Oolite Bulletins

You should be able to simply reverse the order of the "dials" elements in hud.plist in 1.77 to "fix" the bug. I didn't test this yet. If the "fix" does work, I suggest putting up the hud.plist on box.com (or similar) for others to download. The simplest solution, for t...

If I move ship's clock to last, I can set the scanner to second-to-last, and it is still visible, but it is in the "crazy color" mode. The following items can appear before the scanner without causing the crazy color effect: Targeting enhancement scanner zoom indicator stick sensitivity in...

In hud.plist, if I move the scanner entry: Before the speed bar: looks good, no bug After speed bar, but before ship's clock: crazy colors https://dl.dropboxusercontent.com/u/15097351/Oolite/crazy_colors1.png https://dl.dropboxusercontent.com/u/15097351/Oolite/crazy_colors2.png After ship's clock: f...

Offtopic: This was the first time I've ever used git bisect, and I've got to say, that was exciting. It's pretty neat how easily you can find a commit that introduces a bug, which allows you to then just brute-force your way through the changes made to find the specific line that caused the bug.

Ok, on branch master I modified drawDials like so: - (void) drawDials { z1 = [[UNIVERSE gameView] display_z]; // reset drawScanner flag. _scannerUpdated = NO; _compassUpdated = NO; // tight loop, we assume dialArray doesn't change in mid-draw. NSInteger i; const NSInteger dialArray_count = [dialArra...

Commenting out if (scannerUpdated) return; // there's never the need to draw the scanner twice per frame! in - (void) drawScanner:(NSDictionary *)info (line 894) does not fix the bug. It also does not fix the bug when I combine it with cim's suggestion. Found a fix! Replacing for (i = [dialArray cou...

Looking at that commit, the most obvious cause of the scanner disappearing would be if emptyDial was being set incorrectly. Could you try replacing BOOL emptyDial = ([info oo_floatForKey:ALPHA_KEY] == 0.0f); with BOOL emptyDial = NO; in the drawScanner function (around line 900 of HeadUpDisplay.m) ...

I've been away from Oolite for a while and after coming back found this bug affected me too. I did a git bisect and found the problematic commit to be: commit 37bd9b1923cac83ed30da061f971287332623b0a Author: Marc <kaks@oolite.org> Date: Thu Sep 13 14:32:20 2012 +0000 Mass lock refactoring, part I: -...

I'm not getting this behavior. Try unsetting all axis and button mappings, and then set them back to how you want. If you still have this problem, post the contents of ~/GNUstep/Defaults/.GNUstepDefaults

Not yet. School's been real busy lately. I'll get to it eventually, though.

Yeah, it makes sense. Basically you want the option of having different responses besides just linear. I don't think logarithmic is want you want, though. I've a made a few plots of the cubic function y = (1-c)*x^3 + c*x where c varies between 0 (slope at the origin is 0 - maximum non-linearity) and...

I took the following steps to get Oolite to compile for 64 bit arch: Install GNUStep. Copy jsautocfg.h from my 64 bit Firefox package to deps/Cross-platform-deps/SpiderMonkey/js/src. Make SpiderMonkey with "make -f Makefile.ref" cd to Linux_All_DBG.OBJ and move libjs.so to libjs.so.old, so...

I found one other issue related to yaw control (this time not joystick specific). Currently, when the player switches to an "info" screen (F5-F8) while flying, the pitch and roll slowly decrease to 0. Yaw doesn't. This small patch makes yaw decrease to 0, like pitch and roll. This behavior...

Updated code for analog joystick yaw control. I switched the yaw over to using a double, instead of NSPoint. I also didn't include the code that makes default yaw action half as large as pitch action. Also, I fixed what I'm pretty sure was a bug in the code. abs(virtualStick.x) (and y) were changed ...

Ok, I was reviewing the code a bit more, and while that code works, I'm going to change it just a bit. Basically I copied the method of getting the joystick info from the pitch/roll code, which uses getPitchRollAxis, which returns an NSPoint, which is 2 doubles. I used that, so virtualYaw.y was alwa...