Hi there, playing around with base designs recently, I wondered if there might be a reasonable way to actually measure the self-defense capability of a base, being an indicator of how well the base is being built.
In tech/science, you usually try to eleminate variables that may be disturbing the result and standardize whatever you can. This is not easy for a surprisingly complex matter like a base in Tremulous, however, I would love to have such criteria to figure out reproducable and reliable figures in order to be able to tell something about the quality of a base.
here are my suggestions:
attacker specifications:
1a) Human attacker: equipment of choice, walking, not running (to get an easier to measure result) and may use the medkit
1b) Alien attacker: any class of choice, no trampling for tyrants, dragoons may pounce
1c) The penetrator may attack structures in his way as early as getting within range of defence structures, these are turrets, teslas, acid tubes, hives and trappers. Only from the moment such a structure triggers, long range weapons or barbs may be used.
1d) No retreat! The penetrator may not interrupt his attempt to destroy the enemy main structure, this is the alien must not get out of the base and wait for regeneration, the human must not go out of the base and wait until the medkit worked.
base specifications:
2g) All structures are allowed (stage 3), standard amount of build-points,
2h) human base: (the reactor,) two telenodes, one armoury, one medistation, if a defense computer is present, it must be well defended
2i) alien base: (the overmind), three eggs, at least one of them on the ceiling, one booster,
2j) at least one spawn must be covered by defense structures
the rest may be used for defense structures of any kind = all other structures
base penetration specifications:
In general: two things,
3q) one: the time it takes you from the time you first encounter a defense structure until you are able to bring down the main structure (reactor/overmind),
3r) second: how many health points you lose in doing so. The ratio time/healthpoints should be a good value to tell how effective a base is
Scope and results of this penetration test:
(:) Result: Time it takes, dead/health points left the moment the main structure is down, is structure down or not, examples:
Layout xy tyrant, 96 seconds, dead, failed
Layout yz battlesuit with luci and grenade, 23 seconds, 10hp, success
Layout zx goon+, 45 seconds, 1hp left, success
Layout wx battlesuit with saw and grenade, 27 seconds, dead, success
(+) This test should provide you with valuable information on how easy it is to get into a base of a certain layout.
(~) For the human base, half the time a machine gun turret need to make a 180° may be subtracted from the time.
(-) This test must of course be repeated for I would say at least three times if penetrating in the same direction.
(:) Always the best result counts
(-) You need to set up your own local server for this kind of test.
... okay, this is moreless it. If you have any suggestions to make the process more significant or if you see any major flaw in this test, please let me know. If I figured out a usable procedure to perform such tests, I shall configure my own server and run some tests.
