« first day (4852 days earlier)      last day (407 days later) » 

01:33
2
Q: What divination spells (if any) could allow you to learn sigils sequences for teleportation circles?

User 23415Do any divination spells allow you to find out the sigil sequence of a teleportation circle? (For example, the sigil sequence of the circle closest to a specific destination?)

 
6 hours later…
07:12
posted on January 12, 2024 by Alex Helm

How to play big commanding character types – my collection of tips and tricks that I have accrued over 20+ years of larping. The post An Introvert’s Guide to Playing High Status Characters appeared first on Nordic Larp.

 
9 hours later…
15:53
> Ben and Ryan chat about [...] the potential for GenAI to revolutionize game development by creating an infinite array of dynamic game worlds
Using this a jumping off point more than anything - GenAI can be quite good for running an RPG. Especially in play by post/over chat. But even otherwise, it can quickly come up with a description of things that won't sound overly mundane.
Like, generating a tavern. A cave. A village. Useful things to have in your pocket for when players decide to investigate rumours, or go in the wilderness, or try to go to that village you only marked on your map and never expanded on.
Makes prepping for games much easier, since you can just pre-gen these and pull them as needed. Or expand upon if needed.
Now that I think about it, it might also be quite useful at generating rumours. Feed it some fact about the setting and try to generate plausible sounding slightly misinformed descriptions. I'm sure a gen AI can come up with plausible sounding slight BS that's somewhat based on something of substance. Gen AI is very good at that under normal circumstances, but rumours are probably one of the best application there.
 
1 hour later…
GcL
GcL
17:26
Not a fan of AI generated content of any sort.
 
3 hours later…
20:04
@GcL What about Markovian chains? Because that's essentially what I like using an LLM within an RPG for. It's a useful tool to string descriptions together.
 
2 hours later…
GcL
GcL
21:55
@VLAZ I would expect a markov model to choose a series of words to perform even worse than an llm, so I expect I'd like the result even less.

« first day (4852 days earlier)      last day (407 days later) »