Aoslare

There's an analogy here to a broader topic: the difference between manually-generated content and procedurally-generated content. For more than 4 decades now, both types of computer RPGs have thrived. One has never driven the other out of existence. That's because have different strengths and weaknesses. People like alhoon will say that procedurally-generated content is "infinite content," but really it's not. This is because, as you play it repeatedly, the actual game space of the game, aka the actual gameplay experience, shifts slightly from what it is in a static, deliberately crafted game. Consider a traditional RPG that has 6 different weapons, with strengths of 4, 8, 12, 16, 20, and 24. Each weapon also has 1 or 2 special abilities. And they are found or sold in different places at intervals throughout the game. Players will quickly figure out which ones are most useful, which ones are good values for different portions of the game, for different classes or characters, etc. And then those strategies will be the same every playthrough. Now consider a procedurally-generated RPG that has random weapon drops. Their strengths vary from 2 to 28, they have random combinations of special abilities, and the places they are found and sold in will vary. Players will quickly figure out which possible generated weapons are most likely to be useful, where the most efficient places to look for them are, which classes might care more or less about finding the best one, etc. And then those strategies will be the same every playthrough. The strategic game space moves from being about the static values to ranges of possibles values. In a way this is more complex, but it doesn't actually add any strategic depth. It's not richer than a deliberately crafted game, and it's certainly not infinitely rich. It is differently, likely refreshing, and worth playing for a lot of reasons. But the array of possible results doesn't actually end up being richer than a sequence of deliberately crafted ones. And the array is the same every time, the boundaries of possibility are the same every time, even though the exact weapons you see won't be.

Lmao. I don't mean to be rude, but I don't know how else to respond, we are living in such wildly different worlds. I think maybe this is the crux of the issue. Different people have very different standards for what feels "real enough" to not break immersion. It seems that there may be quite an immense gap between the ease of satisfying e.g. alhoon's standards, and mine.

This is not actually what ChatGPT does. It doesn't take your actions and words into account. It creates the illusion that it does that, but it doesn't actually do that. It basically just looks for matches, the same way a painter matches new paint to the color on an existing wall, which is why if you try enough variations of the same prompt it can start to feel like you're playing Mad Libs with a randomizer. It doesn't handle concepts holistically or process logic, has no memory, and has extremely limited ability to even maintain internal consistency within a single output of its own. "Infinite" is also demonstrably false. I'm not going to agree to disagree. You can do that if you want. But facts are facts, and this sort of AI model does not work the way you think it does.

I read your original comment; the "or more" wasn't lost on me. I didn't say "oh it's going to be 11 years, gotcha!" You were emphasizing that "it's coming," and I think that 10 years is a very, very optimistic estimate. That's all. The "or more" doesn't change that.

It's the thing I quoted and was responding to ("advanced AI ... They will so closely resemble real people making real decisions that your actions towards them and their responses in kind will be almost infinitely diverse"). I'm not arguing about this, but I will just say that you have some extremely misguided beliefs about what current AIs are capable of.

You're right. This isn't something ChatGPT is even remotely capable of. On about 10 different levels.

Yeah, those tooltips are frustratingly ambiguous. Jeff is updating some of them in v1.02 of Infestation, probably including that one. Extra spell skill (of whatever sort) does the following things: - increases direct damage/healing done by the spell - increases the duration of status effects caused by the spell It does not increase the odds of successfully landing a status effect, and it does not increase the power of a successfully placed status effect.

This reads almost like you are quoting straight out of Matt P's FAQ for original Geneforge 1. You've been doing this a lot. I can only repeat: - this FAQ is not for Mutagen - this FAQ is a good source for finding out where things are, but when it comes to strategy advice, it is extremely hit or miss, with a few fairly bad recommendations in the mix In Mutagen, unlike OG1, the only difference in how successfully an Agent or Guardian can use Dominate is the extra skill point cost of reaching Mental Magic 3. In OG1, increasing skill further had a huge impact on success rate. In Mutagen, it doesn't. So you could absolutely play a melee or missile Guardian who has some points invested into magic and uses Dominate. This might be simpler for an Agent, but there are still advantages and disadvantages to physical attacks, so the Agent will not be flat-out better in every possible situation.

we are nowhere near what is described here, despite the media misunderstandings of current ai models. this isn’t just something that has to be integrated in rpgs, it’s something that has to be born. 10 years for this to be in games is optimistic at best.

pick one. trying to be good at more than 1 of those 3 will just mean you can do 2 things less well. agent for battle magic, guardian for melee or missile.

The gloves receive an extra 1 die of damage per point of strength, in addition to the regular 5%damage bonus. afaik melee weapons skill works as usual.

Trajkov was not a Trakovite, he had nothing to do with them. Their ideology wasn’t based on a lie, it’s just that their name — and only their name — was based on inaccurate information that everyone in the world thinks is the historical truth.

That's definitely not worth spending skill points on. Control isn't even something you need to worry about much if you have a strong team of creations. But when it does matter, which is mostly just early on if even then, this can be managed by titrating your number of creations versus shaping skill (and not leaping to high level creations right off the bat). Every PC level you gain reduces your control problems, so points you put into other shaping skills are ultimately wasteful -- and not a very efficient way to impact control anyway.

Hey Lorn, can you please add in a warning to make a backup first, since you are directing people to install edited game files? Thanks.

This isn't really an effective way to use battle creations. The ranged attacks of clawbugs are not very strong and also not something you can reliably repeat -- if you are really into AoE poison (as you seem to be) it's an option I guess, but it will be dramatically worse than just using fire or magic. Want you really want to do, with thahds or alphas, is have a leap attack you can use to get into range in round one, and then just stay in close melee range. And then with rots, you really want to start combat as close to the enemy as possible -- Stealth helps a lot here.