GSim version 0.26

[testtubegames]

I present, GSim Version 0.26

“Menu-Mania”

New in this Version:
-Lots o’ bug fixes (center-of-mass key destruction, fixed stars that overstay their welcome, load codes that don’t work on cue, etc)
-ToolTips!
-Title Screen
-Clock (Toggle-able)
-Grid lines zoom well, and actually *mean* something
-Load codes now use the xspot, yspot and zoom parameters
-More popup menus with options:
-Time Menu: Manipulate time, faster, slower
-Quality Menu: Change from Euler to RK4 and more
-Physics Menu: change the value of G
-Visual Menu: a few options here, and a bunch of placeholders
-Barnes-Hut algorithm added in — speeds things up when you have a lot of planets — now can run 2000 planets without much problem
-Hold shift when using the cursor to select multiple planets
-Fling planets in circular orbits by holding down ‘c’, will orbit either the only star/planet around, or the one that is currently ‘selected’
-Navigate the world with arrow keys/wasd
-Zooming changed to +- keys and mouse scroll
-Press ‘backspace’ to delete selected objects

Yet to do:
-Prevent crashes — I’ve done a little, but there’s a long way to go. And the manual mode means you can really crash at will
-Redraw orbits after toggling them (and add in a way to destroy orbits without destroying planets)
-Redraw orbits after changing the tracking mode, right now they just clear
-Figure out the best ‘auto’ mode for the quality settings
-… the rest of the stuff on the to-do list.

So yeah a lot. Have fun and throw some suggestions my way... so that you can have even *more* fun in version .27! Also, there were tons of bug reports for the last version. I think I tracked down all of them (that I didn't mention in the 'yet to do'), but if I missed any others, my apologies!

[wtg62]

Whoa... nicely done!
This version, not only looks awesome, but performs awesome!

My only complaint so far is,
Seems when I open the physics menu, the game doesn't dim, causing the planets (especially the white ones) to make the text hard to read.

Other than that, this version is the best one yet in my opinion.
Keep up the good work, TTG!

Edit 1: I notice you changed the 'View world from it' button to say 'Ptolemy'... what the heck? xD

Edit 2: Are you considering trying to implement a more advanced color picker in the next version? Also, are you considering adding the ability to edit existing planets

Edit 3: Seems I can't put an object in a circular orbit when there is more than 2 objects on-screen.

Edit 4: When clicking 'Learn more' in the quality menu, it closes out. Is this a placeholder button? Is the 'about' button in the main menu a placeholder as well? (it does nothing)

[testtubegames]

Great - glad you like it!

Seems when I open the physics menu, the game doesn't dim

Ah, easy fix. Forgot to copy and paste the background in for that menu. Let me get that right now... and... fixed

I notice you changed the 'View world from it' button to say 'Ptolemy'

Yup, Ptolemy of the Ptolemaic System, where earth was treated as the center of the universe. Replaced by the sun-centered idea when Copernicus came around. Figured it was a fitting historical nod to placing a certain planet as the center-point.

Are you considering trying to implement a more advanced color picker in the next version? Also, are you considering adding the ability to edit existing planets

I would like to add more colors, it'll just take a little rewrite of the code. In fact, I have a 'color picker' all picked out, ready to go. And planet-editing will come -- my plan is to extend the pop-up menu that comes up when you select a planet... so you can change mass/color/speed/position/fixed-or-not-fixed.

Edit 3: Seems I can't put an object in a circular orbit when there is more than 2 objects on-screen.

Ah, if there's only one star around, you center around it. If there's just one planet, you center around that. If there's more, you need to have selected the planet you want to orbit... otherwise it won't work. Your comment, though, made me think there must be a more intuitive system. Perhaps you automatically orbit the closest one? (And you can select another one if you'd rather orbit it.) Could work.

When clicking 'Learn more' in the quality menu, it closes out

Yup, those two buttons are placeholders for now. Forgot to mention that/grey them out. I'll hide them for now to avoid confusion.

[wtg62]

Thanks for your reply!
Although you might wanna make all of that a little more clear in the game. (i.e. add tooltips for buttons in the 'planet selected' menu)
Now all the circularization works much better! (Although yes, I agree, detection of the closest object would be a good choice... unless you got thousands of nearby objects)

Also for circular orbits, could you perhaps make it show the red line as if I had dragged my mouse to the location that would create a circular orbit? That way I can release C, match the lines up and make a similar orbit, but perhaps maybe add some eccentricity if I wanted to.

Also for time, is it possible to add a few more modes (The modes being: 0.01x, 0.05x, 50x, 100x), or will said modes cause performance issues?

Oh, I also noticed... using E notation (a.k.a. computer/calculators form of scientific notation) to input mass doesn't work properly... and will just create a negative mass. I.e. inputting "10000000" gives you 1E+07, which becomes a ridiculously large negamass.

[Stargate38]

What's wrong with this code? Both stars and the asteroid disappear as soon as it's loaded:

Code: Select all

Gravity Fun at TestTubeGames.com: [ForceG: -2,Qual: 1,Zoom: 0.25,xSet: 0,ySet: 0], [x0: -25,y0: 0,vx: 0,vy: -1,t0: 0,who: 2,m: 1000,c: 1], [x0: 100,y0: 0,vx: 0,vy: 4,t0: 0,who: 2,m: 250,c: 0], [x0: 400,y0: 0,vx: 0,vy: 8,t0: 0,who: 3,m: 0,c: 0]

Make sure you set the gravity to 0.4^r before you load it.

[wtg62]

Speaking of which... The ForceG variables need to be used (and have support for x^r). There also needs to be a tag for the new 'G' variable.

[wtg62]

What's wrong with this code? Both stars and the asteroid disappear as soon as it's loaded:

Code: Select all

Gravity Fun at TestTubeGames.com: [ForceG: -2,Qual: 1,Zoom: 0.25,xSet: 0,ySet: 0], [x0: -25,y0: 0,vx: 0,vy: -1,t0: 0,who: 2,m: 1000,c: 1], [x0: 100,y0: 0,vx: 0,vy: 4,t0: 0,who: 2,m: 250,c: 0], [x0: 400,y0: 0,vx: 0,vy: 8,t0: 0,who: 3,m: 0,c: 0]

Make sure you set the gravity to 0.4^r before you load it.

Hmm, works for me.
Found a bug as well. The time menu gets moved out of place after changing the f(r) law or G variable.

[exfret]

Glitch: When Sun's masses become large enough, they turn negative
Suggestion: I think the size should vary more with the mass (it's kind of hard to see now, that is, if you've added that in)
Suggestion: I think the zoom in/out restrictions need to be less restrict-y (that's just my opinion, though)
Question: For the Suggestions forum (this should probably go in there), were you going to put my suggestions on "The List"? I don't see it there.
Bug: When zoomed out, sometimes, planet previews are not accurate (ie, sometimes they are actually smaller)

[wtg62]

Also just noticed, when you load a save that had a 'star' in it, if you select said star, the menu says it's a 'planet'.
I'm guessing this is due to stars and planets sharing the same entity.

[testtubegames]

...add tooltips for buttons in the 'planet selected' menu...Also for circular orbits, could you perhaps make it show the red line as if I had dragged my mouse to the location...Also for time, is it possible to add a few more modes (The modes being: 0.01x, 0.05x, 50x, 100x)...
Oh, I also noticed... using E notation... will just create a negative mass. I.e. inputting "10000000" gives you 1E+07, which becomes a ridiculously large negamass.

Yes to more tooltips (can't believe I forgot those buttons), yes to red line (something I wanted, too, but ran out of time for the prev build), and yes to move time modes. Now, the extra time modes will likely lead to weird stuff (100x, for instance, will probably mean no stable orbits), but I can add it in. The more options the merrier.

The 1E+07 and negamass thing is strange. Presumably one of those issues that computers have when numbers get too big. I'll figure out a way to fix that.

Speaking of which... The ForceG variables need to be used (and have support for x^r). There also needs to be a tag for the new 'G' variable.

Indeed. It'll mean a bit of an overhaul for the code system -- which I'm always wary of (not because it's hard, just because I need to worry about backwards compatibility and that kinda thing). Once I get up the courage, I'll add it in.

Found a bug as well. The time menu gets moved out of place after changing the f(r) law or G variable.

Ha, that it does! Nice find.

What's wrong with this code? Both stars and the asteroid disappear as soon as it's loaded:

Hmm... might be the fact that I forgot to update the Linux version for you. (Whoops!) I'll get that posted shortly!

Suggestion: I think the size should vary more with the mass
Suggestion: I think the zoom in/out restrictions need to be less restrict-y (that's just my opinion, though)
Question: For the Suggestions forum (this should probably go in there), were you going to put my suggestions on "The List"? I don't see it there.
Bug: When zoomed out, sometimes, planet previews are not accurate (ie, sometimes they are actually smaller)

You're absolutely right about planet radius... when I look at it now, it doesn't match with what I envisioned. I must have changed something along the way, because last I recall, planets could get huuuuge. Zooming: I might ease up on the restrictions on it later. But I'm going to wait to do that until I have a better handle on keeping the simulation from crashing due to zooming. I'll look into the planet previews issue.

And I'll throw those suggestions up on 'The List' -- I held off doing it immediately, 'cause I wanted to ruminate a bit on them... but then I just forgot to update it Thanks for the reminder.

Also just noticed, when you load a save that had a 'star' in it, if you select said star, the menu says it's a 'planet'.
I'm guessing this is due to stars and planets sharing the same entity.

Quite right! I'll need to add in a work-around