Have you tried increasing ksmps? Real time spectral processing can be a little CPU hungry. I've no idea if the PI will be able to handle anything but the most basic operations.

If you have the original V1 R-Pi, I would not expect it to support 
real-time pvs. One of the first things I did when I got mine was to 
build and run command-line pvoc, and it didn't ~quite~ have enough speed 
for real-time. I would expect R-Pi 2 to be better. I have just taken 
delivery of the new v3 which should be better still, but have yet to set 
it up.

At a pinch it might just be Ok at a lower sample rate (and the lowest 
analysis rate you can get away with).

Richard Dobson

On 28/03/2016 04:42, Emmett Palaima wrote:
> Hi, I am working on a live audio processor with the raspberry pi, and
> was hoping to use some of the pvs opcodes in my patches, as they make
> some of my favorite sounds. I am running into the issue that I can't
> seem to use them on the pi without causing buffer underrun issues in the
> output. I have tried making the patch as streamlined as possible, as
> well as using a variety of k-rates, sample rates, buffer sizes and
> window sizes. Even when I got into very low sample and k rates with very
> high buffer and window sizes I can't get it to run correctly.
>
> Is there anything else I can do to make it run more smoothly, or am I
> just running up against the limitations of the technology, since pvs is
> very cpu intensive. Here is my code, please let me know if you see
> anything.
>
..

Csound mailing list
Csound@listserv.heanet.ie
https://listserv.heanet.ie/cgi-bin/wa?A0=CSOUND
Send bugs reports to
        https://github.com/csound/csound/issues
Discussions of bugs and features can be posted here