#computertech Bot Logged User list

Network: TechNet

Modes: +ntf

Last Seen: 2 months, 2 weeks ago

Topic:

Users

Channel Log Archive for #computertech

Hide Joins

* All times are UTC

Filtering by user: ski/rizon

Monday, November 7, 2022

[22:13:00] @ ski/rizon meow

[22:14:28] @ ski/rizon heya peeps :)

[22:14:53] @ ski/rizon 'm ok, ty

[22:15:28] @ ski/rizon was just looking at a paper named "Sacrificing the calf of flexibility on the altar of reliability"

[22:15:40] @ ski/rizon (short one, just three pages. from 1976)

[22:16:56] @ ski/rizon well, i was obsessing a bit about a Pythagorean metric idea i had .. which could potentially be applied for selecting a path in a game (although it's not entirely clear whether one'd want to)

[22:17:17] @ ski/rizon ComputerTech : maybe this guy ComputerTech does ?

[22:17:36] @ ski/rizon he seems to be into writing scripts for bot, you could try asking him

[22:18:55] @ ski/rizon anyway .. basically, you know how if you have a right angle (or rectangle) with side lengths `3' and `4', then the diagonal (hypothenuse) is `5', also an integer

[22:19:49] @ ski/rizon so, if you have a game grid, where you can only move vertically and horizontally, then the distance between two points is the sum of the distances of the two coordinates, `dx + dy = (x1 - x1) + (y1 - y0)'

[22:21:16] @ ski/rizon but now i also want to add all these diagonal paths, which are *exact* integers .. so all scalings of `(3,4,5)', iow `(6,8,10)',`(9,12,15)',&c. and also the same for all other pythagorean triples `(a,b,c,)' (where `a^2 + b^2 = c^2') such as `(5,12,13)' (there are infinitely many of them)

[22:22:18] @ ski/rizon one curious thing is that, sometimes, it is more efficient to move along such a diagonal path, and then move slightly backwards, rather than always move further and further away from the starting point .. this might feel a bit unintuitive ;)

[22:22:37] @ ski/rizon s/right angle/right triangle/

[22:25:18] @ ski/rizon .. anyway, i ran out of steam, for a bit, due to it being annoying to check that i didn't miss some possible diagonal composite paths in my analysis. so i wanted to generate them programmatically, possibly using dynamic programming, and also started making a simple library for handling ordered (possibly infinite) streams of data, which i think could help merging all possibilities in order

[22:26:35] @ ski/rizon you're back from Poland, ComputerTech ?

[22:26:50] @ ski/rizon ah, just back to IRC, itym

[22:35:23] @ ski/rizon ah, classic film :)

IRC Log Archive

This is a searchable archive of IRC conversations from #computertech on TechNet. Use the search box above to find discussions on specific topics. All conversations are preserved to help the community learn from past discussions.