bbadger@x102c.harris-atd.com (Badger BA 64810) (03/14/90)
I got jyu.fi!c64/telecomm/kermit2p2a.sda and it garbles lowercase letters. For example, if I type a line of lowercase x's, it comes out like this: XxxXXxXxXxxXxxXXXxXXXXXxXxXxX -rw-rw-r-- 1 bbadger sec 29824 Mar 2 21:48 kermit2p2a.sda I downloaded kermit2p2a.sda to my C64 using XMODEM about three times, and it always does this. None of the other terminal emulators I used have this problem. (XMOBUF, XMODEM, Multiterm, CCGMS6.1, VTERMITE, WTERMITE, MACHTERM). The Self-dissolving archive seems to dissolve without complaint. And I can then LOAD "BOOT",8,1 which loads "KERMIT" and runs it. Is there a checksum program? Could someone mail or post the checksum of the KERMIT produced? Has anyone else used this particular KERMIT? (I.e., gotten it from jyu.fi (tukki) and run it.) By the way, this is from tukki: incoming/cbm/c64.telecom: total 92 -rw-rw-rw- 1 ftp staff 17536 Jan 26 15:08 ccgms.v6p1 -rw-rw-rw- 1 ftp staff 29824 Jan 26 12:51 kermit2p2a.sda -rw-rw-rw- 1 ftp staff 44288 Jan 26 12:52 kermit2p2b.sda So it looks like I have the file length right, anyway. ---- Bernard A. Badger Jr. 407/984-6385 |"Get a LIFE!" --- J.H. Conway bbadger@x102c.ess.harris.com |Buddy, can you paradigm? bbadger%x102c@trantor.harris-atd.com |'s/./&&/g' Tom sed expansively.
scott@max.u.washington.edu (03/14/90)
In article <3313@trantor.harris-atd.com>, bbadger@x102c.harris-atd.com (Badger BA 64810) writes: > I got jyu.fi!c64/telecomm/kermit2p2a.sda and it garbles lowercase letters. For example, > if I type a line of lowercase x's, it comes out like this: > XxxXXxXxXxxXxxXXXxXXXXXxXxXxX > > -rw-rw-r-- 1 bbadger sec 29824 Mar 2 21:48 kermit2p2a.sda > I downloaded kermit2p2a.sda to my C64 using XMODEM about three times, and it > always does this. None of the other terminal emulators I used have this problem. > (XMOBUF, XMODEM, Multiterm, CCGMS6.1, VTERMITE, WTERMITE, MACHTERM). The Self-dissolving > archive seems to dissolve without complaint. And I can then LOAD "BOOT",8,1 > which loads "KERMIT" and runs it. > > Is there a checksum program? Could someone mail or post the checksum of the KERMIT produced? > > Has anyone else used this particular KERMIT? (I.e., gotten it from jyu.fi (tukki) and run it.) I just downloaded the kermit program from tukki and I experienced no problems. As matter of fact this message is being typed using the kermit that I just downloaded. I know this will sound obious, but check the baud rate and all the other configuration settings that is give when you type "show all" in kermit. Perhaps you accident changed on of those values. Or perhaps it was just a case of a noisy line at that particular time. Sincerely, Scott K. Stephen
mat@emcard.UUCP (W Mat Waites) (03/14/90)
In article <3313@trantor.harris-atd.com> bbadger@x102c.ess.harris.com (Badger BA 64810) writes: >I got jyu.fi!c64/telecomm/kermit2p2a.sda and it garbles lowercase letters. For example, >if I type a line of lowercase x's, it comes out like this: >XxxXXxXxXxxXxxXXXxXXXXXxXxXxX > >Bernard A. Badger Jr. 407/984-6385 |"Get a LIFE!" --- J.H. Conway I had the same problem, tried another 64, and it worked fine. Apparently the custom keyboard scanning routine in kermit is sensitive to the timing of the keyboard interface chips. Also apparently, the timing characteristics of these chips has varied over the production life of the 64. Maybe you could trade with someone who doesn't care about kermit. Mat ps. kermit is the only program that I've tried that had a problem with my other 64. -- W Mat Waites | We're sharpening stones, and walking on coals, gatech!emcard!mat | To improve our business acumen. -rem
tronix@polari.UUCP (David Daniel) (03/19/90)
Here's a couple things to consider:
The C64 has a graphics register that gave rise to Pet ASCII. That's what makes standard ASCII invert, e.g.
standard ACSII: Hello looks like this: hELLO in Pet ASCII unles the register is converted.
In using the Kermit program at 1200 bps or higher you might want to set your modem echo to off and the term to half duplex.
Try the x's again and see if that doesn't clear it up.
--
... Women pickled in limbo, their legs chalky and their faces slightly twisted, as if they had been knocked into their thirties by a sideways blow.
John Updike
Rabbit Is Richjbond@escargot.UUCP (Jeremy Shepherd) (03/21/90)
re: Problems with Kermit
The first step is to unplug anything that might be present in either
joyport. If you do this, I believe that your problem will vanish.
Jeremy B. Shepherd
"...tektronix!tessi!escargot!jbond"
or just "jbond@escargot"
"To a new world of Gods and Monsters!" - Hurlbut/Balderstonkentsu@microsoft.UUCP (Kent SULLIVAN) (03/21/90)
Regarding the uppercase problems a few people have been having with Kermit v2.2, our best guess is that it is related to the keyboard scanner routine problems that were discussed here previously. In sum, Kermit is more sensitive to the chips that handle keyboard input than most programs. Apparently your machine has a chip (one of the 6526's does this job, I think) that is just on the edge of tolerance--most programs don't care, but Kermit does. Chips can vary in the ways they operate and still be considered "good" at the factory. I have never personally seen this problem on any computer that I have used Kermit on, and I have probably used at least 25 different machines. The solution to this is a patch to Kermit. Ray Moody, the author, is working on a fix but he won't have it done for awhile--he is in his last semester of school and is very busy. An interim solution is to change programs or computers. Kent Sullivan Microsoft Corporation Author of the C-64/128 Kermit v2.2 User's Guide The opinions expressed in this article are my own and do not necessarily reflect those of my employer.
jgreco@archimedes.math.uwm.edu (Joe Greco) (03/25/90)
In article <53659@microsoft.UUCP> kentsu@microsoft.UUCP (Kent SULLIVAN) writes: >Regarding the uppercase problems a few people have been having with >Kermit v2.2, our best guess is that it is related to the keyboard scanner >routine problems that were discussed here previously. In sum, Kermit is more >sensitive to the chips that handle keyboard input than most programs. It's also sensitive to having a bad SID (or no SID) chip, which causes the same symptoms. My suspicions are that it has something to do with how Kermit is trying to read the keyboard.... perhaps something is not initialized correctly. This machine works fine, with it's defective SID, for all other programs. (I don't use sound anyways...) Kermit however bombs without a functional SID. It took me a long time to discover that. ... Joe
fred@cbmvax.commodore.com (Fred Bowen) (03/27/90)
In article <53659@microsoft.UUCP> kentsu@microsoft.UUCP (Kent SULLIVAN) writes: >Regarding the uppercase problems a few people have been having with >Kermit v2.2, our best guess is that it is related to the keyboard scanner >routine problems that were discussed here previously. In sum, Kermit is more >sensitive to the chips that handle keyboard input than most programs. The only times I have seen this happen is when you setup your terminal for an 8-bit word length and the host is 7-bit, or parity is incorrectly set. An rs232 byte is transmitted as follows: start bit, B0, B1, ... Bn, parity bit, stop bit(s) If your setup is incorrect, the parity or stop bit might be misinterpreted by the system as the most significant data bit, which would cause the rather screwy upper/lowercase problem. While I did report a problem with Kermit's C128 keyscanner, it would not cause a problem like this (it caused "dead" keys in one particular row containing the A,D,G, etc. keys). -- -- Fred Bowen uucp: {uunet|rutgers|pyramid}!cbmvax!fred domain: fred@cbmvax.commodore.com tele: 215 431-9100 Commodore Electronics, Ltd., 1200 Wilson Drive, West Chester, PA, 19380
fred@cbmvax.commodore.com (Fred Bowen) (03/27/90)
In article <3063@uwm.edu> jgreco@archimedes.math.uwm.edu (Joe Greco) writes: >In article <53659@microsoft.UUCP> kentsu@microsoft.UUCP (Kent SULLIVAN) writes: >>Regarding the uppercase problems a few people have been having with > >It's also sensitive to having a bad SID (or no SID) chip, which causes the same >symptoms. My suspicions are that it has something to do with how Kermit >is trying to read the keyboard.... perhaps something is not initialized >correctly. > >This machine works fine, with it's defective SID, for all other programs. The only thing the user port and SID have in common is the 9V supply. If you have a blown fuse (or bad power supply), SID will not work and modems requiring 9V will not work. The newer, 4-dip switch 1670 modem does not require 9V, the older modems do, as does the old VIC1011A cartridge. -- -- Fred Bowen uucp: {uunet|rutgers|pyramid}!cbmvax!fred domain: fred@cbmvax.commodore.com tele: 215 431-9100 Commodore Electronics, Ltd., 1200 Wilson Drive, West Chester, PA, 19380
jgreco@archimedes.math.uwm.edu (Joe Greco) (03/28/90)
In article <10378@cbmvax.commodore.com> fred@cbmvax (Fred Bowen) writes: >The only thing the user port and SID have in common is the 9V supply. If >you have a blown fuse (or bad power supply), SID will not work and modems >requiring 9V will not work. Which is what is very strange. The system is just fine. I might point out that the problem in question happens all the time while in Kermit, and not just while in connect mode. It merrily uppercases commands on the command line, and magically goes away when I leave Kermit. The uppercasing only seems to affect certain letters. This suggests foul play keyboard scanning in Kermit. It was really freaky the first time I encountered it, because I started running a full set of diagnostics and chip swaps on the board in question. Nothing logical corrected it, so I got ticked and replaced the whole LSI chip set. Suddenly it worked.... and it localized down to the SID, which is probably the last component I would have guessed at. >The newer, 4-dip switch 1670 modem does not require 9V, the older modems >do, as does the old VIC1011A cartridge. How did we get into a user port discussion, by the way???? ... Joe
alpert@chovax.enet.dec.com (Enemy of the State) (03/28/90)
In article <10377@cbmvax.commodore.com>, fred@cbmvax.commodore.com (Fred Bowen) writes... > >The only times I have seen this happen is when you setup your terminal for >an 8-bit word length and the host is 7-bit, or parity is incorrectly set. > My C64/Kermit 2.2 combo seems to have the problem even when typing locally. (With no echo from the host I would expect the problem to go away if it were a communications problem.) I have found that my SID seems to be partially dead, someone mentioned that this is one of the chips that might cause the Kermit problem. Have to check and see if it's socketed or soldered in place... ------------------------------------------------------------------------------- Bob Alpert { alpert@chovax.dec.com } Suspicion breeds DEC { ...!decwrl!chovax.dec.com!alpert } confidence. Cherry Hill, NJ { alpert%chovax.dec@decwrl.dec.com } -------------------------------------------------------------------------------
fred@cbmvax.commodore.com (Fred Bowen) (03/29/90)
In article <3113@uwm.edu> jgreco@archimedes.math.uwm.edu (Joe Greco) writes: >In article <10378@cbmvax.commodore.com> fred@cbmvax (Fred Bowen) writes: >>The only thing the user port and SID have in common is the 9V supply. If >>you have a blown fuse (or bad power supply), SID will not work and modems >>requiring 9V will not work. > >I might point out that the problem in question happens all the time >while in Kermit, and not just while in connect mode. It merrily uppercases >commands on the command line, and magically goes away when I leave >Kermit. The uppercasing only seems to affect certain letters. This >suggests foul play keyboard scanning in Kermit. Sorry, I think if you examine which letters are incorrectly cased you will find a correspondence with the parity bit. The shift keys do not occupy positions in the key matrix that would cause this sort of problem, even with the worst keyscan code AND the longest, noisiest cable imaginable. You may not be connected to a host, but are you still connected to your modem? >How did we get into a user port discussion, by the way???? I'm assuming you're interfacing to rs232 via the user port (aka parallel port). I explained one connection between SID and rs232 was the 9VAC supply. The only other connection is through the 4066 analog switch to the key lines, but a problem there would not affect strictly the case of certain letters (as evidenced by occasional problems reported when a mouse or joystick is also wiggling the key lines). Enough. Perhaps Kermit should simply mask bit 8 if it's in 7 bit mode. -- Fred Bowen uucp: {uunet|rutgers|pyramid}!cbmvax!fred domain: fred@cbmvax.commodore.com tele: 215 431-9100 Commodore Electronics, Ltd., 1200 Wilson Drive, West Chester, PA, 19380
alpert@chovax.dec.com (Enemy of the State) (03/29/90)
In article <10407@cbmvax.commodore.com>, fred@cbmvax.commodore.com (Fred Bowen) writes... > >Sorry, I think if you examine which letters are incorrectly cased you will >find a correspondence with the parity bit. The shift keys do not occupy >positions in the key matrix that would cause this sort of problem, even with >the worst keyscan code AND the longest, noisiest cable imaginable. You may >not be connected to a host, but are you still connected to your modem? > I have Kermit 2.2 running on my C64 sitting next to my terminal. The C64 is not connected to a modem; there is nothing connected but the disk drive and monitor. At the local "Kermit-65>" prompt I can repeatedly type any alphabetic key and it will at random come up as upper or lower case with no discernable pattern. (This does not occur if the SHIFT key is pressed.) Since there is no modem involved and the C64 is just echoing local commands I don't understand how it can be a communications problem. The only other known problem with my machine is that the SID seems to be sick. I don't have any other software that displays this problem. ------------------------------------------------------------------------------- Bob Alpert { alpert@chovax.dec.com } Suspicion breeds DEC { ...!decwrl!chovax.dec.com!alpert } confidence. Cherry Hill, NJ { alpert%chovax.dec@decwrl.dec.com } -------------------------------------------------------------------------------
fred@cbmvax.commodore.com (Fred Bowen) (03/29/90)
In article <9719@shlump.nac.dec.com> alpert@chovax.dec.com writes: > >I have Kermit 2.2 running on my C64 sitting next to my terminal. The C64 is >not connected to a modem; there is nothing connected but the disk drive and >monitor. At the local "Kermit-65>" prompt I can [...] type any alphabetic >key and it will at random come up as upper or lower case with no discernable >pattern. (This does not occur if the SHIFT key is pressed.) > >Since there is no modem involved and the C64 is just echoing local >commands I don't understand how it can be a communications problem. > >The only other known problem with my machine is that the SID seems >to be sick. I don't have any other software that displays this problem. Now THIS is the way problems should be reported in the first place! It provides all the info necessary to track down a problem. I stand corrected- the Kermit keyscan in one place checks address $D600 to determine if it is scanning a C64 or a C128 keyboard for the purposes of scanning the C128's CAPS LOCK key. In a C128, this address is a VDC (80-col chip) register which would not normally be zero. In a C64, this is an image of the SID chip, and is normally zero. I can see where it's possible for an unhealthy SID to incorrectly drive the data bus, resulting in random values for this address (you could write a simple BASIC program to PEEK this location and print the results- if my theory is correct the values will be random in your system with a bad SID). This would have the effect of randomly making the Kermit program think the CAPS LOCK key was down. A quick patch would be to load Kermit2.2 (73) into you C64, but before running it do this: POKE 18846,76: POKE 18847,183: POKE 18848,73 This will patch a JMP over the LDA $D600, skipping the check for the C128 CAPS LOCK key. If it works, save it to disk. -- -- Fred Bowen uucp: {uunet|rutgers|pyramid}!cbmvax!fred domain: fred@cbmvax.commodore.com tele: 215 431-9100 Commodore Electronics, Ltd., 1200 Wilson Drive, West Chester, PA, 19380
alpert@chovax.dec.com (Enemy of the State) (03/30/90)
In article <10427@cbmvax.commodore.com>, fred@cbmvax.commodore.com (Fred Bowen) writes...
)
) A quick patch would be to load Kermit2.2 (73) into you C64, but before
) running it do this:
)
) POKE 18846,76: POKE 18847,183: POKE 18848,73
)
) This will patch a JMP over the LDA $D600, skipping the check for the C128
) CAPS LOCK key. If it works, save it to disk.
I tried this and it seems to work fine. Thanks!
-------------------------------------------------------------------------------
Bob Alpert { alpert@chovax.dec.com } Suspicion breeds
DEC { ...!decwrl!chovax.dec.com!alpert } confidence.
Cherry Hill, NJ { alpert%chovax.dec@decwrl.dec.com }
-------------------------------------------------------------------------------jgreco@archimedes.math.uwm.edu (Joe Greco) (03/30/90)
In comp.sys.cbm article <10407@cbmvax.commodore.com>, fred@cbmvax (Fred Bowen) wrote: >>I might point out that the problem in question happens all the time >>while in Kermit, and not just while in connect mode. It merrily uppercases >>commands on the command line, and magically goes away when I leave >>Kermit. The uppercasing only seems to affect certain letters. This >>suggests foul play keyboard scanning in Kermit. > >Sorry, I think if you examine which letters are incorrectly cased you will >find a correspondence with the parity bit. The shift keys do not occupy >positions in the key matrix that would cause this sort of problem, even with >the worst keyscan code AND the longest, noisiest cable imaginable. You may >not be connected to a host, but are you still connected to your modem? Fred, would you please re-read my statement? If you are in Kermit, type "^Y c" to break out to the Kermit command line, and type, the problem STILL occurs. I am _POSITIVE_ that my keystrokes are not going to the modem; my Courier has status lights. It also happens with nothing at all connected to the User Port, I just booted a machine and tried it. Now, I'm not interested in explaining this all day. I can't seem to find a set-up where this doesn't happen. So it is very likely that if you pop your hood, pull your 6581, and boot Kermit that you can try it for yourself. It's probably a stupid program "bug." ... Joe P.S. I just checked maxwell.physics.purdue.edu for the source code, I don't see it.