What are you working on? January 2012

[neos300]

I feel like I may have missed the fad again, but can anyone point me to some articles/tutorials for raytracing?

[Overv]

[QUOTE=neos300;34225181]I feel like I may have missed the fad again, but can anyone point me to some articles/tutorials for raytracing?[/QUOTE]

There isn't really anything to it except for some basic geometry.

[jonnopon3000]

I've been working on my coursework game project all weekend (Maze Blitz in case anyone remembers it), and I've now got two new things to show!
You lucky, lucky people :v:

First, a demo of some new art and the freeze ray and trip mine weapons in their first states:
[video=youtube;Im4xDib8-U4]http://www.youtube.com/watch?v=Im4xDib8-U4[/video]

Second, the first draft of the in-game music, which is going down well with my end-user and some friends.
[video=youtube;fqTwY9xdzVM]http://www.youtube.com/watch?v=fqTwY9xdzVM[/video]

Feedback on the music would make me feel all warm inside :)

[20k]

[IMG]http://dl.dropbox.com/u/33076954/Whynot.PNG[/IMG]

Because why not

[ooo]

-lol whoops how did i get on here-

[Smashmaster]

[QUOTE=Icedshot;34225812][IMG]http://dl.dropbox.com/u/33076954/Whynot.PNG[/IMG]

Because why not[/QUOTE]

Bioluminescent oceans?

[BlkDucky]

[QUOTE=Smashmaster;34226224]Bioluminescent oceans?[/QUOTE]

that implies that the oceans are alive.

[img]http://i.somethingawful.com/forumsystem/emoticons/emot-ohdear.png[/img]

[danharibo]

[QUOTE=synthiac;34226160]okay that's better
[img]http://i.imgur.com/sCFGo.png[/img][/QUOTE]Gold star for effort.

[acpm]

-fucking ninja'd-

[Matte]

[thumb]http://i.imgur.com/jGmPF.png[/thumb]

Still some artifacts, but it's getting better. Everything is done in the shader.

Ray-tracing has so many possibilities compared to polygon rendering!

[neos300]

[img]http://img838.imageshack.us/img838/5348/screenshot2012011514131.png[/img]
Behold, raytracing!
It doesn't handle the z axis well though.

[pollyzoid]

I've been looking into making a skeleton ESP for iris for a while, since I absolutely loved the one I made for [url=http://dl.dropbox.com/u/3760480/pics/oa_camp20004.jpg]a previous hack[/url]. In GMod I'd simply use player:LookupBone and player:GetBonePosition, but the problem is that those are non-virtual methods in C_BaseAnimating, so I can't simply call them. I looked into alternative methods, and found C_BaseAnimating::SetupBones, which would output an array of bone matrices. I considered using it, but that would mean I'd have to deal with the mess C_BaseEntity is with multiple inheritance.
So I just opened client.dll in IDA, dug up LookupBone and GetBonePosition, and wrote signatures for them, then scanned when the method is first called.

[cpp]void
source::player::get_bone_position(int bone, Vector &pos, QAngle &ang)
{
static GetBonePosition_fn GetBonePosition = nullptr;

if (GetBonePosition == nullptr)
{
msg << debug << tag("ply") << "Finding GetBonePosition... ";
iris::memory::signature gbp(iris::g_hack->config()["sig"]["GetBonePosition"]);
GetBonePosition = gbp.get<GetBonePosition_fn>();
assert(GetBonePosition);
msg << "Found! (0x" << static_cast<void*>(GetBonePosition) << ")" << std::endl;
}

GetBonePosition(reinterpret_cast<C_BasePlayer*>(entity_), bone, pos, ang);
}[/cpp]

I had working GetBonePosition! So the first thing I did was find a dump of bone numbers and remake the Hades skeleton ESP.

[thumb]http://puu.sh/duPE[/thumb]

Whoops. I made two mistakes here. 1: I used bone numbers. 2: I used the wrong bone numbers. #1 I did because I thought the numbers were static for all players, and I was wrong. They can change depending on the player's model, which is why different classes have different skeletons in the image. #2 was because the dump I found was for HL2 based games, namely GMod, CS:S and the rest. TF2 changed into a different system with different bones.
Both were easily fixed when I decided to change to bone names instead of numbers. The HL2 format is "ValveBiped.Bip01_Pelvis", the new TF2 format is simply "bip_pelvis". You pass the name as the argument to LookupBone, and it returns the correct bone number if it exists. The different names were easily solved by writing the list twice and having a check whether the game is TF2 or not.

The result works correctly and also works in GMod and CS:S.
[thumb]http://puu.sh/duWw[/thumb] [thumb]http://puu.sh/duWj[/thumb] [thumb]http://puu.sh/duY0[/thumb] [thumb]http://puu.sh/duZ9[/thumb]

The ESP itself added about 30 lines of code:

[lua]local bones = {
{ BONE.Head, BONE.Neck },
{ BONE.Neck, BONE.Pelvis },
{ BONE.Neck, BONE.LeftUpperArm },
{ BONE.Neck, BONE.RightUpperArm },
{ BONE.LeftUpperArm, BONE.LeftLowerArm },
{ BONE.RightUpperArm, BONE.RightLowerArm },
{ BONE.LeftLowerArm, BONE.LeftHand },
{ BONE.RightLowerArm, BONE.RightHand },
{ BONE.Pelvis, BONE.LeftHip },
{ BONE.Pelvis, BONE.RightHip },
{ BONE.LeftHip, BONE.LeftKnee },
{ BONE.RightHip, BONE.RightKnee },
{ BONE.LeftKnee, BONE.LeftFoot },
{ BONE.RightKnee, BONE.RightFoot }
}

local function draw_skeleton( ply )
local col_team = team.get_color( ply.team )

surface:set_color( col_team )
local pos1, pos2
for _, b in ipairs( bones ) do
pos1 = ply:get_bone_position( ply:lookup_bone( b[1] ) ):to_screen()
pos2 = ply:get_bone_position( ply:lookup_bone( b[2] ) ):to_screen()
if not pos1.behind and not pos2.behind then
surface:draw_line( pos1.x, pos1.y, pos2.x, pos2.y )
end
end
end[/lua]

[acpm]

[media]http://www.youtube.com/watch?v=x1TsOHyJPpw[/media]

I know it's not strictly WAYWO, but one thing I have been "working on" is looking for a job. I don't think I'm hirable with the position postings I've seen online, and I can't find any entry level/paid internship positions in Phoenix or Glendale, Arizona.

I got a kick out of this, and it's sorta what it feels like. Dunno if anyone else out there feels this way.

[benji2015]

[img]http://puu.sh/dvds[/img]

Trying to make a horizon curveish shader.

It doesn't stand out too well in the above image, but have a look here:

[img]http://puu.sh/dv8U[/img]
(edit: okay that one's pretty bad)

It still has some weird angles to it though.. hmm

[Nigey Nige]

I love how non-programmers get properly mindblown by parallax scrolling.

[r0b0tsquid]

Took a bit of a break, made a balancing robot with my old NXT:
[video=youtube;4qhJypiS8_g]http://www.youtube.com/watch?v=4qhJypiS8_g[/video]
Measures tilt by measuring reflected light (from the red LED), so it's really sensitive to lighting conditions (sorry). On the programming side of things is some simple PID control, made a bit more interesting by the lack of anything but unsigned integers :) (that's why it oscillates when I push it - if I set the integral gain any lower, it doesn't really work at all.)
Here's a slightly clearer picture:
[quote][img]http://i.imgur.com/o6YW1.jpg[/img][/quote]
Not entirely programming, but mostly :v:

[Shane]

Almost feel ashamed as my projects are always so simple compared to everyone elses genius! Oh well it keeps me busy :)

Still working on a "server" I guess it would be. Has full MySQL support so far, had LUA support but I removed it for now, it was mostly for configuration anyways.
Also working on command entry, so far it works really well and looks professional as it doesn't leave the standard console output.
[img]http://dl.dropbox.com/u/1540469/Server.png[/img]

Also am learning to use XML documentation which I really am liking as it makes the code look "neater" and also outputs all the docs to a file on build!
[img]http://dl.dropbox.com/u/1540469/Code.png[/img]

[Meatpuppet]

Ugh. I'm working on a game in SFML that involves spaceships and planets. I've stopped working on it because I just don't know what the next step is. A lot of stuff doesn't work, and I'm just trying to get the core to work. I have no drive to learn what to do, like making an update function for the rotation of my ship when it changes images (from no thrusters to thrusters), or fixing functions that I broke. Any tips for getting a "drive"?

[Blade]

[QUOTE=amcfaggot;34227365][media]http://www.youtube.com/watch?v=x1TsOHyJPpw[/media]

I know it's not strictly WAYWO, but one thing I have been "working on" is looking for a job. I don't think I'm hirable with the position postings I've seen online, and I can't find any entry level/paid internship positions in Phoenix or Glendale, Arizona.

I got a kick out of this, and it's sorta what it feels like. Dunno if anyone else out there feels this way.[/QUOTE]

"Wanted, Administrator with 10 years experience in Windows 2008"

Really programmer job listings are a joke. I have never seen a real one.

[Robber]

I haven't ever seen ridiculous requirements like that.
For instance this is the job I currently have and I got it straight after school and I'm doing really well [url]http://infonova.at/7_954_ENG_HTML.htm[/url]

[QUOTE=r0b0tsquid;34227763]Took a bit of a break, made a balancing robot with my old NXT:
Measures tilt by measuring reflected light (from the red LED), so it's really sensitive to lighting conditions (sorry). On the programming side of things is some simple PID control, made a bit more interesting by the lack of anything but unsigned integers :) (that's why it oscillates when I push it - if I set the integral gain any lower, it doesn't really work at all.)
Here's a slightly clearer picture:

Not entirely programming, but mostly :v:[/QUOTE]
Using the light sensor as gyroscope is really clever. I love it, the whole project! I've wanted to do something like this forever. I even bought an Arduino, but never really had the time and dedication to do anything with it.

[zenby]

[QUOTE=Maurice;34222444][img]http://i.imgur.com/BMB4s.gif[/img]

Working on the castle repeats. It puts the blocks into the middle of the map, so if I turn it off, it resumes to the original map (those pipes). The problem is that the physic objects don't get moved for the repeat properly yet, so I'll have to look into that.

[editline].[/editline]

Fixed, yay.
[img]http://i.imgur.com/0JiQz.png[/img][/QUOTE]

Finally :v:

[editline]15th January 2012[/editline]

You guys just made my day. It's nice to see so many people trying out raytracing and seeing how awesome it is.

[Jawalt]

Okay I've hacked in a lexer and a parser, both are < 200 lines combined since the languages' grammer is so simple. Some of the shortenedness is from using Lua, but it should clock in similarly in C.

Instead of the current tables I'm using I'll probably do a switch statement in C using the hashed keywords and operators as keys.

Here's the current code, I'm not sure where to go from here. I have tokens and I've made sure the program follows the grammar of the language. I'm using a recursive descent parser. (I based it greatly on the Wikipedia example :P)

WARNING. I am ignoring best practice and easier ways to do things at the moment, and not handling tons of cases I should be because I'm in 'get something on the screen' mode.

[url]http://pastebin.com/iMR9yMLH[/url]

[url]http://pastebin.com/HUp4fvyU[/url]

How it's being used:

require 'lexer'
require 'parser'

input = io.open("C:/Users/Matt/Desktop/script.s", "r")
input = input:read("*a")
output = tokenize(input)

final = parse(output)

And finally the code it's verifying:

main
{
	0 104 101 108 108 111
	hash
	print
}

hash
{
	5381 swap
	0 != while
	{
		pop pop swap dup
		5 << + +
		swap
		
		0 !=
	}
}

Where should I go from here? I'm thinking about doing some kind of two-pass system where I first do inclusions etc. and then recognizing the functions.

And then laying out the 'meat' of the code.

[efeX]

[QUOTE=amcfaggot;34227365][media]http://www.youtube.com/watch?v=x1TsOHyJPpw[/media]

I know it's not strictly WAYWO, but one thing I have been "working on" is looking for a job. I don't think I'm hirable with the position postings I've seen online, and I can't find any entry level/paid internship positions in Phoenix or Glendale, Arizona.

I got a kick out of this, and it's sorta what it feels like. Dunno if anyone else out there feels this way.[/QUOTE]

arizona sucks leave asap

there's really nothing for programming here

[r0b0tsquid]

Whoah, apparently the NBC firmware has float support! :D
That means [i]signed numbers[/i] as well, should get rid of some of the more ridiculous code and let me tune it better.
Unfortunately, it also means I'll have to change these lines:

sum = sum + (err - 100) * 1;
if (sum <= 9000)
    sum = 9001;   //ensure sum is always OVER NINE THOUSAND
else if (sum > 11000)
    sum = 11000;

:v:

[Darwin226]

[QUOTE=Robber;34228496]I haven't ever seen ridiculous requirements like that.
For instance this is the job I currently have and I got it straight after school and I'm doing really well [url]http://infonova.at/7_954_ENG_HTML.htm[/url]

Using the light sensor as gyroscope is really clever. I love it, the whole project! I've wanted to do something like this forever. I even bought an Arduino, but never really had the time and dedication to do anything with it.[/QUOTE]

[img]http://infonova.at/img/careers_sub.jpg[/img]
These people are horrible at high-fiving.

[Dlaor-guy]

[QUOTE=Darwin226;34229266][img]http://infonova.at/img/careers_sub.jpg[/img]
These people are horrible at high-fiving.[/QUOTE]
A high-five with more than 2 people at the same time never works out. NEVER.

[editline]15th January 2012[/editline]

Also, unrelevant but when is the new game programming competition starting?

[neos300]

So I'm comparing different path finding algorithims solving mazes, and something is surprising me.
I'm doing best first, and it takes 87 ms to solve the maze, but breadth first (which should be slower) takes only 5.
I found out it's the sorting the list of nodes is what's causing the speed problem, is there a better way to store the nodes/sort them other then the vanilla std::list/std::sort?

[Darwin226]

[QUOTE=neos300;34229452]So I'm comparing different path finding algorithims solving mazes, and something is surprising me.
I'm doing best first, and it takes 87 ms to solve the maze, but breadth first (which should be slower) takes only 5.
I found out it's the sorting the list of nodes is what's causing the speed problem, is there a better way to store the nodes/sort them other then the vanilla std::list/std::sort?[/QUOTE]

Why are you sorting anything? IIRC BFS and DFS don't need any sorting.
As for the speed, DFS finds A path faster but BFS finds the BEST path faster.

[Xeon06]

[QUOTE=amcfaggot;34227365][media]http://www.youtube.com/watch?v=x1TsOHyJPpw[/media]

I know it's not strictly WAYWO, but one thing I have been "working on" is looking for a job. I don't think I'm hirable with the position postings I've seen online, and I can't find any entry level/paid internship positions in Phoenix or Glendale, Arizona.

I got a kick out of this, and it's sorta what it feels like. Dunno if anyone else out there feels this way.[/QUOTE]

Just apply. I read somewhere that those restrictions are just to weed out people. If you're confident and can show them some stuff, they'll want to hire you. Those listings and position titles are a joke. You might find, after having gotten a junior developer job, that you'll be better than most senior guys at the shop you're at. Either way, you have [i]nothing[/i] to lose by applying and so much to gain.

[swift and shift]

[QUOTE=Shane;34227935]Almost feel ashamed as my projects are always so simple compared to everyone elses genius! Oh well it keeps me busy :)

Still working on a "server" I guess it would be. Has full MySQL support so far, had LUA support but I removed it for now, it was mostly for configuration anyways.
Also working on command entry, so far it works really well and looks professional as it doesn't leave the standard console output.
[img]http://dl.dropbox.com/u/1540469/Server.png[/img]

Also am learning to use XML documentation which I really am liking as it makes the code look "neater" and also outputs all the docs to a file on build!
[img]http://dl.dropbox.com/u/1540469/Code.png[/img][/QUOTE]

what does it "serve"