KestrelPi

Don't read too much into this, because, as he said after the Devolver showcase:

I used the journals a lot, and while they were initially a help they started being part of the problem. It was frustrating that there were lots of items in it that literally couldn't be solved yet because I wasn't at that point in the story yet. I spent so long calling Safely First and trying with different people, and different tools before I finally realised that the reason I couldn't get into the factory was just because i wasn't at that bit in the story yet. I'm not sure I agree. I went into this game with fairly middling expectations and found that they were met. Me being very lukewarm on this game is nothing to do with my expectations. I do have higher expectations for RTMI, because of the approach they are taking to the game - they've talked now so many times about how they don't want it to be seen as a throwback, and that they're doing things they're not ready to talk about which make it something of an update on how the genre approaches interface. But not only that, half the problem I have with Thimbleweed Park as I think it was a really, really bad idea to have so many lead characters. If it were my choice: The agents probably should have just been rolled into one character. Their reactions to each other aren't THAT interesting, and in most situations their dialogue is identical. Dolores and Franklin should stay about the same. Ransome should have been removed from the game and fired into the sun, or turned into a minor NPC so the fairground location could be kept. This would have made the number of characters to juggle much more manageable, and enabled them to work on the characterisation that much more. Heck if it were my choice I would have just rewritten the story to focus solely on Dolores (it practically does, anyway) and have the agents go away and stay away after the first parts of the game. I'm excited for RTMI, much more than I was for this, for several reasons: * One character with one inventory and set of objectives * Collaboration with Dave Grossman, who with his experience I trust to be an asset both in the writing and in the puzzle design * New approach to interface * It's a Monkey Island game and I think Ron and Dave know how to make those * Everything I've read about it makes it feel like this is being approached in a more collaborative way and I think that'll do it favours.

Looking at the sort of results you can get from this stuff, I'm surprised that more don't take this approach now that it's available. Maybe we will see more of it. It certainly improves the prospect of, say, a CMI remaster. But another thing I'm interested in is... how long will it be I wonder until computing power is good enough to achieve this sort of upscaling in real time? I don't know much about the process but I understand that it takes a while and is computationally quite costly, but how true will that be in 10/20 years? Will SCUMMVM be running an auto-upscale algorithm of this quality? 'Cause I'd be into it.

Here's my ridiculous prediction for the interface: Ron's decided to go all the way back to text parser for Return to Monkey Island, with GPT-3 driving it so that it can interpret most of the things you type. 😉

About the hint system... I ended up going to it quite a lot. And it wasn't because the puzzles overall were particularly difficult, it was more that... Well, in the middle of the game the possibility space is just extremely large but because the game is split into so many parts, the particulars of what you should be focusing on at any one time are pretty small. So there ended up being a lot of different locations explorable with a lot of different characters, and more often than not I just wanted to be told the thing the game wanted me to focus on next to progress the story. A lot of times, when I guessed what to do next and got stuck the hint system ended up telling me I should focus on something else first because I don't have what I need yet. A lot of other times, it wasn't that I hadn't been clever enough to work out the puzzle - it was just that I'd not thought to look at one not-obvious thing in one room, and that was the thing that was needed. Or I hadn't used it with the right character. The middle of the game feels very 'padded out' to me. There's not a ton of story happening, and not even any changes in scenery really - you basically have all the locations you're going to go to except for the factory pretty early on. And that had a kind of impact on my relationship with the hint system. I stopped 'trusting the process' of playing an adventure game. That if I wander around enough I'll have an idea and I'll have that 'oh!' moment where I figure it all out. Because every time I'd call up the hintline, it would tell me that it's because a particular character doesn't have a particular inventory item that I picked up 4 hours ago, and forgot about in the bottom of another character's inventory, or because it wasn't time to do something I tried 2 chapters ago, but now it's time (I spent so long trying to figure out how to get that crystal, and it turned out the answer was ... wait until the end). Or because I didn't try to pick up something and walk out with it in the fortune teller's shop because THAT'S NOT HOW SHOPS WORK, geeze, even Monkey Island 1 understand this. Or because I didn't look at something on the 8th floor of the hotel because going up and down that elevator is SO tedious it makes me not want to explore. Or because I hadn't solved a different puzzle that I had no reason to think was important to prioritize yet. Almost every time I called the hint line, I finished the call thinking 'yeah, I would have been wandering for hours before trying that', not because it was a clever solution, but because I really had been given no reason to try it as opposed to the billion other things in the possibility space I could be trying. And that was my core frustration with the puzzles, and the reason I ended up habituallly going to the hint system in the middle of the game. As I mentioned, I had a much better time with the puzzles in the end game. None of them took a LOT of effort to work out, but they all featured a satisfying 'oh! I see!' moment that was desperately missing from most of the rest of the game.

For every plausible synopsis, it came up with 20 more like this with implausible premises and plot details that only half make sense:

Well, you've seen my comments on the other thread now, and, well. I wish I could love it. But I just can't. I think it's... fine. In terms of atmosphere it's there, and in terms of characters, there are a couple of good ones but it's very inconsistent. and over all, I think I'm cool with where the story goes. But I just didn't find it funny. The jokes that were there really just washed over me, and I think that was a combo of ... just not finding the delivery very good and an over-reliance on joke 'formulas'. I enjoyed if for the first 2 minutes that the Pigeon Bros. were talking about the signals, then that being everything they said. Same with all the a-renos and a-boos. It's an okay bit, that is quickly driven into the ground. Same with Ransome's whole character. I found myself frustrated by it in the same way I was frustrated by EMI - there's recognisable jokes here, but I'm just not really laughing. And didn't find very many of the puzzles interesting or fun, I've already covered that in the other post, but basically my gripe is that most of the puzzles felt more like busywork than things I was actually working out. With the exception of the ending parts, that had some nice stuff, and a few of the cleverer bits around the hotel. I -am- glad you had a better time with it, on your second go round, but I couldn't help but be just a little frustrated with it. It's like a game designed to push all my nostalgia buttons, that is constantly juuust missing.

This is OpenAI, using their text playground. Pretty sure anyone can sign up for personal use https://beta.openai.com/

Oh, here, you can have the most plausible Grim Fandango sequel premise it could come up with: In Grim Fandango, Manny Calavera is stuck in the Land of the Dead, working off his debt as a travel agent for the dead. He meets and falls in love with Mercedes Colomar, a newly-deceased woman who is trying to find her way to the Ninth Underworld. In the sequel, Manny has managed to escape the Land of the Dead and is now living in the Ninth Underworld. He is reunited with Mercedes, and they are able to start a new life together. However, trouble soon arises when it is revealed that Manny's escape from the Land of the Dead was not entirely legal. The authorities in the Ninth Underworld are not happy with Manny's presence, and they want him gone. Manny must find a way to clear his name and stay in the Ninth Underworld with Mercedes. Along the way, he will encounter new challenges and face old enemies from his days in the Land of the Dead.

Today I've asked the AI to write an open letter from Ron Gilbert begging fans to stop calling him "Wrong Gilbert" Here are my favourites:

Here:

OK I'll do that but first, some of these are incredible:

Here are my favourite responses to a prompt I did about Ron Gilbert tweeting the end of Return to Monkey Island to spite fans:

I promise this was the first thing that came up:

In the interest of not flooding the AI thread, here is a place for more AI text nonsense fun. To start, a series of responses to the prompt 'Write a snarky multi-paragraph Mixnmojo.com news post titled "Tim Schafer bangs on Ron Gilbert's door and screams 'Wilma!'": '

I have to admit, I was curious: seems legit and... Wow, they really DO know us

I'll admit. Sometimes this scares me a bit: "Write an open letter from Ron Gilbert sarcastically pretending to apologize for ruining the Monkey Island franchise and listing the 3 things he'll do to make amends:" Dear Monkey Island fans, I am sorry. I have ruined your beloved franchise and for that, I am truly sorry. I know how much you all love Monkey Island and I have taken that away from you. But there is still hope! I promise to make things right. Here are three things I will do to make amends: 1) First, I will release a brand new Monkey Island game that is exactly like the first two games. No changes, no new features, nothing different at all. Just the same great game you remember from your childhoods. 2) Second, I will personally apologize to each and every one of you for ruining the franchise. You have my word that I will never again set foot near a Monkey Island game or anything related to it ever again. 3) Finally, in an attempt to win back your trust, I will give away free monkey-related merchandise to everyone who purchases the new game. T-shirts, hats, keychains – you name it! We want you to be happy with Monkey Island once again.

I'm now getting it to generate negative steam reviews, to recentre my expectations: "Return to Monkey Island is a huge disappointment. Not only does it have terrible graphics, but the plot is full of holes. For starters, the game completely ignores the ending of the previous game. In that game, Guybrush Threepwood became the governor of Melee Island, and married Elaine Marley. But in this game, they act like none of that ever happened. Secondly, the villainous LeChuck is somehow back from the dead, despite being killed in a very final way in the last game. And his new plan makes no sense whatsoever. Finally, the whole game feels like it's just going through the motions. There's nothing new or interesting here, and it all feels very stale. Avoid this one at all costs."

OpenAI's Playground using the text davinci 002 model

I'm constantly surprised by how funny this is: Write a sarcastic back of the box blurb for Return to Monkey Island: "What do you get when you cross the fourth wall, a talking pirate monkey, and Tim Schafer? The answer is: a whole lot of nothing. That's right, in Return to Monkey Island, you play as Guybrush Threepwood, a washed-up pirate who has to save his wife from a bunch of monkeys. With gameplay that consists of picking up bananas and saying 'Argh!' a lot, this game is sure to provide hours upon hours of mind-numbing fun."

I'm forced to admit that the AI is a pretty good comedian: "Improvise a letter from LeChuck to Guybrush, apologising for his behaviour and suggesting 3 remedial actions:" Dear Guybrush, I want to apologise for my behaviour. I know that I've been a bit of a jerk and I'm sorry. I hope you can forgive me. Here are three things you can do to help make up for my bad behaviour: 1) Buy me a drink next time we meet. 2) Let me win the next game of pirates we play against each other. 3) Give me a big hug. I hope this makes up for my bad behaviour. I promise to try and be better in the future. Love, LeChuck

Thanks, AI. Both for being no help at all, and also for adding 'LeChuck's Revenge' to my prompt.

I'll add it to the list. In the mean time: Predict the opening ten lines of dialogue between child Guybrush and his creepy brother Chuckie in the Big Whoop Amusement Park in Return to Monkey Island: "Hey Chuckie, what are you doing here?" "I'm just hanging out, what's up with you?" "I'm here to find the treasure of Big Whoop." "That's a myth! There is no treasure!" "Yes there is, and I'm going to find it." "You're going to get yourself killed looking for that treasure." "No I won't. I'll be careful." "Whatever you say, little brother. Just be careful."

I've now asked the AI to "Spoil five major unexpected plot twists in the upcoming game Return to Monkey Island:" 1. Guybrush is revealed to be a robot created by the Voodoo Lady. 2. The Voodoo Lady is actually LeChuck in disguise. 3. Elaine is possessed by LeChuck and tries to kill Guybrush. 4. Guybrush discovers that he is the reincarnation of LeChuck. 5. The final battle takes place on Monkey Island, which has been taken over by LeChuck's zombie army.

In the end, it's still just people. On a personal note, I've had several run ins with companies that employed me getting acquired and becoming bigger, and those have been overall pretty negative for me. I started out my career working for a pretty large media company and at first that was exciting but very quickly you started to see the low-level abuses that come with that - the idea that we should be happy with the prestige of working for a large company in exchange for lower pay, or less individual say, and whatever else. But mostly for me it was just the layers of bull everything became mired in. Being told that new policies were empowering when they felt like the opposite, feeling smaller and smaller and told like this was a good thing. When the companies I subsequently worked for got bought I started feeling similar stuff start to creep in around the edges and I got out as soon as I could, and never regretted that decision. It shouldn't be surprising that I personally have some mixed feelings about Microsoft acquiring Double Fine, even if I can fully appreciate and support the reasons that might motivate that, and am even excited about some of the potential outcomes of it. But the thing is, I suppose, is that these are still all companies that need to create things, and creatives are ALWAYS going to be motivated to put out their best work even if it has to all get filtered through a corporate lens. There's a reason why Disney films are usually still pretty enjoyable and well written and made when they come out, despite the ever expanding presense of the Disney Entity, and I guess it's because through all that, despite all that, there are still creative people, operating within that framework, who really care about the story they're trying to tell. Until the AI gets good enough to do away with THEM too, we'll always have that, at least.