@raynovich Maybe not overloaded but difficult to know. Try relaxing the soundcard Delay (mSecs) ..... maybe 80msecs just to see if that makes any difference.
I have much the same computer and have had nearly 2000 [osc~] running although I use Bootcamp for windows.
Make sure the final output stays below +-1.
Try not to have any visual feedback on screen..... Gui's (even bangs) or arrays updating.
What load is the Mac reporting for the processor, and what size is Pd showing in memory?
Try disabling the Wi-Fi.
I don't know whether OSX gives priority to the audio thread, but interrupts from a Wi-Fi card, USB drives etc. can cause dropouts, and Intel macs connect even to the trackpad, keyboard etc, through USB.
David.

@anna Welcome to the forum...!
Like this......... sync.zip
The Pd Extended version works well..... the Pd Vanilla version I cannot test as I don't have Gem working on my machine. But the only change I have made for Vanilla is to use the right outlet of [soundfiler] to get the sample rate of the audio file. In Pd Extended we used [soundfile_info]
If you need an explanation please ask.
David.

@jakobschauer Here is m-pd.h from pd 0.54.1 .... m_pd.h .... I haven't compared it to older versions so if you are trying to build for another version of Pd please ask.

Here are all the files from the source folder of that version...... Pd source 0.54.1.zip
I hope that helps but I have no idea how to compile correctly. They are from a windows version but I think .h files are the same.
If not then download Pd without installing it and go to the /src folder to find them.
David.

@porres In windows Pd can address 4 devices as you say.
Using ASIO4All more soundcards can be addressed and specific inputs and outputs removed from the final list that is passed to Pd as a single ASIO4All device..
David.

@gcerberv This should be useful....... it used Pd Extended (it was made in 2012) so there are some libraries you need to add to Pd vanilla )listed in the help file).
ulltrasonic sensors Pd.zip
It was posted here......... http://codelab.fr/3842 ...... and is all in French.
But, designed specifically to deal with ultrasonic sensors, smoothing and scaling data, it should be useful to you.
David.

@starpea223 I don't know of course what you are trying to achieve, but here is the full documentation of the Pd file format..... https://puredata.info/docs/developer/PdFileFormat
I hope that helps, but the best way to understand it is to open an existing patch in a simple text editor and read that alongside the documentation.
Not easy though.... using the GUI is much simpler.

I see from your other thread that you had a problem with the Gem window. Have you fixed that?
David.

@EEight The REW wizard is great software..... I would recommend that especially as it is free and powerful with great forum support
Phase inversion of treble frequencies can make you feel dizzy as you move your head...... we are much more sensitive to phase than other audio information.
Just one treble speaker in a stack is instantly horrible...... not so noticeable if the speakers are further apart though.
If you had more than one sub you would need much more tuning...... moving subs around until it sounds good everywhere is an everyday challenge on tour....... a few inches..... or less... can make a big difference.
David.

@EEight My mixer was for in-ear live on stage, the musicians using TouchOSC to control their earphones. So everything was mono. You are welcome to try it of course, and change it as you need. But although there was input and output eq there were no effects. It might be useful though for you to have Wi-Fi control while you are setting up.

I think you will find what you need (mostly) in Automatism..... all the eq and effect modules..... but I haven't seen an automatic room flattener. It is easy enough by ear though using a sweep generator.... sweep-explained.zip or easier still by humming, singing, whistling and "tssss'ing" with a microphone. You will hear the room resonance sing out at the frequencies you need to cut, and it's better than using pink noise as a source. You get the inaccuracies of the speakers at the same time.

You will have to spend some time setting up though. The real problem eq'ing a room is the comb filtering at lower frequencies.... much worse than the room or the speaker response because you can hear a note loud and then take one step and you can't hear it at all. The more speakers you have the worse it will be, and speaker placement is the most important factor. Really you need a room with very little reflective surface and the public should be listening to only one speaker wherever they are. It's ok'ish in a home cinema because you sit in one place and delays can be set to adjust the relative timing between speakers, but you will have noticed that public cinemas have soft furniture and each sound comes pretty much from one speaker as delays don't work for an audience sitting everywhere in the room.

It's probably worth at least slowing down the sweep, or having manual control, as it takes time for the room to get humming sometimes.
The vocal method works really well..... like when you are humming in the shower..... you really hear those room resonances coming back strong....
David.

@jakobschauer Maybe you should PDINCLUDEDIR=../pure-data/src ..... without the forward slash at the end, as it is looking for ../pure-data/src//m_pd.h ..... which presumably doesn't work with the double slash.

I am pretty certain that m_pd.h is in that directory.
David.

@polle Have a look at the doc folder in your Pd folder.
doc/3/audio.examples/G05.execution.order.pd ...... has information about how to use delays.
Also there was a very recent thread about the feedback problem....... https://forum.pdpatchrepo.info/topic/14471/dsp-loop-with-buffer-abstraction

And more...... because it can get complicated if using interpolation..... and the [send~] [receive~] 64 sample fixed block limit is gone in the latest Pd........ https://forum.pdpatchrepo.info/topic/14549/feedback
David.