CrypticSpren
-
Posts
78 -
Joined
-
Last visited
Content Type
Profiles
News
Forums
Blogs
Gallery
Events
Status Replies posted by CrypticSpren
-
Sending this on your personal feed so as to not clog the forum thread.
If we're just doing true bind points then I feel like the code is pretty trivial (just trial and error every set of 3 points as the triangle midpoints lol), though I very well could be missing something. If you've already written something I'll take it though
. Don't worry about the language; pseudocode is fine. As I mentioned earlier though I'm guessing we'll probably end up working primarily in Godot (which can have both GDScript and C# code easily) or QtCreator with C++.
I know we don't have any good theories on % of non-bind point strength, but I thought it would be good for us to come up with some reasonable metric instead of making it all-or-nothing, considering how line strengths in general aren't all-or-nothing.
-
Sending this on your personal feed so as to not clog the forum thread.
If we're just doing true bind points then I feel like the code is pretty trivial (just trial and error every set of 3 points as the triangle midpoints lol), though I very well could be missing something. If you've already written something I'll take it though
. Don't worry about the language; pseudocode is fine. As I mentioned earlier though I'm guessing we'll probably end up working primarily in Godot (which can have both GDScript and C# code easily) or QtCreator with C++.
I know we don't have any good theories on % of non-bind point strength, but I thought it would be good for us to come up with some reasonable metric instead of making it all-or-nothing, considering how line strengths in general aren't all-or-nothing.
-
However, this assumes that the Coppermind statement is true for ideal lines of warding as opposed to actual ones. So something that could account for the relative weakness of larger circles is error in drawing. If the amount of "error" on a Line of Warding is in units of distance instead of being scaled to radius, then multiplying the strength by a small but significant function of the error term could account for the relative weakness of larger LoW.
Also what is Nth-setting? Idk math stuff
-
-
Sending this on your personal feed so as to not clog the forum thread.
If we're just doing true bind points then I feel like the code is pretty trivial (just trial and error every set of 3 points as the triangle midpoints lol), though I very well could be missing something. If you've already written something I'll take it though
. Don't worry about the language; pseudocode is fine. As I mentioned earlier though I'm guessing we'll probably end up working primarily in Godot (which can have both GDScript and C# code easily) or QtCreator with C++.
I know we don't have any good theories on % of non-bind point strength, but I thought it would be good for us to come up with some reasonable metric instead of making it all-or-nothing, considering how line strengths in general aren't all-or-nothing.
-
Sounds good. We could also use text or telegram I guess.
OK I realized I didn't actually say much about what makes a Line of Warding strong in my theory post. My thoughts are quite unorganized.
I think that the first thing we should figure out is, ignoring bind points, how strong is a circular line of warding? There's really only 1 properties that an ideal circular line of warding has: a radius r. Everything else is some function of that radius: perimeter is just kr; surface area of the sphere is just kr^2.
So the question is: what is that function?
I like the area of perimeter, because it lines up well with the charge hypothesis: the "material" that the line of Warding is made up of is of constant charge density that way.
However, we know from the Coppermind that
QuoteA larger Line of Warding is weaker, however, it takes more strength for chalklings/Lines of Vigor/Lines of Revocation to destroy 1/6 of a large circle than 1/6 of a smaller one; the strength of all circles is not the same.
Thus, it appears that if we had some function f that represents the strength were purely a function of radius, it must be the case that 1 < f(r) <= r (ignoring constants). This completely throws kr^2 (surface area of sphere) out the window, and unfortunately appears to make direct proportionality to area of perimeter not work.
-
-
Sending this on your personal feed so as to not clog the forum thread.
If we're just doing true bind points then I feel like the code is pretty trivial (just trial and error every set of 3 points as the triangle midpoints lol), though I very well could be missing something. If you've already written something I'll take it though
. Don't worry about the language; pseudocode is fine. As I mentioned earlier though I'm guessing we'll probably end up working primarily in Godot (which can have both GDScript and C# code easily) or QtCreator with C++.
I know we don't have any good theories on % of non-bind point strength, but I thought it would be good for us to come up with some reasonable metric instead of making it all-or-nothing, considering how line strengths in general aren't all-or-nothing.
-
I have a feeling that if you do triangle based stuff, optimal circles will always end up being special cases: degenerate, right, equilateral, isosceles triangles and the like, thus encouraging the use of circles with fewer than 9 points beyond the already good incentive of relative draw speed. So I'm not a fan of triangle based.
I was originally going to type out my thoughts here but realized there's enough content to warrant its own theory post. So stay posted!
Also, do you have a Discord? 17th shard is really not great for back and forth conversation
-
-
Sending this on your personal feed so as to not clog the forum thread.
If we're just doing true bind points then I feel like the code is pretty trivial (just trial and error every set of 3 points as the triangle midpoints lol), though I very well could be missing something. If you've already written something I'll take it though
. Don't worry about the language; pseudocode is fine. As I mentioned earlier though I'm guessing we'll probably end up working primarily in Godot (which can have both GDScript and C# code easily) or QtCreator with C++.
I know we don't have any good theories on % of non-bind point strength, but I thought it would be good for us to come up with some reasonable metric instead of making it all-or-nothing, considering how line strengths in general aren't all-or-nothing.
-
Ok I’ll read this in a bit! I will say that I don’t like triangle area determining strength since that would favor equilateral triangles/6 point circles over other defenses, which seems antithetical to the fact that 9 point defenses are the strongest though. Unless you mean perimeter, in which case 2 point circle is infinitely strong which is clearly not the case
-
-
Sending this on your personal feed so as to not clog the forum thread.
If we're just doing true bind points then I feel like the code is pretty trivial (just trial and error every set of 3 points as the triangle midpoints lol), though I very well could be missing something. If you've already written something I'll take it though
. Don't worry about the language; pseudocode is fine. As I mentioned earlier though I'm guessing we'll probably end up working primarily in Godot (which can have both GDScript and C# code easily) or QtCreator with C++.
I know we don't have any good theories on % of non-bind point strength, but I thought it would be good for us to come up with some reasonable metric instead of making it all-or-nothing, considering how line strengths in general aren't all-or-nothing.
-
About Haskell: it’s a pure, lazy, functional language, so it’s really different to work with than the impure, strict, OOP/imperative languages that people usually use. Haskell also uses a lot of patterns that pull their names from category theory (functors, monoids, applicatives, monads) so people get intimidated. I can definitely sing more praises about Haskell but l’ll refrain.
About the code not being all or nothing: sounds good! I was thinking something along the lines of if we 90% match, the strength is 90%, and if we dip below 50% then the bins points fail entirely. Or something like that.
About theory of circle strength: I actually don’t know! For some reason I thought that a line of warding gets a “charge” proportional to radius that distributes more to areas of high curvature (kinda like actual electric charge in a conducting shell) but I went on Coppermind and found that larger circles are apparently weaker, so that can’t possibly be the case. Sounds like I need to read up - any chance you can explain the circle strength theories or direct me to a thread that does?
Thanks again.
-
-
Sending this on your personal feed so as to not clog the forum thread.
If we're just doing true bind points then I feel like the code is pretty trivial (just trial and error every set of 3 points as the triangle midpoints lol), though I very well could be missing something. If you've already written something I'll take it though
. Don't worry about the language; pseudocode is fine. As I mentioned earlier though I'm guessing we'll probably end up working primarily in Godot (which can have both GDScript and C# code easily) or QtCreator with C++.
I know we don't have any good theories on % of non-bind point strength, but I thought it would be good for us to come up with some reasonable metric instead of making it all-or-nothing, considering how line strengths in general aren't all-or-nothing.
-
guys, im an idiot, I forgot to mention something super important. I
WROTE
WE DONT TALK ABOUT JASNAH!
-
what is the weirdest place you have met another sharder? just yesterday in Minecraft of all places I was on a server if you get what I mean, and I always say I am a stick, cause why not? and out of nowhere a sharder recognized the phrase, it was good old @Vin's Haircut btw. it was an interesting experience indeed.
-
okay, so it turns out I won't be completely gone, just I won't be posting as often. expect less from me.
-
hey guys, I am going to be gone for a month, so I won't be on here for a while, if anyone asks, I am beating up pineapple.
-
I will silently gather my reputation.
. Don't worry about the language; pseudocode is fine. As I mentioned earlier though I'm guessing we'll probably end up working primarily in Godot (which can have both GDScript and C# code easily) or QtCreator with C++.
