3 ms·
> But GA itself isn't actually that interesting or useful. It's not like GA is being used to prove new math theorems or new calculations That's hardly the only
by modeless 6y ago
> But GA itself isn't actually that interesting or useful. It's not like GA is being used to prove new math theorems or new calculations
That's hardly the only definition of usefulness. I haven't used it yet but based on what I've read GA will be extremely useful in 3D graphics coding by unifying representations for previously distinct concepts and by providing formulas for common tasks with fewer special cases.
- thechao 6y agoI’m a huge GA fan, but until really sophisticated compile-time algorithms can aggressively reduce the computations it won’t be useful for graphics anytime soon. For instance, the “proper” GA setting is 5d conformal, which requires 32x32 matrix computations. The standard operation is “multiply” (even for translation!); that takes an operation from 4 ops to ... 32k ops.
- modeless 6y agoClearly the generic n-dimensional GA libraries out there have little use for real time graphics, but projects like https://www.jeremyong.com/klein/ https://www.jeremyong.com/klein/ that are specialized to 3D with a focus on efficiency look promising.
- thechao 6y agoThis is fantastic; thank you!
- WoahNoun 6y agoDepends on your definition of calculation. From what I've seen so far, the simplifications that GA makes are mostly on the symbolic side. That is, it makes the written equations more compact. But the math "under the hood" is the same. So I don't see how that would yield anything useful from a computing sense. Might make the programming easier to think about in a way similar to monads in Haskell/functional languages.