On 9/12/06, Lourens Veen <[EMAIL PROTECTED]> wrote:

Timothy, did you read my post from Saturday? I proposed something very
similar, but using the VGA nanocontroller to do the YUV->RGB
conversion. It's there anyway, already contains all the addressing
logic, and there's a remarkable conceptual similarity between
converting VGA text surfaces into RGB and converting YUV surfaces into
RGB.

It was a rather long post and there were no replies, but I still think
it's an interesting idea and I would like some opinions from you all...


I may just have not gotten to it yet.  I've been burried in studying
for the qualifier.

The problem with using the nanocontroller is that it'll be way too
slow.  It would take hundreds of instructions and therefore hundreds
of cycles to perform one conversion.  If we're going to do it at all,
it needs to have dedicated hardware, and it seems that we need to
embed it in the drawing engine in order to deal with the various
formats.

I expect I may get the nanocontroller to run at 100MHz in the FPGA.
Don't expect it to average more than about half that rate in useful
instruction throughput.  We'll hand-code all of the assembly, so we
may do better than that, but it's still a hard problem.

The nanocontroller will be most useful for things that it can control
indirectly (initiate N DMA cycles of transfer) or that are not
performance-critical (convert VGA).
_______________________________________________
Open-graphics mailing list
[email protected]
http://lists.duskglow.com/mailman/listinfo/open-graphics
List service provided by Duskglow Consulting, LLC (www.duskglow.com)

Reply via email to