TestTubeGames Forums

robly18 wrote:I have a slight complaint...

OMG that's awful! Sorry to hear that -- yeah, that's gotta be fixed.

exfret wrote:-Words change when you switch from and to Ptolemy (so you no which one you're in)
-A nice button to change to the 'common' speeds more quickly (ie slowest/x0.1, regular/x1, and fast/x10)
-Grid lines strange for tracking
-Zooming w/ lots of orbits is strange. I want a screenshot of what happens before you change it though. (For some reason, I couldn't find a simple screen-capturing software).
-Examples start in default colors (ie, if your next asteroid is going to be brown, the example's asteroids are brown), which is actually kind of cool, except it also starts with default position or something, so now all the examples are starting in the lower right-hand corner for me.
-Circles are only calculated for the r^-2 law. They don't work for other laws (that or gravity is off or something).

-I agree with the tracking thing. Beyond just knowing which you're in, I'd also like to mark the planet you're tracking in some way. (Of course, it's in the middle of the screen, but once you highlight another object, it can still be confusing who you're tracking.)
-You're talking about the fact they're a little jolt-y and wobbly, right? I'll give it a shot and see what I can do
-Yeah, occasionally there are artifacts around. There's also still the issue that the game lags severely when you zoom in real close w/ tons of orbits. Gotta fix both of those problems
-Confused by the 'default' positions issue. It should always reset to the origin at a scale of 1.0. Playing around to see if I can reproduce the problem, if you figure out any other steps of reproducing it, let me know.
-Oh, good catch. It should work for everything of the form r^a, but I forgot to update it for things like a^r and such.

-Yes, especially if you move around.
-You saw my 2nd thing, right?
-It's not just jolt-y and wobbly, they also randomly dim/light up.
-Artifacts means old orbits, right? Yeah, it's really strange and really cool at the same time.
-I was confused, too. I'll tell you if it happens again.
-Yeah, I wanted to look into r^(-r) and I was disappointed when the circles didn't work out.

How about change the merge button into a menu that sets collision style?

What should be in the menu:

• Default - Current behavior with merging off.
• No collisons - Absolutely no collisions.
• Merge - Current merge behavior.
• Shatter - Smaller impacting object breaks into fragments (amount of fragments = mass/2 or mass/4.)
• Disappear - Smaller impacting object simply disappears.

Also considering the save code... if he were to make his own 'programming language', I'm sure a simple mark-up language wouldn't be too hard. He's already got one anyhow! (Although... it's the save code )

TestTubeGames wrote:-'Fixed stars' that move... aka bodies that move yet don't feel the pull of other objects

I suggest that you call these 'Inert stars'

I'd also really like to see the zoom be much less restrictive... (More powers of ten for the grid?)

What's the difference between 'default' and 'no collisions'? You could also define boundaries for which each of those things happens (eg 0-1/15 of bigger disappears, 1/15-1/5 fractures, and 1/5-1 merges). Finally, I think there should be another 'no collisions' that is the same as default/no collisions, except it accurately shows the path of two close-together planets instead of the corner-cutting the current sim does. This'll be laggier, but very useful in some cases. You could also make it so that the sim makes more accurate orbits of some things over others. Oh, and don't forget dust.

Apparently I was going to make this post but I deleted it.

wtg62 wrote:

• Shatter - Smaller impacting object breaks into fragments (amount of fragments = mass/2 or mass/4.)

That would convert the object into all size 2 or 4 particles, which would be laggy even for medium-sized planets. Maybe make it break into 2 to 4 fragments depending on relative velocity and mass?

exfret wrote:-Yeah, I wanted to look into r^(-r) and I was disappointed when the circles didn't work out.

http://codepad.org/HlHw9gKp
Create new paste, change the list of radii of orbits and centralmass, and the force formula. After you load it, you need to pause and run to the force input to set it. (I've set a small delay before loading for this)

exfret wrote:What's the difference between 'default' and 'no collisions'? You could also define boundaries for which each of those things happens (eg 0-1/15 of bigger disappears, 1/15-1/5 fractures, and 1/5-1 merges). Finally, I think there should be another 'no collisions' that is the same as default/no collisions, except it accurately shows the path of two close-together planets instead of the corner-cutting the current sim does. This'll be laggier, but very useful in some cases. You could also make it so that the sim makes more accurate orbits of some things over others. Oh, and don't forget dust.

Difference: Ever noticed how if something impacts a fixed-star, poof, it disappears? Well if there were absolutely no collisions, that wouldn't happen.

I do agree, mass/2 fragments would be laggy, so just 2-8 fragments depending on size should do.

wtg62 wrote:Difference: Ever noticed how if something impacts a fixed-star, poof, it disappears? Well if there were absolutely no collisions, that wouldn't happen.

Oh. Why does it even do that? It seems like we should just eliminate 'Default' altogether.

Fragmenting and shattering? Sounds amazing to look at. I'm in.

Lag of course is the main issue, but if you've got tons of things on the screen, you deserve a bit o' lag

There is indeed a special variant of collisions that happen by 'default', though I'll have to look and remind myself exactly how it's implemented. Something like: if you just have one fixed star, an asteroid can crash into it and disappear. Because if someone is taking an asteroid and just dropping it (the first thing people generally do)... it should hit the star and stop, not fly right through it (and perhaps off to infinity depending on the sim's accuracy). Once more stars are in, the sim eases up, and stops destroying asteroids. So, yeah, I can add in further options there.

On a related note, I would like to improve the near-range behavior of the simulation -- kind of getting at what exfret was saying about the jagged orbit lines. That way 'strange' stuff (like things flying to infinity) would happen a lot less.

Let me add one mode to that list.

Bounce: Where objects bounce off each-other on impact.

Might be an issue coding in and/or with performance, however.
Just a suggestion though!

wtg62 wrote:Let me add one mode to that list.
Bounce

That also sounds fun. I imagine these all being less scientific-oriented, and more just flashy-showy oriented. (Which I am *TOTALLY* down with). But knowing the creativity of all of you, I bet we could get some cool stuff out of them. (An interesting display of conservation of angular momentum or whatnot.)