Difference between revisions of "Glacier/GlacialHawk"

From Robowiki
Jump to navigation Jump to search
(1.1 rating and 1.2 info)
m (Update navbox)
 
(13 intermediate revisions by the same user not shown)
Line 3: Line 3:
 
| parent      = Glacier
 
| parent      = Glacier
 
| page1        = Version History
 
| page1        = Version History
 +
| page2        = Hypothermia
 
| page3        = GlacialHawk
 
| page3        = GlacialHawk
 
}}
 
}}
Line 8: Line 9:
  
 
<!-- {{subst:rumblestats:meleerumble|ags.muse.GlacialHawk x.x|RumbleStatsDefault}} -->
 
<!-- {{subst:rumblestats:meleerumble|ags.muse.GlacialHawk x.x|RumbleStatsDefault}} -->
 +
 +
* '''1.12''' - 24/9/2009
 +
** {{RumbleStatsDefault|link=http://darkcanuck.net/rumble/RatingsDetails?game=meleerumble&name=ags.muse.GlacialHawk+1.12|rumble=MeleeRumble|scorelabel=APS|score=68.3|rank=5th|win=278|loss=5|plrank=5th|glicko2=1637.4|score2label=Survival|score2=88.68}}
 +
** Revert to 1.10
 +
** As the number of enemies decreases, increase the number of predictions per-enemy, up to a maximum of 32. Similar to 1.8 release except without the bulletPower changes.
 +
 +
* '''1.11''' - 24/9/2009
 +
** {{RumbleStatsDefault|link=http://darkcanuck.net/rumble/RatingsDetails?game=meleerumble&name=ags.muse.GlacialHawk+1.11|rumble=MeleeRumble|scorelabel=APS|score=68.7|rank=4th|win=278|loss=3|plrank=4th|glicko2=1641.9|score2label=Survival|score2=90.11}}
 +
** Try out Diamond 1.40's bulletPower logic
 +
 +
* '''1.10''' - 24/9/2009
 +
** {{RumbleStatsDefault|link=http://darkcanuck.net/rumble/RatingsDetails?game=meleerumble&name=ags.muse.GlacialHawk+1.10|rumble=MeleeRumble|scorelabel=APS|score=68.78|rank=4th|win=277|loss=4|plrank=5th|glicko2=1644.8|score2label=Survival|score2=89.31}}
 +
** Revert bulletpower changes in 1.9 back to to 1.5
 +
 +
* '''1.9''' - 23/9/2009
 +
** {{RumbleStatsDefault|link=http://darkcanuck.net/rumble/RatingsDetails?game=meleerumble&name=ags.muse.GlacialHawk+1.9|rumble=MeleeRumble|scorelabel=APS|score=67.7|rank=5th|win=277|loss=4|plrank=4th|glicko2=1633.3|score2label=Survival|score2=89.87}}
 +
** Revert to 1.5
 +
** Tweak bulletpower a bit. Base off of average enemy distance most of the time. Generally lower bulletpower.
 +
** Add a 'distance last 10' and 'distance last 20' segments. Increase weight of reverse walldistance segment.
 +
 +
* '''1.8''' - 22/9/2009
 +
** {{RumbleStatsDefault|link=http://darkcanuck.net/rumble/RatingsDetails?game=meleerumble&name=ags.muse.GlacialHawk+1.8|rumble=MeleeRumble|scorelabel=APS|score=68.35|rank=4th|win=276|loss=5|plrank=6th|glicko2=1642.7|score2label=Survival|score2=89.18}}
 +
** Revert to 1.5
 +
** Tweak bulletpower very slightly smaller
 +
** Instead of always predicting 12 futures per enemy, predict up to 30 per enemy, with a maximum of 100 predictions in total for all enemies
 +
 +
* '''1.7''' - 20/9/2009
 +
** Rating: ''Utter failure...''
 +
*** Early results: {{RumbleStatsDefault|link=http://darkcanuck.net/rumble/RatingsDetails?game=meleerumble&name=ags.muse.GlacialHawk+1.7|rumble=MeleeRumble|scorelabel=APS|score=68.19|rank=5th|win=277|loss=5|plrank=5th|glicko2=1648|score2label=Survival|score2=88.74}}
 +
** Revert to 1.5
 +
** Bias against returned situations which are more dissimilar.
 +
 +
* '''1.6''' - 20/9/2009
 +
** Result: Slightly worse than 1.5. RumbleStats doesn't represent non-current bots well so not bothering to post rating.
 +
** Only put results into the tree after 90 ticks.
 +
 +
* '''1.5''' - 20/9/2009
 +
** {{RumbleStatsDefault|link=http://darkcanuck.net/rumble/RatingsDetails?game=meleerumble&name=ags.muse.GlacialHawk+1.5|rumble=MeleeRumble|scorelabel=APS|score=68.37|rank=6th|win=278|loss=5|plrank=4th|glicko2=1646.4|score2label=Survival|score2=88.89}}
 +
** Revert to 1.2 (for now)
 +
** Increase the number of 'proximity sensor' dimensions and tweak dimension weightings
 +
 +
* '''1.4''' - 19/9/2009
 +
** {{RumbleStatsDefault|link=http://darkcanuck.net/rumble/RatingsDetails?game=meleerumble&name=ags.muse.GlacialHawk+1.4|rumble=MeleeRumble|scorelabel=APS|score=68.22|rank=4th|win=278|loss=5|plrank=4th|glicko2=1651.9|score2label=Survival|score2=88.37}}
 +
** Tweak the new probabilistic model introduced in 1.3, to include a 'doubt' factor based partially on distance. (Maybe make the 'doubt' factor learned in a future version?)
 +
 +
* '''1.3''' - 19/9/2009
 +
** {{RumbleStatsDefault|link=http://darkcanuck.net/rumble/RatingsDetails?game=meleerumble&name=ags.muse.GlacialHawk+1.3|rumble=MeleeRumble|scorelabel=APS|score=67.65|rank=5th|win=274|loss=8|plrank=9th|glicko2=1644.9|score2label=Survival|score2=87.23}}
 +
** Fancy new experimental modeling of bullet-hit-probability at possible angles. Should work better than simply summing the movement profiles in the common 'shadow's melee gun' style. At least, it should once I get it tweaked...
 +
 
* '''1.2''' - 18/9/2009
 
* '''1.2''' - 18/9/2009
−
** ''Rating pending''
+
** {{RumbleStatsDefault|link=http://darkcanuck.net/rumble/RatingsDetails?game=meleerumble&name=ags.muse.GlacialHawk+1.2|rumble=MeleeRumble|scorelabel=APS|score=68.39|rank=4th|win=276|loss=5|plrank=6th|glicko2=1652.6|score2label=Survival|score2=88.72}}
 
** Large improvement in PIF speed, at the cost of reduced wall-intersect checking.
 
** Large improvement in PIF speed, at the cost of reduced wall-intersect checking.
  

Latest revision as of 22:50, 26 September 2009

Glacier Sub-pages:
Glacier - Version History - Hypothermia - GlacialHawk


  • 1.12 - 24/9/2009
    • MeleeRumble ‒ APS: 68.3% (5th), PL: 278-5 (5th), Survival: 88.68%
    • Revert to 1.10
    • As the number of enemies decreases, increase the number of predictions per-enemy, up to a maximum of 32. Similar to 1.8 release except without the bulletPower changes.
  • 1.11 - 24/9/2009
    • MeleeRumble ‒ APS: 68.7% (4th), PL: 278-3 (4th), Survival: 90.11%
    • Try out Diamond 1.40's bulletPower logic
  • 1.10 - 24/9/2009
    • MeleeRumble ‒ APS: 68.78% (4th), PL: 277-4 (5th), Survival: 89.31%
    • Revert bulletpower changes in 1.9 back to to 1.5
  • 1.9 - 23/9/2009
    • MeleeRumble ‒ APS: 67.7% (5th), PL: 277-4 (4th), Survival: 89.87%
    • Revert to 1.5
    • Tweak bulletpower a bit. Base off of average enemy distance most of the time. Generally lower bulletpower.
    • Add a 'distance last 10' and 'distance last 20' segments. Increase weight of reverse walldistance segment.
  • 1.8 - 22/9/2009
    • MeleeRumble ‒ APS: 68.35% (4th), PL: 276-5 (6th), Survival: 89.18%
    • Revert to 1.5
    • Tweak bulletpower very slightly smaller
    • Instead of always predicting 12 futures per enemy, predict up to 30 per enemy, with a maximum of 100 predictions in total for all enemies
  • 1.7 - 20/9/2009
    • Rating: Utter failure...
      • Early results: MeleeRumble ‒ APS: 68.19% (5th), PL: 277-5 (5th), Survival: 88.74%
    • Revert to 1.5
    • Bias against returned situations which are more dissimilar.
  • 1.6 - 20/9/2009
    • Result: Slightly worse than 1.5. RumbleStats doesn't represent non-current bots well so not bothering to post rating.
    • Only put results into the tree after 90 ticks.
  • 1.5 - 20/9/2009
    • MeleeRumble ‒ APS: 68.37% (6th), PL: 278-5 (4th), Survival: 88.89%
    • Revert to 1.2 (for now)
    • Increase the number of 'proximity sensor' dimensions and tweak dimension weightings
  • 1.4 - 19/9/2009
    • MeleeRumble ‒ APS: 68.22% (4th), PL: 278-5 (4th), Survival: 88.37%
    • Tweak the new probabilistic model introduced in 1.3, to include a 'doubt' factor based partially on distance. (Maybe make the 'doubt' factor learned in a future version?)
  • 1.3 - 19/9/2009
    • MeleeRumble ‒ APS: 67.65% (5th), PL: 274-8 (9th), Survival: 87.23%
    • Fancy new experimental modeling of bullet-hit-probability at possible angles. Should work better than simply summing the movement profiles in the common 'shadow's melee gun' style. At least, it should once I get it tweaked...
  • 1.2 - 18/9/2009
    • MeleeRumble ‒ APS: 68.39% (4th), PL: 276-5 (6th), Survival: 88.72%
    • Large improvement in PIF speed, at the cost of reduced wall-intersect checking.
  • 1.1 - 10/9/2009
    • MeleeRumble ‒ APS: 67.72% (5th), PL: 276-5 (5th), Survival: 87.82%
    • Increase bullet power notably, to mostly follow HawkOnFire bulletpower formula
    • Huge SlowBot and skips occasional turns. Will try to fix in 1.2 :(
  • 1.0 - 9/9/2009
    • MeleeRumble ‒ APS: 67.48% (5th), PL: 275-6 (6th), Survival: 91%
    • Movement of HawkOnFire (from here) with Hypothermia (the gun of Glacier).
    • Thoughts: Seems to be scoring how I expected: Above DiamondHawk but below ShadowHawk. But considering this gun was only given a single test run before release and no tuning at all, this is probably quite a good rating. Higher survival score than ShadowHawk indicates that bulletpower could be increased to trade a little survival for score, which doesn't surprise me given that the bulletpower selection is a complete 2-second hack without any real thought or tuning.
Retrieved from "http://robowiki.net/w/index.php?title=Glacier/GlacialHawk&oldid=12528"