having extremely bad time learning DSP

I think it has to do with my issues with math. I haven't done much since I was younger. calculus in high school.

I have some problems with math for some reason. it just feels like it's REALLY hard for me where other subjects are easy and don't take so long for me to understand. math seems quite time consuming to move through where I would like to be able to kind of get where I want ... efficiently.

efficiently. it feels like it isn't as productive a learning experience as it should be honestly.

>>8955446
I know a little bit about electricity not much. But, I do study a lot independently.

Think about the way you use the word "understand". I assume you mean you're gaining the knowledge to calculate and predict electronic signals mathematically on paper.

Are you spending more time reading than doing exercises.

Look at one thing that is confusing you. Diagram out the equation, its variables and there factors. Maybe try utilizing a bit of visual special thinking and finding some good reference materials. Is there a Digital Signal processing handbook. Not some crap one but a no shit reliable document that professionals use. I'll try to find one.

I know machinery's handbook has a guide to go along with it. Heres this

https://www.crcpress.com/The-Digital-Signal-Processing-Handbook-Second-Edition---3-Volume-Set/Madisetti/p/book/9781420045635

>>8955446
so what's your question?

whats your end goal?

>>8955463
>https://www.crcpress.com/The-Digital-Signal-Processing-Handbook-Second-Edition---3-Volume-Set/Madisetti/p/book/9781420045635
thanks

I'm trying to be honest about this as my ego is, frankly, big sometimes. I'm rather used to being able to pick things up quickly and not having to work too hard to understand something.

I'm an artist/musician and I also program. I'm trying to learn DSP so that I can understand sound on a technical basis, to understand how the computer processes sound, produces sound, and changes sound.

On a practical basis, I want to be able to have total mastery over my tools. And to be able to produce sounds using programming code, and understand every aspect of how that works.

I can code basic things in Max/MSP which is sort of feeling like a shortcut through actually learning how digital sample rates, etc work on the side of math.

I don't know how much of that I NEED to, or even MUST learn... I don't consider myself a math person, or engineer really. I do what I do as a musician, really. I just want to be able to say that I understand it well, and to be able to work with it independent of Max/MSP if I WANT to, which means I do kind of need a more solid grasp of it. I can't program my own stuff in C++ if I don't understand what is going on in the low level. And even in Max/MSP this kind of general understanding would be beneficial to me.

I have lots of problems. I don't understand sufficiently the low level math - I think. Like how the unit circle graph translates to the sin wav graph. With radians. Stuff that basic. It's kind of embarrassing. I feel like a fucking moron. I don't want to have to delay learning from these DSP books by having to take a whole math course. So I've tried diving in and I quickly feel like I am just not grasping material fast enough. Rereading chapters over and over, kind of losing focus even.

I'm making this thread in the hope that it will force me to adjust and do a better job.

>>8955453
If you were to write down an equation on one paper for reference.

On another paper write down all the algebraic or mathematical rules that you are going to be using in that equation.

On another paper write down all the variables and what they represent physically as well as maybe some definitions or have a physics dictionary or something.

Now go through the equation mentally and on paper why everything is occurring the way it is. The more time you spend on it and the more disciplined/organized you go about it the better you will understand.

These are just learning theories but, I do know that the only way to overcome every obstacle in life is to be more logical. When you're shit aint working you just need to identify in words why its not working.

Is my advice more annoying than helpful.

>>8955478
that's very helpful. it does seem like it's the concepts that aren't clicking.

like complex sinusoids for instance, I have absolutely NO idea how that would be relevant to the analysis or use of interpreting sound waves. it makes no sense.

e ^ j*the frequency etc...

and sadly I don't even really understand what "e" is.

that said your advice about writing down the equation, the variables, and all that information so that I can look at it more carefully.

I just don't get how the fuck the complex sinusoids or whatever would be relevant, it's small little things like that which compound and confuse, especially when there are tons of them at the same time. I think I just need to continue reviewing basic math.

>>8955477
So instead of making a java calculator that adds two plus two and outputs 4 on a screen.

You want to make a c++ program that is like guitar effects and shit.

I feel like to be halfway decent at that might take about twenty hours a week for a year or so. If you go at sixteen hours a week it'd be like an extra part time job.

If you stay disciplined and focused I think you can pull it off. It's not your ego that's not logical.

>>8955446
OK so here's this. To me this looks like a library but I just glanced at it.

http://www.matthieuamiguet.ch/blog/diy-guitar-effects-python

It looks to me that however this python library or whatever the fuck it is. I don't code. Is actually made with what you are trying to do. Like whoever made that lower level code was doing what you're trying to do. Not the python script you could probably make that.

I think that those weird formulas are actually going to be what you use to manipulate the sounds going into the program through the instruments. Or they are going to be how the sound goes into the digital equipment.

>>8955486
I haven't done math in a long time. It feels on an emotional level like I'm taking this big leap into the most difficult work I've ever done. Learning to code was difficult, but math is even more cool.

I like it. I don't like having to delay results, though. I feel like I should easily pick up a few concepts at will, if I can't even pick up the things I'm working on it feels less rewarding and fun.

I need to do the problems probably.

>>8955484
https://en.wikipedia.org/wiki/List_of_common_physics_notations

there is a better list than this somewhere but I don't remember.

I think these formulas are used to digitally interpret sound over a period of time. Like its being used to make calculations or digital interpretations to the way sounds act. Like when you play two E flats and they ring together type deal.

A computer is digital so it needs math to manage inputs and outputs. the computer can't just take the sound of a guitar and distort it with sensors and outputs without the assistance of processors.

Its not two plus two is four or bill, fred, sue, and jenny equal table five.

Now its yea I can't explain it either but I think that's whats confusing you. I'm looking at it and it seems confusing but I know it's just because I'm not analyzing it correctly. and I'm not attempting to actually do what your doing.

I actually kind of hate sitting down and having to do math problems

As soon as I see math notation I get annoyed. I WANT to enjoy doing math problems because I could so much more. It puts me off or something.

I see the math notation and I'm just like "fuckkkk" lol

I'm not horrible at it, it's fucking hard. It's harder than anything else. I feel like it shouldn't be this hard. It sounds horrible but it's just how I feel.

>>8955495
The fourier transform decomposes a function of time into the frequencies that make it up.

Look at every word that confuses you and identify the correct definition. This might be a multiple layered process because you might not understand some of the definitions themselves. But, I bet that by the fifth layer it will make sense.

The result might look like a list of definitions under an explanation or maybe more like a prime factor tree or whatever it's called. That might be a quick way of doing it.

>>8955519
>decomposes
UGHHHHHHHH
haha

I DONT UNDERSTANDH AHHASDFSA

>>8955516
I used to get frustrated when I couldn't play guitar but, I just pushed through it.

What if you just used references for doing math. You're allowed to use the book now. You odn't have to know everything off the top off your head.

That's like memorizing a dictionary instead of using one.

>>8955519
that's good advice though, I'm looking it up now and I think this should help,thanks

holy shit. OK so I think I understand. decomposition is (basically) taking a single function and breaking it into multiple functions

so the Fourier transform is taking a single sin wav, in this case, and decomposing it into a function which results in a real and complex part of that function, which can represent magnitude and phase.

I feel a click. I think that idea of looking up definitions in sentences that don't make sense is helpful as fuck. I need to do that way more.

>>8955522
that was probably not a good word for the author to use.

If you can't paraphrase something you're just memorizing it. If you can't draw a pictural diagram that displays it you probably don't understand it.

I saw the word decomposed and just assumed they meant fades away, or dissolves. that was kind of confusing in that sentence.

Not to mention when you click on function it starts talking about a bunch of weird shit. But, all that weird shit is probably literally made of a 3 number physics formula.

>>8955529
I just wish I had learnt that when I was an avid studier. Now I just do roofing. As of late I've been studying sales, project management, communication and things of the sort.

Its weird how scientific they are.

also when you don't know why you need to know something either find five theories by someone else that might explain it and come up with five theories of your own.

Then either perform an experiment or ask someone if those theories are correct. The thing is that experiment may take 80 hours or so. (you could get discouraged.

>>8955477
>
I can code basic things in Max/MSP which is sort of feeling like a shortcut through actually learning how digital sample rates, etc work on the side of math.
I actually do that stuff too.

You don't really need that much for Max/MSP unless you want to do really low level stuff. Like publish papers on new synthesis algorithms or something. The hard shit is already done for you, you just have to put it together.

I mean you only need z transforms if you're going to design filters and stuff, even then it's mostly just a matter of applying shit that other people did decades ago.

>>8955484
It's not that hard, you just need to learn calculus first.

Complex exponentials come from the solution to a second order differential equation.

>>8955446
what will help you is really carefully working through the proofs of the different theorems and the derivations of various transforms in the fourier tables.

>>8955529
close but not quite

Fourier isn't about decomposing a sine wave. the Fourier theorem lets us decompose other functions INTO sine waves.

what you described is Euler's formula, which as you said lets us "decompose" sine waves into the sum of complex exponentials

>and sadly I don't even really understand what "e" is.
this is a fairly fundamental piece of knowledge, i'd stop right now and study it before continuing.

cec

>>8955446
Read a book on Fourier transforms like "The Fourier Transform & Its Applications" by Bracewell