Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[misko27]

This ideas been kicking around, finally decided to put it up. When I was vagely listening to watching the olympics, I noted how athlete A, who is generally better then Athlete B, does not always beat athlete B. Simply put, there is fluctuation in the level of skill easch athlete has. I mentally compared a dash to a hypothetical race in DF. What would happen is every single time is that the faster dwarf would win, always by the same margin, with out a single change in speed, every single time. This did not happen in the races, where one racer would pull ahead in one race, fall behind in another, etc.

Basically, I'm asking that skills "fluctuate" from the exact level, and that the top be more of a mean value. A example would be a snatcher having a bad day and getting caught by a dwarf 3 tiles away, pretty far for a ambush. or perhaps a marksdwarf could be on a roll and hit several gobbos in a row when he would have far less chance of that happening at his skill level.

Basically, it would give me a huge touch of realism, especially as I have organised dwarf, and gobbo-dog races, and they always end the same.

[Escapism]

This is actually present in many ways already. Drowziness gives a penalty while crafting, and I think hunger and thirst does as well. Exhaustion gives worse combat rolls. I reckon this is not exactly what you mean, but RNG is there for the other parts of it. Furthermore, I don't think that most players would even notice the changes.

[Di]

Yup, that's called RNG. Apart from running mentioned by you, everything else has varying results: highly trained soldier may lose to recruit, crafter may have luck and create masterful thing while being only novice.

[GreatWyrmGold]

Are there skills where the level always results in result X for skill level Y, but not level Z?

[Chaos Turtle]

To tell You the truth - the variation of dash for example would be very, very heavy for CPU, cause it would be needed to be add to every step of every creature in every frame... That means generating random number, adding it to normal speed, that just makes it two processes more for every unit in the game for making single step.

The idea od hunger, drowsinnes and other factors seems sufficient for randomization of aspects.

I dont think that it should be more complicated at least for most natural activities

[PetWolverine]

To tell You the truth - the variation of dash for example would be very, very heavy for CPU, cause it would be needed to be add to every step of every creature in every frame... That means generating random number, adding it to normal speed, that just makes it two processes more for every unit in the game for making single step.

That's not a very expensive operation. It takes a constant amount of time per step without increasing the number of steps taken. The constant is even relatively small compared to what else already has to be computed for each step - namely, finding a path will remain much more expensive than following the path.

[Damiac]

I don't see what's so complicated about it.
I assume when a creature moves, it just looks at his speed, and moves him every X frames, where X is smaller the higher speed is.
Why not just change the check from looking at speed, to looking at speed * Random number between 0.5 - 1.5

Then do that whenever computing any action.  Just apply some randomization to the deciding stat(s).

Combat needs a similar mechanism, as well as some damage tracking.  I understand bashing a guy in a metal helmet with a wooden crossbow won't do much, but doing it 200 times should result in some damage one way or another.

[GreatWyrmGold]

Term, if there's one more randomized number and one more multiplication every time a creature moves, that might not be a lot per move, but get a fortress of  80-90 dwarves, dozens of domestic animals, wild animals on the surface and caverns, maybe a forgotten beast or two, and a goblin siege comes...that's a big hit on FPS.