I'm after a bit of advice on selecting some parts for a 'hobbyist' project which may eventually turn into a commercial product...
I'm developing a board with an FPGA which will produce RGB digital output ultimately for display on RGB (15kHz), VGA and PAL/NTSC composite video. Not necessarily simultaneously - worst case is a different FPGA image for the former two options.
I've some experience with generating monochrome CVBS and VGA output from an FPGA, so I believe I have a handle on the technical aspects. But I'd like to offload much of the work (especially the PAL) to COTS parts, and this is where I don't have much experience.
First of all, is there a single encoder chip that is capable of outputing *both* RGB (15kHz and VGA) and PAL/NTSC (again, not necessarily at the same time)? For example, an RGB->PAL encoder with a 'pass-thru' mode (which basically acts like a video DAC).
Failing that, I'm looking at the Chrontel CH7003 and Averlogic AL100/128 PAL encoders to do half the job. I'm hoping that they are capable of running in slave mode (they accept the pixel clock) in some funky resolutions approximating but not equal to 'standard' VGA (horizontal) resolutions?
I'd complement that with a simple triple video DAC (like the ADV7125 used on the Xilinx dev boards) for RGB (15kHz) and VGA outputs.
In either case/mode, I'd be doing the video timing (except for PAL output) in the FPGA and so supplying the pixel clock and syncs etc to the devices (they're slaves). Different FPGA images might run at different resolutions, some may target 15kHz monitors whilst others may target VGA, etc, hence the 'flexible' requirements.
As far as my approach to the problem is concerned, am I on the right track here? Did I choose the best/cheapest/easiest-to-get parts? Or is there a better/simpler/cheaper way to do it?
Keep in mind that it is only a spare-time project atm and although I'm willing to make some investment, obviously I'll only be ordering and manufacturing in very small quantities at this point. At best you could consider it a 'niche' product so even if I did eventually realise a commercial product, it would still be low volume.
TIA for any help/hints/suggestions!
Regards, Mark