Resolution problems are first to fix on my lift, thought current workaround would be good enough but it doesn't seem to work. As for frank crash, thanks for letting me know, he's a bit of a troublemaker. Thanks for the video, always thankful for more data!
A great idea, and it's so close to being perfect. Feels like you need just a bit more training data, because right now the dialogues (at least the three that I tried) feel a bit railroad-y.
UI is serviceable, acts a bit weird when you try to resize it - and it also froze on me, so I needed to change the config file.
I think this is the first time I've seen AI used properly. Hope you'll get more data and make the characters more responsive to player shenanigans!
Some of these felt easier than the tutorial. I don't know, maybe I just got lucky with some questions. The premises are getting more memorable.
Here's how it went:
- Life insurance -- Pretty straightforward. I got her to accidentally tell me the full story before I presented all the evidence. Got no answers about why the door was broken in despite it being a suicide. Bumbled around a bit before I realized that she's been ready to confess for 10 minutes.
- Peer Tribunal -- Liked this one the most. Had the most tangled situation so far with all those equally nonsense testimonies that sound like bullshit made up on the spot. Didn't like that all it takes is to just ask about that one key piece of evidence that automatically wraps it up. Could use some more complexity considering the premise has potential.
- Mariachi ninjas -- again, there's only one correct topic to bring up to instantly win. I think it would have been better if the necessary evidence was something that can only be found in [final witness]'s testimony to give the case at least two stages of progression, but still pretty good.
Had data sharing enabled throughout, so I trust you'll make good use of it.
Absolutely appreciate data sharing, it does help with some unexpected ways players approach the subject.
I have made some of the cases much easier because players were confused on how to progress but with the not so subtle hints I could easily revert to more challenging ones.
I honestly appreciate your patience with the game, I know it can easily get infuriating when you approach the case from the angle nobody else yet has
Interesting game. The good news is I am your target audience. The bad news is that I am your target audience.
I only got as far as completing the first three cases for now. First off: it's incredibly difficult to get a game like this right, so I am going to say that I look at everything you attempt to do with great admiration, but there are shortcomings and they need to be covered.
The game is experimental, which is both its strength and its weakness. I'm not a big fan of input-based games. I think there is a reason why text-based adventure games went extinct -- It doesn't mean it's impossible to do, just that it's really really hard to pull off since there's and enormous range of player-system interactions that the developer needs to account for and that takes time and intensive work. Playing this game a lot of the time feels like banging your head against a wall, and when the wall finally starts to crack the next hit just magically makes it fix itself. There were too many moments where telling a suspect about something they already admitted just resulted in the same old denial. I think that's just a flaw inherent in this type of gameplay.
And this is just the tutorial cases. I can only imagine it gets more complex further on.
I think something that needs to be said is that developer intent should never be underestimated and a game that's supposed to be carried by dynamic algorithms loses a lot of that, so it's easy to get lost in the logic.
The UI is very bland and not very attention grabbing. Considering that games like these are supposed to be carried by their UI design I think it needs to be more visually distinct and organized (weirdly the AI dimension is somewhat better at this). Important to note most of the UX bugs are UI-related. Menu buttons (start screen; the city vista screen) very often require 2 clicks to respond and there's no real feedback from the buttons so you're just wondering sometimes if the game froze.
I really want to see more of this, since it's a step in the right direction, but it's a very rough implementation. Player frustration is a big pitfall of all real detective games, so it's always important to account for that. I think a good detective game makes a player feel gratified in his moment-to-moment thinking process, and provides satisfying feedback when he hits the nail right on the head (something that I also struggle with implementing) -- when you make a correct statement or accusation the game usually just spits back a "noted" or "acknowledged" back at you, but I think that this is the perfect place to throw a lead in the player's face (correct action => progression path).
Overall an above average DD game. I like it but I wouldn't call it a great detective experience yet.
Thanks for checking it out, I'm glad you've got patience for it. I do appreciate extensive feedback, it's certainly going to shape following weeks of work. There's a lot to unpack but by the next iteration it's certainly going to get better.
Comments
It started this time around, but the resolution is fucked up. Can't seem to change it in the menu either.
What the fuck, left it running while answering a comment and it magically fixed itself. Did you create a magic game?
Okay, I love this. Gonna try to make a recording. Make way for psychocop.
Pressing hint while talking to forensic frankie crashed the game.
Bugs work in mysterious ways
Resolution problems are first to fix on my lift, thought current workaround would be good enough but it doesn't seem to work.
As for frank crash, thanks for letting me know, he's a bit of a troublemaker. Thanks for the video, always thankful for more data!
A great idea, and it's so close to being perfect. Feels like you need just a bit more training data, because right now the dialogues (at least the three that I tried) feel a bit railroad-y.
UI is serviceable, acts a bit weird when you try to resize it - and it also froze on me, so I needed to change the config file.
I think this is the first time I've seen AI used properly. Hope you'll get more data and make the characters more responsive to player shenanigans!
[SPOILERS BELOW]
Got another three solved.
Some of these felt easier than the tutorial. I don't know, maybe I just got lucky with some questions. The premises are getting more memorable.
Here's how it went:
- Life insurance -- Pretty straightforward. I got her to accidentally tell me the full story before I presented all the evidence. Got no answers about why the door was broken in despite it being a suicide. Bumbled around a bit before I realized that she's been ready to confess for 10 minutes.
- Peer Tribunal -- Liked this one the most. Had the most tangled situation so far with all those equally nonsense testimonies that sound like bullshit made up on the spot. Didn't like that all it takes is to just ask about that one key piece of evidence that automatically wraps it up. Could use some more complexity considering the premise has potential.
- Mariachi ninjas -- again, there's only one correct topic to bring up to instantly win. I think it would have been better if the necessary evidence was something that can only be found in [final witness]'s testimony to give the case at least two stages of progression, but still pretty good.
Had data sharing enabled throughout, so I trust you'll make good use of it.
Absolutely appreciate data sharing, it does help with some unexpected ways players approach the subject.
I have made some of the cases much easier because players were confused on how to progress but with the not so subtle hints I could easily revert to more challenging ones.
I honestly appreciate your patience with the game, I know it can easily get infuriating when you approach the case from the angle nobody else yet has
Interesting game. The good news is I am your target audience. The bad news is that I am your target audience.
I only got as far as completing the first three cases for now. First off: it's incredibly difficult to get a game like this right, so I am going to say that I look at everything you attempt to do with great admiration, but there are shortcomings and they need to be covered.
The game is experimental, which is both its strength and its weakness. I'm not a big fan of input-based games. I think there is a reason why text-based adventure games went extinct -- It doesn't mean it's impossible to do, just that it's really really hard to pull off since there's and enormous range of player-system interactions that the developer needs to account for and that takes time and intensive work. Playing this game a lot of the time feels like banging your head against a wall, and when the wall finally starts to crack the next hit just magically makes it fix itself. There were too many moments where telling a suspect about something they already admitted just resulted in the same old denial. I think that's just a flaw inherent in this type of gameplay.
And this is just the tutorial cases. I can only imagine it gets more complex further on.
I think something that needs to be said is that developer intent should never be underestimated and a game that's supposed to be carried by dynamic algorithms loses a lot of that, so it's easy to get lost in the logic.
The UI is very bland and not very attention grabbing. Considering that games like these are supposed to be carried by their UI design I think it needs to be more visually distinct and organized (weirdly the AI dimension is somewhat better at this). Important to note most of the UX bugs are UI-related. Menu buttons (start screen; the city vista screen) very often require 2 clicks to respond and there's no real feedback from the buttons so you're just wondering sometimes if the game froze.
I really want to see more of this, since it's a step in the right direction, but it's a very rough implementation. Player frustration is a big pitfall of all real detective games, so it's always important to account for that. I think a good detective game makes a player feel gratified in his moment-to-moment thinking process, and provides satisfying feedback when he hits the nail right on the head (something that I also struggle with implementing) -- when you make a correct statement or accusation the game usually just spits back a "noted" or "acknowledged" back at you, but I think that this is the perfect place to throw a lead in the player's face (correct action => progression path).
Overall an above average DD game. I like it but I wouldn't call it a great detective experience yet.
Thanks for checking it out, I'm glad you've got patience for it. I do appreciate extensive feedback, it's certainly going to shape following weeks of work. There's a lot to unpack but by the next iteration it's certainly going to get better.