next »

[iffi]

Not my own code, but I laughed when I saw the fast inverse square root implementation as it appears in the Quake 3 source.
Snippet here:

Code:

i  = * ( long * ) &y;                       // evil floating point bit level hacking
i  = 0x5f3759df - ( i >> 1 );               // what the fuck?

[Waffle]

was checking my code the other day that i used for a school project..... funny thing is i got full credit on the assignment and the teacher said nothing.
check the title line

Code:

public static void main(String[] args) {
    DisplayClock frame = new DisplayClock();
    frame.setTitle("DisplayCock");
    frame.setLocationRelativeTo(null);
    ......
  } 

[mcc]

Two interesting things from the last couple days:

- Found myself typing the comment

Code:

 // FIXME: Guarantee this doesn't somehow cause OH FUCK.

...those last two words sort of coming out involuntarily as at that exact moment I realized (1) that yes, the commented block of code did cause heap corruption; and (2) why.

- So I write what I think is a very nice linear interpolate/decimate filter that works without having to allocate any scratch buffers! I use it to play back PCM samples at various speeds. But then I run it, and everything I play back with interpolation has this nasty hissing noise overlaid on it-- like there's a triangle wave of something like 11025 or 22050 hz overlaid on everything, getting louder and softer with the audio data. In other words, it sounds like classic aliasing. But it's loud, really loud-- it more or less overpowers the sample itself. And aliasing shouldn't be that clear or that audible either with the kind of interpolation I'm using or at the ratios I'm using (like, 2x interpolation)-- or so I thought. I check and re-check the code but there is clearly nothing wrong with it. I begin to doubt everything I know about sample interpolation. Is it possible you really just can't resample a sound without doing the low-pass thing?

Then I poke some more. And I discover something weird.

Apparently, when I prepared my sample files in Audacity, I sort of messed up. I'd thought my samples were all 16-bit mono PCM. It turns out, however, that they were all 16-bit stereo PCM-- and the right channel was always just silence! So all along, effectively, all of the samples I'd been playing back were slowed to 1/2 speed. Since they were all drum sounds this wasn't completely noticeable. And they were slowed effectively by inserting silence gaps between each PCM value-- something which normally wouldn't be noticeable. Except then after this accidental, inaudibly-aliased slowdown, I go and perform a second, intentional, inaudibly-aliased slowdown, and my second slowdown faithfully takes the aliasing from the first slowdown and blows it up in to VERY LOUD triangle waves...

I go and fix my source audio files and everything sounds fantastic!!!

[Theophilus]

was checking my code the other day that i used for a school project..... funny thing is i got full credit on the assignment and the teacher said nothing.
check the title line

public static void main(String[] args) {
    DisplayClock frame = new DisplayClock();
    frame.setTitle("DisplayCock");
    frame.setLocationRelativeTo(null);
    ......
  }

 

Took me a while to get it.

[frederiksen]

i don't know what this was or what it was intended to do, but this made me laugh when i found it in some old code, also it looks kinda nice.

Code:

var pr:Array = [1.49, 5.49, 5.99, 2.98, 3.49, 7.30];
var success:int = 0;
var miss:int = 0;
for (var i:int = 0; i < 6; i++)
{
	var pr0:Number = pr[i];
	for (var j:int = 0; j < 6; j++)
	{
		var pr1:Number = pr[j];
		for (var k:int = 0; k < 6; k++)
		{
			var pr2:Number = pr[k];
			for (var l:int = 0; l < 6; l++)
			{
				var pr3:Number = pr[l];
				for (var m:int = 0; m < 6; m++)
				{
					var pr4:Number = pr[m];
					for (var n:int = 0; n < 6; n++)
					{
						var pr5:Number = pr[n];
						if (pr0 != pr1 && pr0 != pr2 && pr0 != pr3 && pr0 != pr4 && pr0 != pr5 &&
						pr1 != pr2 && pr1 != pr3 && pr1 != pr4 && pr1 != pr5 &&
						pr2 != pr3 && pr2 != pr4 && pr2 != pr5 &&
						pr3 != pr4 && pr3 != pr5 &&
						pr4 != pr5)
						{
							if (pr0 < (pr1 + pr2))
							{
								if (pr1 + pr2 < pr3 + pr4)
								{
								if (pr3 + pr4 < pr5)
								{
									success += 1;
						                       trace(pr0+'__'+pr1+'__'+pr2+'__'+pr3+'__'+pr4+'__'+pr5)
								}
								else { miss += 1;}
							}
							else { miss += 1;}
						}
						else { miss += 1;}
										
										
					}
									
				}
								
			}
							
		}
						
	}
					
					
}			

edit: i remember what it was, i've seen a segment of the price is right. And there was a game where the contestant was given some articles and had to arrange them like this: the price of the first item must be lower than the second and third combined, the second and third must be lower than the fourth and fifth combined, those two must be lower than the sixth (p1<p2+p3<p4+p5<p6). I found it interesting because you had to actually think about when to use your lowest price, how to get above the price of the second and third while still being under the sixth etc. (Of course she didn't even know the prices).
But i'm just glad i hacked this down without thinking about good implementation.

[ஒழுக்கின்மை (Paul Eres)]

it's kind of funny how cluttered gml interface is, here's an example of the code for one of the creatures in my game



it's kind of a hassle to have to constantly keep flipping between the different windows and trying to arrange them in a way that i none hide behind the other

[Triplefox]

Code:

public static function drSeussFindsTheLowestVelocity() : Float
{
	var best = 999999.;
	for (Ball in idx)
	{
		var vl = Ball.GetLinearVelocity();
		
		var test = vl.Normalize();
		
		if (test<best)
			best = test;
	}
	return best;
}

Not the actual function name >.>

Edit: it was buggy.

[NFour]

I used to try to use this as often as possible on computer science tests/assignments

Code:

char zard;

[Average Software]

Code:

glPushAttrib(GL_CURRENT_BIT or GL_CURRENT_BIT);

Whoops.

[Average Software]

Code:

pragma Import(C, Grab_Mouse, "front_end_grab_mouse");
pragma Import(C, Ungrab_Mouse, "front_end_grab_mouse");

Took me a few minutes to figure out why Ungrab_Mouse was doing the wrong thing.

[Triplefox]

I decided to introduce the concept of "high" and "low" layers and joints...

Code:

	for (ent in Entity.gettagindex("tilebody"))
	{
		if (ent.isHigh && joint.startHigh)
		{
			ent1 = ent;
		}
		else if (ent.isHigh && joint.endHigh)
		{
			ent2 = ent;
		}
	}

I should probably wrap Entity.gettagindex into "bongHit()" to make it complete.

[rivon]

I don't get it. What's the problem or funny thing about it?

[BorisTheBrave]

I don't get it. What's the problem or funny thing about it?

Ent n. Slang for somebody who smokes cannabis.

Similarly "joint" and "high".

[st33d]

Good god, is there no escape from reddit anywhere?

[Evan Balster]

Good god, is there no escape from reddit anywhere?

Real life tends to be pretty reliable, depending on your friends.

(Though perhaps the Organization dispatches a token reddit-reader friend to everyone; I have two)

[TaintedFork]

I am working on the menu for a game right now, and I had defined each choice as a variable called xxOption, xx being whatever it represented.

This seemed normal, until I did:

Code:

optionsOption

for the options menu.

[rivon]

Or you can name it Preferences or Settings

[EliteKill]

I just made a method named getDown.

[dasdsad]

I never thought I'd post in here, but this just happened ..and now that I'm posting it, it doesn't seem that funny anymore.  

WELL, WHATEVER

Code:

#define CREATE_SUCCESS 0 /*Read in Borat Voice*/

[David Pittman]

Learning about Python modules and packages while working on a small prototyping project, I wrote the following:

Code:

AI = ai.ai.AI()