Skirmish AI 2.2 RC1 - Post Comments In Thread!
#2
Posted 14 February 2007 - 02:41 AM
Dawn of War Advanced AI Headquarters
Latest DoW Advanced AI Download!
#4
Posted 14 February 2007 - 05:20 AM
LETS GET THIS BETA'ED everyone so we can release!!
Dawn of War Advanced AI Headquarters
Latest DoW Advanced AI Download!
#8
Posted 14 February 2007 - 08:14 PM
Dawn of War Advanced AI Headquarters
Latest DoW Advanced AI Download!
#9
Posted 14 February 2007 - 08:16 PM
Attachments.eldar_squad_seer_council = { "eldar_squad_farseer", "eldar_squad_grav_platform", "eldar_squad_grav_platform_brightlance" } ( Is this used at all ? I found no code using this on first glance).
and since attachements are the only reasons I found for a squad to change default stance I thought plats attached to council made them ranged. Whatever, I'm confused and I sometimes observed even zerks on ranged (some times ago, not related to this version of AI).
So I ask myself : What code can force council to shoot at ranged stance ?
Whatever : I played E v IG on same map as I did with 2.2.b5. With that eldar annihilated IG, with 2.2b6 it was the other way round. Eldar stood no chance.
Edited by LarkinVB, 14 February 2007 - 08:25 PM.
#10
Posted 14 February 2007 - 09:11 PM
WS and falcons jumping to own points, leaving them no jump left during fights.
Seer council again and again doing no damage, fighting ranged.
Watch the replay, lots of strangeness imo.
At the end IG has lots of infantry gathering around neutral relic (the one at top in the river) and ogryns did stay there without capturing until the end.
Attached Files
#11
Posted 14 February 2007 - 10:14 PM
This is hardcoded behaviour. It overrides absolutely everything. I once tried to convince Vespid squads to fight in melee - Hopeless! In general this weird behaviour did exist in WA but in DC it got a lot worse.So I ask myself : What code can force council to shoot at ranged stance ?
Valid point! Fixed!Something seems broken with eldar. No more jump retreat of vehicles ?
Should also not happen anymore except they are attacked or their is a thread (which they can handle) at the target.WS and falcons jumping to own points, leaving them no jump left during fights.
Hmm, I changed the capture command in the tactic file from DoCapure() to DoDefault() like in the capture plan. If this doesn't help... no idea.ogryns did stay there without capturing until the end.
IG infantry has a very low unit strength. I reduced the must attack strength from 4000 to 3000, this should be enough.At the end IG has lots of infantry gathering around neutral relic (the one at top in the river)
This is part of the unit strength system and I can assure you that it's used.Attachments.eldar_squad_seer_council = { "eldar_squad_farseer", "eldar_squad_grav_platform", "eldar_squad_grav_platform_brightlance" } ( Is this used at all ? I found no code using this on first glance).
#12
Posted 14 February 2007 - 10:25 PM
I never saw the WS squad warp at all but I did not watch closely.
Retreating eldar did use FoF, then turned it off for a very short period and switched back on afterwards. Like fast - slow - fast.
Are you sure with council behavior. I remember them fighting CC. Not sure though.
#13
Posted 14 February 2007 - 10:51 PM
Yes, commissars have a min unit strength for attachments. I'll reduced it a bit for the release version.Thanks for fast reply. I forgot to mention commissars only attaching to bigger IG squads. There were 3 commissars running around at gamestart fighting on their own while weak guardsmen squads were around.
Should happen less often than before since they only switch it off if attacking or if they reached their target. And so far I was very satisfied with the test results for the new FoF. Much better than the old behaviour. But FoF effiency gets worse for every additional AI in game.Retreating eldar did use FoF, then turned it off for a very short period and switched back on afterwards. Like fast - slow - fast.
Usually they fight in CC, but on rare ocasions I've also seen them shooting. But it haapens more often with berserkers, at least in my games.Are you sure with council behavior. I remember them fighting CC. Not sure though.
#14
Posted 14 February 2007 - 11:05 PM
I though about writing some simple AI code just doing the thread on path function iterating from endless random x/y coordinates on any map just to show them the inevitable crash. No power left in me to do so.
Edited by LarkinVB, 14 February 2007 - 11:05 PM.
#15
Posted 15 February 2007 - 12:20 AM
I guess go ahead then. Good job and good success on release. Some fine code. Still a shame Relic didn't fix the path problems as the AI sometimes looks stupid trying to build mines/turrets/LP2 at points it can't reach without going through enemies main force or trying to reach LP2 with its army going right through the enemy and losing everything. Stupid Relic.
I though about writing some simple AI code just doing the thread on path function iterating from endless random x/y coordinates on any map just to show them the inevitable crash. No power left in me to do so.
Do you know if Relic has plan to fix this in the "upcoming" patch or did they just decide to ignore this issue completely?
Edited by MaxDoW, 15 February 2007 - 12:21 AM.
#16
Posted 15 February 2007 - 01:38 AM
A Saint's job for everyone!!
Dawn of War Advanced AI Headquarters
Latest DoW Advanced AI Download!
#17
Posted 15 February 2007 - 07:49 AM
Well, so far I've only managed to test Eldar and Chaos a good number of times, and have yet to find any serious issues. I didn't see the builders wigging out as Larkin was mentioning earlier, though I did see a few pathfinding issues at one point on Battle Marshes against Chaos. I'll finish going through the other races hopefully friday (tommorow for me).
Edited by Markoso, 15 February 2007 - 12:39 PM.
#19
Posted 15 February 2007 - 03:20 PM
Dawn of War Advanced AI Headquarters
Latest DoW Advanced AI Download!
Reply to this topic
1 user(s) are reading this topic
0 members, 1 guests, 0 anonymous users