Post by thread:Promaster

Having built my own programmer for all chips (including pics) I decided to use the Promaster when I finally had to program several thousands in "production" quality. I was quite disappointed for the reasons already mentioned: No quick way to program a single chip in a more or less automatic way. Furthermore, I'm used to Unix, both at work and at home (Linux). So this is what I did: 1. Trace all accesses to the UART 2. Figure out (parts of) the protocol 3. Write a C-Program that does what I need I don't have to leave Unix now for a complete developement cycle and can even personalize each chip in a much more flexible way than Promaster provides for. Note that I didn't reverse engineer any of the code, which is explicitely prohibited by the license agreement. Although I'm still not completely happy (the promaster's programming cycle is too long in my opinion), I can live with this situation. (Basically, now the hardware is the problem. I use the SMD-package and the contact gold (sp?) of the fingers seems to wear off. I'm at about 3000 pieces now, which is certainly not too much for a "production" quality programmer.) If someone is interested in this piece of software I can try to upload it to some ftp-server. Obviously, NO WARRANTY. I would be even more delighted if someone took the time to enhance the tool, because of the following deficiencies/limitations: - does not generate proper checksum (the protocol has one) Current solution: 1. send data (will produce NOT OK as acknowlegdement) 2. read data back 3. compare sent/read data 4. start programming if match - does not support anything else than C71 - does not allow setting of programming parameters Regards, Mike -- Dipl.-Ing. Michael Hermann spam_OUTmchTakeThisOuTregent.e-technik.tu-muenchen.de Lehrstuhl fuer Rechnergestuetztes Entwerfen, Postfach 202420 Technische Universitaet Muenchen 089/21053657

Having built my own programmer for all chips (including pics) I decided to use the Promaster when I finally had to program several thousands in "production" quality. I was quite disappointed for the reasons already mentioned in this list, but mainly by: No quick way to program a single chip in a more or less automatic way. The GUI takes forever to just connect and I was unable to figure out how to run a session with a batch-file. Furthermore, I'm used to Unix, both at work and at home (Linux). So this is what I did: 1. Trace all accesses to the UART 2. Figure out (parts of) the protocol 3. Write a C-Program that does what I need I don't have to leave Unix now for a complete developement cycle and can even personalize each chip in a much more flexible way than Promaster provides for. Note that I didn't reverse engineer any of the code, which is explicitely prohibited by the license agreement. Although I'm still not completely happy (the promaster's programming cycle is too long in my opinion), I can live with this situation. (Basically, now the hardware is the problem. I use the SMD-package and the contact gold (sp?) of the fingers seems to wear off. I'm at about 3000 pieces now, which is certainly not too much for a "production" quality programmer.) If someone is interested in this piece of software I can try to upload it to some ftp-server. Obviously, NO WARRANTY. I would be even more delighted if someone took the time to enhance the tool, because of the following deficiencies/limitations: - does not generate proper checksum (the protocol has one) Current solution: 1. send data (will produce NOT OK as acknowlegdement) 2. read data back 3. compare sent/read data 4. start programming if match - does not support anything else than C71 - does not allow setting of programming parameters Regards, Mike -- Dipl.-Ing. Michael Hermann .....mchKILLspam@spam@regent.e-technik.tu-muenchen.de Lehrstuhl fuer Rechnergestuetztes Entwerfen, Postfach 202420 Technische Universitaet Muenchen 089/21053657

Michael Hermann <Michael.HermannKILLspamde.tu-muenchen.e-technik.regent> wrote: ..... {Quote hidden}> If someone is interested in this piece of software I can try to upload > it to some ftp-server. Obviously, NO WARRANTY. > > I would be even more delighted if someone took the time to enhance the > tool, because of the following deficiencies/limitations: > > - does not generate proper checksum (the protocol has one) > Current solution: > 1. send data (will produce NOT OK as acknowlegdement) > 2. read data back > 3. compare sent/read data > 4. start programming if match > > - does not support anything else than C71 > > - does not allow setting of programming parameters > > > Regards, > > Mike > > -- > Dipl.-Ing. Michael Hermann .....mchKILLspam.....regent.e-technik.tu-muenchen.de > Lehrstuhl fuer Rechnergestuetztes Entwerfen, Postfach 202420 > Technische Universitaet Muenchen 089/21053657 > I would be interested, as I would like to do the same thing for my picstart16b firmware 1.5 (are they same/similar?) OR Does anyone know a microchip e-mail address where I could ask for the protocols? BTW either sics.se or nic.funet.fi would be best for an ftp site. =%-) Ian