Need help with two things

[Xuku]

I have a couple of questions about npc scenes and sounds.

My first question is about the response system: once I have the response/criteria/rules files in scripts/talker, will the AI code automatically use it, or is there some way I have to set it up myself? I moved npc_alyx.txt from the Ep.1 gcf to my Half-Life 2/hl2/scripts/talker along with the response_rules.txt which includes it, but when I launch my map with npc_alyx she never says anything (except when she moves out of your way). How do I get Alyx to use her list of responses? Is there a check in any of them that calls for the game to be Episode 1 (I'm running it in HL2, but since Ep.1 response files were copied over that shouldn't matter, right?)

Also another problem is that some of the sounds/scenes files taken from the episode 1/half-life 2 gcf's don't work. For example, when I'm choosing a sound or scene for a logic_choreograph_scene entity, some of the scenes and sounds will preview but others will just be silent when I preview them. When I open up the copied .vcds or .wavs in FacePoser, the character model doesn't move or say anything when I try to preview them. Is there something I'm doing wrong? All I did was use GCFScape to extract the .vcds and .wavs to my scenes and sounds folders, respectively.

All this on top of the other Source SDK bugs makes a really frustrating experience, but I really want to create HL2 maps, so I'm not going to give up yet. It would be greatly appreciated if someone could help me out.