Hi Katja,
I'm looking with great interest at your [helmholtz~] pitch tracking object. I'm not asking to be lazy (I'm going to try it out for myself!), but I'm wondering if you have any general impressions of its performance as to how it compares with [sigmund~]. I'm particularly interested as to how it will do for tracking a fretless electric bass.
It looks like an excellent piece of work, and I've enjoyed reading your detailed page about it.
Phil Stone
hi phil,
what are you trying to do? do you need midi from your electric bass? or just a way to make a synth in pd? i ask because i built a gr-300 emulation for bass that works very well and with almost no latency. you can drive any oscillator within pd from that. it's all signalpath though. ideally you would use such a thing with a hexaphonic (or quadraphonic) pickup, because it's monophonic. but the same is true for helmholtz i guess.
cheers,
simon
On Feb 14, 2013, at 1:24 AM, Phil Stone pkstone@ucdavis.edu wrote:
Hi Simon,
I've been using [sigmund~] with pretty good results, tracking the bass and using it to drive various things in a complex Pd setup. I'm always interested in alternative pitch trackers, though. I play a Steinberger XL, so I won't likely be carving it up to put in a hex pickup; that's kept me a way from Roland's approach (that plus the cost!).
I'm still quite intrigued by what you've done; do you have any documentation about it?
Thanks for writing,
Phil
On 2/14/13 12:48 AM, Simon Iten wrote:
I'm not home at the moment but I will send you the snippet I used for pitch tracking when I get home.
Have a nice day On Feb 14, 2013 5:13 PM, "Phil Stone" pkstone@ucdavis.edu wrote:
here is an old version. the interesting part is in the synth subpatch on the lower left side. it's messy but you get the idea. it's for a five string bass tuned e a d g c. the new version is on another laptop, but the frequency conversion is the same. it's a straight copy from the gr-300 principle.
cheers, simon
On May 25, 2013, at 10:32 PM, Alexandre Torres Porres porres@gmail.com wrote:
but don't use it! the osc2 part is really cpu hungry in this form. i did a rewrite which is much easier and works better. however as said before the frequency-extraction is the same. On May 25, 2013, at 11:15 PM, Simon Iten itensimon@gmail.com wrote:
dont know if I got how it works, but anyway, can you share the less hungry one? thanks
2013/5/25 Simon Iten itensimon@gmail.com
[helmholtz~] is probably more accurate and will show you all of the pitch fluctuations live. [sigmund~] with the argument "notes" is probably preferable, if you only want one simple value per note.
On Thu, Feb 14, 2013 at 1:24 AM, Phil Stone pkstone@ucdavis.edu wrote: