Page 1 of 7

Thoughts about data control

PostPosted: Mon May 25, 2009 10:31 pm
by DST
I want to explain that i'm not trying to be a professional telling everyone else what to do, rather i'm still learning how, and often debating the very methods i post.

This is a learning game we're all playing together, and though we may be playing different levels of this game, we're all playing the same game.

So it seems to me that there are 3 main issues that keep appearing over and over again on the forum.

    Controlling physics in Platformers
    Handling data over multiple actors - especially assigning/retrieving actor variables from other actors/scripts.
    Errors with crashing/freezing.

These forces are not simply obstacles to be overcome and left behind, they're more like mountains to climb - you are always climbing up one of the mountains, and you cannot ever reach the top, because the top would be perfection. There is no such thing, all the best company in the world can do is be "higher on the mountain than any other person has been".

The overall concept of game structure will assist you in all three; as you learn one, parts of the others become more obvious. There is a super sensible method for handling nearly everything, if we could only find out what that method is, and be able to comprehend it.

First let me point out that Rome was not built in a day. What this means in learning, is that you can't learn C++ in a day. To do so would damage your brain beyond repair, we generally refer to something like this as "Trauma".

Not only does it take time for your brain to build pathways(just like it takes time build muscles), so too the data is only useful if its relative to everything else; For instance, what good would it be to learn how to rebuild an automobile transmission if you haven't yet learned how to remove the housing and take the transmission out first?

So its gonna take time. That's all there is to it.

One of the problems with trying to learn these things when you don't have much money, is that generally, the more successful a product is, the less willing the authors are to share the details of it with you. Any of you who have looked at flash tutorials on the web know what i mean - thousands of incomplete, outdated tutorials, or ones that simply don't make any sense.

You can go to a university to learn flash, but that costs more money - you can google free tutorials for about 50$ a month, or whatever your internet connection costs. But college costs you 1,000$ a month.

And of course, cheaper schools pay their professors less, so they get crappier professors. The top dog programmers at Rockstar games don't have time to teach; they're too busy getting rich making games.


It would be nice if we could develop a few branches of ge programming and establish them as a standard. Perhaps there could be a community project, whereby teams take the existing demos and scripts and develop the smoothest established template for these;

Then the teams could post their final script/ged on a top-link page or post, where new users could find out things like anti-moonwalk and view follow actor right from the start.

And what i mean by branches is that there could be multiple versions of a script with multiple teams on it; if one team wanted to make an rpg setup by arrays while another set it up thru actor scripts - we could end up with multiple methods. And I think the methods would sort themselves out - especially if people posted about each one as they used them, so that methods that don't work in the long run could be dumped.

binarytree.png
A binary coordinate map

A binary tree can be used to locate screen areas; notice how you can get to the corner with only a few 1's and 0's - by the time you've narrowed it down to a 32x32 square, you've only used two bytes of data - or less, depending on how exact it needs to be. A targeting or pathfinding system could really benefit from this. (A* pathfinding uses 3 integers for each path cell - which could number in the hundreds!)

Its not that there's anything wrong here that needs to be fixed; its just that things could always be better. We could all be learning faster. This forum is a catalyst for learning, and we could develop it further and make it ....neater.

The reason animals, reptiles, fish, and insects all breath air is that oxygen is used in metabolism; Inside the cells glucose is converted to energy. It is literally set on fire. And of course, oxygen helps fire burn.

The difference between an anaerobic creature (one that doesn't need oxygen like mold and bacteria) and aerobic creatures like us is a final energy return of 19 to 1!!!! For every nineteen calories we can get from a piece of sugar, the bacteria can only get one!

Is your current game 19x better than your first one?

Re: Thoughts about data control

PostPosted: Mon May 25, 2009 10:48 pm
by makslane
DST, you are right. Is very important have templates for games and for actors.
For each type of game there are a set of common actors. These actors can be actor templates.

I hope see this feature in a future version of Game Editor.
Will be a day that you will make a game by just adding existing actors and customizing things like animation, velocity, size, ...
And everyone will create new templates to help the other users :-)

Read more here:
http://code.game-editor.com/ticket/6

Re: Thoughts about data control

PostPosted: Tue May 26, 2009 12:48 am
by jimmynewguy
DST wrote:Is your current game 19x better than your first one?
probably :lol:

But to be serious i completley agree with you. I want to help more, but there aren't that many people to help since 98% of the people on the forum are the smart ones :lol: . If i ever find an effecient method i'll tell you....

Re: Thoughts about data control

PostPosted: Tue May 26, 2009 3:29 pm
by Kalladdolf
As long as the scripts aren't too restricting (I mean, being able to customize them fully and helpful "//remarks" would be nice).
Good idea!

Re: Thoughts about data control

PostPosted: Wed May 27, 2009 7:57 am
by Fuzzy
I grinned when I saw the binary tree.

DST and I have lots of long talks about stuff like that!

There should be no doubt that he loves this community. From what I see, he shares everything he learns. He is not secretive at all.

I cannot add very much, he phrased it so well. There is indeed always an easier way.

If you are working on something and are finding that its complicated.. then you are doing it wrong. It probably means that you need to simplify or look at the problem in another way.

The best games you will make will be within your ability to understand. If you find yourself doing something and you dont know why... you shouldnt be doing it that way. That can be depressing, but its true. Learning takes time.

There are ways to enhance your understanding, to learn more quickly. I can suggest a few rules.

1. Plan it all out. On paper or in a document. Its critical in doing anything complicated.
2. Break things down into steps. Once you are at small steps, find a way to make it even smaller steps. This is why rule 1 is important.. so you dont lose your way.
3. Think generically. Walking = running = flying = shooting. All of these are just styles of moving, and they will be very similar in coding. You need to stereotype actions. Jumping is the same as a cannon ball launching.
4. Share what you learn. You will benefit more from helping than succeeding on your own. Look who shares most around here? They have the best games! Be like them.

Re: Thoughts about data control

PostPosted: Wed May 27, 2009 8:03 am
by Fuzzy
Having thought about it, we could list several game types and work on this. Each game would Start with a base type, and a version code. When people write modules for that game, their version number would indicate how and what it would fit in.

Re: Thoughts about data control

PostPosted: Wed May 27, 2009 11:00 am
by jimmynewguy
Fuzzy wrote:we could list several game types and work on this.

do we have an easy game type for me to start with? I'd be happy to help

Re: Thoughts about data control

PostPosted: Wed May 27, 2009 12:34 pm
by DST
Fuzzy wrote:
1. Plan it all out. On paper or in a document. Its critical in doing anything complicated.

I highly suggest FreeMind for fast document creation - give it a try, you'll be amazed!
http://freemind.sourceforge.net/wiki/in ... /Main_Page

Fuzzy wrote:3. Think generically. Walking = running = flying = shooting. All of these are just styles of moving, and they will be very similar in coding. You need to stereotype actions. Jumping is the same as a cannon ball launching.


If i could give people one piece of advice in learning to code (after 'share your info with others') this would be it - Its all the same to computer - and it should be all the same to a professional programmer.


Here's a sample document made with freemind - in this one i'm mapping out a game that involves hundreds of units - like Harvest: Massive Encounter. Definitely the sort of thing you want an outline for first!

Massive Encounter Game.jpeg

Re: Thoughts about data control

PostPosted: Wed May 27, 2009 9:44 pm
by Hblade
Does anyone know if DST Is human? O.o. He's way to frekin smart lol.

Re: Thoughts about data control

PostPosted: Wed May 27, 2009 10:57 pm
by DST
Hblade wrote:Does anyone know if DST Is human? O.o. He's way to frakin smart lol.

Well, I'm smarter than your average bear, but that's only part of what being smart is.

Listening to other people and learning from them - that's what smart is!

Programming the computer to work for you - that's what smart is!

Never putting personal pride ahead of learning - that's what smart is!


These things are all really hard sometimes! the last one is probably the hardest!
(especially when you get old!).

Re: Thoughts about data control

PostPosted: Thu May 28, 2009 3:28 am
by BlarghNRawr
them old people
*shakes fist*

i agree tho
the parts of that i understood seemed very helpful and useful in ge

Re: Thoughts about data control

PostPosted: Thu May 28, 2009 12:40 pm
by Kalladdolf
Programming a computer that works - that's what smart is!

Re: Thoughts about data control

PostPosted: Wed Jun 03, 2009 5:43 pm
by Bee-Ant
I think we need to make a new subforum called "Game Tutorial"
In this forum, you cant post anything. The content managed by admin only. And the topic is the chosen topic, chosen by admin. Sorted alphabeticaly. Anyone who want to share tutorial post it in Game Demos or anywhere. So that it wouldnt fulled by unnecesary topic and we can find it easily :D
I would like to share i found then :)

Re: Thoughts about data control

PostPosted: Wed Jun 03, 2009 5:56 pm
by Bee-Ant
Anyway, my current game is way better than the first one.
But its not 19x better.
I think about 10x.
Im not growing so well :?

I think the top of the mountain is imposible to reach. And we shouldnt reach it. Because we must avoid perfection. Perfection doesnt make us grow. If something is perfect, then theres no room left. Room to dream and make something better. Because youre already at the top, and wont go higher anymore. Thats why lets imagine that the top is unlimited, so that we can go higher, higher and higher...

Re: Thoughts about data control

PostPosted: Wed Jun 03, 2009 7:21 pm
by makslane
Bee-Ant wrote:I think we need to make a new subforum called "Game Tutorial"
In this forum, you cant post anything. The content managed by admin only. And the topic is the chosen topic, chosen by admin. Sorted alphabeticaly. Anyone who want to share tutorial post it in Game Demos or anywhere. So that it wouldnt fulled by unnecesary topic and we can find it easily :D
I would like to share i found then :)


For this kind of content, I think is better use the Wiki!