This thread is permanently archived
abs(x)+abs(y) but
| math g/u/rl, go home, you're drunk
| >>1018271 seriously? i am not actually, more beautiful than beautiful? just exactly like something else?
| You playing a Zachtronics game or something? Sounds like the kind of thing you'd have to do to write a TIS-100 program.
| If you can use If statements, I would do:
If X <0
X= 0-X
And the same for y
| May be OP means something like to use binary shift operations , etc like it. Like in C? Or this is pure mathematic question?
| >>1018286 no, it's coding.
>>1018329 ...i can but i dont wanna
>>1018378 huh??
basically this is what went on and i need to work it out.
https://imgur.com/a/G1LeIWJ
i dont even know how or what. it kinda is better but i am not even sure why.
| this is a classic X, Y problem
| >>1018264
I think this in Stackoverflow post will help you "Absolute value abs(x) using bitwise operators and Boolean logic":
https://stackoverflow.com/questions/60601908/absolute-value-absx-using-bitwise-operators-and-boolean-logic
| >>1018600 urgh i really dont want to use abs or
anyway i got this far and it looks ok but i dont get it at all
https://imgur.com/a/RXK1xcf
pls explain
| imgur game dev op
| >>1018700
Um, I do nott understand, is imgur here as a smiley?
Let us assume that this is the case, but it is not clear on the very issue with abs(x). Here hard to explain it well, and in general, I didn't really get into bitwise operations.
| >>1018700 to continue>>1018764
In short, the bottom line is that using the internal representation of a number, for example int, namely in sequence(s) of bits, you can perform division, multiplication, etc. operations using, for example, shifting bits in a sequence, changing the initial bits, etc.
| >>1018813
What do you mean? We were looking for a way to achieve the goal without using the above methods. As a variant of "manipulation" with bit operations, it was proposed. Are you not interested in mathematics? But nowhere without it.
| >>1018983 Hmmmm, well i guess it's not "without mathemathics" anymore
https://zzzchan.xyz/v/thread/231536.html i just need to understand this (the last 3 post)
i avoided abs operations,
but it dont work as i want, and it seems so...very closely.
as for bitwise operations>>1018765
aint that just math but done using bits? it just makes things longer,but i can see now that the bits dont mean much in math. good to know.
| >>1019264
hm, I can't open this link.
Yes, that bit operations work with bits of number representation in computer on machine level. Example on C. As you may know the variables of int type are take 4 bytes (32 bits, of course it depends on CPU architecture) in memory (RAM), that's mean you have a sequence of 32 bit, in what a number is "encoded", and any manipulation on these bits will change a number, sometimes cardinally. So, yes it is more informatics, rather than mathematics.
| >>1019449 how about this. well nothing about informatics nor math. i guess it is tech https://zzzchan.xyz/file/f71200d7c865053ffb81e2341d4693ba05eb314bf0758e878a83a85c09f90816.mp4
it's more on solving graphics.
| >>1019703
Eh, apparently the whole zzzchan.xyz site is not available to me. Do not use vpn. Maybe on imgur?
| >>1019715
Interesting, I guess it recalculate positions if points depending on nouse position. So you like as it: no math, no informatics?
| >>1019724 yes, basically coordinate projection, if you have been followin up
but problem is im kinda stuck as it scales to 0, when naturally projections should stop at a certain point, but so far i cant do so other than diving it by abs(x)+abs(y), which is also jittery as fuck
or changing the entire math
| >>1019724
not exactly "no math"(but idont know how informatics work), as it still codes abit, but basically something else than what i am doing now, which according to the system means only collission and events
| >>1019725
>>1019725
perhaps you should change the coordinate system, perhaps also change the coordinate representation itself, for example, if you use vectors, try quartenions.
| >>1019726
>collission and events
That's recalled to me Particle Swarm (Horde) Optimization method (PSO). Where you encode certain solution (that to make equation system come close to identically) as particle (a point) in "space of solutions" , and then "interaction" between them make them to move, to find optimal solution of system.
| >>1019753 I am not sure we are on the same page right now. I used the usual coordinate system just without matrices. So I should get solution without anything too fancy
>>1019754 ...the name don't sound very correct but I almost got you, I mean I do but you need to give me example here than synopsis and movie summary lel
| >>1019873
You have an equation(s), its solution turns its expression into an identity. Consider the space (of solutions), the points of which determine the possible solution of the equation. Let's define a function F in this space, let it be the absolute value of the difference between the left and right sides of the equation. The minimum of this function means that its arguments in this case, the value turns the original equation into an identity.
| >>1019873
>>1019904
Let's define a set of points in the solution space. Consider the analogy with gravity. We assign a "gravitational mass" (charge) to all the points under consideration. Let the points move according to the law of motion under the force of attraction (gravity) like physical particles, but at each moment of the motion simulation time we will add/subtract an additional charge depending on the function F.
| >>1019873
>>1019904
>>1019906
So the particles moving will be "attracted" to the areas in which the function F is minimal. We found the minimum or so, stopped the search for a solution.
So you can adopt electromagnetism, etc.
That is PSO method.
| >>1019873
You mean Cartesian coordinates in Euclidean 3-dimensional space. And also without rotation matrices, maybe you didn't use them directly, but say the rotation functions that you use (I assume) use rotation matrices (read Euler angles). And yes, all this is done by representing a point with radius vectors.
| >>1019873
> So I should get solution without anything too fancy
I'm afraid no solution will come to mind without a complete picture of the problem.
Maybe there are some tricks related to the language, in which you write your project?
This thread is permanently archived
| so i need something that values like abs(x)+abs(y) BUT, in another way without abs or log, or multiply it by itself,
so it has to be only using + - and / and not even *, or maybe only + and - and gives the same result because i need it to be 0+0 to still be 0, but for x = -1 and y = 1, x=-y vice versa it must not add to 0,
trippy ain it?