| Is minefield hit % based on number of ships? |
Tue, 26 July 2005 05:03  |
|
 dreadlordnaf |
 | Crewman 1st Class |
Messages: 27
Registered: April 2005
Location: Bulgaria | |
|
Does a really really big fleet have a higher chance of detonating a minefield than a really small one?
My situation is a i have a big enemy fleet 1 turn away from a major world of mine. The thing is he has to travel warp 8 to get there in one turn and he has to fly through 35ly of my minefields.
If im doing the calculations right: 4 warps over = 1.2% chance to hit * 35ly = 42% chance to hit the field. This is lower than I'd like and he still have a 58% chance to make it just fine. Im worried he'll gamble and try for it. But if he fails and hits the field he will lose all his support ships in the fleet (mini bombers, fuel tankers, transports, etc) and will still have his combat ships but with no way to hurt the planet at all.
I need some advise from someone thats been in a lot of games. This guy seems like an experienced player. You think he'll gamble and try it? Or you think most people would see a 42% chance to get all thier support ships wiped as a no go?
Report message to a moderator
|
|
|
|
|
|
|
|
|
| Re: Is minefield hit % based on number of ships? |
Tue, 26 July 2005 11:17  |
|
| LEit |
 | Lt. Commander |
Messages: 879
Registered: April 2003
Location: CT | |
|
Depending on the actual number of mines in the field, it will take 57 or 58 hits to remove, wich required about 170 chaff sent at warp 10 completly through the field. If they have scout chaff, they cannot go warp 10 (50mg fuel isn't enough to go warp 10), and will need about 10% more chaff on average.
Note that these numbers are 'on average' and you could need a lot more depending on luck.
I've used mini-bombers as chaff sweepers before too, but only going to the planet of course. And for much smaller fields.
Also, note that for mine field hits, speed is calculated based on the distance traveled, so if you say to go warp 10 for 61ly, it will only go warp 8 for mine field hit chances. An easy test (what I did) to verify this is to send 500 SD ships at warp 10 49 ly into a speed trap field - none of them hit. If they went warp 10, each would have less then a .5% (1 in 200) chance of making it.
Interestingly, 55 of the ships died due to engine failure from going warp 10, so the speed there is based on requested speed, not needed speed.
- LEitReport message to a moderator
|
|
|
|
|
|
|
|
| Re: Is minefield hit % based on number of ships? |
Wed, 27 July 2005 22:44 |
|
| crr65536 |
 | Chief Warrant Officer 3 |
Messages: 180
Registered: June 2005 | |
|
That the game is implemented in this way makes sense, especially in light of LEit's test. However, it does not seem like it would be hard to make the game behave in the way the help file suggests. I could do it as follows:
- Determine the distance traveled in the minefield. Call it n. This is the length of the intersection between a line and a circle, so it can be calculated with some geometry.
- Determine the maximum distance that can be traveled in the minefield at the fleet's speed. Call it max.
- Get a random real number between 0 and 1, p. p can have a precision of only 2 decimal places and still work, since the maximum distance a fleet can travel in a minefield would be 100ly (warp 10).
- The fleet will hit a mine after traveling p * max light years. If (p * max) > n, a mine is not hit. The location of the hit is (p * max) ly along the fleet's path.
Test: a fleet travels 10ly at 1 speed over in a heavy field. It's chance of hitting a mine is 10% (according to the help file, not the actual behavior). The maximum safe distance it could travel would be 100ly (of course you can't travel 100ly at w7, but bear with me). It has a 1% chance of hitting a mine after one light year (because if p <= .01 then it hits the mine before 1 ly, and p has a 1% chance of being <= .01). It would also have a 10% chance of hitting after 10ly, etc., as can be calculated in a similar manner.
Sorry to get off-topic, but I'm deadly curious why the creators of Stars didn't do this. While it may be harder to implement than a simple loop, it seems that it would be quicker to run (one calculation per fleet, instead of one calculation per light year per fleet). It would also make the helpfile and game behavior consistent.
If there is a flaw in my math, please point it out.
If there isn't, then I think using this could be a fine option for FreeStars 1.1.
Report message to a moderator
|
|
|
|
|
|
|
|
| Re: Is minefield hit % based on number of ships? |
Mon, 01 August 2005 05:01 |
|
| dreadlordnaf |
| Crewman 1st Class |
Messages: 27
Registered: April 2005
Location: Bulgaria | |
|
| iztok wrote on Wed, 27 July 2005 14:33 |
Hi!
I'm affraid you're not doing your calculation right. If it would be done your way there wouldn't be possible to travel W-10 through a minefileld larger than 56 LY. The right calc is
(1 - prob_of_hit_per_LY) ^ distance_traveled
In your case it is 0.988^35 = 0.655.
BR, Iztok
|
Can you explain this more. Does that .655 represent a 65.5% chance to hit the field? or a 0.655% chance per ly to hit? Also you sure on that formula? The way you set it up it would suggest going faster through a minefield would actually lower your chances of getting hit. Did you mean 1 - (prob_of_hit_per_LY ^ distance_traveled)
[Updated on: Mon, 01 August 2005 05:14] Report message to a moderator
|
|
|
|
| Re: Is minefield hit % based on number of ships? |
Mon, 01 August 2005 06:50 |
|
|
|
ah no...it is the chance of NOT itting a mine IOW the chance of making it thru the minefield.
Report message to a moderator
|
|
|
|
|
|
| Re: Is minefield hit % based on number of ships? |
Tue, 02 August 2005 03:40 |
|
| dreadlordnaf |
| Crewman 1st Class |
Messages: 27
Registered: April 2005
Location: Bulgaria | |
|
| gible wrote on Mon, 01 August 2005 06:50 |
ah no...it is the chance of NOT itting a mine IOW the chance of making it thru the minefield.
|
Ah this clears it up then. Thanks.
Report message to a moderator
|
|
|
|
Members
Search
Help
Register
Login
Home
Surprizingly its only 87.5%.
So ... all add-on ideas are dumped for unforeseeable future. Can you propose more efficient algorithm than ly-by-ly that implements valid (current) behaviour and differs from faulty (help file) one only by (few) parameters?
!
!
You must be reading my mind. That's exactly what I've been thinking when reading request for more explanation. Please remind me on that when we get in the same game.