|
|
|
|
| Re: Battle Engine breakdown |
Sun, 14 July 2013 17:04   |
|
|
|
Hint:
My biggest mistake with the battle engine usually is that I see tactical behaviour where none is but just chance.
If you run tests, use 20+ tokens of the same ship design. This helps a lot to see where and when ships move according to a strict rule or just randomize their movement and actions.
Bystander wrote on Sun, 14 July 2013 18:06
"Each token has an attractiveness rating. This is used in both working out where ships move to and which ships are shot at first."
I have been told by some expert players that battle attractiveness does not enter into the movement decision, or at least cannot override a damage ratio order.
I have the same impression.
Quote:I ran some simple tests that seem to agree with their opinions. But now I have a recent real-game example of a stack of range two battleships that have a range 3 enemy cruiser stack pinned up against the edge of the board and ready to be killed when the battleships move to the center of the board to get next to a piece of chaff. They then get shot by range 3 cruisers and cannot return fire with them. All ships had max damage ratio orders.
The battleship behaviour might have been caused by the battle engine giving priority to targets which are primary targets AND can be killed with 1 shot.
Quote:But I have been told by some expert players that max damage ratio orders can allow ships to move out of range of enemies in anticipation of enemies moving forward later.
I doubt this. More likely they stayed out of the battle because ANOTHER ship engaged the enemy and by staying away "maximise damage_done/damage_taken" was optimized.
Quote:Is a damage ratio of 0 given / 0 taken better than a ratio of 10 given / 500 taken?
See above. The battle engine seems to calculate the ratio not per token but with all warships together.
So the answere to the above question is: 10 given / 500 taken... but not because of the ratio but to fulfill their battleorders to engage the enemy. It looks like at least 1 token must engage the enemy, then the other tokens can follow a "ratio"-behaviour.
The other questions... I'd like to know, too.
Report message to a moderator
|
|
|
|
| Re: Battle Engine breakdown |
Mon, 15 July 2013 15:00 |
|
 Bystander |
 | |
Messages: 141
Registered: June 2003
Location: Tampa, Florida, USA | 
|
|
Thanks, Altruist.
I am willing to possibly chalk up two bizarre battle incidents I have seen recently to your first observation, that sometimes the small random factors built into the game look like tatical decisions.
1) Clarification on situation where stack of range 2 battleships had range 3 cruisers pinned at range 2 but moved away to go after chaff:
The battleships had whittled the cruisers down from 39 to 18 on previous round, so had enough firepower (and initiative) to completely kill cruisers next round without taking any damage. Instead, BBs took two moves toward center of board, getting next to single chaff and killed it. Then the cruisers took one move toward center to get back in range 3 of BBs and shot them for 1824 of armor damage, but no BB kills.
Possible explanations:
A) Two simultaneous random moves - or random selection of new target
B) BBs predicted no loss of ships to themselves, so didn't care what was targeted next
C) Battleships with Enigma Pulsar Engines just might act weird because they move so fast
2) Regarding another weird situation concerning an undergunned navy where every token had max damage ratio orders:
I believe that all but one token followed the definition in the Stars! help file about movement for damage orders. That you have to get in range AND stay in range of at least one enemy token of primary (or failing that secondary) target category before you can start getting conservative with your movement. All beam tokens went forward into risky territory to get into at least sniping range of some enemy token(s), except the largest, most powerful (in terms of shield damage), and one of the least attractive tokens. That token took one step forward and two back. Consequently, it ended the round incapable of shooting ANY enemies with its range 3 beams or sappers. Next round it moved in, and did fairly well - considering it had much less protection.
A) Battleboard was too crowded to move forward on 1st round. This despite Stars! manual saying that any number of tokens can occupy a location. Stars! help file is somewhat inaccurate/outdated.
B) No targets found. Sapper/beam ship was high on fleet number. Stars help file does mention there is a PREFERENCE for ships to target enemies that have not previously been targeted. Perhaps every enemy token was already targeted multiple times and the game just gave up finding a target.
C) Random movement
D) The game anticipates that more enemies will move their full allotment movement points forward, but because of their battle orders, some enemies hang back for sniping. This sounds reasonable, excepts for the tests that I set up that indicate ships are ignorant of future enemy movement/initiative and this also goes against Stars! manual description of movement for damage battle orders.
E) Token was held back because of a fleet determination of max damage ratio. But it seems to me that if damage ratio is determined at a fleet level, an undergunned collection of ships would OFTEN stop sending ships forward as soon as one fleet got in range. That would be pretty timid movement.
F) A large stack of battleships with an unbalanced design of 14 sappers and 6 range 3 beams will rarely not engage, similar to the known bug where range3 sapper / range2 beam designs NEVER close to range 2 if the sappers alone can bring down all the shields of an enemy stack. Never able to recreate this, but recently uncovered a similar observation about mostly sapper with range3 beam design staying out of battle for a round from a very old post.
Report message to a moderator
|
|
|
|
|
|
| Re: Battle Engine breakdown |
Tue, 16 July 2013 13:40 |
|
| skoormit |
 | Lieutenant |
Messages: 665
Registered: July 2008
Location: Alabama | |
|
Two thoughts:
a) I wonder if ships are aware of initiative firing order, and therefore sometimes move to maximize damage/ratio knowing exactly who will fire before whom.
b) I have seen a 1-on-1 ship battle that indicates that ships do in some instances anticipate enemy token movement. The scenario (I do not recall precise details):
1) Ship A: Range 3 beamer.
2) Ship B: Range 2 beamer.
3) Ship A is about 40% heavier than ship b, and hence will always move last (both ships have movement of at least 1).
4) Ship A has more movement than ship B, and hence will get an extra move in some rounds.
5) Both ships have some form of "do some damage" orders (not disengage or min damage, etc)
In one round, ship A moved from range3 to range4 with it's first movement of the turn, apparently anticipating (correctly) that ship B would move closer.
Again, I don't have the precise details, but maybe others have seen this as well and can corroborate?
[Updated on: Tue, 16 July 2013 13:41]
What we need's a few good taters.Report message to a moderator
|
|
|
|
|
Members
Search
Help
Register
Login
Home
