So lets set the stage here it's your basic path based tower defense game, but the twist I wanted to do was instead of haveing a collection of set towers to use. You would build you own towers out of a collection of parts. I'm going for a sort've modern day warfare approach to graphics, so imagine the turret on a battle ship. Each tower is broken down into 3 parts. The turret (base of the tower), The Barrel, the Munition it shoots, and possible a scope/tracking. What each part does: -The turret would determine how fast the tower can rotate. -The barrel effects rate of fire and range. -The munition effects what type and how much damage is caused. -The scope would effect accuracy, but still not sure if this would be too much work for what its worth. Game play would play as anyother td game does allowing you to pause at anytime to build new towers. So I'm wondering what would you like more. 1: Having a set amount of towers you can have per stage, like Plants vs Zombies. You would put together several towers that you want to be able to use before the stage even starts, and after the stage starts you would be unable to develop new towers. 2: Each time you go to place a tower on the map, you would put it together then, choosing the pieces as you see fit.

Perhaps each component could have an effect on all attributes, rate of fire, effective distance, accuracy, rotation etc.

Accuracy:
Barrel - length, rifling, material.
Turret - material, rotation speed.
Scope - magnification.
Munitions - shape, weight, charge.

The barrel length would have a definite effect on accuray, along with whether or not it is rifled. Perhaps a turret made of wood would be a less stable platform than steel, thus lowering accuracy. Rotation speed could also effect accuracy, the faster it rotates, the more difficult to get a good aim, especially if the scope has a high magnification, as the target wouldn't be seen until it is right in line of sight, so the turret would have less time to slow down for a good aim in the target. A bullet shaped munition would be more accurate than a grenade shaped munition. The bigger the charge(power used to accelerate the munition down the barrel) the faster and more accurate it is.

Rate of fire:
Barrel - rifling, material, cooling system
Munition - Charge

A rifled barrel may overheat faster than a non-rifled barrel. The material of the barrel could affect the rate at which it over heats, A good cooling system for the barrel could enable a faster rate of fire. The charge of the munition could also affect rate of fire(A low charge could be fired faster because it doesnt heat the barrel as fast.

Effective distance:
Barrel - material, length
Munition - Charge

A barrel material may effect frition on the fire munition, so less friction would mean it could go further. The charge of the munition and length of barrel would affect how fast the projectile is sent out of the barrel. Longer barrel means it is subjected to the preasure pushing on it toward the end of the barrel for a longer time.

Rotation:
Barrel - Length, material
Turret - Material, Power system

A longer/heavier barrel or heavier turret would take more energy to start and stop rotation. Rotation speed/acceleration could be effected by the power system as well.

As for when to design your towers, maybe you could allow a directory of previously created towers to choose from and the option to create a new tower at any point in the game. That way, the player wouldn't be limited to only certain types of towers made before the level started, and also wouldn't have to rebuild every singlr tower placed.

Anyway, just thought I'd share some design opinions. Hope it's useful! :)

I really like the idea of keeping track of previously built towers and will probably do something along those ideas. I like your other ideas but this game I want to keep as simple as possible as it is just a means to test and develop new features for my engine, So I'm not attempting anything spectacular. Just sticking to the casual, easy, to start type game, but with a little flexible play with not too many worries of losing. I've always liked sandbox type games that let you choose what ever you like, and don't confine you to a certain path of doings, just to beat a level.

I belive the falw of TDs are that you can usually find one perfect build order and the result will always be the same. And the choises aren't really that strategical.

Quote:

Original post by Kohake I belive the falw of TDs are that you can usually find one perfect build order and the result will always be the same. And the choises aren't really that strategical.

I completely agree that's why I want to give this one a sandbox feel, other than some completely outrageous choices most any tower can win the stage. I'm going for more of a visual appeal to the game. So far I've just been building on the mechanics of a single tower and it looks cool just waching the different types of towers firing randomly at stuff.

That would be cool. I think you need a slightly richer tower-construction system than just 3-4 sets of interchangeable sets of parts. Different munitions may require different barrels, or only able to be fired at certain rates. Different barrel/platform configurations may allow it be fired faster, or more accurately or farther. Don't be afraid to make it a little complicated and interesting.

As for your last question, it might be interesting if I had to pick a limited set of tower types before the game, which might add some strategy if different stages/enemies required different towers. Really though, I'd like to be able to pre-construct towers, and also pause in the middle to modify my designs or create new ones, especially if you set the stage up to be a long-term affair. Implement both, and just try them out on whoever you can to see which works better.

what are some of the classic tower defense games out there? for comparison.

You want to ensure that the TD has replay value and that there isn't a 'best build' scenario?

1] Give a point value system to mobs.
ie: 50 dmg = 1 pt, 1 armor = 1 pt etc etc (the sum of all the mobs stats are it's point value)
2] Give a point value for each level.
ie: lvl 1 = 50 pts, lvl 2 = 100 pts
3] Randomize the mobs but ensure their summed point value is always that of the level. This way you don't even have to have all the same mob type running the lanes on that level if you don't want.