Finished. It had me hooked. I think the length was perfect for me. I found a deck and pace I felt comfortable with and it was enjoyable and pretty fun when you demolish a level's target on the first try.
What's the problem with the MP4s? The decoding can't take that much of your GPU's resources. And if it's a latency thing, you could start loading the next video before the previous finishes... Maybe grabbing a few of the first frames if there is some kind of difficulty in switching between the two.
I liked your progress and approach in more than words so thought I'd take a look at this too. It seems like a good start. The mini-game is pretty brutal. You're having a normal conversation and then are forced to suddenly blurt out some incredibly awkward attempt at flirting. It's hilarious, but you just have to hope that the AI interprets it well.
I was impressed by the AI loading speed in this one, it was faster than in More then Words! I was a little disappointed in that as far as I could tell, the videos were just random and not associated with either her mood or the conversation. I apologize if they were responsive. I also hope that you can get the speech synthesis up and running for this game too.
I've seen someone else working on a similar kind of continuous AI video concept. Check out David Browne on Youtube. He also has some interesting stuff with making an anime style android alongside his AI video experiments.
No worries at all. It's better that you get a game out than get sucked into scope creep. I appreciated that you could finish the game while acting in a sensible way, even if there was no benefit to it.
I didn't open the door that said "Don't open" and it seems Mr. Rabbit still had his way. Nuts. Creepy little game. Too bad there seems to be only one ending.
Seems like it's a good starting point. I was able to defeat the boss once in a couple playthroughs. I encountered one bug where if at the beginning you kill the enemies too fast, the remaining ones don't spawn and you can't progress. The art is nice.
After booting up my computer, the program works, but after restarting it it seems to not work any more. It appears that the LLM doesn't start sometimes but it is hard to tell. All I know is that I apply the settings but then get no responses even after typing/speaking prompts. This includes also getting no auto talk.
I can see with the original GGUF file, for example, that my GPU's VRAM is loaded with something. So it's not simply just not loading.
I figured out what the text to speech issue was. It seems that my GPU didn't have enough VRAM for both models to run simultaneously. So, I found a 4B parameter LLM and now the speech to text works!
Edit: It doesn't seem to be so shrimple... I can't tell which microphone it is actually using. The default, I would assume, but changing to another microphone that is kind of far away seems to work just the same as one right up to my face.
I am just getting started with this, I am able to get generations from the LLM, but the microphone seems to only produce "the" over and over rather than any real words that I am speaking. Thank you.
P.S. Is there any way to move the avatar around on the screen? Huge and in the right in the middle is exactly where I would rather it not be.
Once the fire rate bug kicked in and almost crashed the game with the number of bullets at around wave 30, it was just a matter of ending the day over and over.
Nice concept with a cohesive art style. The high risk die doesn't seem to actually appear after buying it. Also, the two upgrades on the far right are inaccessible (right of bullets and harvest)
Fun little game. I will say that there seems to be a cumulative effect of the fire rate buff that gets very silly very quick. I don't know what causes it, but it results in things like this:
I like that there were multiple "challenges" to get through this time instead of just the one. I hope you'll add more interactivity and/or goals to accomplish in future releases.
A couple issues I'm having.
Controls are leaking through while typing. For example, space still jumps while the text box is selected, WASD also move around still when they should only enter text.
The first door is sometimes not triggered by the AI saying the exact expression "Cortana, open the door." so I had to ask if there was another way and then it was able to open.
The backspace key sometimes only works once in the text box. So, I have to use the delete key instead to fix mistakes.
Random crash after being kicked out of the map to stare at a wall, stuck.
"3rd party game template" doesn't stop running after quitting the game.
I'm just downloading now and am curious to see the new improvments. I'm not sure about the change of environmental tone away from the futuristic, but that's not too big of an issue for me.
