[…]INTRODUCTION, REFERENCES, AND A DISCLAIMER

This text is designed to accompany your study of introductory digital signal processing.1 It’s an eccentric piece of not-so-rigorous literature with a preoccupation for explaining things using interactive visualizations, animations and sound.2 My goal is to explain the Discrete Fourier Transform using a miniature curriculum which leverages your ability to learn concepts and absorb information visually instead of linguistically.3 My hope is that these glyphs become slightly more comprehensible and slightly less intimidating after reading the subsequent 30 or so pages.4

1. Almost always referred to as "DSP", especially if you’re into the whole brevity thing.
2. I’ve attempted to avoid the gratuitous use of interactivity after reading the section titled Interactivity Considered Harmful in Bret Victor’s "Magic Ink" essay. If you have strong feelings about interactivity and computer-centric presentations of information, I highly recommend this essay.
3. This sentiment and wording might be familiar to you if you’ve read Bret Victor’s Kill Math . If you haven’t already, you might also be interested in reading Evan Miller's critique of Kill Math.
4. See Seymour Papert’s exceptional Mindstorms. Chapter two is all about "Mathophobia" where he writes, "It is not uncommon for intelligent adults to turn into passive observers of their own incompetence in anything but the most elementary mathematics." If you’re a mathophobe (like me), I hope this text feels like a welcome respite."[…]

Source: https://jackschaedler.github.io/circles-sines-signals/index.html

 

Last edited by a moderator: Nov 4, 2015

I'll be honest.... I have absolutely no idea what you're talking about.

 

either i am high or i feel dumb

 

It's more for people that are learning DSP. If you want a challenge, or if you're gonna program in audio (soecifically spectral-based filters).

It goes over a lot of things relating to FFT/DFT. You can think of it like this: every audio signal can be decomposed into a series of various frequency, amplitude, and phase sines.

 

This has been a complete waste of time.

 

@davea - Next time please don't mark half of the article as single link, and if possible avoid uppercase-only titles. When you're quoting something, it's useful to add source as well.

 

This really is an excellent homepage. Incredibly well made!

Yeah how's that gonna work?

 

Math isn't for everyone.

 

Hi guys!
I've found this great article/presentation/workshop on the web, and naturally, found it precious to share at AS.
I was aware that some of people here wouldn't get it but some other will ;-)

@Andrew : Sorry mate, my brain was in the "foggy/shadows" at this time yesterday, after a long session day …
Thanx for your help and understanding

Btw, I'm not really a "math" guy but this kind of article excite my curiosity enough to stop by. Here at AS, I'm sure there are guys who could help us to better understand some obscure lines

 

Last edited: Nov 4, 2015

Don't know what is about....But if you want to say "seeing DSP effect on simple sines/sweeps is waste of time", as mentioned in thread title, I say you next...This shows you what happens to simplest signal after one processing, and know you must think of what 15 processors do with complicated signal (which consist of millions of simple sines etc.), and what happens when you sum it up with other processed complicated signals, and finally when you process it on mastering stage..It's not like "aliasing + aliasing + aliasing", it's like "(aliasing x aliasing x aliasing + aliasing x aliasing x aliasing) x aliasing".... it's not only about aliasing, but as well as intermodulation distortions, even/odd harmonics etc....
It's like fundament and material for build a house...cheaper material, worse fundament, worse the building work, bad workers means the house will be a shit...cold, crackling walls, uneven, bad sound treatment, and it will be on the ground just 20-25 years, not 70-150 and more...
And your track...if there are high-quality algorithms, high-quality internal processing, high-quality oversampling filters, close-to-analog emulations and more other accurate/precise/HQ stuff, it means higher quality results will be...
If you use shitty digitally-harsh stuff on each channel and masterbus, no oversampling/low-quality oversampling, lower sample-rates, no dithering, digital clipping, peaks/true-peaks over 0 dB - you will get anti-analog LO-FI harshest sht...

When I have no ideas to create music, I just update my stuff and check several plugins, and put them on different folders in REAPER's FX Browser folders, like that:
HQ Master EQs (Equilibrium, Pro-Q2, Sonoris Mastering EQ, PLParEQ, PSP Neon HR2, TDR VoS SlickEQ, DDMF...)
"Analog" Gear (Nebula, Acqua Plugins)
Bad Stuff EQs (Crysonic...)
HQ Analog Synths (Diva, Lush-101, TAL, OP-X Pro 2, Arturia, imposcar2...)
LQ Analog Synths (...)

Check all and sort your stuff, if you have a lot of them..

 

Last edited: Nov 4, 2015

Hahaha ! You're killin'

Nope isn't a waste of time at all, thanks for your inputs . Obviously, the sound chain needs to be checked.

 

Crysonic: eg., see what happens to sound when you use SpectraphyHD... Your quiet tails will be destroyed....it sounds like 12-bit quantization or smth like that...LO-FI...I don't like stuff from Crysonic....everything shitty you can find without any good monitoring...

And about checking: Yes! Our ears may be fooled, analyzers can show you something you can't hear...distortions @ 20 kHz or noise at -80 dB...

Do you know that Pro-Q 2 algorithms are worse that Pro-Q1 ???? Use LP / HP filters (as I remember - Linear Phase mode) and see on Voxengo SPAN....you can see that Pro-Q has no distortions higher than -180 dB, but Pro-Q 2 has distorions and noise higher than -140 dB....Everytime you use Pro-Q2 on each channel will add noise...And it summing and summing - so it means noise+noise+noise.....
It was discussed on FabFilter Forum, so FabFilter developers say that they will fix it...but I don't know if they fixed it or not in latest 2.03..

 

Last edited: Nov 4, 2015

Discrete Fourier Transform is used in DSP and used to analyze frequency which is present in a signal. Think Voxengo Span which uses Fast Fourier Transform. There is lots of cryptic specialist YouTube stuff to explain:

It is not really usesfull to study this as long as you know it's appplication, so try this one: