[Eylrid]

On the scale of Planck's constant to Avogadro's number, how is your day today?

[Blue Ocean]

On the scale of Planck's constant to Avogadro's number, how is you day today?

I would say me day is 6^10. I don't understand why my images are being cropped rather than scaled, but other than that, fairly reasonable.

[Josie Suttle]

Well, I kind of think this depends on the person in question.

With me, if I am actually to the point of wanting to slam a door out of anger, hurt, or whatever else is compelling me to slam a door then I probably mean it.

If we are talking a physical door, then I don't intend to walk through that door again that day or any significant amount of time.

Now if the door in question is referring to an emotional passageway between two people, then that's a different question entirely.

If you're having a fight with an INTJ, then I think that you just need to be honest with them. I know that people lying to me is one way to destroy ANY friendship.

If you want to redeem yourself in the eyes of an INTJ I personally think you should just kind of prove to them that you're there for them, and that you're not going to betray them again. Although if your INTJ friend is anything like me or other INTJ's I know, they will never forget if you betrayed them. There will always be that little mistrust.

I say this because I had a friend betray me before, and ever since then I didn't trust them completely, and I never forgot.

I hope all goes well with you though

[Josie Suttle]

Isn't that too hard for an INTJ to be often misunderstood by your peers because of how many people thinking you're some closed-minded super patronizing fellows that sould "need to get out more often or something" ? I can't help but getting irritated myself when I see these somehow insidious attacks

Well, I decided that most people don't truly know me. And those people who are passing judgments on me that don't know me are obviously incorrect.

But the people the truly know me don't pass judgments on me, and even then only a fraction of those who truly know me matter.

I just decided that only the people who matter can judge me, and all of the rest are null and void :)

[Josie Suttle]

I have learned a lot from the INTJs I have known. They are the only type to see through my BS...even when I don't realize I am BSing they call me out on it. Essentially, they put a mirror in front of me and I see truth in myself and I am able to remove my facade and start living a more honest life. But when I turn that mirror around and try to help them see another side of themselves, they get angry and reject my Fi magic. I don't understand how people who are so good at helping me are so bad at understanding that I could help them, too, in a totally different way. We sort of have different symptoms of the same disease. And I feel bad when I can't help you when you have helped me so much.

Well, if what you're trying to help them out with has anything to do with emotions, then good luck. Because a part of being an INTJ is being emotionally retarded, but I'm just going based on my own personal experiences.

What I mean by that is, emotions are alright to analyze when it's someone else's emotions to analyze because INTJ's can detach from those emotions and turn them into something to be observed.

But when those emotions are their own emotions, they freak out because it's too real and they can't exactly run away and hide from those emotions. So maybe that's what your friend is doing, attempting to detach from their own problems and emotions.

If what you're trying to help your friend out with has nothing to do with anything I said, I apologize

But I tried lol

[Josie Suttle]

im not religious just wanted to know if you guys knew what hed do

Lol, I'm sorry, but I had to do it

You're not religious but you ask people what he'd do, which makes it seem like you believe he existed?

This doesn't prove you're a religious person, but I'm curious about your response

[Eylrid]

I'm working on a space exploration game in which each sector is generated when you arrive, so space is theoretically infinite. Right now, the number of stars in a sector is independent of it's neighbors, so a sector with thousands of stars may be next to one with zero.

I'm trying to come up with a good way to impose some kind of consistent form to the whole of space.

I was thinking of using something like the Mandelbrot set, but it's of finite size. Do any of you know of anything similar of infinite size?

[Falan]

I'm working on a space exploration game in which each sector is generated when you arrive, so space is theoretically infinite. Right now, the number of stars in a sector is independent of it's neighbors, so a sector with thousands of stars may be next to one with zero.

I'm trying to come up with a good way to impose some kind of consistent form to the whole of space.

I was thinking of using something like the Mandelbrot set, but it's of finite size. Do any of you know of anything similar of infinite size?

Okay, so some questions for clarification:
1) Are you programming the game as opposed to playing it?
2) Are you concerned about the finite dimensions of a Mandelbrot set because your sectors do not have boundaries?
3) Or are you generating one large star map before dividing it into sectors?
4) Are you working in three dimensions or two?


Some suggestions.
1) Use a random number generator (which you're likely doing anyway) but define a range when deciding how many stars to use. Such as 1,000 to 10,000 or something similar.
2) If you can define the number of sectors bordering the current sector, then calculate (and save in memory or disk) just the number of stars in those sectors: (Only need the number not position) then a more consistent number at least, could be calculated within the current sector. Similar to Minesweeper.
3) Or you might look into Random Walk.
Or Diamond-square algorithm
Or The cluster growing method used in fractal dimensions on networks.

Basically, I'm assuming that each sector is bordered by other sectors, the amount of space is defined within each sector and finite. However, the number of sectors would be unlimited and the player able to move infinitely in any direction.

And if you define a 'galactic center' from which all star positions are calculated, then a two or three-dimensional grid around it expanding into infinity. Then by using the 'sector' position relative to the 'galactic center' a consistently generated number and position of stars could be generated. Well, depending on your algorithm; but much the same way as zooming in on a Mandelbrot.

*Added: Actually thinking about it some more, and the spiral arms of the galaxy. You might try something similar to (or a variation of) a Dragon Curve.

[Eylrid]

Okay, so some questions for clarification:
1) Are you programming the game as opposed to playing it?
2) Are you concerned about the finite dimensions of a Mandelbrot set because your sectors do not have boundaries?
3) Or are you generating one large star map before dividing it into sectors?
4) Are you working in three dimensions or two?

A.
1) I'm programming it.
2) Each sector is finite and bounded, but there are an infinite number of sectors.
3) I've thought about various approaches involving dividing space into super sectors, and creating a density map over that. However, no matter how big I make the super sectors, I still have the consistency issue between super sectors.
4) Two dimensions.

Some suggestions.
1) Use a random number generator (which you're likely doing anyway) but define a range when deciding how many stars to use. Such as 1,000 to 10,000 or something similar.
2) If you can define the number of sectors bordering the current sector, then calculate (and save in memory or disk) just the number of stars in those sectors: (Only need the number not position) then a more consistent number at least, could be calculated within the current sector. Similar to Minesweeper.
3) Or you might look into Random Walk.
Or Diamond-square algorithm
Or The cluster growing method used in fractal dimensions on networks.

B.
1) I'm currently generating a random number between 0-1, taking the reciprocal of it, and multiplying by a scalar. Numbers are more likely to be low than high, but there is theoretically no upper limit. The PRNG is seeded with the sector's coordinates before the stars are created, so the sectors are consistent between visits.

2) That would probably be the easiest way to do it. The draw back is that the size of "formations" is limited to the size of the neighborhood I look at. (My grand vision is to have a universe with clusters of unlimited size.)

Note: I don't care about positions of stars within a sector, just how many stars there are.

3a) I'll have to cogitate on random walk. I may be able to make it work.
3b) I thought about the diamond-square algorithm. I run into the issue I mentioned in A3 above.
3c) I'm not familiar with "The cluster growing method used in fractal dimensions on networks." I'll have to see if I can figure it out.

Basically, I'm assuming that each sector is bordered by other sectors, the amount of space is defined within each sector and finite. However, the number of sectors would be unlimited and the player able to move infinitely in any direction.

And if you define a 'galactic center' from which all star positions are calculated, then a two or three-dimensional grid around it expanding into infinity. Then by using the 'sector' position relative to the 'galactic center' a consistently generated number and position of stars could be generated. Well, depending on your algorithm; but much the same way as zooming in on a Mandelbrot.

*Added: Actually thinking about it some more, and the spiral arms of the galaxy. You might try something similar to (or a variation of) a Dragon Curve.

I would need a quick and reliable way to determine where a point is in the curve.


Thank you for your response. Bouncing things off of someone can really loosen the gears of thought.

[Falan]

1) I'm currently generating a random number between 0-1, taking the reciprocal of it, and multiplying by a scalar. Numbers are more likely to be low than high, but there is theoretically no upper limit. The PRNG is seeded with the sector's coordinates before the stars are created, so the sectors are consistent between visits.

2) That would probably be the easiest way to do it. The draw back is that the size of "formations" is limited to the size of the neighborhood I look at. (My grand vision is to have a universe with clusters of unlimited size.)

What about using the surrounding clusters to base a range for number generation, just to create a better 'grouping' between sectors. Such as including the random value when calculating the average of the other sectors.

Example: (I suck at reducing to mathematic formula. I'm better at coding it)

Foreach (neighboring_value as x) {
    y *= reciprocal(x);
}

x = rand(sector_coordinates % 1);
y *= x;
y /= 10; //10 for 9 surrounding values and rand

new_sector_value = reciprocal(y) * scalar;

(okay, so it's a mishmash of coding, but I'm only doing an example not an actual program)

Something along these lines should be an average of the surrounding values, modified by a random value. You'd still maintain the possibility of unlimited cluster size, while maintaining some form of cohesion.

*Added: Okay, so saying it could give unlimited cluster size is like saying we can travel at the speed of light, but you could get progressively denser clusters.*