8bitfiles.net/archives

home *** CD-ROM | disk | FTP | other *** search

/ 8bitfiles.net/archives / archives.tar / archives / genie-commodore-file-library / Information / CAT5-MSGS.ARC / 1.CAT5-930915 next >

Wrap

Text File | 2019-04-13 | 33.5 KB | 818 lines

rea 8-9 nor ************ Topic 8 Mon Jan 25, 1993 R.KNOP1 [Rob Knop] (Forwarded) Sub: 512K and 256K Thoughts on programming machines which have been expanded as per Richard Curcio's article in TC128 #30 and #31. 3 message(s) total. ************ ------------ Category 5, Topic 8 Message 1 Mon Jan 25, 1993 R.KNOP1 [Rob Knop] (Forwarded) OK, here's a brainstorm as to something very useful that one could do with a 512K 128: One could write a mongo machine language debugger. Symbolic even. Or, even a source level debugger for a C complier, though one would probably also have to write the complier in order to have it output the right sort of debugger files (with labels and symbols and line positions and blahblahblah). Here's how it would work: you load the debugger into the 1st 256K of RAM. The debugger then loads the program to be debugged into what _it_ sees as bank 5, and installs alternate 256K switching code in both bank 0 and bank 5. In bank 5, the NMI vectors (for a RESTORE hot key) and the BRK IRQ vectors are redirected to code that switched back to the 1st 256K. The program is then executed full speed in the 2nd 256K. The only part of memory that would be offlimits would be the small amount of code necessary to deal with switching back to the first 256K. The debugger has full access to all of the code and variable space of the program being debugged, in the form of banks 5-7. Other than the switching code, the debugger uses none of the system resources so far as the program running is concerned - as long as the debugger is careful to save and/or leave alone registers in the I/O block. The debugger would use a 40 column screen, thus leaving the VDC completely alone (except for maybe VDC accessing user commands), and using 40 column memory in a range of RAM inaccessible to the debuggee. The debuggee has full access to 256K of RAM... meaning, however, that one can't test a program's access to 512K using this scheme. How's that for a brainstorm? Ed, when can you have it written by? :> Maybe I'll do this someday in my copious spare time. (I know, how do I know RTCM doesn't do it already? <G> ) -Rob ------------ Category 5, Topic 8 Message 2 Mon Jan 25, 1993 CBM-ED [e.g.bell] (Forwarded) Actually, Rob, your message was so long, I had time to write it as an overlay for RTCM while reading your message. I'm going to type it in now.... should be ready by the weekend. I'll upload it on Monday and you can debug it... The only problem I see with a project that size is that there are so few people with the memory expansion, and so few of them who develop ML that you might be investing an enormous amount of time for a miniature audience. Steven Schnedler has sold a symbolic debugger for years that I don't own, but I LIKE your idea. If you ever do write one, it is a utility I *will* use, and that is as great a compliment as I can pay! ------------ Category 5, Topic 8 Message 3 Tue Jan 26, 1993 R.KNOP1 [Rob Knop] (Forwarded) Here's another brainstorm I had for 512K, this time in CP/M. Though I'm not anything of a CP/M guru, from what I do know I believe there is no reason why CP/M couldn't use, with slight modification, banks 2 or 3 (and perhaps banks 4 through 8) as the RAM for TPA instead of bank 1. Now, with some patching to the CP/M system (and space has been left by James Waltrip and Randy Winchester's modified version that slashes out the 40 column support and otherwise improves the system), one could envision a utility like the GateWay Switcher, which would allow one to choose, at the press of a hotkey, which bank should be the forward TPA. I've bounced this off Randy Winchester, have yet to get a reply from him. (Incidentally, I also will be uploading Randy's "BIOS-R6" which has system improvements for CP/M, and includes the BDOS replacement ZPM3N08.LBR, once it is stable enough that Randy gives me the OK to spread it.) So many interesting programming projects, so little time.... -Rob ------------ ************ Topic 9 Fri Dec 11, 1992 CBM-ED [e.g.bell] (Forwarded) Sub: RAMDOS - In MY programs... How can I use RAMDOS in my own programs... those ML GEMS that would be just that much better if they could use RAMDOS transparently to my DOS routines? 3 message(s) total. ************ ------------ Category 5, Topic 9 Message 1 Fri Dec 11, 1992 CBM-ED [e.g.bell] (Forwarded) When I did BellTerm, I used existing RAMDOS routines and all I really had to do was re-divert the diverted NMI vector back to George Hug's code. I took a stab at doing that with the RAMDOS II routines for the C128 by Andrew Mileski today, and the first attempt failed miserably. I put the interface page at $0F00 and set it to device 13. I am using 4K common RAM if that matters. The default setting in the RAMDOS.BAS for this incarnation of RAMDOS is page $0E00, so I'm not that far removed. I am sure it will work... but if I can save myself a lot of development time, I'm defintely not too proud to get that help. Anyone have anything that can shed a little light? Thanks! egb ------------ Category 5, Topic 9 Message 2 Mon Feb 01, 1993 B.ECKARD (Forwarded) I am currently trying to make ramdos compatable with a program that I recently compiled. Could anyone shed some light on what an interface page is? How does one calculate the page number that Andrew Mileski's program uses to interface the ramdos. : My program when kept in basic is compatable with the ramdos at page 14 When I run the compiled version and set up the ramdos the machine locks up. I think that my compiled program is interfering with ramdos. Could anyone help with this? ------------ Category 5, Topic 9 Message 3 Tue Feb 02, 1993 CBM-ED [e.g.bell] (Forwarded) The interface page is a 1 page (256 bytes) of RAM. The 'page' designation refers to the high byte of the address of that page. For example, to find the address of 'page' 14, multiply 14 * 256. That is the default address for A. Mileski's interface page. Since you are using a 128, can I also assume you are using Abacus BASIC 128 compiler? If so, try adjusting the memory configurations to protect low memory. I don't know if you will have success with this as I have never used RAMDOS with a compiled program but it may work. I know this is a common problem with compiled programs. Incidentally, the way RAMDOS works is by swapping out memory in the ram area between $2300 and $3FFF or somewhere in that vicinity. Does your program allocate that area for anything? That would not make a big difference really, but if you can, allocate the graphics area in your program. That may also help. If you are using a different compiler, however, none of this may be any help at all. ------------ 5 rea 12-15 nor rea 19-20 nor ************ Topic 12 Sun Jan 10, 1993 F.CANANZI (Forwarded) Sub: wanted protocol shell Does the library have a protocol shell for the 128? I'm looking for a shell that will work on my 128 at 2400 baud using xmodem 65 message(s) total. ************ ------------ Category 5, Topic 12 Message 1 Sun Jan 10, 1993 F.CANANZI (Forwarded) Still trying to get used to this. Anyway i'm looking for a working protocol shell for my 128 at 2400 baud using xmodem ( 1k ) and ymodem. ------------ Category 5, Topic 12 Message 2 Sun Jan 10, 1993 C128.JBEE (Forwarded) Do you mean a shell like CS-DOS? ------------ Category 5, Topic 12 Message 3 Sun Jan 10, 1993 CBM-ED [e.g.bell] (Forwarded) If you are talking about a shell like theold Punter program (The shell was the BASIC part), no one who has written Ymodem protocol has released anything like that. If you want these protocols, get Ed Parry's term or DesTerm. Ed Parry was the first to accomplish Ymodem on 8 bit machines on the 128. If it is source code you are looking for, you can get that too, but it will cost you $30, again from Ed Parry, and is the new rate for registration of his EBBS128. The upside is that you get a very good BBS program for the C128 that supports Ymodem, a terminal that does also, and source code for the protocols. And you get to support a most excellent author and person! ------------ Category 5, Topic 12 Message 4 Tue Jan 12, 1993 F.CANANZI (Forwarded) I already own and run EBBS128 by Ep. I already tried using his source code and buddy , to make my own ml . However i just can't get it to work properly at 2400 baud .. It works fine at 300 baud but not at 2400 baud. I may have some sort of timing error or perhaps i'm missing something. Anyway .. i just thought that we might have something in house that could be used . JBEE .. This might be a great addition to the library eh? ------------ Category 5, Topic 12 Message 5 Tue Jan 12, 1993 CBM-ED [e.g.bell] (Forwarded) What you may need is a copy of George Hug's newnmi code from the February 1989 issue of the Transactor. Though I have used Ed's terminal at 2400 to test my own, so I know his works. What is the problem you are having? ------------ Category 5, Topic 12 Message 6 Tue Jan 12, 1993 C128.JBEE (Forwarded) What would make a great addition to the library? ------------ Category 5, Topic 12 Message 7 Tue Jan 12, 1993 F.CANANZI (Forwarded) JBEE.... I thought something like 'already developed' shells for various applications would be nice. Like in this case i need a protocol shell .. would be nice if we had a shell in the library to use. You know like te IBMers have. Something like this under CS-dos would be even handier. Anyway ... the problem i'm having... I tried to use a home made simple xmodem protocol and the 2400 baud code from the Transactor. Did not work at 2400 baud .. worked just fine at 300 baud ... no good ... Next i tried to use Ep source and the 2400 baud code.. this did not work at all the two just don't mix. Here i started to suspect the timing problem. Next i tried Ep code alone and this did not work at all ... suspected problem with the timing again .. looks like the computer or the modem is not at 2400 baud. My next trial will be to try Ep code and use his poke codes to adjust for 2400 baud. I wasn't doing this before. If we had a simple ' even compiled ' shell we could just call the send or receive routine .. what protocol ... what drive number ... and that would be it. ------------ Category 5, Topic 12 Message 8 Wed Jan 13, 1993 CBM-ED [e.g.bell] (Forwarded) There are protocol source files in our libraries for Xmodem CRC, Wxmodem, and Punter. We may be considering shells differently though. To me, a shell is the BASIC part of the program which calls the ML, though that is not really important. I know George Hug's routines work on both the C64 and the C128 because I have used it on both. I think Ed Parry used Steve Punter's 1200 baud kludge to get 2400, which is do-able but not really the way you want to go. This may not apply, but I have found that when I had problems like yours, the problem was always my code and never my computer. Again, that is my experience and may not apply to you. Why do you want to re-write Ed's code? Or are you trying to add it to something that did not come from Ed? If you upload your source code I will at least look at it, though I would not attempt to correct it. I may be able to see what is going on. Maybe not. But if it is your computer working against you, will a shell help you at all? You are the first person I have heard of for whom George Hug's code did not work. ------------ Category 5, Topic 12 Message 9 Wed Jan 13, 1993 F.CANANZI (Forwarded) I just made my trials and it works ... sort of. I'm trying to make my owm term .. that's all. I used Ep source for lack of something better or short of trying to re-invent the wheel. Btw Hugs code does work in my other 'projects' just fine. I was talking about the combination of his and Ep codes .. that doesn't work. I was right it was a speed error in my machine .. to get it to work (Ep) code .. i had to use 2424 as a baud rate. At least it works. My next problem is that on uling it adds 4blocks (using XM 1K) and the same on dling. THe file i sent to the bbs was 33 blocks it received 37 and i got back 41. Think something is wrong? ;] I'm too tired to look at the files with a disk editor but i bet it's adding 1 packet ( 1024 ) each time .. for what ever reason. ------------ Category 5, Topic 12 Message 10 Wed Jan 13, 1993 CBM-ED [e.g.bell] (Forwarded) Your problem is the padding. There are probably 2 problems you are dealing with too. First, as I remember, Ed never supported the step-down to 128 byte blocks for the last blocks if the file did not end on an even 1k boundary. Second, your download does not strip the padding. I don't know what to tell you on that. I didn't support stripping in BellTerm but in the program I am doing now for the *FlagShip* I did in the Xmodem protocol, and will also in the Ymodem when I get around to doing it. However, as long as there are programs that don't, there will always be the situation you are experiencing. I was leery of doing any kind of padding strip when I did BellTerm because it seems no one, including Chuck Forsberg who authored Ymodem, sticks to a standard. What seems to be popular is to pad with 0's unless the last character in the file is a 0, in which case the file is padded with CHR$(26) characters. This is in line with the old CP/M standard for eof. GEnie pads with zeroes. If the last character is a zero, GEnie sends 1 chr$(26) then pads the rest with zeroes. Forsberg's spec said to use CHR$(26), but he also did the Ymodem here on GEnie. Matt Desmond and Gary Farmaner apparently do it differently too. Most of the terms that strip do so successfully most of the time, failing only occasionally so it can be done. You will have to do the code on your own though. If you are using Ed's code, you might want to get permission before you release any kind of public domain shell. You may have already handled this, I don't know. Nothing is implied by that suggestion! Good luck! Be interesting to see what you come up with! egb ------------ Category 5, Topic 12 Message 11 Wed Jan 13, 1993 C128.JBEE (Forwarded) For what it is worth we no longer release files that are effected by xmodem padding (Abacus compiled programs). We usually ARC them. ------------ Category 5, Topic 12 Message 12 Thu Jan 14, 1993 F.CANANZI (Forwarded) Jbee..... Sorry i don't understand? Are you sayng we don't allow compiled programs? Does the xmodem padding have an effect on Abacus compiled programs? EGB... I think you are correct in that he does not strip the padding. However i Just don't understand how the BBS does it .. if a user ul to the bbs there is no problem. I ran into this problem by using his code. I have no intention of releasing anything , not just yet anyway. This project is something that i'm trying to put togethere , for my own use , to work under CS-DOS ( wish list here ). Anyway will be talking to Ep about this glitch and see if he has a solution or point me in the right direction. I did find the timming error on my own ( re my BBS ) .. perhaps other EBBS owners should take note here .... if you are having any kind of problems with users at 2400 baud try using 2400 + - 25 or so for speed. ------------ Category 5, Topic 12 Message 13 Thu Jan 14, 1993 C128.JBEE (Forwarded) We allow compiled Abacus programs, we just don't allow them to be uploaded without some type of compression because the padding nukes the file. I believe the Abacus files look for the end of the file to compute a SYS call and if you move the end of the file it gets mixed up. ------------ Category 5, Topic 12 Message 14 Thu Jan 14, 1993 CBM-ED [e.g.bell] (Forwarded) I have wondered aloud before if there were not some way to determine the original end-of-file and restore it on Abacus' compiled programs but I am not sure that is the only compiler that suffers this problem. I guess I should pay more attention to the cutesy line used to start those programs. I am sure Ed's code does not strip padding in any of his protocols for a couple of reasons. First, I have it. Second, his BBS uses the same protocol shell that his term does. Padding is only a 'problem' on Abacus compiled programs and programs whose size is critical, i.e. overlay programs that must load and not disturb memory immediately after them. Ymodem is about the only protocol that someone has no code in pd for, and Ed Parry is the only one I'm sure anyone knows of whose code is available for, albeit for registered owners of EBBS. GEnie minimum file size is like 1260 characters. Upload an 1 block file and you will see that. I guess the header data is all with the file, though you don't get that back when you download. However, I don't think GEnie strips padding for this reason. There is a mechanism in Ymodem to send the file length. I do it in BellTerm, based on the size in disk blocks * 254 bytes. This also was Ed's technique. However this part of the Ymodem header is optional. It was intended as a way to eliminate padding, or more correctly, to insure the file received was the file sent. However, with CBM files and disk drives, this is a lot more iffy. Note a DesTerm Ymodem upload... the blocks register 0. But I know what you must deal with. One thing I never did in Bell Term, tho I have done in RTC Master, was to send a block of nulls as padding for files that end on even 128 byte boundaries. But I never experienced the problems you originally had w/the growing files. I assumed always that it was not doing that that accounted for that. ------------ Category 5, Topic 12 Message 15 Thu Jan 14, 1993 HOWIE-CBM (Forwarded) A Term within a shell? A shell something like CS/DOS... I believe the CS/DOS shell already has an ansi Term. :) What would be nice tho is giving it SwiftLink capability for faster bps. Howie ------------ Category 5, Topic 12 Message 16 Fri Jan 15, 1993 F.CANANZI (Forwarded) Cs-dos ... has a term? News to me. Care to enlighten us .... like file name? ------------ Category 5, Topic 12 Message 17 Fri Jan 15, 1993 CBM-MARK (Forwarded) If I remember right, CS-DOS first came out it included a TERM module but when Chris Smeets released CS-DOS to bulletin boards as shareware the TERM mod was left out for some reason. To those who know more than I do about this, wasn't CS-DOS originally available only by ordering directly from Chris Smeets? Or did he market it through someplace else? Mark ------------ Category 5, Topic 12 Message 18 Fri Jan 15, 1993 CBM-ED [e.g.bell] (Forwarded) It was only the last release of CS-DOS that did not include Term. And contrary to what Matt Desmond would have people believe, Chris' Term was the first program to achieve 9600 baud on the 128, even though he did it using Geoff Welsh's nmi code. Originally CS-DOS was a shareware kind of deal for $30 from Chris. Chris was not one to acknowledge registrations from what I hear. I know he never said anything about mine, tho I did so after he dropped the price to $10 or $20. I do remember the stir he caused tho, when he announced that he had done 9600 with his program using the code someone uploaded to his bulletin board. I bet the thread is still around here somewhere, unless it got purged when the new BB was created back in 1990. ------------ Category 5, Topic 12 Message 19 Fri Jan 15, 1993 C128.JBEE (Forwarded) Term was released only to registered owners of versions later than 1.1 and wasn't released to the public because it still had bugs. I might release it later this year just for kicks. BTW:Chris may not have responded to registrations but he did keep fairly good records. I will probably contact all the people that registered sometime later this year once I have all the names typed into the database. ------------ Category 5, Topic 12 Message 20 Fri Jan 15, 1993 HOWIE-CBM (Forwarded) The CS/DOS term? It is called Term. :) It was released with the commercial release version of CS/DOS. Howie ------------ Category 5, Topic 12 Message 21 Sat Jan 16, 1993 F.CANANZI (Forwarded) Just think .... I wouldn't have know that a term exists for CS-Dos if i hadn't asked... Just think. JBEE ... Are there anymore goodies in the package? ------------ Category 5, Topic 12 Message 22 Sat Jan 16, 1993 HOWIE-CBM (Forwarded) Well, if I recall correctly, TERM was described as being in a pre-release format. Anyway it is not very user friendly, a la Dialogue128. In fact, the user must know exactly what he/she wants to do since there are no menus, prompts, etc. That said is is/was nice as a means of communicationg from within the CS/DOS shell, since you could use batch files for auto activity, etc. I used to have it logon to GEnie and do routine activites. I believe TERM is the only one that will let you play back an ANSI session once off-line, and all the graphics play back in their original way. I think it is TERM, or am I confusing it with some other term (smallcase)? There was a time, when CS/DOS had been upgraded, but not to its present version that Chris took it 100% commercial. It was with this version that I got TERM, included on the set of disks, along with a considerable professionally printed Manual of docs. As I've said, giving TERM SwiftLink capability now would bring it up to date. Altho adding a few user nice facitlities might make it even nicer, but then it would be the same as Dialogue. :) Altho Dialogue cannot play back an ANSI session. Or if it can, I do not know how to do it. Howie ------------ Category 5, Topic 12 Message 23 Sat Jan 16, 1993 C128.JBEE (Forwarded) There are other goodies :) ------------ Category 5, Topic 12 Message 24 Sat Jan 16, 1993 F.CANANZI (Forwarded) Howie... Tell me something.. Am i correct in this ... Xmodem 1k .. Is it Xcrc with 1024 byte packets? In other words is it CRC with a buffer till 1024 bytes received with 128 packets? Looks like i got to reinvent the wheel if i want to get this thing done. I'm trying to get information on protocol specific. The one i realy want is 1k as this will be compatible with most systems. Still looking for a protocol shell or something like it. Perhaps Q might have it . ------------ Category 5, Topic 12 Message 25 Sat Jan 16, 1993 B.VRIELING1 (Forwarded) Flavio, Hi. :)) Xmodem 1K is 'almost' Xmodem CRC, with 1K blocks. I say 'almost' because there is a difference: the character used to indicate 'start of block' is different. In this way, the recieving term knows whether an incoming block is 128 bytes or 1024 bytes. That really is the only difference according to the official specs. ...Bruce GEnie: B.VRIELING1 Internet: bvrieling@undergrad.math.waterloo.edu Fido: Bruce Vrieling @ 1:229/15 ------------ Category 5, Topic 12 Message 26 Sun Jan 17, 1993 CBM-ED [e.g.bell] (Forwarded) Although, to make things seem even more confusing, an Xmodem 1K implementation must, by spec, support any combination of 128 and 1024 byte packets. Ideally, it should step down, for example, if the line is noisy and causing lots of errors. Another example would be a file which is not an even multiple of 1024 bytes. There could be up to 7 blocks that are not 1K that would be sent as 'normal' 128 byte Xmodem CRC blocks. Other than that, Bruce is exactly correct about the difference between 128 byte and 1024 byte packet Xmodem CRC code. ------------ Category 5, Topic 12 Message 27 Sun Jan 17, 1993 HOWIE-CBM (Forwarded) Thanks for the answers Bruce and Ed! I didn't really know all of this. The only hing I can add is that the blocks are the 1024 bytes, so if there is an error all 1024 get retransmitted. Howie ------------ Category 5, Topic 12 Message 28 Sun Jan 17, 1993 F.CANANZI (Forwarded) Hello Bruce.... How is U.W. going? I was going to call you but i thought you might be busy with school things. Yes i knew that .. i was just trying to see how much 'co-operation' I would get on this topic. As usual getting information out of established programmers is rather difficult . I guess i should of done the term from scratch in ML from the start . BTW .. i have a simple working version with the 'Hug' code. Installing a protocol is not easy .. Because of the lack of this code in the PD area it's not as simple as one might think. if the 8 bit is to continue to be the machine of choice ( read hobbie here ) then there has to be more material 'SOURCED' . This booty should be made available to as many users as possible. You of all people know this all too well , you supported this theory in all of conversations in the past , as well with support on the XP and the latest mods to CS-dos. I guess what i'm trying to say is ... we ( read all ) should be going out of our way to fill the libS with as many moduals as possible. Like going to the hobbie store ... one of these , one of those and two of that and it will fly. This is the only way that the 8 bit can and will survive into the future. After all this is what goes on in the IBM world. Assembling bios is what it's all about. Any rebuttle here ? ------------ Category 5, Topic 12 Message 29 Mon Jan 18, 1993 CBM-MARK (Forwarded) Having source code available would also help those who are first learning how to code in ML, C or whatever. :) Learning by example, to me, is far easier than learning on your own. A good book or two helps too ;D Mark ------------ Category 5, Topic 12 Message 30 Mon Jan 18, 1993 C128.JBEE (Forwarded) The only way for the 8 bit to survive, is for companies that have the code, is to use it to make money :) We publish source code in TC-128/64, there is source code in Lib#5, and source code in the BBS libraries uploaded by the shareware authors. Plus, there are information files on protocal formats in Library #1! Of course there is nothing preventing you from buying source code from someone like Ed Parry (which you did) and other authors. ------------ Category 5, Topic 12 Message 31 Mon Jan 18, 1993 CBM-ED [e.g.bell] (Forwarded) I think that is an important point. The problem is frequently not the lack of source code, but the lack of the source code WE WANT. Stuff like Ymodem is obviously not that easy to develop. Thus the people who have done the leg work want an advantage for their program. But I thought whoever started this thread had Ed Parry's source code. If that is the case.... ------------ Category 5, Topic 12 Message 32 Mon Jan 18, 1993 B.VRIELING1 (Forwarded) Flavio, Waterloo is going the same as normal: school stinks. ;) We're learning methods for categorizing sorting algorithms into 'families' do we can compare them. Fun, wow. :P Yeah, I remember the fun I had playing with the ML with the Spence XP BBS. My, that was a lifetime ago. However, I had an advantage: I had a working Xmodem protocol to start with. Taking that to Xmodem 1K and eventually to Ymodem was not that difficult. It seems to me that the 'Hacking Magazine' is currently the best place to gets ones hands on good source code. They've had some real gems lately. Reminds me, I have a submission to write up... :) Give me a call some time. ...Bruce GEnie: B.VRIELING1 Internet: bvrieling@undergrad.math.waterloo.edu Fido: Bruce Vrieling @ 1:229/15 ------------ Category 5, Topic 12 Message 33 Mon Jan 18, 1993 C128.JBEE (Forwarded) I have been toying with the idea of making all the source to Bellterm available under license ($50 plus 3% on commercial software). (except the ARC source) If the uploaded files work out okay maybe I will if enough people are interested. ------------ Category 5, Topic 12 Message 34 Mon Jan 18, 1993 CBM-ED [e.g.bell] (Forwarded) That might generate some other people doing work on it AND provide people some reference material on how to (or not to) do things. ------------ Category 5, Topic 12 Message 35 Thu Jan 21, 1993 F.CANANZI (Forwarded) That' a great idea JBEE. A little steep but a good start. Learn by example ... JUst like TC . But real practical things. Like making a Database or Flexible Spreadsheet with macros. To tell you the truth i hate the ' well let me think this out' stage. However tinkering with something fits more my time schedual. I like the idea... Bruce , Howie , Ed .. what do you think? ------------ Category 5, Topic 12 Message 36 Thu Jan 21, 1993 HOWIE-CBM (Forwarded) I am not sure... I think, but about what? :/ Howie ------------ Category 5, Topic 12 Message 37 Fri Jan 22, 1993 C128.JBEE (Forwarded) FC, what would you consider a reasonable price? ------------ Category 5, Topic 12 Message 38 Sat Jan 23, 1993 F.CANANZI (Forwarded) It all depends on the demands. Like i paid $60 us for the BBS from Ep. I now understand it goes for less. Let me put it to you this way. If you had say 10 users that wanted to learn how to construct a term you could negotiate a fixed price .. however ... some sort of course should go with it , some would call support. In that case $50 us is fine. Maybe one or two of the ten would need real support. That's the sort of thing that sells. LIke in my case ... all i want is to implement 1k in my term. A ready to use , xmodem 1k, ymodem source code i have but getting it to work is another thing. Source code for this project would be worth $20 to me. ------------ Category 5, Topic 12 Message 39 Sat Jan 23, 1993 CBM-ED [e.g.bell] (Forwarded) This may be way out of line, but if my source code is being referred to here (from BellTerm), there is no way you would be able to simply assemble and call it from your code. It is too entwined with the main program. It and all of the modules were written as though a part of the main program. Ed Parry's on the other hand I believe is written to do exactly what you say you are looking for. ------------ Category 5, Topic 12 Message 40 Sat Jan 23, 1993 C128.JBEE (Forwarded) FC, since you have been so nice and reasonable I will upload the source code for both versions in Feb just for you under the same conditions I did for Bellterm itself. All I ask is that people don't pirate it and respect the copyright and distribution limits. ------------ Category 5, Topic 12 Message 41 Sat Jan 23, 1993 CBM-ED [e.g.bell] (Forwarded) Is that all of the source code, John, or just the Ymodem? Don't care.. just curious. Kudos one way or the other! ------------ Category 5, Topic 12 Message 42 Sat Jan 23, 1993 C128.JBEE (Forwarded) Everything but the ARC code :) ------------ Category 5, Topic 12 Message 43 Sat Jan 23, 1993 F.CANANZI (Forwarded) JBEE.... What can i say .. You left me speachless. However on second thought .. i have one thing to suggest. Don't get me wrong .. perhaps i gave you the impression that i 'WANTED' the source to Bell Term , JB .. that's not what i was after... sorry to have missled you. What i was looking for was source for the 128 . Specifically the 1k protocol. I have the code that Ep wrote , but it too is dependent on the main ML source. I just thought that we had or should of had something like that in the library. As for my suggestion. Why not create a 'class' on how to build a term in ML , i found out the hard way that compiled basic just doesn't cut it at 2400 baud. I managed to put together a vanila ascii term . After giving it a clock with an alarm function and a menue line .. my next step is the implementing of 1k protocol. It can do this .. slowly but it can be done.. I was just looking for a simple way out. ;] ------------ Category 5, Topic 12 Message 44 Sun Jan 24, 1993 R.KNOP1 [Rob Knop] (Forwarded) The concept of a term programming class is an interesting one... I wonder if the "E. G. Bell RTC class in Machine Language Programming" would go over well? -Rob ------------ Category 5, Topic 12 Message 45 Sun Jan 24, 1993 CBM-ED [e.g.bell] (Forwarded) I would be willing to do something like that on as regular a basis as I could, but I think there are some closet ML programmers around here too that could be just as involved, including JBEE. Don't you program in ML Rob? I'm sure it is a need, because I learned totally on my own and know what I wished for all along. But how would such a thing be done? What kind of format? I have tried stuff like this in our local user group and it is popular for awhile, but hard to plan for. ------------ Category 5, Topic 12 Message 46 Sun Jan 24, 1993 F.CANANZI (Forwarded) If i may... Ed.. Why not create moduals .... something like .... 1 ... How to build a dumb terminal explanation of Kernal calls required sample source 2 ... Installing translation tables explaining how to access tables in ml sample source 3 ... Menue line explains how to window in ml