OPENCORES.ORG

Mike Delaney wrote:

There is an NTSC composite sync generator in some code I wrote.  It
sounds like that's the problem you are having.  Sync is a tough issue
and it is not well documented.  Note though that my sync in the file
linked is not perfect.  It is operational, but things shake a bit.  I
don't think I have the post-equalization phase correct right now.  It
works on several tvs though.

As for generating color, that's a bit trickier, though not impossible.
 I have been considering developing a color generator, but I haven't
gotten to it yet.  For the project below, I had a standalone color
encoder.  I just had to give it the three colors (rgb in this case)
and the composite sync and it took care of the color for me.

http://www1.pacific.edu/~a-cox/SeniorProject/graphics.rar

You may use my code as you will, but I cannot support it.

-Arlen

On 7/13/05, Mark McDougall <markm@v...> wrote:
> Mike Delaney wrote:
> 
> > Has anyone seen a free core for NTSC video generation (or something
> > like a tutorial that I could at least use as a starting point)?  I
> > haven't had much luck yet, I can get a nice zigzag pattern that's
> > supposed to be verticle lines ....
> 
> There's a site where someone has done some NTSC generation in a PIC - I
> used that as a basis for my experiments with video generation at the end
> of last year.
> <http://www.rickard.gunee.com/projects/video/sx/howto.php>
> 
> > Also, it seems like most of the designs or eval boards I've found
> > online use an external NTSC encoder, typically from Phillips.  Is
> > generating color NTSC (with a parellel DAC) that hard, or is it just
> > people being lazy? :)
> 
>  From what I understand, you need quite a high clock to give you the
> resolution to generate the colour information for more than a few
> colours. It determines how 'fast' you can change the raster colour (as
> it is encoded as a phase shift IIRC) and hence limits your effective
> horizontal resolution.
> 
> You'll also need an accurate clock to generate the burst signal as
> displays tend to be quite fussy when it comes to locking onto the burst.
> Even experiments with a programmable clock didn't produce satisfactory
> results for me - I bought some PAL/NTSC crystal oscillators but am yet
> to plug them into my design...
> 
> Regards,
> 
> --
> Mark McDougall, Software Engineer
> Virtual Logic Pty Ltd, <http://www.vl.com.au>
> 21-25 King St, Rockdale, 2216
> Ph: +612-9599-3255 Fax: +612-9599-3266
> 
> _______________________________________________
> http://www.opencores.org/mailman/listinfo/cores
>