Psudo-3D RPG possible with gamesalad? (like Undercroft)

DigiChain

Ok, does anyone think this could be possible -

A RPG similiar in design to Undercroft (http://www.undercroft.com/) or The Quest?
Basically these games update the scene image every time the player clicks to move forward, or turns left or right (in 90degree steps).

I've been playing around trying to make this work - but am getting bogged down with a HUGE amount of attributes and condition to check for, and GS just ends up freezing (even with very small scenes)!

Does anyone have any idea how this could be possible in GS? And how you could go about designing some sort of grid reference system that would indicate where the player is standing, and in which direction they face (North, East, South or West) that uses very few attributes and can be referenced by all scene images - to determine whether they should be visible or not?

Or am I just deluding myself in even thinking this could be possible?!

Asobu_Games

Hmm well I am fairly new to gamesalad so not really sure, but I just had an idea for how to create this 'grid' system. For example say you want a 100 square grid, you could make 2 game attributes called something like 'grid position X' and 'grid position Y' which both go up to 10 each. If the player clicks 'forward' you would add +1 to the X or Y attribute (depending which axis they are on). If they click left or right this will change the active attribute (axis) from x to y or vice versa. Then clicking forward would add to that attribute.

Example: You are at position x=1,y=1 and want to move to x=3, y=2. You would press forward twice to add 2 to the x attribute, then press 'turn right' to change attribute to y and forward once to add 1 to the y attribute.

Then you could have rules like if x attribute = 3 and y attribute = 2, spawn image (whatever image you want here).

Does this make sense. Like I said I'm a newbie, so maybe a GS pro will be able to tell you that my idea is ridiculous. :P

DigiChain

Yes, it does make sense - and is a variation on what I have been doing. The problem seems to be that as soon as you have multiple images all checking for multiple attribute conditions, the whole thing just grinds to an agonising halt!

It seems that I need a more effective way of doing it. But maybe GS just isn't geared up for creating this kind of game?

Asobu_Games

Well it could be that your images are really big? Make sure the dimensions and resolution are no larger than necessary. If it's a solid background image (which doesn't need transparency) you can use a jpg instead of png. If it's for the small iphone screen you could probably get away with a pretty low quality (low file size) image. You could even have a very small amount (1 or 2 or 3) main background images to cut down on size and just add a few props to add variety.

It might be possible in GS, but keep it simple. Just a thought.

DigiChain

Well - my initial test was just using blank actors created in GS (not imported gfx) and then chucking on the rule that specified all the posible conditions searched for to see what it could handle. But after a few actors were created with these rules - game over man!

Basically, if you were to work on fairly small 10x10 grid, on which you could be facing in one of 4 directions it would mean 10x10x4 = 400 possible views needed in that particular scene.

But that then means that every image need to be either visible or not visible for all of those 400 possible condition. As soon as you have say 20 scene elements each checking watching/checking upto 400 conditions it seems unfeasible.

Unless someone knows something I don't?

Asobu_Games

You are right even for a small 10x10 game, 400 different views is still heaps. That's why I was suggesting having minimal views/images and repeating them often so GS won't explode. And to spice up the different views, add some small detail actors (also repeatable) like say a bush, tree or rock etc. But yes even with fewer game elements this doesn't seem ideal, so unless someone else can help I think GS might not be best for this type of game.

Maybe you could use different scenes to break it up into smaller grids so it's not trying to read so many conditions at once?

old_kipper

For myself I would hang on until arrays are available within GS, then use them to set up the map conditions and an actor to generate the environment on the fly. My knowledge is not that broad but from my understanding this might work.

kipper

DigiChain

Do any of the GS experts out there have any other ideas on how this may be possible?

I'm not a programmer, and know nothing about tables and arrays. How do they work - and how could they help?
Ultimately I'd still have to set the conditions for every possible viewpoint - wouldn't I?...