There''s one other issue that''s going to crop up later on down the track if you succed with this, AI. The more flexability that you give to the player the harder time you will have doing the AI.
What if the AI used predefined moves, perhaps with a little variation, while player has full control?
AI doesn''t seem to be the issue to me, it is the modelling that I can see being a problem. -- but the graphical side seems to be bugging me the most (I am not sure how smooth the fighting would look).
Skeletal animation and skinning?
One big problem I see is preventing the player from cutting himself with the sword , and preventing the sword from going to unnatural positions.
Basically, the way that I see the AI side of things, the creature has a set routine (maybe every creature of that race has a number of different ''routines'' that they could use) and basically they tell exactly what to attempt to attack and under what circumstances. (ie. Move: "Kick in Balls" When: "After Player Attack" Likeliness:"1%" [too nasty for more often ] or Move: "Headbut" When:"After successful block" Likeliness:"10%")
Then, the player can learn the routines, as well as there can be random events... Hmm.. Fun...
Not if it''s all PRERENDERED... I am doing 2D ISO remember? .
You can still use polygons. Actually, I don''t know how you could allow full control with prerendered graphics... Or are you even thinking of full control, or just a complex set of predefined moves?
I am thinking of BitMasks but for the mouse. Different numbers in a Bitmasky type of array (same size as the enemy and yourself) and by clicking on certain numbers, you can find the attack. From there, you find the current position, and try to animate towards the starting position for the next move (or wait until the move has completed). From here, the animation goes to the start of the next move, and the next move is carried out. You don''t need polygons, or skeletons... Just a hell of a lot of sprites
As far as animating it goes: With a bit of thought you might be able to separate the animation into separate "body parts": upper body, lower body, left arm, right arm. Combining these parts in a single character animation would reduce the total number of sprites necessary.
For more complex animations you could look at skeletons and skinning ( as has been suggested ) or voxel technology for relatively small voxel densities, i.e. like in some of the recent command and conquer titles.
People might not remember what you said, or what you did, but they will always remember how you made them feel. ~ (V)^|) |<é!t|-| ~
It's only funny 'till someone gets hurt.And then it's just hilarious.Unless it's you.
To do with the body part images, I was thinking this myself. Diablo II did this... It might help if that was just how you stored it, but you did a composition back-buffer render to have the actual images in a useful form before the blit.
That is getting technical on the graphics side though, let us just stick to the theoretical fighting side until we have this thing designed to a Tee
, and preventing the sword from going to unnatural positions.
] or Move: "Headbut" When:"After successful block" Likeliness:"10%")
