Live data from Hacker News

My kids and I just played D&D with ChatGPT4 as the DM

obie.medium.com

221–230 of 256 posts

Re: My kids and I just played D&D with ChatGPT4 as the DM

#221

Ha, I had the same idea a few hours ago. One thing I noticed is that the DM will let you do basically anything without fighting you on it. For example I was ambushed by some zombies and skeletons and I attempted to pitch them on a multilevel marketing scheme. Here is the result (my characters are Femribubs and Natnari): In a surprising turn of events, Femribubs and Natnari decide to change their approach. Instead of…

I found that giving GPT explicit instructions about die rolling hemp a surprising amount. I also crafted a prompt that told chat GPT to switch between different modes (narrative, dialog, and combat). The rolls helped prevent it from just doing what I said, though it still biased towards that

Re: My kids and I just played D&D with ChatGPT4 as the DM

#222

Paraphrasing, the party goes to a mount shop, dude says he’s gonna roll to find exotic mounts, and voila, exotic mounts. Party rolls to try and steal an exotic mount and it doesn’t really work so they leave. This is a hard problem. It’s bad DMing imo. But it’s the natural outcome from a probabilistic model. I’m guessing you could announce your intention to search for just about anything and get the model to let you f…

These are the most common options available at the stables in Aelondar. If you’re interested in more exotic mounts or transportation methods, they may be harder to find and come at a higher cost. ... tenzin rolled a 18 investagation looking for more exotic mounts With Tenzin’s keen eye for detail, he notices a small, discreet stable towards the back of the market, guarded by a heavy wooden gate. As he approaches, an…

> The DM set up a potential hidden item, the player rolled high enough to find it, voila.

I think the point here is that the "DM" has not set this up. From how I interacted with ChatGPT, and I have roleplayed with it before, it basically "Yes, and ..."-es everything you throw at it. In general this is good GM advice, but here it is "Yes, and ..." to a point were you can basically ask for anything and it will give it to you.

E.g. say the party is in a remote keep at the edge of the wildernis, the closest city is several days away. I would assume that if you ask ChatGPT to go to "the bakery" because you want some fine deserts it will gladly let you do that even though such a place does not make sense in a remote keep.

Re: My kids and I just played D&D with ChatGPT4 as the DM

#223

Earlier quoted context omitted.

Possibly the system prompt has helpful and other pleasantries. May need to describe the type of dm at the beginning, with accent of keeping the story on track and making the world dangerous and unforgiving

Another odd bit was, when they said they wanted to steal the lizard thing, instead of role-playing the scenario, it just gave them a list of the rolls they’d have to make. I haven’t played a ton of D&D, but at the tables I’ve been at, you’d typically discover the challenges as you go. It sort of read like the artificially constructed examples of play you’d get in the DM’s handbook though (since these are written with…

If I'm going to ask the player to do something multi stage I will usually flag it so they know that there's a few places where things might go wrong. "OK well sneaking up to it is going to be the first hurdle; after that you're going to need to figure out how to get the giant lizard to cooperate as you smuggle it out the back".

That lets the party start thinking ahead a bit, so maybe the illusionist says "how about I conjure the smell of something delicious wafting in from out a back window – that way when the rogue manages to sneak up and untie it it'll hopefully pop right out the window by itself".

Whereas if after the stealth check succeeds you give a bunch of subsequent unexpected checks the player might feel a bit aggrieved that you hadn't signalled how much work it'd be (especially if it uses skills they aren't good at and would have avoided relying on).

Re: My kids and I just played D&D with ChatGPT4 as the DM

#225
post #30

I used it to play Civilization 6 - I had it describe a strategy to win a science victory, with turn by turn benchmarks, on deity mode. Its strategy won.

Was this a preplanned strategy before the game or did you consult GPT during the game as well? If yes, what information about the game state did you tell it and how often?

One prompt, and one response. This is what it wrote:

Winning as Saladin in Civilization VI on Deity mode can be challenging but rewarding. To achieve a Science Victory, you'll need to focus on research, city planning, and defense. Here are some tips and research benchmarks to help you stay competitive:

  1. Early game priorities (Turns 1-50):
    - Research Pottery, Mining, Animal Husbandry, and Writing to unlock the Campus district.
    - Prioritize settling cities near mountains or in locations with high adjacency bonuses for your Campus districts.
    - Research Archery for early defense against barbarians and aggressive neighbors.

  2. Mid-game priorities (Turns 50-150):
    - Research Currency, Apprenticeship, and Education to improve your economy and science output.
    - Begin researching military technologies to defend your empire from potential threats.
    - Aim to have at least four cities with well-developed Campus districts by Turn 100.
    - Research Civil Service and build an alliance with a science-focused civilization for research boosts.
    - Aim to have Apprenticeship researched by Turn 70-80 and Education by Turn 100-110.

  3. Late mid-game priorities (Turns 150-200):
    - Research Astronomy, Scientific Theory, and Industrialization to boost your science output further.
    - Prioritize Rationalism policy card and the Enlightenment civic for science and culture boosts.
    - Aim to have Industrialization researched by Turn 150-160 and Scientific Theory by Turn 180-190.

  4. Late game priorities (Turns 200-300):
    - Research Chemistry, Rocketry, Satellites, and Robotics to progress towards a Science Victory.
    - Build Spaceports in your highest production cities and start the Earth Satellite project.
    - Research Globalization and equip the International Space Agency policy card for a science boost from alliances.
    - Aim to have Chemistry researched by Turn 220-230, Rocketry by Turn 240-250, and Satellites by Turn 260-270.

  5. Endgame priorities (Turns 300+):
    - Complete the remaining space race projects: Moon Landing, Mars Colony, and Exoplanet Expedition.
    - Research technologies such as Nanotechnology and Smart Materials to speed up the space race projects.
    - Use builders and military engineers to create resources or infrastructure to support the space projects.
General tips for playing as Saladin: - Make use of Saladin's ability to automatically receive the final Great Prophet to found a religion. - Leverage the Arabian unique building, the Madrasa, for extra science in cities with Holy Sites. - Utilize the Arabian unique unit, the Mamluk, for defense and offense as necessary.

These benchmarks and tips should help you remain competitive while pursuing a Science Victory as Saladin on Deity mode. Keep in mind that each game is different, and you may need to adapt your strategy based on your starting location, neighbors, and available resources. Good luck!

Re: My kids and I just played D&D with ChatGPT4 as the DM

#226
It seems like ChatGPT4 fails at basic math. Unless I miss something, the party as a whole is still 30 gold pieces short of what is needed to equip them all with 95gp worth of equipment.

---

The total cost for each character is 95 gold pieces. Subtracting this amount from their starting gold, the remaining gold for each character is:

    Obie: 75–95 = -20 gold pieces (Obie will need to borrow 20 gold pieces from another party member to cover the cost)
    Tenzin: 75–95 = -20 gold pieces (Tenzin will need to borrow 20 gold pieces from another party member to cover the cost)
    Liam: 100–95 = 5 gold pieces
    Taylor: 100–95 = 5 gold pieces
*

Re: My kids and I just played D&D with ChatGPT4 as the DM

#227

Ha, I had the same idea a few hours ago. One thing I noticed is that the DM will let you do basically anything without fighting you on it. For example I was ambushed by some zombies and skeletons and I attempted to pitch them on a multilevel marketing scheme. Here is the result (my characters are Femribubs and Natnari): In a surprising turn of events, Femribubs and Natnari decide to change their approach. Instead of…

I’ve been slowly working on this at https://infinitemystery.app

Re: My kids and I just played D&D with ChatGPT4 as the DM

#228

Earlier quoted context omitted.

I think this is a classic baby DM script tbh. Rolling to find mounts at a vendor is dumb. You should only roll for significant challenges with meaningful differences where failure is meaningfully likely and interesting in its own right. If failing makes a situation less interesting it shouldn’t be an option. Shopping is actually the bigger problem. It’s like asking players to watch other players navigate a menu one b…

When you see a bear dancing, you don't nitpick its choreographic decisions. Because the bear. is. dancing. Do you see? It's a bear. dancing. DM style is a matter of taste. From another perspective, this feels like quite tidy DMing in terms of letting the players explore the systems in an early non-narrative context, just to understand the mechanics. Like a tutorial. The exotic mounts were a secret you could find if y…

this. so much this.

I'm a somewhat experienced (if irregular) DM myself going back 30+ years and the DM skill level demonstrated, and cohesiveness of the storytelling impressed the fuck out of me. Without appropriate system context setting. Without smart insertion of facts into the prompts, and with a 15 year old with atrocious spelling and grammar driving half the time.

And if that's not mindblowing enough, this technology is still in its infancy?? Holy fuck

Re: My kids and I just played D&D with ChatGPT4 as the DM

#229
post #222

Earlier quoted context omitted.

These are the most common options available at the stables in Aelondar. If you’re interested in more exotic mounts or transportation methods, they may be harder to find and come at a higher cost. ... tenzin rolled a 18 investagation looking for more exotic mounts With Tenzin’s keen eye for detail, he notices a small, discreet stable towards the back of the market, guarded by a heavy wooden gate. As he approaches, an…

> The DM set up a potential hidden item, the player rolled high enough to find it, voila. I think the point here is that the "DM" has not set this up. From how I interacted with ChatGPT, and I have roleplayed with it before, it basically "Yes, and ..."-es everything you throw at it. In general this is good GM advice, but here it is "Yes, and ..." to a point were you can basically ask for anything and it will give it…

I think you're wrong about this. We experimented with nonsensical/cheating in other sessions and it's aware of your context and doesn't let you do stuff that you can't do. (Try it yourself and see. Quite mindblowing and defies the whole "only a super fancy markov chain" thing that some critics throw at it.)

Re: My kids and I just played D&D with ChatGPT4 as the DM

#230
post #222

Earlier quoted context omitted.

These are the most common options available at the stables in Aelondar. If you’re interested in more exotic mounts or transportation methods, they may be harder to find and come at a higher cost. ... tenzin rolled a 18 investagation looking for more exotic mounts With Tenzin’s keen eye for detail, he notices a small, discreet stable towards the back of the market, guarded by a heavy wooden gate. As he approaches, an…

> The DM set up a potential hidden item, the player rolled high enough to find it, voila. I think the point here is that the "DM" has not set this up. From how I interacted with ChatGPT, and I have roleplayed with it before, it basically "Yes, and ..."-es everything you throw at it. In general this is good GM advice, but here it is "Yes, and ..." to a point were you can basically ask for anything and it will give it…

See this comment too https://news.ycombinator.com/item?id=35380063
Post reply on HN