On Fri, 31 Dec 1999 Picsend@AOL.COM wrote: > For an embedded application, I have to digitize signal from a color CCD module > and to multiplex it with 4 conditioners signals (binary) using a Pic. > Informations from this mobile system will be transmited to PC using HF > transmitter with 9600 only capability. > In first thinking I have to acquire image, store it and transmit each line > with > a start and stop "bit" and a synchronize bit for next page. Is it wrong ? Frankly, you would be far better off transmitting the video directly to the PC using any one of a huge number of available RF transmitters, and using a cheap video digitizing card in the PC. > For a standard CCD module (300K pixels with 400 lines) can someone tell me in > PRACTICAL, with a 20MHz pic frequency, how much image I can really transfer > par > minutes ? Depends on how much of the image you can throw away. If you really want to send a full COLOR RS170A image the limiting factor is your ridiculously low comm rate. Simple math: 9600 baud/10 bit/character = 960 cps. One character per pixel gives 300K/960=313 seconds/image not counting forward error correction, etc. Then multiply by 3 because you have 3 video components to deal with (Y,B-y,R-y) > Wich component can I use for this project to make it very simple ? Can I find See above. Use a different method. > a > lsi wich can handle all video acquisition and storing part ? Yes, there are many PC video digitizing cards that you could use. There are also number application specific cards out there (mostly from security suppliers) that can give you a serial output of a video signal. I stongly suggest that you don't try to reinvent the wheel. > I think (may I wrong ?) that any PIC ADC are not suitable for this > application. None will intrinsically do video. However, you can use an external sample and hold to allow you to use the low rate A/D's on board the PICs if you must. Essentially you digitize the video from top to bottom, with a few samples per line, and moving the start point linearly across the video scan line. You must have an identical image for many frames for this method to work well. Look up SSTV (slow scan TV) to get a better idea of what I'm talking about. Good luck. Robert --Robert.Rolf@UAlberta.ca "If 'debugging' is the process of removing errors, then 'programming' must be the process of putting them in".