I love these tutorials. I am re-watching this again. This is the best tutorial for people who have trouble keeping these concepts in the head (maybe one day I will stop forgetting this).
Watched this and didnt totally understand, so I went and studied more math and kept programming with max, now I'm watching it again and it's starting to fall together!
16:35 and that's the reason why Reaktor was so much ahead of the game , it was always capable of doing 1 unit sample feedback loops (even before core ) , and the introduction of the core environment in 2005 ( that's 20 yeas ago ) really opened up lo level dsp . Max-mps wasn't capable of anything like that untill gen ~
Ok so I guess I'm 5 years late but I have a sort of simple question. If I have more A's and B's do I still want them to all sum to 1? And is it like all A's added together = 1 and all B's = 1 or is it, all A's + all B's = 1? Or do we just think of that as the sort of output gain? Like if they sum to 2 for example then the output is twice as loud? Also (sorry I have a lot of questions, I guess this is my most pressing question) is there a way to make our own version of filtergraph without the GUI? So basically something that takes in basic filter specs like cutoff and resonance and spits out the 5 coefficients? I know this is sort of useless since filtergraph does a great job but just as an exercise?
I feel like i followed everything exactly as you have done yet when i change the values in my number boxes to 0.5 both red and blue lines are exact and have no slope. I set my input to 1.66 and output to 1.12 to see any sort of filter slope. Great video though!
I'm probably gonna try these on Reaktor for now... Because it'll probably take me at least to the end of this year to get Max, but even then I might as well just continue on using Reaktor for prototyping DSP algorithms before porting to max anyway. (Well, I can just use PD I guess, but that thing is kinda awkward for not being very good at loading samples quickly.)
The pass~ object is in a class of objects used for managing the computational load of audio processing. Some of these techniques and their associated objects are not only deprecated but no longer supported. In particular, the begin~ object no longer does anything at all. That said, muting entire patchers (e.g. with the pcontrol object) still works fine and I will take a look at clarifying the documentation. A further, but subtle, benefit of using the pass~ object here before an outlet is that the connections will all be summed (added together) only once and prior to the outlet. Without some sort of object to do this, multiple connections outside the object would result in multiple summing operations, which uses more cpu. To adapt for poly~, the summing happens at the out~ object and there is no need to manually zero the output. So in this case removing the pass~ object is fine.
![Max 7 Tutorials: 8.3 Effects: Resonant filters ([reson~] and [fffb~]](/img/n.gif)
I love these tutorials. I am re-watching this again. This is the best tutorial for people who have trouble keeping these concepts in the head (maybe one day I will stop forgetting this).
Watched this and didnt totally understand, so I went and studied more math and kept programming with max, now I'm watching it again and it's starting to fall together!
16:35 and that's the reason why Reaktor was so much ahead of the game , it was always capable of doing 1 unit sample feedback loops (even before core ) , and the introduction of the core environment in 2005 ( that's 20 yeas ago ) really opened up lo level dsp .
Max-mps wasn't capable of anything like that untill gen ~
Thanks so much for this, I think there's weeks or coming back to this as I slowly understand more. Please keep these up! Happy Patching.
I don't know what to say. Really great explanation. I hope you will show us more gradually tutorials. Than you.
Ok so I guess I'm 5 years late but I have a sort of simple question. If I have more A's and B's do I still want them to all sum to 1? And is it like all A's added together = 1 and all B's = 1 or is it, all A's + all B's = 1? Or do we just think of that as the sort of output gain? Like if they sum to 2 for example then the output is twice as loud? Also (sorry I have a lot of questions, I guess this is my most pressing question) is there a way to make our own version of filtergraph without the GUI? So basically something that takes in basic filter specs like cutoff and resonance and spits out the 5 coefficients? I know this is sort of useless since filtergraph does a great job but just as an exercise?
Thanks! Very useful implementation indeed.
I feel like i followed everything exactly as you have done yet when i change the values in my number boxes to 0.5 both red and blue lines are exact and have no slope. I set my input to 1.66 and output to 1.12 to see any sort of filter slope.
Great video though!
Me too.
I'm probably gonna try these on Reaktor for now... Because it'll probably take me at least to the end of this year to get Max, but even then I might as well just continue on using Reaktor for prototyping DSP algorithms before porting to max anyway. (Well, I can just use PD I guess, but that thing is kinda awkward for not being very good at loading samples quickly.)
why switch to max if you already know Reaktor?
wow , amazing explanation!!!!!!
Docs say pass~ is deprecated: docs.cycling74.com/max7/maxobject/pass~ how would this patch change using poly~ as suggested there?
The pass~ object is in a class of objects used for managing the computational load of audio processing. Some of these techniques and their associated objects are not only deprecated but no longer supported. In particular, the begin~ object no longer does anything at all. That said, muting entire patchers (e.g. with the pcontrol object) still works fine and I will take a look at clarifying the documentation.
A further, but subtle, benefit of using the pass~ object here before an outlet is that the connections will all be summed (added together) only once and prior to the outlet. Without some sort of object to do this, multiple connections outside the object would result in multiple summing operations, which uses more cpu.
To adapt for poly~, the summing happens at the out~ object and there is no need to manually zero the output. So in this case removing the pass~ object is fine.
Brilliant. Thank you.
🙏🏻🤩
best