Teh Pizza Lady

I'm sorry, maybe I'm misunderstanding you entirely then or we're using different definitions for "burst damage". Spears still hit harder per throw, but burst isn't just about a single hit of damage. It's about how much damage you can do in an amount of time. If I can land two arrows in the time it takes to throw a single spear (and hit the target since spears are harder to aim and less accurate), then the bow is effectively doing more burst damage than a single spear toss which makes it the better ranged option despite having a lower per-hit damage. Also if you are able to accurately land any sort of ranged hits during a temporal storm, please teach me your ways. I almost always have to resort to melee to do anything during a storm because the visual distortion is too high to land any hits further than 10 yards.

A steel arrow (+2.5 dmg) fired from a recurve bow (4.0 base dmg) does 0.5 damage less than a steel spear (7.0 dmg) when thrown, stack way higher, can shoot further, faster, and with better accuracy. The bow and arrow IS your burst damage. They just require a bit more work to get there. The spear only does as much damage as it does because it's a larger weapon, but is slower, cannot be thrown as far or with the same accuracy as the specialized ranged weapon. To me, that's not worth the extra 0.5 damage per hit, especially since I can fire off two shots in the time it takes me to throw a spear and ready a 2nd one. You can check the wiki pages for the bow, arrow, and spear if you dont' believe me.

I believe the changes were made because players were just defaulting to spears and not taking the time to invest in other forms of weaponry, like the falx or the bow and arrows. Even as a hunter, I was favoring the spear for melee because it could also function as a ranged weapon in a pinch. Now you can fire off more shots from a bow and arrow, shoot further and faster with more accuracy, making it more valuable than a spear for long ranges. The spear was just supposed to be a dual-use weapon but it was powerful enough that it functioned better than the weapons that were specialized for those purposes. The nerfs were 110% necessary. Yes they went a little hard, but in rc.6, the spears feel a LOT better.

I generally classify these as stepping stones to Iron/Steel, so the general player is not encouraged to keep using them. The problem I see with this is that the instant you are able to collect a resource, you are also able to use it. Take for example stone. The only way to collect stone is either by picking it up or mining it. If you pick it up, you can combine it with clay to make cobblestone. Clay is easily acquired with a flint shovel, for example so you are able to use stone the instant you gather it. Same goes for copper, bronze, iron, etc, for the most part (though there is a prerequisite for using mined copper: having a hammer to crush the ore bits). The reason this is a problem is there is no way to gate the player by requiring them to have certain buildings for forging because the instant the player spawn into the game with nothing but a couple pieces of flint, a rock and two sticks they are instantly able to start constructing sturdy buildings. The requirement of crafting and using a certain number of bronze tools also presents a problem in that bronze is, again, a stepping stone on the path to iron and steel. You aren't supposed to use it any longer than you need to. For the new player, this makes sense, but for the seasoned veteran, this is only a nuisance for their desired playstyle to be dampened by a meaningless requirement to make and use a tool for which they have no desire. I agree that there could be some sort of checklist for newer players to follow to make sure they are prepared for advancing to the next ages, but those advancements only really come when the players already have what they need. If they are missing a requirement, for say... Iron, then they won't be able to advance until that requirement is met, be it a supply of coal-like fuel, or the fire clay needed to make bloomeries. And even so without a bronze anvil and other tools, that iron will be pretty much useless until those requirements are also met. Just build double the forges and refineries #fivehead

They just do that. Even in wetter areas we still end up having to water our crops every so often to keep them growing at 100% growth rate. You could plant your garden next to a pond and it would still need to be watered.

I think that link you found is more for forum support. They only accept bug reports through the official trackers now.

No need for that. All I'm saying is if you're going to make a stance on a public forum, at least get your facts straight. That's basically how I see it, too. However the stigma against it comes from the fact that the programmers stole images that somebody else owned and used them to create their programs. That's where I draw a hard line. The AI I used to generate my PFP here doesn't use stolen art to the best of my knowledge because it was created by a derivative of nVidia's Cosmos 2 model. I can't say the same for other AI models.

NO need... unless it wasn't a real story...

Also Salty's floppa is a vector image, not AI-generated.

Except: Salty and Proto created one of the best mods on the ModDB currently: Temporal Symphony. Granted not all of Salty's mods have been easy on performance, but I don't think he uses AI to create them. The things he does with the game, its engine, and the mod features are just too complex for AI to do the way he does it.

Thankfully survival goods traders are prone to selling it.

Oh no... someone already suggested a diving bell and that thread did NOT go well. Spoiler: it was me who suggested it.

I was going to reply "AGAIN??!?" but then I remembered it wasn't ever actually fixed so maybe it's intentional. At least we can see temporal rifts again. Those were scary before!

I thought that was just the standard experience when milking a goat. TIL this was a bug.

The only drawback is that the bloomeries require fire clay to make, which still requires calcinated flint unless you happened to get lucky with a clay deposit.

yeah from my understanding of the code it's going to take roughly 64x more fuel to heat up a full stack of flint. The closer to the max temp you get of the fuel, the slower it will heat up. The mod I made simply patches around that code and forces it to cool down at the same rate so that you don't have to keep the fire burning for nearly as long. I also noticed that the cooking temperature for flint is now 1000C which means I think that you would be better off just making a bunch of charcoal and getting it HOT as quickly as possible and then just placing a bit in every now and again to keep it from cooling off too much and resetting all your progress. So it's a bandaid, for sure, but I think I would accept some feedback on how to tune the whole firepit situation so that the thermal management of the fire isn't a needless and arduous task that takes up all your time. Heating things up should take more fuel for a larger stack for sure, but I don't think it needs to be quite so linear.

mostly agreeing. Anyone can still make use of the tool. I think the nuance here is less how much the tool was used or even who used it, which is why people ask, and more how it was used. I wouldn't trust a carpenter with a monkey wrench to cut a board in half, but the electrician using a saw will find your water pipes faster than you think. I think if you understand the code that the tool produced, then you shouldn't have to put a disclaimer that you used the tool to create it. The idea there is that if the correct person is using the proper tool for the job then everything will go okay, but if you're trying to use it incorrectly or for a different task, then things can go sideways fast. Visual Studio comes with CoPilot garbage anyway nowadays so it's likely that nearly every mod produced will have some AI-generated code in it now. I particularly prefer the Windsurf IDE if I'm doing any LLM-guided exploratory programming since it allows me to make use of Claude Opus 4.6, but that is another conversation for another time and the keyword there is "exploratory" in the sense that I'm just noodling around with code, exploring concepts and trying to see what is even possible before committing to a larger project.

I wanted to respond, but I was waiting for time to give you a proper response. I think you're mixing up whether something works with whether or not it exists. "Vibe coding" is 100% a thing. Early on, there were even cases of people trusting AI blindly and doing real damage, like wiping production data because the model “decided” to rebuild a system from scratch without even asking because it couldn't understand how to work with the existing code to make the necessary edits. I've even tested the best coding models and they can barely figure out how to fix issues even when I explain to the LLM exactly what is happening and why. Personally, I have reviewed PRs on GitHub from people doing exactly that, prompting their way through entire program features without understanding any of the output and blindly trusting the AI knows what it's doing. The result was dozens of bloated files, duplicated and inconsistent logic, and the product barely worked after weeks of "effort". So yes, AI needs oversight. But the failure mode you have described is exactly what people mean by vibe coding and not only is it real, but it produces horrid results when not guided by a real developer who actually understands how to program. I also think that calling it "agentic coding" is going to create it's own problems. It can make a workflow sound more deliberate and reliable than it actually is, which risks normalizing over-reliance on AI over good, old-fashioned human ingenuity and creativity. A bad developer will treat it as a crutch and never seek to gain understanding needed to maintain what it is they're creating. A good developer can fall into an entirely different trap, letting standards slip and trusting the AI's output when they haven't fully reviewed it because "it looks right and worked in testing". Both lead to the same outcome: poorer code quality and a code base that is harder to maintain. And back to the PRs I've reviewed on GH. I've used the products behind those PRs and the common issue wasn't bugs. It was that the end product felt flat. The individual pieces worked individually, but the design around them didn't make sense. It wasn't intuitive. It didn't feel human. The difference between generated content and created content is stark. That’s also why you see people asking on ModDB how much of a mod was coded by an LLM. It’s not always anti-AI sentiment. A lot of the time, it’s a practical concern: if something breaks, will the person who made it actually be able to fix it?? I don't want to use a mod that doesn't work and won't get fixed, would you?

I feel like that flax number is a little high. Have you tried the rapids and waterwheels yet that are coming in 1.22?

Welcome to the forums! Glad you're enjoying the game. I'm really hoping to see some sort of advancements in weaponry that we can unlock as we play through the story chapters. Maybe discovering a blueprint left behind that we can use to craft a firearm or something that will be useful as the game progresses, something that we cannot just "discover" through normal gameplay. Overall it's a fun game, it's ONLY in early release, so I hope you continue to enjoy it as it improves and gets more and more fleshed out over time!

I mean I just got an email on my work laptop today that they're now using AI to write the quarterly reviews. Yesterday they published metrics on how AI is being used and just last month GitHub's CoPilot wrote over 50,000 lines of code for my organization. It's pretty sad when you think about it.

Not entirely true. Xskills allows you to earn skills via gameplay, yes, and it is easy to fall behind in multiplayer. However there is a cap on skills because of how the mastery system works in that mod. You can learn skills, but you can only master one or two of them. Maybe three if you take the right talents. I'm not sure.

The problem is that people are mostly distrusting of AI and don't realize just exactly how long it's been around. Facebook moms got wind of it and demonized it without truly understanding what it was they were even saying. Remember Cleverbot from the late 90's and early 2000's? It was the first primitive chat AI program and people LOVED it. But even older was ELIZA from the 1960's which even made an appearance in the Young Sheldon TV show though they never actually said that's what it was. As @LadyWYT said, it's a tool, which can be very powerful when correctly used. I think where a lot of people are drawing the line is when someone who doesn't actually know what they're doing comes along and types a prompt "Create a Vintage Story mod that does xyz and interfaces with these mods. Create a description for it for me to copy to the ModDB website and also generate a README document for it. Also summarize any changes made so I can copy and paste into a git commit message" and sit back and let the AI do all the work. Compare this to "Here's the link to the VintageStory modding API. Find the class/method that allows me to access XYZ and give me the link so I can review it myself." Then while it works, you're free to think about the next steps. It's allowing you to do two things at once. Then you start typing some code, and get something workable and ask it, "Am I crazy or will this code work?" It crunches away and points out a syntax error that would prevent your code from working properly. Or you're trying to remember for the umpteenth time how to setup a Harmony patch. Or you just ask it, "Is it possible to do xyz in C#??" and after it crunches away it says, "yes and here's how". It's one thing to have a vague idea and obtain something passable. It's another thing to have a truly inspired idea and use the tools available to bring it to life in a way that no one has done before. One is thoughtless and uninspired. The other is thoughtful and shows that the modder actually has two brain cells to rub together. One is having your 3-year-old splash paint on a canvas and trying to sell it as an original Picasso. The other is letting your 3-year-old rough in a pencil sketch and you go back over it and clean up the lines and make something truly original, unique, and fun. The AI gave you the rough sketch. It was up to you to figure out how to connect the lines. There's no shame in that. Take pride in your work. The only person whose opinion should matter to you is yours. 99% of the detractors aren't even mod authors anyway.

AI = Artificial Intelligence Currently AI cannot exceed a quarter of what a human mind could achieve. This is why your most powerful computers in scifi games and TV Series use human brains at their cores. That said, it can help organize things, help reason through processes, act as a sounding board, etc, but it should never be in the driver's seat and certainly shouldn't ever perform the majority of the work when developing a game or mod or creating a picture. AI can only replicate that which has already been created. It cannot create anything new.

so it WAS a bug.... interesting. so no more cooking the space trip out of the mushrooms.