Poll: Worst Unit Of The (time interval) Award (VI)

[keith.lamothe]

quick test . fist wave at 10/10 with support corps:

12 speed boosters (cap 78)
22 decoy drones (cap 9)
101 decloakers (cap 4)



been a while since i actually played vs support corps, but from what i recall this is pretty standard split for it, never thought of checking the cap for the units before.
Something bit amisss here.

Odd, do you have the wave-calc logs for that?

[rabican]

what do i need to do to enable that?

[keith.lamothe]

what do i need to do to enable that?

Settings Window -> Advanced Tab -> check the toggle labeled Enable Advanced Logging.

Then each time a wave is announced it will log the relevant stuff in two (one for the main thread, one for the AI thread) log files in your AIW installation's RuntimeData directory

[rabican]

7/18/2012 4:47:53 AM (5.041)
-----------------------------------
Receiving AddInboundWave from AI Thread at Game Time: 0:10:00
WaveSize factor: 1
Raw Units Dictionary Entries:
BomberII => 32
DecoyDrone => 12
SpeedBooster => 7
Flagship => 1
AIRaidStarship => 1
Decloaker => 11
MissileShipII => 1
ZenithChameleonII => 1
BomberStarship => 1


CheckWave: populating count of Decloaker with base magnitude of 11
numberUnits = kv.Value * this.WaveSize : 11
after applying the ship-type-specific cap multiplier (which includes the unit-cap-scale multiplier), numberUnits : 11
after applying UsefulnessInAIWaveMultiplier if any, numberUnits : 11
after applying Mark-based multiplier if any, numberUnits : 16.5
after applying at-least-one rule, numberUnits : 16.5
after applying difficulty-based multiplier (if <= 3 then 1, <= 4 : 1.5, <= 5 : 1.75, <= 6 : 2, <= 7 : 2.25, <= 9 : 2.5, <= 9.3 : 2.75, <= 9.6 : 3, <= 9.8 : 3.8, 10 : 4.5), numberUnits : 74.25
numberUnitsInt = numberUnits.IntValue : 74



Wave total ships: 184
TypesForDirectAdd count by type:
BomberII => 51
DecoyDrone => 37
SpeedBooster => 11
Flagship => 1
AIRaidStarship => 1
Decloaker => 74
MissileShipII => 4
ZenithChameleonII => 4
BomberStarship => 1

Mark based multiplier thingie is bonkers

and yes i have nothing better to do at  5am

[keith.lamothe]

It may be due to the decloaker being effectively a mark-0 ship in one sense.  Can you attach the logs themselves to a post so I can see the numbers for the other types too?  Mainly interested in seeing the math for how the BomberII entry started at 32 and the Decloaker entry started at 11 and they wound up at 54 and 71.

[keith.lamothe]

It's also due to their ship cap multiplier not being used; not sure if that's due to mark-0 or the fact that we set the cap directly instead of by multiplier.

[rabican]

here

[keith.lamothe]

Thanks

[TechSY730]

Of course, I can't anaylze these to the same level of depth you can, but I can pick out a few things.

The decloaker's ship specific multiplier doesn't seem to be touching its count for some reason, even though it has a tiny ship cap of 4 (is there a large ship specific multiplier for decloakers?)

Code: [Select]

CheckWave: populating count of Decloaker with base magnitude of 11
numberUnits = kv.Value * this.WaveSize : 11
after applying the ship-type-specific cap multiplier (which includes the unit-cap-scale multiplier), numberUnits : 11

Compare this with the speed booster

Code: [Select]

CheckWave: populating count of SpeedBooster with base magnitude of 7
numberUnits = kv.Value * this.WaveSize : 7
after applying the ship-type-specific cap multiplier (which includes the unit-cap-scale multiplier), numberUnits : 3.5

Where even though its cap is, while not huge, is larger than the decloaker's, it is getting its number cut in half by that step. Is there a severe ship specific wave count penalty for speed boosters?


The decloaker's "mark 0" wave count multiplier seems to be 1.5.

Code: [Select]

after applying UsefulnessInAIWaveMultiplier if any, numberUnits : 11
after applying Mark-based multiplier if any, numberUnits : 16.5


[keith.lamothe]

Yea, I'd already mentioned the mark-0 and the no-ship-cap-multiplier things.  The mark-0 thing turns out is pretty harmless: it just uses the mkI multiplier which is fine in this case.

The no-ship-cap-multiplier thing was the problem, and I've fixed it for 5.043