Can anybody recommend some MATHS videos?

I’m interested in Maths video series’ on Youtube - starting from Class 11-12 level and going up, covering AI and game programming/procedural art-type of stuff. Anybody got any? :slight_smile:

1 Like

Depends on what you like, really, but since you mentioned procedural arts stuff: vihart does really interesting stuff with geometry and can explain them in a way that’s easy to understand, though (at least in her videos) she usually paints these and doesn’t do them on a computer (she does do that in a couple papers, though I guess “reading maths papers” is not really a helpful answer to your question); Henry Segerman does similar things, though I’m not sure how understandable much of this stuff is if you don’t already have a background in geometry (admittedly, I might not be the best person to ask on this, since I do happen to study geometry, so sometimes my sense of “easily understandable” might be skewed …).

Apart from that there’s always numberphile or Matt Parker for more general intersting maths-bits which are also explained really well, on all kinds of topics.

Hi,

General math videos I’ve watched in the past and enjoyed (unsure though whether these lay within your grade level or not):

If you want to start doing procedural art or videos you might want to explore Animation Nodes addon and Sverchok addon. Both of which can do many things from motion graphics to math art to architectural stuff. Lots of tutorials on both of these can be found.

On a side note, this site has mind bending videos on Cosmology, Time and Space (Dr. Matt O’Dowd is very good):

More advance math topics:

General good science related videos:

For perhaps one of the best magazines for science and math that I’ve found so far which explains things plainly and provides background also is Quantamagazine. I highly recommend this site.

Another thing I would recommend is if you are very interested in mathematics (especially if you wish to pursue it as a career or just to dive deeper) is to grab a student version of Mathematica by Wolfram (it’s not free but it’s certainly powerful math software): https://www.wolfram.com/mathematica-student-edition/

Cheers and have fun blending and learning.

1 Like

…I was kind of hoping for a “what they teach in class but-made-fun-n-cool” kind of series - that 3blue1brown is the one that comes closest, from what I can see (just, a cursory look @ the youtube list). Is there any way to get it, or the other ones, graded, i.e. from least difficult video to most difficult one? Youtube doesn’t do that… (by itself) :frowning:

Any other ones?

…and isn’t Octave the free version of Mathematica?

mathematica comes free and preinstalled on raspberry pis. don’t know if there are any limitations though except for only running on the raspberry pi. :slight_smile:

GNU Octave is a free rebuild of Matlab (licensed under the GPL); both are special languages (and programs built around them to make them easier to use), or rather two almost identical versions of the same language, just implemented in different ways by different people, built for numerics (roughly “solving mathematical stuff that’s too hard to do by hand by using approximations and lots and lots of iterations to make them better, and get a result that’s hopefully close enough to the real thing to be useful” — lots of physics simulations work like that, from global climate models to the smoke simulator in blender).

afaik, while Mathematica (I’ve never used it) can also do numerics, it’s also more focused on symbolic computation (roughly “take lots of symbols like x or y for variables, then move them around until we have something that looks useful” — in other words, you can use it to actually solve equations in general, instead of “just” getting a numerical approximation, though of course your result will also be composed of symbols, i.e. it might contain an infinite series or an irrational number, so if you actually need a number it might not always be helpful), which I’m not aware Octave/Matlab can do. Mathematica is build by the same company which also does Wolfram Alpha, which is a website that solves equations for free (and which has been said to have helped more than a few students to get through the more mathematical parts of their studies :joy:)

Edit: for the problem of “grading” these videos: I’m not sure if it makes sense to grade many of these videos, since most of them have very little to do with each other — e.g. numberphile videos are almost always self-contained, and just present interesting tidbits from various sub-fields of mathematics that have often almost nothing to do with each other, so their “difficulty” mostly depends on either how familiar one already is with the field, or the field’s “intuitiveness”, which also varies from person to person. Still, it would be nice if there’d be something like a “crash course” for the sort of basic maths you might learn in school / the first few semesters of university, but I’m not aware of one — there’s a Crash Course Statistics on youtube, which I guess comes closest, but also has pretty much nothing to do with either procedural art or AI stuff; I also found a Crash Course AI, but it seems to be pretty short …

Hi,

Perhaps, I got carried away. Did you try searching for Math Grade 11 on YouTube? You’ll find a lot of stuff. I’d suggest looking through the various suggestions and seeing what interests you. You might find a topic/subject you like.

Also something that might be more geared to you is Mathantics. Seems popular and targeted towards kids.

For something more structured one of the learning sites might be better like Kahn Academy, Courseca, etc… For a list of free sites see https://www.forbes.com/sites/zackfriedman/2019/05/29/free-online-education/#6b9a0557342b

Cheers and have fun.

I am going to assume you want Math as used in Node Trees in procedural Materials…and how to use them…there are a scattered few that are good…

Here is one that is great for reproducing the effect…but short on explanation ( and the mouse click sound will drive ya nuts!)


Check out his other videos as well, they are well worth the time!
Check out and links on the web as well as Videos with the search term, Nodevember since we are almost there!!

Though I dislike mathematics, I like programming ten times better. Which means that I find it easier to understand programming math because the implementation is far more expressive in terms of design and look. Even if something doesnt make sense code can be rearranged into smaller and cleaner steps until it becomes far easier to understand. While math notation looks very dull and cryptic at sometimes and uses abbreviations and other shorthands, that make the solution too much of a high-level approach and harder to understand.

So with programming in Python (since it fits the Blender ecosystem) is a chance to get really good into deep waters and see how everything works, also having another benefit is to validate the solution and see if the calculation was correct. Then just by looking at the math notation solution you will get something like: “oh, so this is how is done in math?” which is practically having the same knowledge but represented in a different way (one in programming and other in math notation).

1 Like

Hey, thanks for all that guys :slight_smile: , yeah, I dislike maths and like coding too - hence this post! :slight_smile: (why do people always say they’re related? -puzzled- )

1 Like

What kind of AI math are you interested in? (AI for games or other kinds of AI algorithms?)

hey, I was just gonna post about that right now :slight_smile: - well, with Machine Learning, I think it’s about finding functions to fit data, and then doing stuff with those functions - and I find that Damn Cool! How does one do that??

DeepBlender, never you mind what I’m interested in, just sling all the videos you can at me (let me be the judge :slight_smile: )

The art of code channel is filled with videos relating to math and procedural shaders (not Blender specific):

There are way too many resources available and even though I might be aware of many, it still takes time to find them again. That’s why I asked what specifically you are looking for.

Is my assumption correct that you are looking for this kind of video?


I think you would like this as it will show you what every material node is for and how to use them>>>

Hey - I’ll take a look at those. Generally speaking, I wanted videos which show “how it’s done”, not “how to use library X”.

Something really impressive regarding math.

If you try to code this in Python is a good starting point.

When it comes to understanding the math of deep learning, I am not aware of short versions, but only of (university) lectures. In my opinion the easiest to understand one is this course where a deep learning library is built from scratch while introducing all the mathematics behind it:

1 Like

thanks, deep :slight_smile: