Hello everybody:
I developed the patch to know the notes (name, octave and tuning) of an audio signal, nothing too complex, just a nice tool.
I wanted to make use of the [switch~] object to avoid extensive cpu usage, limiting the calculus to the time required by [fiddle~] object. If I'd put the [switch~] object and connect it as shown on the patch by comment lines, it will crash Pd as soon as I open the patch.

1. Would anyone give me a hint on this, please?
2. Furthermore, how can I make the GOP part less ugly? I thought about concatenating the midi note, octave and tuning as texts, I don't know how to make a string (or text) out of the three of them.

Thank you very much.

Sumidero.

PD: when this issue is solved, I will post it on patches~ section.

http://www.pdpatchrepo.info/hurleur/NoteNames-sumidero.pd.bz2

I solved the ugliness issue, but still I can't figure out how to solve the first one. Here is the latest version. Best regards to everyone.

Sumidero.

http://www.pdpatchrepo.info/hurleur/testNoteNames-Sumidero-V2.zip

It seems that fiddle~ does not like switch~. Maybe the fault is with fiddle~ and not switch~. You could try to use the sigmund~ object which can analyse frequency contents and pitch as wll. As far as I can see, sigmund~ does not have this problem with switch~.

Katja

Thank you very much, Katja! I couldn't have guessed that issue between objects. I'll check on sigmund~ soon and fix my abstraction.
By the way, your slicycle~ stuff is amazing, thank you again.

Sumidero.

PD: (To Maelstorm) Once the issue is solved, should I post it again on patch~ or abstract~ sections?

If you just want to show how you fixed it, post it here in this thread. I think it helps for people to be able to see the previous posts leading up to it to get a better understanding of the problem and what to do when they encounter something similar.

Hello everybody:
Finally I solved the issue of using fiddle~ with switch~, which caused Pd to crash. The idea of using switch was to use fiddle~ just as needed. As Katjav proposed (read the previous posts), I turned to sigmund~ instead of fiddle~. This brought a new advantage: sigmund~ can wait for a number of samples until the next calculus of frequency, so there is no need for an additional metro and switch~ control. Just using the output as a bang to the calculus worked just fine.
The abstraction is set to calculate the pitch every 0.19s, more or less, when using 44100Hz of samplerate. Higher sample rates would make this output rate decrease.
Yes, I could have used sigmund~ right away to calculate the midi pitch and then calculate just the cents and that could have made the abstraction less cpu consuming. I will leave it to others to improve it, as it took me more time than I thought to make it work (I'm quite new to Pd). Ten instances of this abstraction just connected to the phasor~ on the test patch I attach gave 3% of cpu usage (as read on htop, text based Debian utility for cpu processes monitoring). I hope it works well with smaller computers too.
I also added a datum in the packed output that gives the error percentage in the frequency calculus. I did not test it on vocals or instruments yet, just osc~ and phasor~.
Well, I hope this could be useful to someone. Thanks to everyone who contributed.

Sumidero.

http://www.pdpatchrepo.info/hurleur/testNoteNames-sumidero-V3.zip

Hi everyone, very interesting.

I would like to try your Patch . Please, would you share the .pd file .

I´m trying to convert my guitar audio signal (note) to a Vsti midi note.

Please help me

here are those 2 patches

http://www.pdpatchrepo.info/hurleur/testNoteNames-sumidero.pd

here are those 2 patches

http://www.pdpatchrepo.info/hurleur/testNoteNames-sumidero.pd

Thanks mod !! (once more).

Sumidero