gnupic: Re: CUMP and programmers for Linux


Previous by date: 14 Nov 2001 21:26:04 -0000 Re: SDCC unsigned char arithmetic test, Scott Dattalo
Next by date: 14 Nov 2001 21:26:04 -0000 Re: CUMP a very good idea, lionel guidez
Previous in thread:
Next in thread: 14 Nov 2001 21:26:04 -0000 Re: CUMP and programmers for Linux, lionel guidez

Subject: Re: CUMP and programmers for Linux
From: Carlos Nieves Onega ####@####.####
Date: 14 Nov 2001 21:26:04 -0000
Message-Id: <1005773501.514.57.camel@otilio>

Hi all,
Well, I think CUMP is a nice project. It started with the wishlist of
the community, so it has all people expects from something like that.
I'm doing something simpler, but a bit more flexible, and which is not
by its own an stand-alone programmer, but it could be, adding the right
hardware to control it.
I think, like Alessandro, that you have to approach the design of such a
thing with layers. I'm doing the bottom layer of a programmer, and if
you want a stand alone programmer, it could be the next layer, but the
bottom should be designed to be used both with a PC and with stand-alone
hardware.

I think I have solved the low-level hardware of a programmer. It allows
you to control each pin of a 48 pin (or less, if you want) socket
independently, with programmable sources, from 1.4 to 25V, and read the
voltage at any pin you want. It is controlled with a serial port, with
text commands, so it can be controlled by any terminal (idea from the
CUMP and other programmers). It doesn't need an adapter for every device
family you want to program, well, at least, if it has a dual in line
package... and an adapter for any other device can be easily done, since
you can control any pin of the socket.
With that, I think you will be able to program anything, even new
low-voltage devices, using 3.3V or even less, and of course, use it for
any other thing, like a function generator, or an analyzer, but I'm not
sure what speed you could expect for that.
I have finished the design, and I have to start testing it, and program
the firmware, so I have nothing working yet.

Get it working will take some time, but keep in mind that if you can
keep it simple and modular, like Alessandro says, you could build an
extremely good programmer. This is the low-level hardware, and could be
used by whataever you want.

The software to control all programmers should be build also in simple
layers, so if any change you want to do could be easily done. If not,
your software will require some rewriting in some time.
All of you are thinking of porting some existing programmer you know
well, like Ic-prog, because that is what you want to get. Well, that's
right, but I will take a look at some other ways to do that, like the
one proposed by James Newton, the Engine (the webpage can be found on
www.piclist.org ). It seems a good way to have an independent layer to
program any devices, and which could be easily programmed for each new
device. This could be used by the layer regarding the programming
protocols.

I don't know if all these mails regarding the programmer and the
software could be annoying other people in the list (not me). James have
already suggested to open a new one, or use the CUMP mailing list, so
maybe it's the time to reopen it...

Regards,

Carlos

P.D. If anyone knows of a device requiring more than a supply voltage
and a programming voltage to program it, please, let me know.


______________________________________________________________________________
mensaje enviado desde http://www.iespana.es
emails (pop)-paginas web (espacio ilimitado)-agenda-favoritos (bookmarks)-foros -Chat


Previous by date: 14 Nov 2001 21:26:04 -0000 Re: SDCC unsigned char arithmetic test, Scott Dattalo
Next by date: 14 Nov 2001 21:26:04 -0000 Re: CUMP a very good idea, lionel guidez
Previous in thread:
Next in thread: 14 Nov 2001 21:26:04 -0000 Re: CUMP and programmers for Linux, lionel guidez


Powered by ezmlm-browse 0.20.