Mark Knecht wrote: >On Thu, 18 Nov 2004 18:19:45 -0800, Brad Fuller <brad@xxxxxxxxxxxx> wrote: > > >> >>Mark Knecht wrote: >> >> > > > >>>Instead of a $400 2 channel PCI card we might end up with a $600 >>>16-in/16-out device with hardware signal processing on board. To me >>>this is probably a better place to go. If we do all this work ten we >>>want to start working towards an architecture that will last. >>> >>> >>> >>> >>Taking ladspa and mapping it to FPGA: how? and how would you do this >>efficiently, if you could do it? A C function to VHDL function >>convertor? (it's been a long time since I've worked with FPGAs. I'm sure >>there are advances) >>It might be more cost effective to use DSPs -- that is: more cost >>effective in the long run for everybody -- mostly the end user. >> >>brad >> >> >> >> >Sure - that's a fair comment and a design decision once some project >like this gets started. I just brainstorming. However, even with an >onboard DSP, which is most likely what Pro Tools does, we'd still need >to map from LADSPA C code to DSP code. Is that easy? > I would think it's easier than mapping gates. I have not ever looked at LADSPA code. I assume most people write in C. Today's DSPs, even 10 years ago, have a full compliment of C programming tools. Bingo. Of course, the goal would be to strike a proper balance with making it easy and cheap to design and mfr with making it easy and inexpensive for the end user. From what I've read, that is your profession! I have always said that engineering is an exercise in compromise ;-) It would be a cool and fun project! Have you seen plugzilla or receptor? http://www.plugzilla.com/overview http://www.museresearch.com/receptor_overview.php Kinda what you're talking about? I think they created a VST wrapper to run in Linux. I would bet the creators of one or both of these products visit this site. brad
