JointedRail stuff [screenshots/renders]

I'm sorry man, but posting this comment out of the blue in a thread where some of the most talented creators in any gaming community are showcasing their efforts for us to enjoy? That is... Brutally disrespectful, to say the least. Over the past 15+ years, I've watched the quality of content available for this game progress to a level that I (nor any of us, I'm sure) never could've imagined to be possible back then. And every step of the way, it has been a purely human effort. Never once have we, as a community, required the tools of AI programs in order to realize our visions, nor to push the envelope of what is possible, and there is absolutely nothing to suggest that we are even close to reaching the limits of what can be created for the game before needing AI assistance. It's always been a passion project of people who love trains putting their mind to it. Why do we now suddenly feel the need to suggest doing away with the very thing that makes this game so great? Using AI tools fundamentally goes against the spirit of this community, especially since it doesn't actually fill a hole in content creation. I really don't care how "good" AI content can be; I simply don't want it coming anywhere near this game.
There's nothing disrespectful about asking if AI can assist content creation in Trainz. Automation is coming to Trainz and just about everything else you can think of in real life, like it or not. For people like me who don't have the knowledge or wherewithal needed to engage in full-fledged 3D modeling this is a way to help out. I certainly don't have the money to spend on 3DS Max which is very expensive and my attempts to pay 3rd party sites to teach me as an apprentice have so far been rebuffed. Besides AI is already in Trainz. How do you think one can dispatch all these different trains without having to babysit them constantly? Your comment without even bothering to read between the lines is brutally disrespectful.
 
There's nothing disrespectful about asking if AI can assist content creation in Trainz
You literally just said, to Justin's face (unprompted I might add), "Hey I wonder if a computer can do your passion project for you". If this was me, it would feel akin to being spit in the face. You're knocking on his door and asking him to consider throwing away all the hours of research, studying reference material, visiting the prototype, honing skills in 3D modelling, texturing, sounds, scripting, etc. in favour of a computer program's interpretation of a few sentences typed up for it to waste a bunch of water whilst processing. You're right that automation is everywhere. And guess what? Being replaced by it sucks. Like, it really freaking sucks. Again, whether it can equal JR's work is irrelevant. Ask him if they want to be replaced by it.

Your comment about being too unintelligent to grasp one of the most well-documented programs available anywhere is also not an argument in your favour (Blender is free and there are endless free tutorials available online for it. Like, there's an entire community built around just 3D modelling). When the only barrier to entry is your brain, then that is a form of quality control (in other words; skill issue). Removing that barrier with regards to trainz with AI may yield some gems, but over time, this will absolutely become a needle-in-a-haystack situation. Remember those wonderful AI generated buildings that were uploaded to the DLS recently? "Well if you don't like it, then don't download it!" You may say, but I don't want to be forced to sift through an ocean of garbage on the DLS while looking for something actually properly put together. If AI content creators don't want to receive subpar feedback, then maybe they shouldn't upload subpar content.

I also don't understand what your point is in bringing up Trainz driver AI, as that is in no way shape or form related to the AI generation programs we are discussing here. Maybe you should learn to read between the lines of a blender tutorial?
 
Last edited:
You literally just said, to Justin's face (unprompted I might add), "Hey I wonder if a computer can do your passion project for you". If this was me, it would akin to being spit in the face. You're knocking on his door and asking him to consider throwing away all the hours of research, studying reference material, visiting the prototype, honing skills in 3D modelling, texturing, sounds, scripting, etc. in favour of a computer program's interpretation of a few sentences typed up for it to waste a bunch of water whilst processing. You're right that automation is everywhere. And guess what? Being replaced by it sucks. Like, it really freaking sucks. Again, whether it can equal JR's work is irrelevant. Ask him if they want to be replaced by it.
Comparing my question to spitting in someone's face just goes to show how much of an idiot you are. That's a textbook case of the false equivalent fallacy. By the way, it's not about replacing anyone's work. It's to assist in the amount of time needed to do the work. It's called efficiency. I don't particularly care for AI myself but in this world where people with BILLIONS of dollars basically control EVERYTHING, anyone who tries to regulate or even ban AI will be deemed as antiquated, stuck in the past, unable to compete in the real world amongst other things. That's what the lobbyists will say on Capitol Hill and will spend millions of dollars to get their way because it's profits over people. That's the price you pay for free market capitalism. Justin can NEVER be replaced by AI because someone has to tell it what to do. There's a lot that goes into making the models but unlike him I don't have unfettered access to live locomotives because he did or does work in the railroad industry. I applied and got a start date but was turned down due to my criminal record being longer than a CVS receipt.
 
Your comment about being too unintelligent to grasp one of the most well-documented programs available anywhere is also not an argument in your favour (Blender is free and there are endless free tutorials available online for it. Like, there's an entire community built around just 3D modelling). When the only barrier to entry is your brain, then that is a form of quality control (in other words; skill issue). Removing that barrier with regards to trainz with AI may yield some gems, but over time, this will absolutely become a needle-in-a-haystack situation. Remember those wonderful AI generated buildings that were uploaded to the DLS recently? "Well if you don't like it, then don't download it!" You may say, but I don't want to be forced to sift through an ocean of garbage on the DLS while looking for something actually properly put together. If AI content creators don't want to receive subpar feedback, then maybe they shouldn't upload subpar content.
Now you want to insult my intelligence. I'm probably way smarter than you'll ever be. By the way, I used GMax 15 years ago when still in high school. It was courtesy of Phil Campbell and World of Trainz. There was one particular step I had to follow but due to the lack of clarity in the instructions it was almost impossible to figure it out. Now that he's gone along with all the tutorials he made, I haven't been able to go back and figure out where things went wrong. Now if someone has a link to those tutorials that isn't compromised with anything that would harm my computer, then I'm all ears. As for it being a needle-in-a-haystack situation, that's already the case given how much of the DLS content is mostly 10+ years old and the tools needed to create modern up-to-date content has changed drastically. Furthermore, it continues to evolve so that makes your premise invalid. I've had to dig through a pile of garbage numerous times to find my desired items, but you don't hear me complaining although I'd be justified in doing so.
 
I also don't understand what your point is in bringing up Trainz driver AI, as that is in no way shape or form related to the AI generation programs we are discussing here. Maybe you should learn to read between the lines of a blender tutorial?
AI is actually related to this topic because Auran/N3V Games had to come up with a program to create that feature which also includes coding used for generating the commands users want computer-controlled trains to execute. Perhaps you should read between the lines and do some more research before speaking. There's more than just Blender you know.
 
You're knocking on his door and asking him to consider throwing away all the hours of research, studying reference material, visiting the prototype, honing skills in 3D modelling, texturing, sounds, scripting, etc. in favour of a computer program's interpretation of a few sentences typed up for it to waste a bunch of water whilst processing. You're right that automation is everywhere. And guess what? Being replaced by it sucks. Like, it really freaking sucks. Again, whether it can equal JR's work is irrelevant. Ask him if they want to be replaced by it.
By the way, AI can't replicate the studying of reference material, research or visiting such prototypes content creators want to build for Trainz. Now for 3D modelling, texturing, sounds, scripting, etc, you mean to tell me Jointed or RRMODS wouldn't use AI if it could substantially decrease production needed for their respective projects provided that it's reliable?
 
I wonder if AI and ChatGPT can actually produce the same quality of work you guys do with some skills and knowledge with automation.
No. AI will never be able to replicate something of this quality. Basic buildings and cars, sure (with error), but anything more complex than a locomotive requires human input. AI is far too inaccurate and unreliable to do anything that complicated by itself, and if it did a human would spend weeks fixing errors and cleaning up the model, UV mapping, fixing texture errors, assigning materials, etc., enough time to go in and create a model by hand. It is not efficient in the slightest.

I say, instead of spamming the JR suggestions threads with locomotives every week and insulting users for their human-centric opinion on Trainz content creation, use your hot-shot AI to make those C30-7s, B30-7s, what have you, or get off your high horse.

You thank that individual for it temporarily getting off track.
You started it, not him.
 
I mean, if you're openly and immediately endorsing AI because you like the idea of it being a content creation megaphone for people who are not intelligent enough to understand content creation, whilst simultaneously identifying as one of those people (I'm not insulting your lack of intellect if you're proclaiming it right off the bat, and don't try to pretend like you didn't), then I can't imagine you're intelligent enough to understand the enjoyment of said content creation either. However, I'm going to refrain from stooping to your level any further, and will instead try to elaborate my thoughts on the AI question as best as possible. Past this sentence, this comment is no longer directed specifically at you, and is meant moreso as food for thought with regards to the AI-Trainz debate for the wider audience reading this thread, so you also need not reply any further (Sorry to JR for derailing your thread, and sorry for the impending wall of text; this will be my last comment on the subject in this thread).

So just to preface this, and to give folks reading this thread an idea of where I'm coming from with this; I have a route in this game that I've been working on for 3-4 years, which is absolutely not at all efficient. And it's not close to being finished either. But I enjoy working on it. In fact, I have enjoyed all 3+ years of putting the route together, despite having stripped it bare of all scenery and redoing everything from the ground up multiple times. And I went through all that effort because I have a creative vision, and I find the tools in this game to be a near-perfect platform for me to realise this vision. I believe the ease of content creation that comes with this game is likely what has kept me around for so long (Will have first installed UTC 20 years ago before the end of this decade), and I find myself spending significantly more time reskinning stuff, building routes, etc. than I do actually just running trains. I would like to really emphasize this point: I personally find it more fun to create trains than to drive trains. You would be utterly shocked how many people actually use this game primarily as a creative outlet/form of artistic expression. I am not personally incentivized by the idea making a profit off my route whatsoever, even after all the blood sweat and tears I've invested into it. I continue chipping away at this project simply because I enjoy doing so (I have posted previews of my work elsewhere on this forum in the past). Nobody is paid to keep their mind busy in their free time, but we all do it anyways.

So having said all that, this is the fundamental point here: People create content for this game because it is fun to do so; efficiency/profit motive be damned. I don't care how long my route takes for me to finish, because it's not just about finishing it. Quality over quantity is absolutely essential. But when I do eventually finish it? My plan is to make a second version set in the steam era. That will likely take me another several years' worth of work, and that prospect excites me greatly. I have stressed over the feeling that I'll never finish the route, and that I need to get it done and released as fast as possible, as if I have some arbitrary deadline to meet. Whenever I get into this mindset, I don't find myself becoming more productive; rather the opposite happens. It stops being fun to work on (which is the whole point) and progress slows down and sometimes even grinds to a halt. And yet, theoretically, were a day to ever arrive where a new version of trainz includes a feature that automatically generates content based on a prompt, that will most likely also be the day where I, along with most other dedicated content creators, pack my bags and leave.

This whole thing inherently devolves rather quickly from "Should we use AI to generate content for trainz", to "Should we automate the things that we do for our own enjoyment", and I believe the answer to both questions is "No". The reason for this, is that a fundamental aspect of art and creativity is that it's about the journey as much as it is about the finish line. AI generation is forever at odds with this reality because the entire basis of it's existence is to completely eliminate the journey. Trying to frame the debate of AI generated trainz content through any lens other than personal enjoyment in the form of art/creativity/self-expression is irrelevant here, because those things are why people create high-quality content for this game in the first place. 99.9% of us devoted content creators are not turning a profit from our contributions to this game; not that that was ever the intent to begin with. You realize very early on that trainz content creation is not viable as a primary source of income (JR/RRMods are exceptions to this rule, not examples). You can use AI programs to create content without a profit motive, but again... What is even the point of automating it if I enjoy doing it? Like that's the whole damn reason I'm still here in the first place.

Having a creative vision and setting your mind to it and seeing it through to completion without a profit incentive is a fundamental human experience. No AI program will ever be able to subjugate the immense satisfaction that comes with actually doing something yourself.
 
No. AI will never be able to replicate something of this quality. Basic buildings and cars, sure (with error), but anything more complex than a locomotive requires human input. AI is far too inaccurate and unreliable to do anything that complicated by itself, and if it did a human would spend weeks fixing errors and cleaning up the model, UV mapping, fixing texture errors, assigning materials, etc., enough time to go in and create a model by hand. It is not efficient in the slightest.

I say, instead of spamming the JR suggestions threads with locomotives every week and insulting users for their human-centric opinion on Trainz content creation, use your hot-shot AI to make those C30-7s, B30-7s, what have you, or get off your high horse.


You started it, not him.
Well instead of interjecting yourself into my business and everyone else on this forum why not do something meaningful or get off your high horse. All I did was ask a question. A simple no would've sufficed but no he had to go over the top and I'm tired of people always jumping on my back every single time I make a comment. As for as insulting users for their human-centric opinion, I don't insult people unless they insult me or it's warranted. Also, the "WISHLIST" section is just that. A place where users can say what they'd like to see be made. Whether or not its done is beyond my control. The whole reason for me asking is because I have projects and was curious if that was a remote possibility. Obviously based on your response, the answer is no. That's all you people needed to say. All of that extra crap is not needed and I won't be dealing with it anymore. I already gotta deal with that from RetroRails12 and it gets old FAST. By the way he started it and you are continuing it, not me.
 
Having a creative vision and setting your mind to it and seeing it through to completion without a profit incentive is a fundamental human experience. No AI program will ever be able to subjugate the immense satisfaction that comes with actually doing something yourself.
You could've just said there's no way it can come close to emulating the quality of such models and that would've been it. All I did was just ask. But you wanted to attack me and make ridiculous comparisons to spitting on people. That alone discredits your argument. Whether or not AI gives lazy or unintelligent people a megaphone is debatable and case-by-case.
 
I wonder if AI and ChatGPT can actually produce the same quality of work you guys do with some skills and knowledge with automation.
this might be one of the worst comments on this site atm
all of the AI made models i have used have severe issues like overly high poly count and lack of LOD's
let the humans make models, we dont need soulless bots making stuff
 
@AdvancedApproach @krpkk @DARTrider
I think there are a few key takeaways because this is going nowhere fast. I do think AdvancedApproach, that asking your first question with the wording you used, and at the same time assuming people would read between the lines, is simply not going to work. It really does come off as a spit-in-your-face "well, maybe AI can just replace your work." Maybe you should realize that most people are of the opinion against it, and maybe explaining your question with that filter would be helpful. krpkk and DARTrider, I've read many of AdvancedApproach's posts, and I do think he tends to be very straightforward and, no offense, blunt. I do think giving him the benefit of the doubt or asking for clarification goes a lot further than the tangent you guys took. In the end, do I agree with AI replacing some of the work Justin and his team do? Absolutely not. An algorithm will never have the personal touch a human will. This is the truth that most don't want to admit about AI; simply put, because its not human and can do acceptable work sometimes, its become companies' favorite way to fire real people. After all, AI doesn't ask for a salary or medical insurance, right? Companies are saving millions all the while destroying people's jobs. I think something like Trainz, that is made for real people, should support real people's work to make it better and make money.
 
Last edited:
Back
Top