Hey, I recently found this game and I ABSOLUTELY love it, it is an incredible concept and neat integration of LLM technology. I do, however, have a few notes, if you would indulge me.
- The CUDA approach does not work in my case, as I have an AMD gpu. I don't know if there are any alternatives, but it would be nice to know if there are as the game works quite slowly for me. I find myself needing to tab on and off while doing other things instead of focusing on the game.
-A toggle for the vfx & Vignette would be great, it does get annoying after a while
-It would be amazing if we could toggle the LLM to not speak in old english, as it can be annoying, but I'm not sure how possible that would be since it appears to use a lot of old English tokens. Apologies if this is a big ask.
-When a prompt has 'click to continue' you could already start preparing the next prompt, instead of having the user wait
-It would be nice to be able to hover over locations you can't travel to and be able to see what they are
-The way skills affect rolls is sort of unintuitive, I think there's too much random, or it feels too random, independent of what your skills should lead to. It is also unintuitive since its hard to contextualize what all the numbers mean, so maybe like a dnd d20 approach + a modifier, at least on the surface? Might just be me tho.
-The left half of the 'confirm' button on the selection screen can't be clicked on, a different UI object probably overlaps it.
-This is more of a question, what's the reason you can't reenter the same location twice?
-After clicking on menu stuff its sometimes easy to misclick where you want to go
That's all for now, thank you for reading through :P