[Knight26-77]

Stumbled on this one today:

VICTORY — a Wing Commander III fan remake

A free, non-profit Wing Commander III fan remake in development. Explore the ships and the rooms of the TCS Victory.

staragent.site

It reports that it is a full remake. But all the models look off, almost like someone fed an "agent" the designs and various references, and that is what it kicked out. The cockpits especially look wrong, and that informs the rest of the design. The Kilrathi ships suffer hard with their Agent/designer. The video upscales look pretty good, at least the ones they showed. The only thing I can say I actually like is the animated weapon bays.

He also posted a couple youtube videos:

I wonder if the 30th anniversary remaster/remake will kill this remake.

 

[Knight26-77]

Looking at the ship descriptions, they read like something you would input into an AI agent.

 

[Pedro]

The WCIV remake actually supports WC3, or did. Just lacks the models. It’s interesting seeing so many vibe coded remakes and no one so much as asking for any info.

But what I’m learning from the number of likes and views is people as a rule are generally fine with this? That’s a good motivation killer.

 

[Knight26-77]

The WCIV remake actually supports WC3, or did. Just lacks the models. It’s interesting seeing so many vibe coded remakes and no one so much as asking for any info.

But what I’m learning from the number of likes and views is people as a rule are generally fine with this? That’s a good motivation killer.

Don't worry, your project will be worth the weight.
I asked the channel for more info, especially once I recognized how AI-generated it was.

 

[Col.SE-mACE 59]

We want the one and only WC4 Remastered that was created with real passion—not some AI-generated, bullshit WC3 remakes

 

[Exteban]

We want the one and only WC4 Remastered that was created with real passion—not some AI-generated, bullshit WC3 remakes

I think someone with passion can make a good fan project of Wing Commander (or whatever) using AI.

Anyway, I'm also eagerly awaiting the wonderful remake of Wing Commander IV. I don't know how many times I've played the demo. My record is 12 kills.

 

[ChrisReid]

I think someone with passion can make a good fan project of Wing Commander (or whatever) using AI.

That is the crux of it though - the passionate fans are using AI as a tool to automate repetitive elements while handcrafting the parts that matter. You don't even notice. On the other hand, there has been an explosion of projects this year that throw a bunch of familiar WC art into an AI blender, pop out some flashy YouTubes to show it off and then pat themselves on the back. We absolutely welcome fans who want to invest the time and effort to make something (with or without AI) that they care about, but a lot of these projects don't demonstrate that they do.

 

[Knight26-77]

That is the crux of it though - the passionate fans are using AI as a tool to automate repetitive elements while handcrafting the parts that matter. You don't even notice. On the other hand, there has been an explosion of projects this year that throw a bunch of familiar WC art into an AI blender, pop out some flashy YouTubes to show it off and then pat themselves on the back. We absolutely welcome fans who want to invest the time and effort to make something (with or without AI) that they care about, but a lot of these projects don't demonstrate that they do.

You're right. Using AI to help with basic, repetitive tasks is a good use; using it for creative tasks is not. It is pretty obvious this project I stumbled upon used AI to make the models, or at the very least to "enhance" the original models. They would have been better off reaching out to the community and using fan-made models; then, if they really needed cockpit interiors, either adding them themselves or tasking AI with adding them. Though I am not sure how much you can make AI do that.

I think someone with passion can make a good fan project of Wing Commander (or whatever) using AI.

Anyway, I'm also eagerly awaiting the wonderful remake of Wing Commander IV. I don't know how many times I've played the demo. My record is 12 kills.

Making a game/project from scratch using only AI? No way. Using it to help, sure. I even use AI in my writing workflow to clean up my transcriptions and do basic editing, but that's it. AI cannot create, and when it is used to create, it is always obvious and wrong-looking.

 

[Vidmaster]

I wish people (and the media) would stop sharing this crap, it is not worth anybody's time. It looks wrong, feels wrong and is very likely vibe-coded.

If someone would truly care, that someone would wait until the basic were right until sharing it. This Victory project is fishing for views and likes and, later, money. Just like that WC Reloaded business a few weeks back (just look at how wrong the AI is in those videos, especially after they spawn).

The AI vibe coder crowd is demotivating experts of all fields right now, but seldom do they produce anything worth the effort and our time, be it GameDev or else. People are begging Valve for a No-AI filter for a reason, it is a tidal wave of swamp water into our wells right now, and many are sick of it.

We all need to vote with time, wallet and interest about what we want. Giving a WC-free example, I payed/gifted money to people like Jorio a while back because he poured sweat and time into porting games that mean a lot to me (the Pangea Games, see https://github.com/jorio). And he made sure it was done right.

It is my belief that passionate talented people who pour in half a life (see https://openbioforge.com/) should be able to continue to do that, unmolested by the AI scamers. Because it is worth doing and people recognize when a thing is done right. Always.

 

[Dishwasher]

Hi all, I am the one making the project, yes I use AI tools. I was making this for myself. Wing Commander III is the first game I bought for myself and I was experimenting with Astra and Claude Opus 5.5 to see if I could get this thing running in Unreal engine and with VR. It worked and my personal experience was pure joy. I found people who like what I did an put a quick website together so they could download and play with the models.

A few days ago I put a video online to show my friends how my remake handles sophisticated missions like the Laconda mission. This video got way more traction than I ever expected. Obviously the badly lit Thunderbolt cockpit with a hole in the radar is not my proposed endpoint for quality, the thing was even rendered at only 75%.

I understand that people can be demoralised by the insane speed at which AI tools allow people to generate new content. Calling people AI scammers, or "not real passionate fans" is gate keeping and honestly insulting. Anyone who would take a small amount of time to talk with me about this project would notice I am extremely passionate about it. I will never recover the investment I put into this 2 billion tokens today alone, I am doing this because I love Wing Commander, and because I love seeing my vision for it being turned into something I can actually play and enjoy. If you do NOT like what I am creating that is fine, I am not forcing it on anyone.

I am not interested in community models, not because I do not like them, but because I want to learn how to get good results on my own, it's my process. I think the WCIV remake looks amazing.

Rejoice in the fact people are investing time, money, tokens into creating awesome content. If you do not like the content then don't play it. But it is not fair to question someones fandom or devotion because you do not like the tool they are using.

 

[Pedro]

I understand that people can be demoralised by the insane speed at which AI tools allow people to generate new content.

That's very much not what I meant. The existence of the technology isn't an issue, anyone can use it so it's not like looking on in awe at something which is out of grasp. It's the the fact that despite the limitations it's largely being accepted and even embraced that is the demoralising aspect. That doesn't seem to have been the same for AI video.

 

[Knight26-77]

2 billion tokens, with a B? Depending on which market you use, that is minimum, what, $100? I get wanting to make something cool and to show your love for the IP, even if just for yourself, but damn, if you are spending tokens on this, then use them to be fiscally responsible; requesting the usage of community-made models would make more sense. You could use the tokens better to do things like insert cockpits, animate weapon fire, and add weapon mounts.
I applaud the effort to port it into Unreal Engine, but also, if you don't like the criticism, why share it? If you wanted to only share it with friends, you could have made the videos private and just uploaded the WIP materials to Google Drive. The fact that you posted it on a public platform and made the website suggests you wanted some form of attention.
If you wanted to "learn how to get good results on my own," then why are you spending tokens to have AI redesign the models? Learning to get good results on your own would imply making the models yourself, using the originals as a base, sure, but making them yourself. If you then used AI to do things like bake and apply the textures, add greebles, that sort of thing, that is understandable and is an enhancement of your work, but simply uploading the original model with a prompt and spending tokens is not getting good results on your own.
That being said, the AI video upscaling looked awesome, and I will be curious to see how it compares to what EA releases in their WCIII remaster using the original master tapes.

 

[Dishwasher]

2 billion tokens, with a B? Depending on which market you use, that is minimum, what, $100? I get wanting to make something cool and to show your love for the IP, even if just for yourself, but damn, if you are spending tokens on this, then use them to be fiscally responsible; requesting the usage of community-made models would make more sense. You could use the tokens better to do things like insert cockpits, animate weapon fire, and add weapon mounts.
I applaud the effort to port it into Unreal Engine, but also, if you don't like the criticism, why share it? If you wanted to only share it with friends, you could have made the videos private and just uploaded the WIP materials to Google Drive. The fact that you posted it on a public platform and made the website suggests you wanted some form of attention.
If you wanted to "learn how to get good results on my own," then why are you spending tokens to have AI redesign the models? Learning to get good results on your own would imply making the models yourself, using the originals as a base, sure, but making them yourself. If you then used AI to do things like bake and apply the textures, add greebles, that sort of thing, that is understandable and is an enhancement of your work, but simply uploading the original model with a prompt and spending tokens is not getting good results on your own.
That being said, the AI video upscaling looked awesome, and I will be curious to see how it compares to what EA releases in their WCIII remaster using the original master tapes.

All you are doing is projecting your morality, beliefs and dislike for AI on others. Obviously a lot of people do like what the project looks like.

I have taken all constructive criticism to heart, things like the font, cockpit and scale being off. But remarks like

"is my belief that passionate talented people who pour in half a life (see https://openbioforge.com/) should be able to continue to do that, unmolested by the AI scamers."

These are just your views and not very constructive.

Also please do not dictate what skillset I should pursue and what constitutes a "good" or "useful" skill or how I should spend my income.

This is the last I will contribute to the topic. If you like my project, great, if not that is too bad, if you have constructive feedback on what should be improved and can be more specific than "this looks like slop" I would greatly appreciate it.

 

[Bob McDob]

That's very much not what I meant. The existence of the technology isn't an issue, anyone can use it so it's not like looking on in awe at something which is out of grasp. It's the the fact that despite the limitations it's largely being accepted and even embraced that is the demoralising aspect. That doesn't seem to have been the same for AI video.

I've noticed that these sort of making-of videos often get very big very quickly, and as far as I can tell it's because they're reaching an audience that don't really care about game development, or even games at all. They want something cool-looking to watch and don't really care about the technical details. Often they're not even interested in playing the game at all: I can think of cases where games whose videos got tens of thousands of views launched to single-digit concurrent player counts on Steam.

That's if any of these numbers are even real at all, which I have reason to doubt. Surely, the attention of random internet strangers, who possibly don't care all that much and might not even be real at all, shouldn't matter? I'm always happy when real people leave a comment or make a video about something I've done, but I really don't think we should care about metrics like this.

 

[Exteban]

All you are doing is projecting your morality, beliefs and dislike for AI on others. Obviously a lot of people do like what the project looks like.

I have taken all constructive criticism to heart, things like the font, cockpit and scale being off. But remarks like

"is my belief that passionate talented people who pour in half a life (see https://openbioforge.com/) should be able to continue to do that, unmolested by the AI scamers."

These are just your views and not very constructive.

Also please do not dictate what skillset I should pursue and what constitutes a "good" or "useful" skill or how I should spend my income.

This is the last I will contribute to the topic. If you like my project, great, if not that is too bad, if you have constructive feedback on what should be improved and can be more specific than "this looks like slop" I would greatly appreciate it.

I think it’s a fantastic idea. Go ahead with your project!
I’m one of many who get excited about every new project, and yours looks absolutely great.

Once it’s finished, it’s sure to look incredible.

I only wish I had the knowledge to help you.

*Wing Commander* isn’t an IP that draws huge crowds these days. Anyone who embarks on a project of this scale, investing their own time and resources, clearly loves the franchise.

AI is just a tool—it’s neither good nor bad; it depends entirely on the person using it. It’s much like a knife: it’s just a knife, whether you use it to slice a steak or to kill someone.

*TRON* was denied an Oscar for Best Visual Effects on the grounds that using CGI was "cheating."

Many people who post videos on Instagram criticizing AI forget that they themselves use tools that "cheat." With just the press of a button and a few clicks, they can create a flawless video—technology allows them to bypass the need for camera operators, sound technicians, lighting crews, focus pullers, film lab technicians, editors, sound mixers, and so on.

Even modern-day illustration involves "cheating"—tools for cropping, filling, erasing, and more can reduce a task that once took hours down to mere seconds.

For a professional, AI serves to enhance their work.

Technology has made it possible for anyone to create photos, films, songs, or illustrations using only a fraction of the resources that were once required. Now, it’s programming’s turn—and this in no way diminishes the hard, magnificent work done by those who program.

 

[cd1188]

I have seen this video too and I saw that he took the upscaled videos from my HD video pack, but in one video he has noted that fact in the video description... BTW, that reminds me to fix the audio glitches during the transitions in the hangar 🥲

 

[Pedro]

Technology has made it possible for anyone to create photos, films, songs, or illustrations using only a fraction of the resources that were once required. Now, it’s programming’s turn—and this in no way diminishes the hard, magnificent work done by those who program.

I think it's worth noting that all those things you can now just give a prompt and say "make me a song and don't stop until you have one", ditto on a film, and now a game.

Used as a *tool* game development should speed up significantly. So much of the remake was working out old formats which now claude can give you in a single prompt. I'm actually not too bummed about that part being easier personally because that was no fun for me. But that has been true for years.

However in the last three months some significant models came out that really let you just say "go make game". And low and behold the last two months the modding chat has been flooded with projects. Maybe one of them pre-dated those models, it'd be a coincidence, but maybe. There is a reason everyone is cynical though. Also a big recompilation project rocketed up this month, so again, flood of specifically remakes. Between the two you have a combination to make a remake with zero input.
AI for a while was already good enough to be a great accelerant - but we didn't see a wealth of projects in that period, only now.

Some projects are slop on a single prompt and burnt tokens, some people will be empowered by AI and produce high quality work more quickly.

How I will differentiate between then will be quality not quantity. Games developers have always tried to get one level right before moving on, if you see a project that tries to do everything and then go back and fix it that's a red flag.

Remakes are the easier thing for AI too, you don't even need to create the prompts - just tell it to keep going until it's recreated the game.
Lets ignore this project - to identifying one worth watching look for a mission statement - what line between old and new, what visual target is the game looking for.
It will be easier to spot worthwhile original projects than remakes for some time to come whilst the novelty persists.

 

[Dishwasher]

I have seen this video too and I saw that he took the upscaled videos from my HD video pack, but in one video he has noted that fact in the video description... BTW, that reminds me to fix the audio glitches during the transitions in the hangar 🥲

I think it's worth noting that all those things you can now just give a prompt and say "make me a song and don't stop until you have one", ditto on a film, and now a game.

Used as a *tool* game development should speed up significantly. So much of the remake was working out old formats which now claude can give you in a single prompt. I'm actually not too bummed about that part being easier personally because that was no fun for me. But that has been true for years.

However in the last three months some significant models came out that really let you just say "go make game". And low and behold the last two months the modding chat has been flooded with projects. Maybe one of them pre-dated those models, it'd be a coincidence, but maybe. There is a reason everyone is cynical though. Also a big recompilation project rocketed up this month, so again, flood of specifically remakes. Between the two you have a combination to make a remake with zero input.
AI for a while was already good enough to be a great accelerant - but we didn't see a wealth of projects in that period, only now.

Some projects are slop on a single prompt and burnt tokens, some people will be empowered by AI and produce high quality work more quickly.

How I will differentiate between then will be quality not quantity. Games developers have always tried to get one level right before moving on, if you see a project that tries to do everything and then go back and fix it that's a red flag.

Remakes are the easier thing for AI too, you don't even need to create the prompts - just tell it to keep going until it's recreated the game.
Lets ignore this project - to identifying one worth watching look for a mission statement - what line between old and new, what visual target is the game looking for.
It will be easier to spot worthwhile original projects than remakes for some time to come whilst the novelty persists.

I have seen this video too and I saw that he took the upscaled videos from my HD video pack, but in one video he has noted that fact in the video description... BTW, that reminds me to fix the audio glitches during the transitions in the hangar 🥲

Hi I, in no way want to take credit for your amazing upscaling work. To install the mod you are redirected to your download link, I also kept the credit in the starting frame of the opening cinematic so it is the first thing you see every time you start the game. Is there something else I can do to give you the credit you deserve? I have a little trouble remembering your nickname when I type a reply, but when someone notes the upscaled videos I try to respond and redirect the credit.

 

[Exteban]

I'm not a die-hard AI advocate either. AI does the jobs it's capable of very well, but not everything. I'm not in favor of someone writing a prompt and publishing the first result the AI throws out. Someone passionate about a project will use AI to enhance their creativity, not replace it. Just as I don't like to immediately reject something if it's made with AI, we shouldn't embrace the first thing it produces either. As has already been said, it should be judged by its final quality.

 

[Dishwasher]

CD1188, I added the credit for the Video Upscaling to the title of both video's