For a paper I'm writing: What's your math background?

[vsr600]

no Fourier transforms would be good... maybe just do the discrete transform though and mention the continuous version and briefly mention FFT.

[RefD]

I will not read papers on acoustics because they bore me (personal preference) which is perhaps why my recordings sound like they were made in somebody's sinuses...

get out of my head!

[kweis7]

I have a BA and MM in music but was originally gonna do EE and took calc 1-3 but dropped differential equations when it became clear I would not be happy doing anything but music. Once upon a time I was pretty sharp with that stuff (most everything on the list except the wave form stuff) but now I would have a hard time with a double integral.

Regardless, the thinking that it took to get good grades in those calc classes has really helped me repeatedly and often; I don't think I'd have the comfy technical job I have right now (applications development for a C.S. dept) if I had not done that stuff (I'd also report similarly on challenging reading and writing long papers). Learning to deal with complexity translates to many things. Anything that gets folks more math literate and appreciative is good for the world, imho. Good luck with the paper(s)!!!

[oudplayer]

I think that my 2.5 years of calculus, year and a half of college physics, and fft programming background have been helpful assets for being an audio engineer. Acoustics and digital signal processing actually make sense for the most part, rather than relying on "faith" to steer me thru my sessions.

But, still no idea what the hell a "phason" is.

[river]

I've seen and used all the terms mentioned. I went to IAR in NYC, which started out as an AES workshop. The math courses there were an amalgam of algebra, trig, calculus, etc......A lot of emphasis was put on the reactive phase component of a signal path from input to output, since manufacturors in pro audio could never agree on any standard for phase component of a piece of equipment's impedance. Their main goal was to beat it into our heads that it was worth the time to determine and correct any phase angles in a studio's signal chain to get it to zero, by building T pads and the like.

[vsr600]

But, still no idea what the hell a "phason" is.

that's in solid state physics... you know how you have quantized acoustic energy called phonons well you can also quantize the phase... i think that's it i mean. I doubt that would be helpful here anyway.

[Skrasms]

Thanks for all the input!
It will probably be a while before it's finished, but I'll get it online once it's done.

[bradb]

I think that explaining how continuous waveforms in the time domain transform to functions in the frequency domain would be quite enlightening. I'd avoid the math and just explain the concept. Thats all that sticks with you in the end anyway.

Phase cancellation also applies at RF... It's called multi-path fading... When you are sitting at a stoplight and your FM station is fuzzing out and you inch forward and it comes back... you were in a null.

recording IS everywhere

[tret-lo]

This is a very interesting post. I have spent a lot of time thinking about this question of how much math understanding is necessary to work with audio.

In my view, Fourier analysis lies at the heart of the matter. It is easy to tell someone, "an audio signal can be represented as the sum of pure sine waves." They nod and smile, and if they can play with a filter and start to get the idea. But unfortunately, Fourier analysis itself is a pretty sophisticated concept, requiring solid knowledge of calculus, linear algebra, and analysis. In my eyes, the difference between the conceptual idea and actually being able to calculate the transform of some given signal, like a square or triangle wave, is equivalent to the difference between seeing a picture of the grand canyon and actually seeing the grand canyon with your own eyes. In either case, without the first-hand experience, you're taking a lot on pure faith.

Does understanding the deep mathematics behind audio processing make you a better musician? Probably not. But as someone who has "seen the light" of mathematical science, I have no doubt that it makes you a better engineer.

[trodden]

damnit, this was fun until math showed up. wow, just like science class!

[vsr600]

In my eyes, the difference between the conceptual idea and actually being able to calculate the transform of some given signal, like a square or triangle wave, is equivalent to the difference between seeing a picture of the grand canyon and actually seeing the grand canyon with your own eyes.

and just to confuse everyone... if you try to take the Fourier transform of a square wave or a triangle wave you get the Gibb's effect where your periodic function "rings" at the discontinuities...

[EarlSlick]

All the concepts you mentioned are quite accessible to people if they are willing to spend some time figuring them out. But trying to explain Fourier Transforms to someone who doesn't have any formal education past the high school level is quite an uphill battle. Most high schools do a very poor job of teaching math. Infinite series in general are very hard to explain to lay people. Furthermore, people just don't care, because they think the math is either too hard, or irrelevant. However, just because many will ignore the math is no reason to exclude it. In order to fully understand the physics, you must fully understand the math. I would reccomend that everyone interested in recording at least take 1.5 years of calculus, so you have a vocabulary (although still somewhat weak) for understanding the basic Electricty and Magnetism, and Electronics. Although it is certainly not necesary to even understand the physics, many would find it very helpful and also very interesting.

Bart

[therethere]

i'm in Chem Eng, 4th year.....but my calculus skills are not exactly what you would call strong, as i barely got by in first year.....but i'm getting B's in my English electives