What Are You Working On? V13

[Darwin226]

[QUOTE=Richy19;25829508]But the feeling you get once it is complete is great as you have had to do EVERYTHING yourself and not have half the language do it for you[/QUOTE]

C++ is hardly a language that makes you do everything yourself...

[Richy13]

[QUOTE=Darwin226;25830179]C++ is hardly a language that makes you do everything yourself...[/QUOTE]

Well its the closest thing without going to assembly

[Darwin226]

Why?
What makes it closer than C or D or any language that compiles to asm?

[Richy13]

Touche
What i meant with this anyway, is that you get everything done like with a managed language

[Darwin226]

The only good feeling after making a successful program in an unmanaged language I'd get is because the language gave me more opportunities to screw up and I didn't.
The garbage collector for example. It takes care of memory management for you (for the most part).
Why do it yourself? You won't manage your memory that much better manually while there are infinite memory leak possibilities.

[Jallen]

I've just spent the last 2 hours on my presentation for "Introduction to Software Engineering" for my second year of my Computer Science degree. We have to pick a subject relevant to software engineering and give a 10 minute presentation on it. I decided on Object Oriented Programming - Design, Implementation and Testing.

SO BORING.

But at least I have to do pretty much no research because I know most of this shit already.

[Liquid Helium]

[QUOTE=NovembrDobby;25829450]Most times I see people posting their code here they just get ripped open and it all seems so elitist[/QUOTE]
I don't think anyone is ripping on the guys who post this code as much as the books that teach them shit practises. It kinda scares me because I am learning c++ myself now and how am I supposed to know if the stuff I am being taught is utter rubbish or not? I have probably avoided it by reading a book that was recommended by reddit rather than just picking any old one up, but still.

[Darwin226]

Rendering as points / Rendering as triangles
[img]http://lh3.ggpht.com/_ZoH0ul2PquE/TNGk0ZckQYI/AAAAAAAAArE/vWJV6R4GCaw/teapot.png[/img]

Any idea what could be wrong?

The triangles are properly connected. Even when I set the fragment shader to output white, I get a black screen.

[Overv]

Try disabling culling.

[20k]

Err.. that wasnt meant to happen
[IMG]http://i53.tinypic.com/2luqs9g.png[/IMG]
But it looks awesome

[Darwin226]

[QUOTE=Overv;25831110]Try disabling culling.[/QUOTE]

It's not enabled :(

[Chandler]

[QUOTE=Icedshot;25831121]Err.. that wasnt meant to happen
[img_thumb]http://i53.tinypic.com/2luqs9g.png[/img_thumb]
But it looks awesome[/QUOTE]

I think you just accidentally created some fractal geometry. congratulations! :D

[Darwin226]

[QUOTE=Darwin226;25831287]It's not enabled :([/QUOTE]

I'm glad you guys think that's funny but this is one of those problem where I have no idea where to start...

[Jallen]

[QUOTE=Chandler;25831464]I think you just accidentally created some fractal geometry. congratulations! :D[/QUOTE]

No, looks like he just looped through colour values without a limit, so it looped round back to white from black etc.

[20k]

I added in a variable when trying to fix it. Now it just gives me a slider of how fractalicious it is

[IMG]http://i52.tinypic.com/ih44g3.png[/IMG]

[editline]3rd November 2010[/editline]

[QUOTE=Jallen;25831576]No, looks like he just looped through colour values without a limit, so it looped round back to white from black etc.[/QUOTE]

Yeah im pretty sure ive done something along the lines of that, though oddly enough the only modifications i made (before) to the colour components were dividing

[efeX]

[QUOTE=Richy19;25829508]But the feeling you get once it is complete is great as you have had to do EVERYTHING yourself and not have half the language do it for you[/QUOTE]

haha. oh wait you're serious?

[likesoursugar]

[QUOTE=Darwin226;25831090]Rendering as points / Rendering as triangles
[img_thumb]http://lh3.ggpht.com/_ZoH0ul2PquE/TNGk0ZckQYI/AAAAAAAAArE/vWJV6R4GCaw/teapot.png[/img_thumb]
[/QUOTE]

0.45584 frame time :O
That's 2.2 FPS

[Jookia]

What's that? std::list::iterator doesn't support binary addition you say?

                menuEntries.insert((++currentEntry)--, newEntries.begin(),
                    newEntries.end());
                
                menuEntries.erase(currentEntry--);

[Darwin226]

[QUOTE=likesoursugar;25832127]0.45584 frame time :O
That's 2.2 FPS[/QUOTE]

That's milliseconds and it's around 0.01-0.03 the most of the time.
It's just a teacup model.

[Jallen]

[QUOTE=bootv2;25832200]could be the devcpp 5.0 beta that's bitching with me then.[/QUOTE]

DevC++ is oldddddddd.

Get Visual C++ Express 2010.

[likesoursugar]

[QUOTE=Darwin226;25832461]That's milliseconds and it's around 0.01-0.03 the most of the time.[/QUOTE]
Explains a lot!
[QUOTE]
It's just a teacup model.[/QUOTE]
Yeah, that's why I was worried :P

[BlkDucky]

[QUOTE=bootv2;25832200]could be the devcpp 5.0 beta that's bitching with me then.[/QUOTE]

DevC++ is outdated, switch to Code::blocks or better yet Visual Studio Express 2010.

Edit: Didn't notice Jallen's post, somehow. :v:

[backwardspy]

[QUOTE=BlkDucky;25832758]DevC++ is outdated, switch to Code::blocks or better yet Visual Studio Express 2010.[/QUOTE]
I rated you late, then looked up and realised that you even rated Jallen "agree"
I am now confused as to why you made that post :v:

[Koridas]

I'm currently working on figuring out Visual Basic. Yes I know its a noob language, I just want something easy to be my first.

[20k]

[QUOTE=Richy19;25827401]-Rawr-[/QUOTE]
Just spotted your pm, no idea if my reply has worked

[Frugle]

I'm working on Invaders demake. It is last exercise on my C# book. I got drawing (gdi+), invader and player movement already done. For tomorrow I have shooting and enemy waves left :v:.

[backwardspy]

I know there were a couple of people who wanted help with tile-based platformer collision physics.

I finally worked out a nice method that seems pretty solid to me, and I did so by drawing up a little picture. I added some text because my friend wanted to know my method. I dunno how helpful this'll be to anyone else, but it's here if you want it.

(click for full size)
[img_thumb]http://dl.dropbox.com/u/6661951/Images/platformer%20collisions.png[/img_thumb]

That's only for a downwards collision with the top of a tile. All other sides are the same, just modify the coordinates accordingly.

[grlira]

I'm working on getting familiar to XNA and currently have a LinkedList of renderable objects setup. The basic setup is
The update() function of the main game class goes trough each element and calls its update() method.

The draw() function of the main game class does the same, but calling draw() instead.

Now, I want the objects to be removed from the LinkedList if they go offscreen, and can't for the love of god get this to work (probably out of unfamiliarity with C#). Here's what I have:

[cpp]
//bla bla bla code
// we're in the update() function

foreach (RedBall redBall in RedBalls) { //RedBall is my renderable object, RedBalls is my LinkedList
if (redBall.getPos().Y > 300) { //300 hardcoded for testing purposes atm
RedBalls.Remove(redBall);
}
redBall.Update(gameTime);
}

// bla bla bla more code
[/cpp]

Now, what this does, is complain that "Collection was modified after the enumerator was instantiated".
I know what it means, but can't figure how to solve it. Any ideas?

Also, dumb question, with XNA, how on earth do you get the dimensions (resolution) of the game window (or better even, the draw area)?

thanks for any help

[Darwin226]

[QUOTE=Chris220;25833877]I know there were a couple of people who wanted help with tile-based platformer collision physics.

I finally worked out a nice method that seems pretty solid to me, and I did so by drawing up a little picture. I added some text because my friend wanted to know my method. I dunno how helpful this'll be to anyone else, but it's here if you want it.

(click for full size)
[img_thumb]http://dl.dropbox.com/u/6661951/Images/platformer%20collisions.png[/img_thumb]

That's only for a downwards collision with the top of a tile. All other sides are the same, just modify the coordinates accordingly.[/QUOTE]

I think that people have a problem with side to side collisions.

Also, what if the falling player ends up in the floor tile, but more than half the way in.
If you included the head-tile collision as well, he should snap to the bottom of the tile instead of the top.

[Robber]

[QUOTE=Chris220;25833877]I know there were a couple of people who wanted help with tile-based platformer collision physics.

I finally worked out a nice method that seems pretty solid to me, and I did so by drawing up a little picture. I added some text because my friend wanted to know my method. I dunno how helpful this'll be to anyone else, but it's here if you want it.

(click for full size)
[img_thumb]http://dl.dropbox.com/u/6661951/Images/platformer%20collisions.png[/img_thumb]

That's only for a downwards collision with the top of a tile. All other sides are the same, just modify the coordinates accordingly.[/QUOTE]
I would imagine a line from each of the moving block's points that represents movement direction and speed and intersect them with each of the static block's lines. That way it can't go through small blocks at high speeds and it can probably be optimized and simplified quite a bit.