Geep

HMart, I'm guessing that for Bienie, manipulation of a forward vector would be needed, but probably not the dot product. He wants the AI to spawn more or less in front him. When that happens, I imagine the AI should be looking at the player. That can be done by taking the difference between the two origins, normalizing it, and making that into the AI's forward vector.

In all fairness to yourself, trigger_look is a fairly sophisticated piece of coding. If you do decide to pursue it further, the wiki sheds some light, and includes near the end a link to Obsttort's video describing the math. I think, if this script was (re)written today, it could use calls to now-provided angle conversion functions, instead of doing all the math itself. Regarding the AI that appears, maybe another approach is to have AI spawn at some fixed "station" location, and run like crazy to where the player is... you could have some special music & smoke effects play while waiting for him to arrive

Don't know exactly, but if I was doing it with scripting, I'd think about using $player1.getOrigin() and probably $player1.getViewAngles(), calculate the position I want the AI to appear, and then use setOrigin() to place it there. For the calculation, you could hardcode the distance, or create a more general script object and pass the distance as a spawnarg; the "trigger_look.script" might serve as an example of some of this. However, you have to arrange your trigger location environment to avoid the risk of trying to move your AI, say, to the middle of a wall. Maybe just living with a one or a few named teleport targets would be easier.

I'll do that now.... done

Hmmm, intermitant behavior, everyone's favorite. Good workaround, tho... I'll add it to my list of things to wikify, if no DR fix gets queued up.

One of the things I left unresolved when I wrote up the wiki entry for Model Scaling, is how to make duplicates. If you just hit the space key, it creates a new unscaled object, which is really not what you want. I just recalled this problem again, when I was set to make a dozen scaled items that I wanted to be identical in size. I don't know if there's a workaround.

@taaaki , could you contact me about use of TDM SVN for a FM collaboration? I guess your PM mailbox is full.

Do you think using monsterclip as the texture, instead of the original tdm_nodraw_solid_liquid, has made the setup more fragile?

The wiki page "Ragdoll Corpses in Water" is updated, and now with pix. @PranQster, let me know if you'd like any changes (or make them yourself). Similarly, @angua, I recently made some tweaks and extensions to your "Swimmable Water" page.

Thanks. I'll rework the wiki page soon. Your related AI-sitting experiments are off-topic for that, so won't be included there. But there will be a pointer back to this thread, though.

@PranQster, I'd like to add a precis of your method to the "Ragdolls floating on water" wiki, including the 2 photos above. OK? Also, I wasn't sure that the monsterclip suggestion was going to work... won't ragdolls just fall through monsterclip? Did you try it?

I'm not sure where to post this, but this group seems relevant... In a 3-month time frame, a nude or semi-nude male AI is needed for a new FM. He would be paired with an existing nude female AI in a brief cutscene. I am unaware of any pre-existing TDM nude male. Please PM me for more details if you might have the expertise and inclination to help.

Yes, we're good now. Working with fflood and others, 2 works were retrieved, made runnable under 2.07, and considered. One seems to have particular promise. It's way too early to say much more about that.

Done. In my FM, I found i could live with the T pose. My wiki piece links to ways to pose a ragdoll, but I didn't really pursue them.