If you want to feed your signal back to the same object, you must use a [s~]/[r~] pair, otherwise Pd will correctly complain about a DSP loop. Also, I don't see any real reason for using [fexpr~], as you only reference the current sample of the input with x1[0]. If you did something like x1[-1] to reference the last sample. it would make sense, but the way you wrote this, [expr~] with $v1 should suffice.

If I'm not mistaken, iemguts is a collection of objects and doesn't come as a signle binary, which means that you only need to set its path to Pd's search paths. If you have it in your externals directory, which in turn is in Pd's search paths, then you can load an object with the library name prepended, like this: [iemguts/oreceive].

The image you posted is a bit too small to really figure out what's going on. Both for the clickiness and reading backwards, you can check this very old but still valid tutorial http://pd-tutorial.com/english/ch03s04.html
Actually, I would suggest going through the entire tutorial and not only this section. It covers a lot of ground.

You can find information on abstractions and the dollar arguments here http://pd-tutorial.com/english/ch05.html. Go to 5.1.1.2 which is focused on abstractions.

You'll might find the cyclone library useful. It clones a bunch of Max objects in Pd. Go to Tools->Find externals and type "cyclone" to get it.

The heavy compiler will behave the same way, whatever hardware you use. Using a Pi though, you can use Pd as is, without heavy. That being said, you can use any objects you like, whether native or externals.

PureVST-0.8 is out. This is Miller's message on the mailing list:

PureVST version 0.8 is out on the usual site:

msp.ucsd.edu/tools/purevst/

This might fix some of the bugs in previous versions. It's available as source, or compiled for "M" series processors in MACOS and for fedora linux (Intel processors), or you can compile your own binaries from the source tarball.

"pdnam" returns no results in deken (I'm not hidding foreign architectures, so it's not only the Linux binaries that don't show up).

gain is just a multiplication of a signal with some constant value in digital audio.