Quote:
Originally Posted by tater
Great post.
The thing is that many things can STILL be abstracted, but modern designers seem incapable of absorbing this. In WW2OL, there is a player who has as a day job modeling the effects of high explosives for the US military. He has for ages suggested a statistical approach—since that's what the pros use. Many players, and devs too, have the idea of tracking shrapnel through the air and checking to see where it hits you, etc, as the holy grail of modeling when "rolling a die" apparently does far better at predicting real world outcomes.
|
That makes a lot of sense, why bother wasting tons of development time and computing resources in creating a damage model that is not really more accurate than a statistical approach.
I am not pretending to be a game programmer or to know much about game development but my idea of a perfect simulator would be an open model, where developers and modders work together. The developers should create a basic simulation that would appeal to a larger audience. For example, instead of developing a submarine simulation, they should develop a martime warfare simulator with basic functionality for most platforms available in the sim. A working destroyer, sub, airplanes, and even merchant shipping. The modders then would use a released SDK and open programming hooks into the sim (LUA ???) to develop much more technical accurate models that are more interesting to the hard core sim gamer.
This approach would lead to lower development costs and larger community participation. Well, it's not like this model is anything new and it might not even work because most gamers expect a finished and polished product, which is quite obvious in reading the forms here.