I have to make an animated explainer for Fourier transformations and sadly it looks like I might need to properly mathematically understand Fourier transformations?
-
Like even though I have a script and diagrams from someone who knows what they're doing, I'm going to need to work out accurate measurements and speeds and so on. There's probably a way to code some of this stuff into expressions but I have no idea how to go about that. Damn you Joseph Fourier
I cover loads of different subjects but mostly I have a lot of leeway with how the concepts are visualised. The most complex ones have been some biological stuff where the detail was important, but that was 2D enzymes floating around, not 3D protein folding or anything. But this is the first topic that has me thinking, ah shit, I might not actually be able to do this.
-
I have to make an animated explainer for Fourier transformations and sadly it looks like I might need to properly mathematically understand Fourier transformations?
@internetsdairy I did stuff with Fourier transforms for years without understanding them. Maybe it's kinda like a Taylor series but better? This was 3D proteins and stuff btw.
-
@internetsdairy I did stuff with Fourier transforms for years without understanding them. Maybe it's kinda like a Taylor series but better? This was 3D proteins and stuff btw.
@internetsdairy Do you want input from random Mastodonians?
-
@internetsdairy Do you want input from random Mastodonians?
@drgroftehauge thank you, I need to explore and experiment a bit before I even know what I need to know tbh. Apparently it doesn't have to be accurately worked out, it just needs to kind of look clear enough to get across the concepts.
-
@internetsdairy Do you want input from random Mastodonians?
@internetsdairy Okay. I'm gonna summon @folfdk and @iris
My example is going to be the transform of time. Say we have some data on when things happen during the day. If something happens approximately at midnight then it will appear both around the maximum and minimum value of time. But we can do a cos and sin transform of what time it is with a 24 hour frequency and that makes it much easier to reason about. We can continue with 12, 6, 3 hour frequency indefinitely. -
@drgroftehauge thank you, I need to explore and experiment a bit before I even know what I need to know tbh. Apparently it doesn't have to be accurately worked out, it just needs to kind of look clear enough to get across the concepts.
@internetsdairy Oh, you liked it so I went ahead. Like sounds like yes to me.
-
@internetsdairy Okay. I'm gonna summon @folfdk and @iris
My example is going to be the transform of time. Say we have some data on when things happen during the day. If something happens approximately at midnight then it will appear both around the maximum and minimum value of time. But we can do a cos and sin transform of what time it is with a 24 hour frequency and that makes it much easier to reason about. We can continue with 12, 6, 3 hour frequency indefinitely.@drgroftehauge @folfdk @iris thanks, yeah I don't know what cosines etc are, I'll likely be using something that doesn't involve significant maths. It just needs to look roughly right, apparently.
-
@drgroftehauge @folfdk @iris thanks, yeah I don't know what cosines etc are, I'll likely be using something that doesn't involve significant maths. It just needs to look roughly right, apparently.
@internetsdairy @drgroftehauge @iris
Kathryn Cowtan has some nice visualisations on their site.
In particular this one (to me) is really informative, although it relates to Fourier transforms in x-ray crystallography and is a bit simplified, the basics are sound
http://www.ysbl.york.ac.uk/~cowtan/fourier/ftheory.htmlI'm not 100% sure what you're looking for will read the other comments now
-
@internetsdairy @drgroftehauge @iris
Kathryn Cowtan has some nice visualisations on their site.
In particular this one (to me) is really informative, although it relates to Fourier transforms in x-ray crystallography and is a bit simplified, the basics are sound
http://www.ysbl.york.ac.uk/~cowtan/fourier/ftheory.htmlI'm not 100% sure what you're looking for will read the other comments now
@folfdk @drgroftehauge @iris thank you, it's more of an After Effects problem than a math or visualisation problem it seems.
-
@folfdk @drgroftehauge @iris thank you, it's more of an After Effects problem than a math or visualisation problem it seems.
