There's a difference between asking questions about specific problems and asking questions that are covered in one or more chapters in a book. The question above is of the latter variety. Expecting a useful answer in a forum post about something that pages dozens of pages to go over is kinda silly. Yes, you answered his question, but how useful was it really? Did your answer give him the ability to go off and start using sprite sheets? Probably not, so was your post worth making?

Maybe it's just because I see dozens of these kinds of posts every day and the laziness of people these days pisses me off. Usually the next post in this kind of discussion is "Can you give me some source code to do this?"

_________________
Bored? Head on over to my blog and see what I'm up to.

Microsoft XNA MVP

I'm inclined to side with Machaira. If you're going to post a question on a forum, please show evidence of having looked for the answer elsewhere first. Explain as much as you know about the background of your question. Even if you have looked for information first but don't mention any of your prior research, it looks lazy for everyone else who is meant to help you.

Hope I don't sound too gruff.

_________________
Whatever the mind can conceive and believe, it can achieve

Maybe my post didn't give him any specifics, or working code, but really -- was your "RTFM" post any more helpful?



Well, is it too much to give the person the benefit of the doubt until they ask for such a thing?

The forum rules talk about things like erring on the side of friendliness. I know I'm not a shining example of friendliness most of the time, but yeesh...

_________________
What most people don't understand about "enlightenment" is that it is not an end-goal; but where you find yourself just before taking a new "first step."

I'm all about the tough love. A person learns nothing by being spoon-fed.

I'm happy to answer specific questions.

_________________
Bored? Head on over to my blog and see what I'm up to.

Microsoft XNA MVP

I never ask 4 source codes, I only ask about the algorithms.

But I must say...that spritesheet thing was my fault. I should have asked a more specific question that wasnt answered elsewhere. Sorry.

so my question about spritesheets version 2 is:
In Metal Assault, a 2d MMO side-scrolling game, there's a bunch of outfits, armors, weapons. Do they have a ton of spritesheets? Or is there a trick to do it? (Just wanna know)

I know. I said that in my experience, general questions are usually followed by a request for source.


Probably a bunch. It's a much used technique.

_________________
Bored? Head on over to my blog and see what I'm up to.

Microsoft XNA MVP

Never played it, but most likely there are sprite sheets for every type of outfit/armor/etc., and they are overlaid on top of the sprite image. This way, you can have characters with different skin colors, different outfit configurations, etc. without needing to make each combination in images. Think of it as a "dress-up" game.

_________________
What most people don't understand about "enlightenment" is that it is not an end-goal; but where you find yourself just before taking a new "first step."

I agree, but I still don't see the point of insulting someone, when it would work much better to just point them in the right direction.

Anyway, I'll get off the subject... this thread is already plenty derailed.

_________________
What most people don't understand about "enlightenment" is that it is not an end-goal; but where you find yourself just before taking a new "first step."

Yeah this is how it usually goes down. This is best example I could find of a sprite sheet with separate pieces for hair, weapon, clothing, etc.

_________________
Whatever the mind can conceive and believe, it can achieve

Thnx. I am having a problem with the fps timing and sprite changing.

In 10 fps how many times should I change the sprite to give the illusion of motion?

and I think I will code the gameLoop function later after I get the other technical stuff right so can you pros tell how to make it work in every scenario? Or maybe I am taking a bad step here. Maybe I should think small now.

It depends on the range of movement. It's generally accepted that 16fps is enough for most animation to appear as smooth movement. (http://en.wikipedia.org/wiki/Persistence_of_vision) However, you can get away with less for small movements.

Consistent motion is probably more important, using time based modelling rather than frame timing adds complexity, but provides a better result.

_________________
10 PRINT "Bad Monkey ";
20 GOTO 10

As Codehead suggests, this is not something that has a hard-and-fast "correct" answer... it depends on many factors -- the complexity of the animation, the number of frames, and so on. Then it still takes a subjective eye to find the motion that "feels" right.



The game loop is pretty much the first thing I always work on -- in fact, it was the first feature I coded for my JavaScript game library (which I plan to release soon). The game loop is one of those parts of a real-time game that is pivotal, and should not wait, IMO.

To make it work in "every scenario" is pretty simple. The main idea is to separate the game loop from the game logic-state, and have the game loop control the logic-state. Also, you will want to measure the time between updates, and feed that to your logic-state so that it can "adjust" itself according to how much time has passed.

Let's say your logic-state has a function update(elapsedTime), which moves a sprite according to time. You would do something like this:



For example, of course -- you would do things like collision handling in your own code, but that gives you an idea how it works.

_________________
What most people don't understand about "enlightenment" is that it is not an end-goal; but where you find yourself just before taking a new "first step."

thnx.


Looking forward to it! Maybe I would dump my project and use your framework.

BTW what's the difference between a game engine and a game framework/library? I am confused.

The differences between them are a bit fuzzy, but a game engine has more functionality geared to a specific game/genre, while a framework or library has more generalized game-related functionality.



I'll post something on the forum (maybe in the Announcements section) once I have something available to download. I mainly am working on it because I got tired of writing the same/similar code for different projects, as well as having bits and pieces of that code in separate scripts. I already have the game loop/logic, rendering, input, assets management, and audio parts done, but I still need to add a couple more features (mainly related to images) before it will be released.

_________________
What most people don't understand about "enlightenment" is that it is not an end-goal; but where you find yourself just before taking a new "first step."

Then I will keep an eye on it.

I am trying to keep mine simple. But I have rewritten my code twice all over again, even though each renewal was faster than the last. I think it's becoz I didn't sit down n plan my game engine. Either way, html5/javascript coding is addictive.

Me: Need help.....
You: again!! COME ON!!

I was looking at other game engines one of them is http://love2d.org/wiki/love.timer, and the thing that my project needs is the timer class. I read the documentation of those engines. But I can't seem to grasp the concept. And can't answer this questions I keep coming up with:

- Why do I need all those properties?
- What methods would use them?
- How would this class make life (game coding) easier?

They are pretty much the same question.....

You need a timer to do plenty of things in a game. Most importantly, to keep your program running the same across all different types of hardware. If someone is running the game on a better rig, the game's logic should still run at the same pace and the animations should run at the same pace.

To achieve this, the best way is to separate out your drawing code from your game logic code. Decide on a number of "game" fps you want, like 100, and then keep the game logic running at that rate. The draw code should run without modifying the game state in any way.

_________________
Whatever the mind can conceive and believe, it can achieve

Here's a more JavaScript-friendly game timer --



You'll need to have an update(elapsedTime) function that handles your logic's input/AI/drawing/etc., but other than that this function is pretty complete. (I pulled this from my library... if anyone has any suggestions or "gotchas" to make it any tighter, let me know!)

_________________
What most people don't understand about "enlightenment" is that it is not an end-goal; but where you find yourself just before taking a new "first step."

Algorithm would have been just fine. Now Machaira got something to tease me
But I do appreciate the code. Thanks


Got it. But I got some questions...why limit the game time to 0.5? Shouldn't we reset the game time once in a while?


Appreciate the javascript syntax, one thing...it seems to have more of a function structure than a class, so how do I get access to the privates variables? Don't I need them for calculation?

I see, basically you use the game time together with some logic to update state/data and animate.
So I can just add this algorithm to the game loop. And I thought I had to write another class.

The code I pasted here is slightly modified, but the only thing you need to do with the function in its current state is define an update(elapsedTime) function that does your drawing/input and other stuff, and a desiredFramerate variable. The desiredFramerate variable is 1000 divided by how many frames per second you want.

Oh, and I suppose having a variable called 'running' and setting it to a non-false value would help. Almost forgot about that one. Once you have all that set up, just make one call to cycleFunction(), and the rest handles itself.

_________________
What most people don't understand about "enlightenment" is that it is not an end-goal; but where you find yourself just before taking a new "first step."