Print

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

[Draco18s]

Spreading the points to broader categories is an interesting thought.

I don't know how well it would work, and would require a LOT of retooling to even test.

[Akhier the Dragon hearted]

   If it would require a lot of retooling then try something
easier first. I would hate for you to go through all the
trouble and in the end find out you just had some bit of
data out of whack that you could have easily fixed.
   Also I thought of a way to compare the ships if you
are using any kind of point buy. Just keep where the
points are put and figure out how much is spent in each
of the 3 category's or whatever way you want to
differentiate it and you would have useful numbers to
compare instead say trying to compare what 500 HP
means versus 11 Speed. Really even if you do not use
point buy as long as you can say that ##HP equals
# points you can compare them.

[Draco18s]

   Also I thought of a way to compare the ships if you
are using any kind of point buy. Just keep where the
points are put and figure out how much is spent in each
of the 3 category's or whatever way you want to
differentiate it and you would have useful numbers to
compare instead say trying to compare what 500 HP
means versus 11 Speed.

Not exactly doing that, but I am saving the random values which go into the function, and those could be compared easily.
And would be a 5 axis distance function on a grid of [0-1)

[Akhier the Dragon hearted]

   As long as you have a starting point to begin with. Once
you start figuring it out then who knows what you may
think of. The first step is always the hardest.

[squeakyReaper]

This is a pretty inspiring read.  I'm really enjoying it!  Here's a question though.  Say we go the route someone mentioned earlier, and introduce flying units, special armors, etc.  How do you find balance between Unit A, which cannot hit flying units, and Unit B, which flies?  Or will you have multiple triangles, possibly overlapping at some points?

[Akhier the Dragon hearted]

This is a pretty inspiring read.  I'm really enjoying it!  Here's a question though.  Say we go the route someone mentioned earlier, and introduce flying units, special armors, etc.  How do you find balance between Unit A, which cannot hit flying units, and Unit B, which flies?  Or will you have multiple triangles, possibly overlapping at some points?

   Probably by making it a triangle and not a line. The
flying unit beats the ground only unit, a new unit that
can attack air units but is a ground unit kills the flying
unit and the regular ground unit kills the ground unit
that can hit fliers.

[squeakyReaper]

Aye, just pointing out that it's a hard-stop way to have balance if you need it.  I know he wanted to avoid the "2x damage vs certain types" earlier, but I figure this bit is an easy and cheap way to make a starting triangle.

[Akhier the Dragon hearted]

   Lets worry about when we get there. At the moment
there is a problem with getting a actual rock paper scissor
thing going. At the moment if a fight goes one way it
should always go that way and if the "should" did not clue
you in it does not. There is no randomness in the actual
fights themselves yet they come out differently some times.

[squeakyReaper]

   Lets worry about when we get there.

Haha, well said.  Sorry, I like to speculate- I'm still wondering what sort of wonders I'll do with bees in the next dwarf fort after all!  The latest version I've found, after running it several times, seems pretty solid.  Usually takes ~17 rounds for me, though I only played through four times.  1 beat 2 and 3, 2 beat 3, in the first two generations.  It was balanced in the last two iterations though.  I did see 2 and 3 trading victories though.

[Akhier the Dragon hearted]

   I think that to find out whats wrong there needs to
be a bigger sample size. Like my earlier suggestion
about having it dump results to a text file and restart
so you could leave it run overnight and see the results.

[squeakyReaper]

   I think that to find out whats wrong there needs to
be a bigger sample size. Like my earlier suggestion
about having it dump results to a text file and restart
so you could leave it run overnight and see the results.

Dunno about overnight, but yeah.  Getting a huge sample size would be awesome.  A text dump would be nice.

Also a silly speculation, feel free to dismiss.  Should the goal of this be "generate a new set of units each time", or "generate units instead of having them pre-made"?  The first is an unoptimized system where you get something different every roll, the second is an optimized one where the game spits out the unit despite it not being hard coded.  I see perks to both!

[Akhier the Dragon hearted]

   I think we need to wait till Draco chimes in on this
seeing as we are not the one programing it and have
no actual control over it. But with how the units are
generated I think that as random as possible would be
good as long as the game is fun. I spend way to much
time generating Dwarf Fortress worlds just to see how
they turn out and not even playing them. If you can
make the units unique enough and fun enough people
could end up doing that with this.

[squeakyReaper]

We do have a degree of control over it.  This is less of a personal diary for him, and more of a small public blog; we're feeding some opinions in.  Still, you're right, his opinion matters a lot more than yours or mine on this matter.

Good point there.  "The Planet of Winds" and "The Eternal Plane" never look the same in Dwarf Fort, and screwing around with world gen is interesting...  I'm just considering the balance between efficient RTS units and random ones.  I do remember him mentioning earlier that he may try to apply archetypes to his units; that is, the high armored ones being named something Tank-related and the high damage ones being bombers.  As long as we get something within each archetype, or most archetypes fulfilled, I'm happy.

Funky idea; in some games, remove an archetype.  That is, deliberately find a sequence of triangles that work, and remove one of the factors.  Let's say my Tank had two checks, and we took out one (not both) of them.  Then the game changes as we try to find a new balance, wanting to create the now sole check for the Tank while protecting them.  Iunno, I don't think I'm wording this correctly.  Perhaps making one particular unit that much more important than others is unhealthy.

[Akhier the Dragon hearted]

   The thing is that we do not make the units at all.
They are generated at random though I am partial
to some kind of break up of the attributes but I already
went over that.

   HP and Armor; Dmg, AP, and RoF; and Speed
and Range are 3 categories of stats

   First the defense attributes then attack and finally
there is the tactical. I guess you could use that for
naming units. Defense oriented are tanks, attack ones
are DPS and Tactical would be a mixed bag, especially
depending on whatever ability's end up being added.

[squeakyReaper]

Seems straightforward.  Damage comes in different varieties and whatnot, though perhaps an...  obstacle course could be made?  Humor me for a bit.  What does a DPS unit do?  Kill stuff.  The game could figure out which is the most efficient at just straight up dealing damage over time, ignoring other factors, and make that the DPS related unit.  It's a jumble between damage, bullet multiplier, ROF and possibly even reloading if you can shoot fast enough.  Same for tanks; in one game it can figure high HP is what you need, resulting in sponges who can fight anti-armor units, but in another it thinks a high armor unit is the best tank, resulting in anti-mob types.