beginners first project, questions before i start

hello, i was wondering about some things for my upcoming first game project. I was thinking i would use a radar sensor for the ai, then it would start casting rays to random points around the area the radar sensor covered, if one then hits the object with the “player” property it would go hostile. Is this a good method or is there better ways?

also, can shape keys be used to make generators? like a character generator, for example?
just curious!

i dont know how many lights a scene can take but i read somewhere that its limited quite heavily, is there an occlusion method for lights?

also, as for custom mouse cursors, it seems obvious that all you have to do is map an object to the cursors location, so could i make an animated mouse cursor like this?
i was thinking the mouse cursor would change from blue to red if the mouse is over an enemy.

p.s does the bge have rag-doll physics? i want to make it so the player can move around the unconscious enemies.
p.s can you code shaders in python? for the mini-map i think i would need to either draw it all out and map it over the same areas or have a custom shader and use the texture module?

answers to these questions are much appreciated : )

It sounds like a very complex first project. Try something easier first like Tunnel Runner or a top down simple shooter. My first proper game was a simple one, Grid. I learnt so much from it, so so much. If you still think this is a good first game, then I’ll try to answer your questions. I’m not trying to put you off, just to give some advice!

i have been following tutorials , like the turret defense and the tunnel runner from goran, but this would be my first game, that i design and make : )
i will build it step by step, and follow tutorials to learn new things as i go along, but i probably am being overly ambitious, as always :stuck_out_tongue:
still, i find its easier to perform your best when forced into the corner with no escape! so i really want to just jump in and start making progress and learning. even if i have to put it on hold to do other things, or if i have to throw it out all together : )

I guess that’s a nice attitude. :slight_smile:
Do you know Python? (I don’t!)

Anything bigger than (even including) platformers can become complex very quickly without the global scope that Python provides. If you are serious about using the engine, I would suggest learning it to be a priority!
I agree that starting a big project forces you to learn a lot! But, small projects are useful to get used to the workflow, so start something very basic!

@RossBlenderArt i would say on some level i do! i don’t know much about the blender modules and i still have much to learn, but i think i am an intermediate python user @_@ hard to say, i know the basics at least :stuck_out_tongue:

@agoose77 well i thought i would make different parts of the game into smaller learning projects, then just polishing it up into the final game? i think my first steps should be getting the custom cursor and character movement, which should all be stuff i know anyway : ) well, not the cursor bit! i don’t know if there is an actual recommended method, but i thought it sounded pretty reasonable to map the object to the mouse? o.O i could map the object to the cursor , use a mouse over sensor to play the transition action which will change the cursor cross-hair from blue to red in an aesthetically pleasing way.

would it be best to use a mix of logic bricks and python? i think logic bricks are faster to run and also aster to set up, but there is a lot more flexibility with using python scripting? so if it can be done easily with logic bricks, i should, else i should use scripting ?

what would you suggest i should do before beginning this project?
i should probably finish the tutorials i was doing, but in goran’s tunnel runner tutorials he uses an f-curve and i don’t know how to get it working in 2.63! i guess that’s a start, figuring out how to get it working?
i have a lot of time to spend learning!

Logic Bricks are faster at the tasks that they are designed for, as they are written in C++. When things get complex, it can take more steps to acheive the intended function, and it as this point that Python is faster. However, you rarely notice any speed difference at all, so don’t be concerned with speed for now.

Thanks for that Angus, I always thought it was like a Python shortcut, because when you hover over a Logic Brick, it tells you the Python name, but I guess it’s just API help. Thanks for that.

@ RossBlenderArt: yeah i noticed that after searching the interwebz : )
the generator wasnt something i had in mind for my current game project idea, but in future i want to work on a turn based strategy with unit types, and i would want to vary the units. So i was thinking about it and at the time i read about shape keys, and wondered if using shape keys to create variation between the units would work. Im new to all of this (blender and programming) and dont know how its conventionally done, but it sounds plausible to me at least. that way i could add slight variations and randomize certain features and colors etc, and turn one character with a shape key into as many as i could need without having a clear repetition. With that i would just have to model one or two characters (male and female?) for each unit type, add some shape keys, have it randomly select, sticking to reasonable …whats the word im thinking of (been up all night again, not healthy!)…hmm…proportions! took a whole minute to think that one up, haha. so yeah, that was really just out of curiosity and looking forward to my future projects! But you can see why i would want to have it generate instead of having to model each manually, and then still being limited. It would be far less time consuming and give far greater results.

thats cool, when is 2.64 coming out anyway? will it be within a few months? p.s how could i get the features into 2.63 already? can i add it in myself? or is that a lot of work?

thanks for the link , i’ll check it out :smiley:
will do!
good to know, but i imagine it will be beyond me!

@agoose77: i see, thanks : ) i watched your “Introduction to python coding” video, at least i think its yours. I already knew it all but it was still entertaining. If im not mistaken you know c and c++? how hard is it to pick up c and c++ once you know python? python being multi-paradigm language shares a lot of similar concepts with c++ right? i briefly scanned over a pdf file i downloaded the other day titled ‘python2c’ and it seemed pretty fairly simple. Obviously i didn’t see everything though! but c has arrays instead of lists, you have to include libraries like you have to import modules in python right? you have to tell the compiler what type the variable will contain, whereas in python, you can just start assigning. you end every statement with a semi-colon? instead of indentation it uses curly braces? in python when you make a variable , say a = 10, then if you were to make b = a , and then change what a is, b would also change, because in python they refer to a location in memory, instead of actually holding the value? in c is it different? sorry for all these questions, just curious : )
much to learn.

p.s is it possible to bake textures and alpha-maps from a higher poly model , onto a very low poly object in such a way that it appears the same at a set distance, for level of detail? instead of having to model something several times with vastly different detail levels? is that possible? if it is, would it work or would it take up to much texture space? o.O

p.p.s would it be fairly simple to make it so that level of detail only switches meshes when not rendered by an occlusion culling algorithm? i dont like it when you see meshes switch so clearly, i would much rather have it switch outside of the view so that its not so apparent.

The two languages are similar in respect that they are both object orientated of sorts, and follow a programming workflow, but that aside they are quite different :s
The last two questions are both possible. The LOD is supposedly available to become part of blender soon.

do you think it would be possible for me to pick up c and c++ and be able to use it within the year? when do you think is the right time to pick up a new language? obviously once i have started and completed my game project at least :stuck_out_tongue:

that’s good to hear!

Python FIRST!!! C++ is a good language, but can be complex to grasp!

yeah, but when will i be ready?
i want to be able to contribute to blender, and i dont think (especially with the game engine) i can do much with python?
oh well, one thing at a time : ) still just learning the bge after all!