Sb: #14445-Rogue under Multivue Fm: Bert Schneider 70244,427 To: Mike Guzzi 76576,2715 Let me know if it works. It should, I have not had any problems and I am not running any of the new Shell+ programs. Bert Schneider #: 14446 S1/General Interest 29-Feb-92 23:22:51 Sb: #14092-About AR Version 1.4 Fm: Mike Guzzi 76576,2715 To: Wayne Day 76703,376 (X) ok that thing you showed to find all my files is fine but what I asked is I am not sure which ones could have been done with 1.4. obvoisly the files before 1.4 came to be are OK but afterwards i am not sure 1.4 and 1.5 do such a good compression and faster then pak I use it alot. Well if anyone does say something I will correct it. Mike #: 14447 S10/OS9/6809 (CoCo) 29-Feb-92 23:25:27 Sb: #14099-#E.A.R.S. Fm: Mike Guzzi 76576,2715 To: Lee Veal 74726,1752 (X) There was an article in BYTE magazine on how to build one. Finding the SP-1000 chip is the hard part (i have 2 spares) but the rest is common parts. The article was called "listner 1000" i think. I have some OS9 drivers but can't distribute them without permission. I had done a project using ears to make a lock that opened on voice command. Mike There is 1 Reply. #: 14482 S10/OS9/6809 (CoCo) 04-Mar-92 13:42:47 Sb: #14447-E.A.R.S. Fm: Lee Veal 74726,1752 To: Mike Guzzi 76576,2715 How difficult will it be to get permission to distribute the OS9 Drivers? What address(es) does/do the E.A.R.S. device use? Thanks for the info on the Byte Mag. article. Lee #: 14448 S10/OS9/6809 (CoCo) 29-Feb-92 23:27:42 Sb: #14334-OS9 support Fm: Mike Guzzi 76576,2715 To: Bernard Brooks 72411,2172 (X) I also reccomend looking on BBS's. sometimes there is someone selling a copy and thats usually cheap. when RS blew out there coco stuff I got two copies of os9 level II for $10 each and dynacalc for $2. (i bought a few copies in case someone needs one) Mike #: 14449 S10/OS9/6809 (CoCo) 29-Feb-92 23:31:50 Sb: #14332-In memoriam Fm: Mike Guzzi 76576,2715 To: Kevin Darling 76703,4227 (X) I agree, seems AFTER tandy dropped support is when an upsurge of os9 users came to be. I just found a few local coco users and helping them along. One just got CoCo max 3 and max-10 and then bought 512K. he is looking at os9 now and hopefully will get interested. my message bases on my bbs are inactive but it happens in surges. but I don't feel it is dead Miker #: 14451 S15/Hot Topics 01-Mar-92 01:24:48 Sb: #Secret 6309 Modes? Fm: Kevin Darling 76703,4227 To: all Hold onto your hats, folks! Several days ago, someone on Usenet asked if there were any faster/better replacements for the 6809. A student from Japan replied that there had been an article years ago about new modes in the Hitachi 6309... and he also said he had tried many of them himself. After exams, he's going to post more info. If this pans out, someone may have to locate a good source of 63C09E's :-) In any case, these are some supposed features he has mentioned... 1. More registers (additional two 8 bit accumulators, 8 bit register, and a 16bit register), 2. Two modes (6809 emulation mode and native mode), 3. Reduced execution cycles in native mode, 4. More instructions (16 bit x 16 bit multiplication, 32 bit / 16 bit division, inter-registers operationa, block transfer, bit manipulating operation which is compatible with what 6801 has, etc) 5. Error trap by illegal instruction, zero division. Some of this jibes with rumors I've heard previously. So keep your fingers crossed... this could be the fun scoop of the decade for 6809 fans! There are 2 Replies. #: 14468 S15/Hot Topics 02-Mar-92 23:26:48 Sb: #14451-Secret 6309 Modes? Fm: Brother Jeremy, CSJW 76477,142 To: Kevin Darling 76703,4227 (X) Dear Kevin: I looked through several catalog tonight..Allied, Jamesco, Newark, etc, but culd not find the 6309 or 63C09E chip listed. If I can find a source for them, I will post it. I may check with Marty over on Delphi. WIth all best wishes, Br. Jeremy, CSJW #: 14489 S15/Hot Topics 05-Mar-92 15:00:02 Sb: #14451-#Secret 6309 Modes? Fm: Lee Veal 74726,1752 To: Kevin Darling 76703,4227 (X) I've found a source for Hitachi 63C09EP microprocessors, where I can get: 1 - 24 @ $9.95/ea 25 - 99 @ $8.75/ea 100 - 999 @ $7.80/ea I'm going to try to get over to there tomorrow and pick up a specifications book on it. Lee There are 2 Replies. #: 14490 S15/Hot Topics 05-Mar-92 15:17:34 Sb: #14489-#Secret 6309 Modes? Fm: Kevin Darling 76703,4227 To: Lee Veal 74726,1752 (X) HEY! The guy posted the secret 6309 opcodes on Usenet today... AND THEY WORK!!!!!!!!!!!! (at least, the ones I tried, which included a block transfer mode, and a 32 by 16-bit division!!!!) Details forthcoming... my gawd, if only we'd known about this a couple of years ago! And now my hands are tied up with 68K stuff. Gentlemen, you're going to have loads of fun with the 6309, I can tell already! With its bit manipulation, fast copies, extra 16 and 32-bit register modes, division, etc... you could do some really hot things!! Can you tell that I'm excited? kev There is 1 Reply. #: 14495 S15/Hot Topics 05-Mar-92 21:00:24 Sb: #14490-#Secret 6309 Modes? Fm: James Jones 76257,562 To: Kevin Darling 76703,4227 (X) Well...if there are extra registers, you need to be vewy, vewy caweful (as Elmer Fudd would say). Remember, we're multitasking! To really do it up right, I fear that the process descriptor would have to be modified and that the kernel would have to save any extra registers at context switch time. That might make for an interesting project, were there people with skill and time and permission. There is 1 Reply. #: 14505 S15/Hot Topics 06-Mar-92 09:49:45 Sb: #14495-#Secret 6309 Modes? Fm: Kevin Darling 76703,4227 To: James Jones 76257,562 (X) Right - the kernel would have to know about the extra register (W) saved, but the process descriptor wouldn't need modification. I've put Kent onto looking at what needs to be changed... . He's been pretty bored for the last year. I think this'll change things ;-) First up: changing F$Move, and GRFDRV scrolling, to use the block move instructions! The CoCo oughta hum right along with those. Disk drivers could be updated, too. Fun times ahead! kev There is 1 Reply. #: 14522 S15/Hot Topics 07-Mar-92 08:04:24 Sb: #14505-#Secret 6309 Modes? Fm: Mark Griffith 76070,41 To: Kevin Darling 76703,4227 (X) Kevin, I sincerely hope that you have finished with the MM/1 windows code before you go diving into "fixing" Level II to work with the secret 6309 codes. If this is true than can you PLEASE upload it into DL16 so the developers can use it to start coding some applications? Mark There is 1 Reply. #: 14529 S15/Hot Topics 07-Mar-92 12:59:28 Sb: #14522-#Secret 6309 Modes? Fm: Kevin Darling 76703,4227 To: Mark Griffith 76070,41 (X) Mark, As I've already mentioned, others will have to "fix" Level II... I'm too busy on the 68K stuff right at the moment. kev There is 1 Reply. #: 14536 S15/Hot Topics 08-Mar-92 07:47:50 Sb: #14529-Secret 6309 Modes? Fm: Mark Griffith 76070,41 To: Kevin Darling 76703,4227 (X) Kevin, Well, judging from your messages on the BITNET CoCo list, you have already started playing around with it on your system. While I can understand the exitement this causes you, you can see by reading the BITNET messages that there are a number of people having trouble with the earlier versions of the window driver and the differences between them. Getting a stable version out to everyone is extremely important. Paul Ward mentioned here last Monday that your window driver would be uploaded here in 7 days. That makes tomorrow the day. Even a few hours messing with the 6309 code can cause this deadline to be missed so please try and keep yourself on the more important tasks. Mark #: 14491 S15/Hot Topics 05-Mar-92 15:19:01 Sb: #14489-#Secret 6309 Modes? Fm: Kevin Darling 76703,4227 To: Lee Veal 74726,1752 (X) PS: can you post the source for 6309E's that you found? thx! - kev There is 1 Reply. #: 14510 S15/Hot Topics 06-Mar-92 17:18:41 Sb: #14491-#Secret 6309 Modes? Fm: Lee Veal 74726,1752 To: Kevin Darling 76703,4227 (X) Kevin, The source that I found for HD6309 microprocessors is: Marshall Electronics 2045 Chenault Dr. Carrollton, Tx 75006 (214) 233-5200 I called them yesterday (Thursday), and the guy that I talked to said that they did in fact carry them, but were currently out of stock. However, he added that it looked like a new supply was coming in in a couple of weeks. I guess if you wanted to order a few, then you'd just be backordered for a little bit. I went over there today and picked up a data book that includes most, if not all, of Hitachi's 8/16 bit microprocessors. The book's in the car now, but I seem to remember that there were 6301s through 6809s (which includes the 63B09, 63C09, 63B09E and 63C09E microprocessors). And Hirotsugu was right on in his "Secret Feature Memo", the new features are not documented in the data book. It looks like they just took the MC6809 specs and retitled the document for the 6309. I'm going to call Hitachi America to see if they now have official (or at least "semi-official) specs written for the heretofore secret 6309 features. The data book that I got today also included some MC68000 counterparts, too. I wonder of some of them have "secret" features, too, that could blow the doors off stock 68000 operations. Hmmmm.... Lee There is 1 Reply. #: 14512 S15/Hot Topics 06-Mar-92 17:44:17 Sb: #14510-#Secret 6309 Modes? Fm: Kevin Darling 76703,4227 To: Lee Veal 74726,1752 Thanks Lee! Yah, I have old 6309 docs... not a word about the secret stuff. I can't believe we've gone **4** years without knowing about it. Auuugh. See if you can look up in the 6301 manual, and find out what the AIM, OIM, EIM, and TIM commands do, woulda? thx - kev There is 1 Reply. #: 14519 S15/Hot Topics 07-Mar-92 06:39:55 Sb: #14512-#Secret 6309 Modes? Fm: James Jones 76257,562 To: Kevin Darling 76703,4227 (X) Just thought of something the other day: with the 6309 in native mode, wouldn't the layout of the register packet (see Technical Manual, page 6-15) have to change? Maybe not; guess it depends on how that's saved. There is 1 Reply. #: 14528 S15/Hot Topics 07-Mar-92 12:55:57 Sb: #14519-Secret 6309 Modes? Fm: Kevin Darling 76703,4227 To: James Jones 76257,562 (X) Right... the register packet would have to change, which could be critical to some drivers (and of course, the kernel and file managers). A simple way around that would be to have the interrupt kernel code move the W register someplace else, and also diddle with signal intercept routines, but it'll take some thinking on what is the best method (whole hog or hack?) to handle things. Another note is this: if I recall correctly, Basic09 *cheats* in its intercept routine, and looks for the DP reg on the intercept stack. That'll have to be patched as well. kev #: 14454 S10/OS9/6809 (CoCo) 01-Mar-92 19:48:10 Sb: #CARMEN/MV Fm: Brother Jeremy, CSJW 76477,142 To: All Dear Friends I recently got Where in the World is Carmen Sandiego up and running under MV, using the files here on CIS. I forget the library that they are in. Is there any way to get the hi-res mouse to work? The program is set up to either use the keyboard or the lores joy stick. Can it be patched or is it a lost cause? With all best wishes, BR. JEREMY, CSJW There are 2 Replies. #: 14462 S10/OS9/6809 (CoCo) 02-Mar-92 18:57:06 Sb: #14454-CARMEN/MV Fm: Erich Schulman 75140,3175 To: Brother Jeremy, CSJW 76477,142 (X) With software, anything your hardware can handle is possible. However, supporting the hi-res 'stick requires that special analog pulses be generated through the cassette port and requires the co-ordinates for hi-res be handled. So you can change joystick support from lo-res to hi-res but about the only viable method is disassembly, changing of the source, and reassembly. No simple patch can do it. You might instead look into the hi- / lo- res joystick adapter that will let you switch between modes without all the plugging and unplugging. Erich Schulman (KTN4CA) #: 14488 S10/OS9/6809 (CoCo) 05-Mar-92 11:55:28 Sb: #14454-CARMEN/MV Fm: Lee Veal 74726,1752 To: Brother Jeremy, CSJW 76477,142 (X) I disagree slightly with Erich because very simple patches are all that's required to make Deskmate/3 use the Hi-Res Adaptor instead of the stock Lo-Res ports. It would probably require a disassembly to figure where to patch the Hi-Res Mouse Call in, but unless the logic in Carmen is extremely convoluted, then I think it'd be quite possible to effect the change. Lee #: 14455 S7/Telecommunications 01-Mar-92 19:51:35 Sb: #STERM Fm: Brother Jeremy, CSJW 76477,142 To: All Dear Friends: I am trying to get STERM 1.5 up and running. I am using the PHONE program by Bob Van der Poel. I cannot seem to get the login to work. I tried his example, but I have a problem after I receive the CONNECT reply. Then I have to complete the logon manually. Several times I was unable to recieve anything onto the screen. Help! Thank you, with all best wishes, Br. Jeremy, CSJW There is 1 Reply. #: 14464 S7/Telecommunications 02-Mar-92 19:47:26 Sb: #14455-#STERM Fm: Bob van der Poel 76510,2203 To: Brother Jeremy, CSJW 76477,142 (X) Bro, are you dialing up CI$ through a direct node, or a network. Some of the prompts might be different. I suggest you capture the logon stuff to a file and compare the prompts to what PHONE is expecting. I use this prog. all the time, and it works without fail for me. There is 1 Reply. #: 14467 S7/Telecommunications 02-Mar-92 23:23:21 Sb: #14464-STERM Fm: Brother Jeremy, CSJW 76477,142 To: Bob van der Poel 76510,2203 (X) Dear Bob, I will do a capture and let you know. --Br. Jeremy, CSJW #: 14456 S12/OS9/68000 (OSK) 01-Mar-92 22:12:03 Sb: #C help Fm: Bob van der Poel 76510,2203 To: all I'm not sure if what I'm trying to do is possible, but it should be . Maybe someone with a bit more C-knowledge can help... I want to get the size of one element of a structure. Problem is that I don't have an actual structure to work on, just a pointer or typedef. Assuming typedef struct{ char a1[40]; int i1[10]; }foo; What I'd like to do is this: int t=sizeof(foo.a1); Which should set t==40. I just get an error. I tried it without the typedef and used sizeof(struct foo.a1) but got the same problem. If you actually create storage, things work fine: foo dummy; int t=dummy.a1; I've worked around this by using #define ASIZE 40 and then using this to set the size in the struct. But there should be a way to do it my way! Any ideas? There is 1 Reply. #: 14458 S12/OS9/68000 (OSK) 02-Mar-92 07:25:30 Sb: #14456-#C help Fm: James Jones 76257,562 To: Bob van der Poel 76510,2203 (X) Maybe there should be, but there isn't. Look again at the permissible operands of sizeof: you can either have an expression, or a type. foo.a1 is neither. The nearest you can come to what you want is to use yet another typedef for the type of the structure member; then you can hand sizeof *that* type. There is 1 Reply. #: 14479 S12/OS9/68000 (OSK) 04-Mar-92 00:36:41 Sb: #14458-C help Fm: Bob van der Poel 76510,2203 To: James Jones 76257,562 (X) Well, I was hoping I'd missed something in the sizeof() operand specs. Too bad I didn't -- always worth asking. But using a typedef to define the structure member is a neat trick. Thanks. #: 14457 S10/OS9/6809 (CoCo) 01-Mar-92 23:44:22 Sb: #Sterm Fm: Ian Hodgson 72177,1762 To: All I have a question about Sterm on the CoCo. I have just, for what seems like about the thousandth time, accidentally dumped myself out of Compuserve by hitting the BREAK key, without thinking, to abort an Sterm operation (in this case opening a capture file). Now, BREAK is a good way to abort most operations within CIS, but to have the same key, under slightly different circumstances, abort Sterm itself is ... shall I say counterproductive (5 minutes ago I used a much stronger word). I don't have the source, but even if I did I don't understand C well enough to track it down. Does this bother anyone else? If so, how much of a change is involved to fix it? I have an uncomfortable feeling that the BREAK is being handled by Shell rather than Sterm, but am certainly not sure. Suggestions? Also, does anyone know how long it takes CIS to recognize that I have suddenly disconnected and stop the meter running? (Actually, with the new flat rate I am no longer even sure when the meter _does_ run, but that's another story). There are 2 Replies. #: 14473 S10/OS9/6809 (CoCo) 03-Mar-92 07:51:22 Sb: #14457-Sterm Fm: Steve Wegert 76703,4255 To: Ian Hodgson 72177,1762 (X) Ian, Interesting stuff! I'll try to duplicate your situation on my CoCo and see what I come up with. Steve #: 14480 S10/OS9/6809 (CoCo) 04-Mar-92 00:36:54 Sb: #14457-#Sterm Fm: Bob van der Poel 76510,2203 To: Ian Hodgson 72177,1762 (X) Ian, I just tried hitting BREAK in sterm and nothing happens--it sure doesn't abort sterm. What version are you using? Hmmm, another trick would be to set abort and quit to 0 before starting up sterm (but I suspect that it already does that). There is 1 Reply. #: 14481 S10/OS9/6809 (CoCo) 04-Mar-92 07:35:29 Sb: #14480-#Sterm Fm: Steve Wegert 76703,4255 To: Bob van der Poel 76510,2203 (X) I found the same thing, Bob. Mark tells me he traps the break key so Sterm shouldn't be doing what Ian says it is. Steve There is 1 Reply. #: 14483 S10/OS9/6809 (CoCo) 04-Mar-92 21:23:35 Sb: #14481-#Sterm Fm: Ian Hodgson 72177,1762 To: Steve Wegert 76703,4255 (X) Bob, Steve, OK, try this: while in Sterm, hit escape C to open a capture file. When Sterm asks for a filename, hit BREAK. Unless my version 1.5 is different from yours, you will find yourself immediately back at a shell prompt, and no longer in Sterm or connected to Compuserve. This happens to me when I hit esc-C by mistake, perhaps intending to hit esc-T or something else. My knee-jerk reaction is to hit BREAK to abort. Well, it aborts, all right ... Let me know if you have the same result. Ian There is 1 Reply. #: 14484 S10/OS9/6809 (CoCo) 04-Mar-92 21:27:09 Sb: #14483-#Sterm Fm: Ian Hodgson 72177,1762 To: Ian Hodgson 72177,1762 (X) Oh, that was neat (and expensive)... When I was dumped back to OS9 I went off to another window and composed my reply. When done (about 4 or 5 minutes later) I loaded Sterm again and tried to get the attention of my modem. Rather to my surprise, I found that I was still in the OS9 forum! Even though Sterm disappeared, my modem and CIS were apparently happily conversing during the interim. Good thing I didn't decide to do some long computing task that took hours! Could be expensive. Ian There is 1 Reply. #: 14485 S10/OS9/6809 (CoCo) 04-Mar-92 21:54:06 Sb: #14484-#Sterm Fm: Ian Hodgson 72177,1762 To: Ian Hodgson 72177,1762 (X) Steve, I have just been on the phone with Bob van der Poel and have this Sterm thing more or less sorted out. First, there is no problem with Sterm 1.3, only with 1.5.1. With 1.3 if I hit BREAK it just tells me that there is no filename, with 1.5.1 it dumps me out. Now the cure; since I call Sterm with a procedure file (I can't remember all the darned parameters) I just added to the procedure file: tmode quit=0 and this effectively disables the BREAK key for the duration. Ian There is 1 Reply. #: 14498 S10/OS9/6809 (CoCo) 05-Mar-92 21:51:17 Sb: #14485-Sterm Fm: Mark Griffith 76070,41 To: Ian Hodgson 72177,1762 (X) Ian, The problem with Sterm breaking out when you are prompted for a filename is due to setting the screen parameters back normal (meaning turning on break again) to allow SCF to echo charcters and allow command line editing while inputting the filename. The actual code has not changed from several versions back. What I need to do is change things around so the echoing an editing features of SCF can still be used, but the break and abort signals are still disabled. Thanks for finding this. Mark #: 14459 S7/Telecommunications 02-Mar-92 16:32:04 Sb: #STERM Query Fm: Ches Looney 73016,1336 To: STERM users I have a problem related to MultiVue and STERM. I can use STERM with no problems if I don't use MV (the OS9Boot modules include VDGInt in both cases, but the MV bootfile uses WindInt and the other boot uses GrfInt). In my use of STERM, it appears that a ramdisk must be used for B+ file downloads to be successful since (apparently) my hard disk is too slow and causes lockups. If I start a ramdisk and then try to use STERM, OS9 tells me it can't load T2. If I deiniz R0, STERM runs fine. Running the boot without MV, the ramdisk and STERM work fine together. Any ideas? Ches. There is 1 Reply. #: 14461 S7/Telecommunications 02-Mar-92 18:52:09 Sb: #14459-#STERM Query Fm: Erich Schulman 75140,3175 To: Ches Looney 73016,1336 (X) It is possible that your hard disk is too slow. I would recommend you download to RAMdisk then immediately copy the file to your hard disk. If t2 is your normal terminal you need to either have it in memory before you launch Sterm (preferably as a part of your boot file) or LOADed manually. You can put /t2 in your execution directory and have it loaded as needed but putting it in OS9Boot is the best thing to do. You can use just one boot disk for both MultiVue and non-MV. Prevent MV from starting on boot by deleting autoex and run MV manually with Multistart. For all your OS-9 purposes, just let WindInt completely replace GrfInt: just put in WindInt and forget you even have GrfInt. It is OK to have VDGInt in your boot file along with WindInt (or GrfInt) as long as you use VDG screens, otherwise get rid of it and use the memory for something useful to you. If you have both 512K and a hard disk you have a lot of room for a big bootfile that will cover all your text and graphics needs and serve them well. Your OS-9 can stand a tuneup and one boot is proabably the best way. Go for it, and good luck! There is 1 Reply. #: 14465 S7/Telecommunications 02-Mar-92 19:53:59 Sb: #14461-#STERM Query Fm: Ches Looney 73016,1336 To: Erich Schulman 75140,3175 (X) Erich, I'm afraid I didn't make my problem clear. T2 is in memory, along with R0, ACIAPAK and RAMMER as shown by mdir. If I iniz R0 and then try to initiate STERM, the system tells me it can't open T2. This is when I boot my MV system which has WindInt in the OS9boot. If I deiniz R0, STERM initiates and works fine (but I don't have a ramdisk). If I boot my non-MV system (which has GrfInt instead of WindInt) and iniz R0, STERM initiates fine and works well (and I have the Ramdisk). The question is why won't the MV (and WindInt) system let the ramdisk be started and have STERM run? ??? Ches. There is 1 Reply. #: 14470 S7/Telecommunications 03-Mar-92 07:46:51 Sb: #14465-#STERM Query Fm: Steve Wegert 76703,4255 To: Ches Looney 73016,1336 (X) Fun problems, Ches! Maybe Mark will has a suggestion ... I'll mention you're having problems when I hit the office today. In the meantime, I'll also try to duplicate what you're doing and see what I find. Sounds suspiciously BLOB like. Steve There is 1 Reply. #: 14474 S7/Telecommunications 03-Mar-92 08:00:44 Sb: #14470-#STERM Query Fm: Ches Looney 73016,1336 To: Steve Wegert 76703,4255 (X) Steve, please refresh my memory. BLOB??? Thanks, Ches. There is 1 Reply. #: 14475 S7/Telecommunications 03-Mar-92 08:14:45 Sb: #14474-#STERM Query Fm: Steve Wegert 76703,4255 To: Ches Looney 73016,1336 (X) Ches, BLOB = BootList Order Bug can manifest itself in a number of ways. Most frequently seen is the BOOTFAILED message at boot. But it can be subtle as well. I had a bootlist combination that gave the outward appearance of all being well. The system booted normally and everything seemed fine. Until, that is, I tried to copy or format. Those two commands just flat out refused to work. 'Re-ordering' my bootfile and making a new boot fixed it right up. There's no one perfect list order. It seems to varry from one machine to another, so all I can suggest is to play around and see what happens. Keep a copy of what you presently have, tho! :-) Steve There is 1 Reply. #: 14477 S7/Telecommunications 03-Mar-92 12:22:40 Sb: #14475-STERM Query Fm: Ches Looney 73016,1336 To: Steve Wegert 76703,4255 (X) I wondered about that. Thanks for the nudge; I'll try a few things. Ches. #: 14476 S7/Telecommunications 03-Mar-92 10:09:27 Sb: #sterm/xmodem bug Fm: Carl Kreider 71076,76 To: MMU users I found the xmodem bug on mmu protected systems and notified Mark. Perhaps a fix will be forthcoming. - Carl There is 1 Reply. #: 14500 S7/Telecommunications 05-Mar-92 22:07:01 Sb: #14476-sterm/xmodem bug Fm: Timothy J. Martin 71541,3611 To: Carl Kreider 71076,76 jeeeez, I just recently got the source to complie (compile) but couldn't get the bus trap to occur, and gave up. Good for you! #: 14478 S7/Telecommunications 03-Mar-92 18:01:47 Sb: pointer for new Kermit Fm: James Jones 76257,562 To: All By request of the folks at Columbia University, the folks who coordinate Kermit development, rather than upload C Kermit here, I point you at them: Kermit Distribution Columbia University Center for Computing Activities 612 West 115th Street New York NY 10025 USA Telephone: +1 (212) 854-3703 Email: kermit@columbia.edu (Internet), KERMIT@CUVMA (BITNET/EARN) Columbia University has Kermit for an amazing range of computers, from the VIC-20 to the GE/Honeywell computers that run (ran?) Multics, available on a variety of media. (I'm not associated with Columbia, though long ago I did some minor work on a port of a stripped-down Unix Kermit to OS-9, and the other weekend I made C Kermit 5A on my MM/1. The people involved with Kermit development have done a humongous amount of work to help everybody move data among an incredible number of computers, large and small, and IMHO they deserve support and appreciation.) #: 14496 S10/OS9/6809 (CoCo) 05-Mar-92 21:48:53 Sb: #R0/T2 Conflict Fm: Ches Looney 73016,1336 To: Steve Wegert 76703,4255 (X) Steve, I found a solution to my problem, but I don't fully understand why the problem occurred in the first place. The symptoms are that using GrfInt permits simultaneous operation of T2/ACIAPAK and R0/RAMMER when each pair is loaded separately but that WindInt does not permit T2 to be opened after R0 is INIZd. The solution is to merge all four modules in one file and then load the single file. I can sorta understand why a driver and its descriptor should be merged, but why must both pairs be merged into a single file? And why does WindInt insist on such a merge when GrfInt does not? Thanks for your encouragement. Ches. There are 2 Replies. #: 14504 S10/OS9/6809 (CoCo) 06-Mar-92 07:24:20 Sb: #14496-#R0/T2 Conflict Fm: Steve Wegert 76703,4255 To: Ches Looney 73016,1336 (X) Ches, Your work around further suggests that the BLOB is the problem. You've effectively changed your bootlist by removing the modules and opting to load them manually. But ... you're also robbing yourself of valuable system space if you're loading the merged file after boot. Might wish to reconsider. Steve There is 1 Reply. #: 14508 S10/OS9/6809 (CoCo) 06-Mar-92 13:39:55 Sb: #14504-#R0/T2 Conflict Fm: Ches Looney 73016,1336 To: Steve Wegert 76703,4255 (X) Well-l-l-l, I'm not directly removing modules and then manually loading, but I think I see your point. (Kevin kinda rubbed my nose in it, too, by Email. Isn't there a limit to the size of the bootfile? I thought that was supposed to be 8k or less, but your comments now and earlier make me wonder whether I'm remembering incorrectly. Thanks for your help. Ches. There is 1 Reply. #: 14525 S10/OS9/6809 (CoCo) 07-Mar-92 11:48:19 Sb: #14508-#R0/T2 Conflict Fm: Steve Wegert 76703,4255 To: Ches Looney 73016,1336 (X) Ches, it's 8K multiples. My boot file is around 32K. The other place where you may be rembering 8K limits is when you load a module. It take a fulk 8K. Eats up the space real quick. Steve There is 1 Reply. #: 14532 S10/OS9/6809 (CoCo) 07-Mar-92 14:58:12 Sb: #14525-#R0/T2 Conflict Fm: Ches Looney 73016,1336 To: Steve Wegert 76703,4255 (X) Yep, I think what I'm remembering is the limit to the size of a file called shell which has shell plus other modules totalling less than 8k. I suppose one should make the boot file as close to a full multiple of 8k. For example, if the boot is approx 33k, maybe it takes 40k and wastes almost 8k. Is that reasonably correct? Ches. There is 1 Reply. #: 14538 S10/OS9/6809 (CoCo) 08-Mar-92 19:04:45 Sb: #14532-R0/T2 Conflict Fm: Steve Wegert 76703,4255 To: Ches Looney 73016,1336 You got it, Ches. I'm thinking I'm just under the 32K boundry. I'd think the same would be true of any file. When it's loaded into memory, it's gonna take 8K. So you mays well make use of the space. Steve #: 14506 S10/OS9/6809 (CoCo) 06-Mar-92 09:51:33 Sb: #14496-#R0/T2 Conflict Fm: Kevin Darling 76703,4227 To: Ches Looney 73016,1336 (X) Ches! You're running out of system space. Put all those drivers/descs in your bootfile. Loading them after boot is okay for some testing, but also really eats up the 64K kernel map. Use "smap" and "pmap" to see. best - kev There is 1 Reply. #: 14509 S10/OS9/6809 (CoCo) 06-Mar-92 13:41:19 Sb: #14506-#R0/T2 Conflict Fm: Ches Looney 73016,1336 To: Kevin Darling 76703,4227 (X) I'm off to redo the boot file, but why do I seem to remember that the bootfile was supposed to be less than 8k?? Ches. There is 1 Reply. #: 14511 S10/OS9/6809 (CoCo) 06-Mar-92 17:42:03 Sb: #14509-#R0/T2 Conflict Fm: Kevin Darling 76703,4227 To: Ches Looney 73016,1336 (X) Umm.. the bootfile has to be less than, umm.... gee. Dunno. Mine is ~33K. Didn't mean to rub your nose, btw :-) Just excited! (Pete thinks I use too many "!"s, and he's right. Hehe kev! There is 1 Reply. #: 14513 S10/OS9/6809 (CoCo) 06-Mar-92 18:47:07 Sb: #14511-R0/T2 Conflict Fm: Ches Looney 73016,1336 To: Kevin Darling 76703,4227 (X) Hey, I didn't take it bad - I enjoy the exchange - what I don't enjoy is having to re-learn lessons I thought I had learned earlier. Ah well, the alternative to advancing age and memory loss is less pleasant than an occasional embarassment. Somewhat more seriously, I continue to marvel at the capability of this "little" computer. Just about the time I think I understand it pretty well, it stings me. Great fun - Thanks again, Kev, for your guidance and encouragement. Regards, Ches. #: 14497 S10/OS9/6809 (CoCo) 05-Mar-92 21:50:32 Sb: R0/T2 Conflict Fm: Ches Looney 73016,1336 To: Mark Griffith 76070,41 (X) Mark, I've found a solution to my problem in using STERM and a ramdisk simultaneously. Please see msg 14496. Thanks for your encouragement. Regards, Ches. #: 14499 S10/OS9/6809 (CoCo) 05-Mar-92 21:51:57 Sb: R0/T2 Conflict Fm: Ches Looney 73016,1336 To: Erich Schulman 75140,3175 (X) Erich, I found a solution to my problem in using STERM and a ramdisk simultaneously. Please see msg 14496. Thanks for your encouragement. Ches. #: 14507 S15/Hot Topics 06-Mar-92 09:54:34 Sb: #Secret 6309 Opcodes Fm: Kevin Darling 76703,4227 To: all The secret opcodes and registers of the Hitachi 6309 have been uploaded to the Reference library (Lib 1). Fascinating reading! There is 1 Reply. #: 14517 S15/Hot Topics 06-Mar-92 23:29:28 Sb: #14507-#Secret 6309 Opcodes Fm: Brother Jeremy, CSJW 76477,142 To: Kevin Darling 76703,4227 (X) Dear Kevin: I had a problem downloading the file. My OSTERM2.08 would not accept the file name 6309.txt. I changed it to txt_6309 and it worked. Might just be me or could be a problem. --With all best wishes, Br. Jeremy, CSJW There is 1 Reply. #: 14518 S15/Hot Topics 07-Mar-92 00:01:45 Sb: #14517-#Secret 6309 Opcodes Fm: Kevin Darling 76703,4227 To: Brother Jeremy, CSJW 76477,142 (X) Jeremy - ummm, oh yeah. OS9/6809 (unless patched) won't accept a number as the first character of a filename... best - kev There is 1 Reply. #: 14534 S15/Hot Topics 08-Mar-92 01:07:40 Sb: #14518-#Secret 6309 Opcodes Fm: Brother Jeremy, CSJW 76477,142 To: Kevin Darling 76703,4227 (X) Deaar Kevin: How do I patch OS9/6809 to accept a number as the first character of a filename... You knew I'd ask, didn't you? --With all best wishes, Jeremy. There is 1 Reply. #: 14535 S15/Hot Topics 08-Mar-92 02:48:52 Sb: #14534-Secret 6309 Opcodes Fm: Kevin Darling 76703,4227 To: Brother Jeremy, CSJW 76477,142 Dear Jeremy, Yeah, I knew you'd ask... and you should know that my answer will be: Drop into the CoCo library (Lib 10) and do a "bro /key:kernel" :-) Up comes KRNL.AR, which is an AR'd ipatch file for making a new kernel which'll accept numbers as the first character of a filename. Hopefully, it also tells you how to install the new kernel. best regards - kev #: 14514 S12/OS9/68000 (OSK) 06-Mar-92 19:09:26 Sb: #CD-ROM Modules?? Fm: Keith H. March 70541,1413 To: All Help, Does any one know of/are writing drivers/descriptors for use with a SCSI CD-ROM drive on a 3 Meg MM/1? Keith There are 2 Replies. #: 14523 S12/OS9/68000 (OSK) 07-Mar-92 08:04:29 Sb: #14514-CD-ROM Modules?? Fm: Mark Griffith 76070,41 To: Keith H. March 70541,1413 (X) Keith, As far as I know, no one is writing any drivers for CD-ROMs. The problem is getting the drive manufacturers and especially the CD-ROM makers to release the specifications on how it is done without paying them large sums of money. If you can do this, then please do and let us know. It is just like when I decided to do UUCP for the CoCo. There were no specifications written down anywhere. There are now after several people got together and reverse engineered it which is more or less when I had to do with the help of some already existing source code. Mark #: 14530 S12/OS9/68000 (OSK) 07-Mar-92 13:02:19 Sb: #14514-CD-ROM Modules?? Fm: Kevin Darling 76703,4227 To: Keith H. March 70541,1413 (X) Keith, Mostly, you'd need an ISO-9660 file manager. I'd guess that MW has one, but I don't know the price or availability. Alternatively, someone could write one, but I don't know anyone capable of it right now. Check around on nets/BBSs tho... there may be some Unix C source for such things, or at least rough readers. thx! - kev #: 14515 S12/OS9/68000 (OSK) 06-Mar-92 19:42:33 Sb: #Focus 3001 Keyboard Fm: Keith H. March 70541,1413 To: 76703,4255 (X) Steve I have a FOCUS 3001 keyboard with the two DIPS set as follows: 1 = ON 2 = OFF Sometimes when I turn on the MM/1 the following lights come on and then go off Caps Lock Num Lock Scroll Lock if these lights do not come on and then go back off the keyboard will not execpt UPPERCASE letters by using the SHIFT key. Is their a way to fix this problem?? I paid 60+ bucks for this (Second) keyboard. The first one did the same problem but only cost me $20 HELP Does the keyboard need modifided??? Keith There are 2 Replies. #: 14520 S12/OS9/68000 (OSK) 07-Mar-92 06:42:37 Sb: #14515-#Focus 3001 Keyboard Fm: James Jones 76257,562 To: Keith H. March 70541,1413 (X) I'd kinda hate to reboot just to make sure, but I think my keyboard flashes those LEDs at powerup, too. Doesn't seem to affect its operation. There is 1 Reply. #: 14521 S12/OS9/68000 (OSK) 07-Mar-92 08:02:04 Sb: #14520-Focus 3001 Keyboard Fm: Keith H. March 70541,1413 To: James Jones 76257,562 (X) James If they (the leds) do not come on then I can not get the SHIFT keys to work. They have to come on and then go back off. It did it Ii'Za few minutes ago, I have to turn the MM/1 off wait a few seconds and then try again, it worked the second time. Keith #: 14526 S12/OS9/68000 (OSK) 07-Mar-92 11:50:34 Sb: #14515-Focus 3001 Keyboard Fm: Steve Wegert 76703,4255 To: Keith H. March 70541,1413 (X) Keith, I had no luck with the FOCUS keyboard I bought for my MM/1. Worked just fine on the PC's at the office, but no go on the MM/1. Returned it for a Chicony 5191 and have been very pleased. Steve #: 14516 S10/OS9/6809 (CoCo) 06-Mar-92 19:47:04 Sb: EARS & CoCo 3 Fm: Keith H. March 70541,1413 To: All HELP How can I get E.A.R.S. from Speech Systems to work with a CoCo 3? Keith #: 14527 S12/OS9/68000 (OSK) 07-Mar-92 12:19:25 Sb: #14246-Mshell Discount? Fm: edward gow 71670,3274 To: John Semler 70324,633 The GNU C compiler has not stifled the marketplace. On the contrary, it has done much to improve it. First, the Microware C compiler is the one with outdated technology and lack of support - look at the K&R Microware offering vs. GNUs ANSI, not to mention GNU C++. Also compare bugs in GCC against Microware C and see who loses! Second, look at what GCC has done for companies in the marketplace - Next, Wind River, and next DEC (with Wind River) all use it as their standard compiler. This way they don't have to waste their resources creating and supporting what will likely be an inferior C compiler. Instead they can apply their effort to generating customer features, such as POSIX functionality. Note the lack of POSIX offerings from Microware. Note also that Wind River uses GDB as the debugger for VxWorks and thus provides it for free while Microware wants $850 for its source debugger. GNU software is a boon both to individual users and to the industry at large. Microware will learn their lesson the hard way as OSK systems in the hands of home users lead to the porting ond development of replacements for all but the kernel of OS-9/68000. Then ALL anyone will buy from them is an industrial license at $29 per system. Even in Des Moines you can't live on that. #: 14539 S12/OS9/68000 (OSK) 08-Mar-92 19:19:03 Sb: #MM/1 help Fm: Steve Wegert 76703,4255 To: Kev 76703,4227 (X) Kev, I've been tracking down a problem with my MM/1 and it leads me suspiciously toward either the windowing software and or tsmon. I've got a 3meg MM/1 with 5 serial ports, tho /t2, /t3 and /t4 are all that are in use at the moment. /t2 is a local terminal while /t4 is set up to be my dial in. (tsmon /t4&). Within a 24 hour period, /term and /w1 will have locked up. /t2 is still fully functional. Rebooting has been the only way out of this fix. Sometimes, with /term and /w1 locked up, /t2 will demonstrate what I've named the 'just one off' effect. For example, if I wanted to issue the 'dir' command, pressing the 'd' would net no echo. Pressing the 'i' would net a 'd' echo,. Pressing the 'r' would net the 'i' echo. Hitting would complete the command and again to execute. Any thoughts on what's going on? I've taken to just running a shell on /t2 rather than using tsmon to see if that makes any difference. (By the way, using mtsmon doesn't make any difference. The same things happen.) I'm using version 30 of wcf and windio but the same problems appeared with 22. I've also noted unexpected values for xon and xoff in the window descriptors. WHat's behind $0F and $02. CHanging them to 00 has caused an interesting result at boot. Black border, black background, black type and a white cursor. Blindly using the color command sets things right but it's still got me wondering. (Thought the xon/xoff parameters might be related to the locked probelms). Anything you can suggest would be appreciated. Steve #: 14540 S12/OS9/68000 (OSK) 08-Mar-92 19:21:12 Sb: TOP mmon Fm: Steve Wegert 76703,4255 To: Ed 76576,3312 Ed, Have you been able to get the TOP mmon package working? I've spent a few hours with it this weekend without much luck. Any pointers? Steve Press !>