I'm really enjoying this series. As a sound designer/engineer without a strong math background it's been difficult to find accessible content about the fundamentals of DSP. This is really helping to demystify the subject. Thank you!
Your videos are actually making be better understand the Book Designing Audio Effects plugin by Will P. You have no idea for a niche audience you are a life saver god. That being said I can't imagine you taking your time to craft these amazing videos for which you definitely need more recognition, but don't be disheartened akash... The way you explain stuff (teaching) it feels like you love to do because it reflects by the delivery of these concepts. I will definitely donate when I am capable of.. that's the thing about niche audience it's loyal
Thank you for the feedback! I do like making them, and as you say, it's a niche audience, so I don't expect the viewership to grow exponentially. But it's good to see that it's proving itself useful for people who do watch it.
@@akashmurthy I am only a musician with passion for C++ programming too ( Juce ) and I found your video so useful and beautiful. Thank you very much !!
Correction needs to be done for explaining Nyquist and 1/2Nyquist, they don't fit as per the sampling thing taught, why Nyquist signal x[0] = 1 and not 0?
Brilliant series of videos. A couple of questions though... I can't find the video number one of the Digital Filter Basics series, is there any? Is the filter mentioned in this video "Feedforward Filter" any similar to Simple Moving Average? Regards.
Thanks very much! Video number one is just the introduction video where I'm on camera. A moving average filter is a particular type of feedforward filter of order n, where every coefficient has the same value of 1/n I was going to do a video on simple implementations of these filters, including moving average, leaky integrator, etc.
@akash murthy. can you please kindly explain why the phase response for Nyquist pattern is 90 degree? From my understanding, after going through the filters, everything gets cancelled out ... Thank you
Good question. If you look at 10:30 I talk about how a 1 sample delay causes 90deg phase shift in the 1/2 Nyquist input signal, but the output signal as a result of the combination of original input and 90deg phase shifted input is 45deg phase shifted. It's the same case here. The phase shift represented in the graph is the phase shift of the "output signal" of the filter, not the phase shift introduced just by the delay element.
It really nice explanation, can I ask the symmetric coefficients you mentions in phase response: what is "symmetric" means in this case? I guess it means abs(a0) == abs(a1), right? And how "symmetric" extents to more than 2 coefficients (n coefficients)?
That's what it means, yes. The same formula applies to any number of coefficients. I know for certainty that when coefficients are equal, the phase is linear, like in a moving average filter for example. I'm not a hundred percent sure about how sign of the coefficients affects linearity for anything more than 2 coefficients. But I don't think it does. I'm not in front of my computer for many days to check, unfortunately.
~10k views?! This series deserves so many more.
I love the explanations. well done!
This is the finest digital filter explanation video i ever seen. So helpful and thank you!
Oh you're too kind, thanks a lot!
ridiculously clear. thanks for your service man
Outstanding DSP tutorials, please keep making these and audio programming ones too!
Thanks so much! I'll continue this series soon hopefully!
I'm really enjoying this series. As a sound designer/engineer without a strong math background it's been difficult to find accessible content about the fundamentals of DSP. This is really helping to demystify the subject. Thank you!
I'm glad you're enjoying the series!
Wonderful explanation of digital filters. Videos can be watched many times
Thanks great to know, thanks !
I've been trying to understand this for weeks and I finally get it, thank you! Such a clear and thorough explanation.
I'm glad you found the channel! :)
Outstanding. This is how anyone should be tought this topic
This series is amazing, the animations are top notch. Thank you for making them.
Thank you for including your patreon, this is incredible incredible stuff
Thanks very much! :)
This is masterfully explained. Thank you very much!
Thank you! :)
Old brain grind slowly. Thanks Akash.
It's a slow burn alright!
Finest lecture on digital audio.... ...thank you
Thanks so much!
This is my new favorite channel.
:)
Had I known you had a Patreon, I would have signed up sooner! Thanks for all the hard work on all your content.
You're welcome mate, and thanks a lot for your support on Patreon!
Your videos are actually making be better understand the Book Designing Audio Effects plugin by Will P. You have no idea for a niche audience you are a life saver god. That being said I can't imagine you taking your time to craft these amazing videos for which you definitely need more recognition, but don't be disheartened akash... The way you explain stuff (teaching) it feels like you love to do because it reflects by the delivery of these concepts. I will definitely donate when I am capable of.. that's the thing about niche audience it's loyal
Thank you for the feedback! I do like making them, and as you say, it's a niche audience, so I don't expect the viewership to grow exponentially. But it's good to see that it's proving itself useful for people who do watch it.
Excellent tuto ! Animations are great and help a lot to understand 👍
Thanks! Glad you think so..
In the animation circa 3:30, the bottom triangle should show a_1, not a_0
Good spot!
super teaching anna 👏
Great Video thank you.
Just fantastic!!!
I wish your my DSP teacher in my college.Anyways Thanks for sharing your knowledge on youtube.
I'm glad you find these videos useful!
So how would you do a 2-Pole (stage) FIR LP filter?
Best Lesson !!¨Great Video !!
Thank you!
@@akashmurthy I am only a musician with passion for C++ programming too ( Juce ) and I found your video so useful and beautiful. Thank you very much !!
Correction needs to be done for explaining Nyquist and 1/2Nyquist, they don't fit as per the sampling thing taught, why Nyquist signal x[0] = 1 and not 0?
you are so awesome!!!
Brilliant series of videos. A couple of questions though...
I can't find the video number one of the Digital Filter Basics series, is there any?
Is the filter mentioned in this video "Feedforward Filter" any similar to Simple Moving Average?
Regards.
Thanks very much!
Video number one is just the introduction video where I'm on camera.
A moving average filter is a particular type of feedforward filter of order n, where every coefficient has the same value of 1/n
I was going to do a video on simple implementations of these filters, including moving average, leaky integrator, etc.
@akash murthy. can you please kindly explain why the phase response for Nyquist pattern is 90 degree? From my understanding, after going through the filters, everything gets cancelled out ... Thank you
Good question. If you look at 10:30 I talk about how a 1 sample delay causes 90deg phase shift in the 1/2 Nyquist input signal, but the output signal as a result of the combination of original input and 90deg phase shifted input is 45deg phase shifted.
It's the same case here. The phase shift represented in the graph is the phase shift of the "output signal" of the filter, not the phase shift introduced just by the delay element.
It really nice explanation, can I ask the symmetric coefficients you mentions in phase response: what is "symmetric" means in this case? I guess it means abs(a0) == abs(a1), right? And how "symmetric" extents to more than 2 coefficients (n coefficients)?
That's what it means, yes. The same formula applies to any number of coefficients. I know for certainty that when coefficients are equal, the phase is linear, like in a moving average filter for example. I'm not a hundred percent sure about how sign of the coefficients affects linearity for anything more than 2 coefficients. But I don't think it does. I'm not in front of my computer for many days to check, unfortunately.
thanks for your info.
🤗👌🏻