TEW IX Developer's Journal Discussion Thread

[Clairo]

I like the heights & weights update. It's a complete positive for the C-Verse, but it does sound absolutely brutal for (real world) mod makers, especially those who strive to be 100% accurate. I certainly don't envy them, but my appreciation and respect for them (which was already quite high) has risen even more.

AI renders for the C-Verse... I'm willing to see it play it out. I personally don't have anything against the use of AI, but (likely far more important) ethical concerns aside, I feel most AI renders, even on these forums, look way too samey– no offense to the "creators". Obviously some of the archaic C-Verse renders needed an overhaul, and I'm happy I won't have to spend, what feels like, more time browsing through the various re-render threads than actually playing the game, but this does run the risk of destroying the vision many people had in mind for the characters. I'm happy an easy option to revert to the old style renders will be included.

[EdJames]

52 minutes ago, foolinc said:

As it's already been pointed out there are a lot of issues with AI, especially Midjourney, from standpoints of legal and ethical concerns. 

On top of that using AI renders also makes it near impossible to keep a level of quality and consistency of an image of a particular wreslter. Like I understand TEW X isn't really on your radar, but like if you wanted to produce that game or even a mid TEW C-Verse mod to get us to say 2025 and have a wrestler with a gimmick change (or even just changing their expression from a smile to a frown) - the AI is going to split out an image that at best looks like a relative of the wrestler from the previous image.

Feels to me like if you wanted consistency from the C-Verse the easiest method would have been just to decide on a style of render and make sure anyone that was paid to make them followed that style guide. The Thunderverse has been around from multiple games and has always kept the level of consistency I'm guessing you were going for.

You realise you can literally feed Midjourney an image of someone and have it re-render that exact same person in different outfits, facial expressions, ages etc? It's called character weighting.

[Self]

AI art is a tricky one. I have issues with it, but also use it a lot for my non-commercial projects. It's a fun tool which can produce good results, especially compared to my old Daz3d work. For TEW, it feels like a natural progression to go from using free images produced by the community, to the free images generated by a machine. I'm not entirely sure those Daz3D assets were all licensed for commercial use as it is  (EDIT: corrected by Markw below) although I may be confusing it with some photoshop plugins I used for my diary work. 

I'm more disappointed the traditional 150x150 style pictures are still being used. I've always wanted a more graphically ambitious TEW, with full body images to really show off the characters. I understand the benefits to keeping the smaller style. Just my personal preference. 

Edited July 3 by Self
Rightly Corrected

[Markw]

10 minutes ago, Self said:

AI art is a tricky one. I have issues with it, but also use it a lot for my non-commercial projects. It's a fun tool which can produce good results, especially compared to my old Daz3d work. For TEW, it feels like a natural progression to go from using free images produced by the community, to the free images generated by a machine. I'm not entirely sure those Daz3D assets were all licensed for commercial use as it is, although I may be confusing it with some photoshop plugins I used for my diary work. 

I'm more disappointed the traditional 150x150 style pictures are still being used. I've always wanted a more graphically ambitious TEW, with full body images to really show off the characters. I understand the benefits to keeping the smaller style. Just my personal preference. 

Daz3D assets don’t need a commercial license if you’re only using a 2D image of the character as is the case with TEW, the commercial licenses are for 3D use in games.

[Wonk]

I have no major issue with AI art but equally as a Brit hadn't heard of this Midjourney case otherwise perhaps a different platform would have been chosen.

 

https://docs.justia.com/cases/federal/district-courts/california/candce/3:2023cv00201/407208/67

[Self]

1 minute ago, Markw said:

Daz3D assets don’t need a commercial license if you’re only using a 2D image of the character as is the case with TEW, the commercial licenses are for 3D use in games.

Thank you. I am mistaken. It's been a while since I've done that kind of work.

[JrSquared]

I hope workers retirement is more of a thing. I never really liked seeing workers without the attribute have 10 stamina and still work. 

I hope attributed in general are way more fleshed out. Trainees especially rarely got attributes. 

Super pumped!  Thank Adam appreciate this series so much. 

[D.W.]

My line with AI is that personal use is fine (writing a story for fun, asking ChatGPT to make you a recipe, messing around to see what a house made entirely out of doughnuts would look like) but commercial use is where it gets sketchy. I'm someone that doesn't interact with the Cornelverse at all, so this doesn't really affect me, but I can't fault anyone that sees this as a deal-breaker.

Geo Tags are one of those small things that I never really thought about but should be a nice quality of life improvement, especially for someone like me that mods and watches more than I actually play. I'm all for anything that allows us to create more parameters to make the AI behave. I really hope we get something that stops the mass slaughter of stables, brand splits, etc that happens with new saves. 

Heights and weights are going to be a LOT of work, but it's going to look good when it's all done.

[EdJames]

1 hour ago, foolinc said:

As it's already been pointed out there are a lot of issues with AI, especially Midjourney, from standpoints of legal and ethical concerns. 

On top of that using AI renders also makes it near impossible to keep a level of quality and consistency of an image of a particular wreslter. Like I understand TEW X isn't really on your radar, but like if you wanted to produce that game or even a mid TEW C-Verse mod to get us to say 2025 and have a wrestler with a gimmick change (or even just changing their expression from a smile to a frown) - the AI is going to split out an image that at best looks like a relative of the wrestler from the previous image.

Feels to me like if you wanted consistency from the C-Verse the easiest method would have been just to decide on a style of render and make sure anyone that was paid to make them followed that style guide. The Thunderverse has been around from multiple games and has always kept the level of consistency I'm guessing you were going for.

 

On the left is Adam Matravers as a 30-year-old man in TEW 2010.
On the right is Adam Matravers as a 40-year-old man in TEW 2020.

This is what you're currently playing with 🤷

Below is a random wrestler image I found online. It took me 30 seconds in Photoshop to change his gimmick. ANYBODY can do this with a copy of Photoshop and generative fill. About ten people on the forums can re-render an Adam Matravers and have the technical skill in Daz3D, let alone the paid assets that you need to accumulate and needing the original character model. 

 

I'm not sure keeping things as they are is the easier option whatsoever.

 

 

Edited July 3 by EdJames

[JrSquared]

Feel like the game world economy could be more balanced. Especially as bigger companies they just print money hand over fist. There should be more of a fall. TNA, WCW, ROH, and ECW all really were operating at a loss for periods. 

[Mammoth]

The issue is less about licensing (as I understand it, you can use Midjourney results commercially as long as the person who "created" them was a paying member) and more about ethics anyway. As someone who does graphic work, it leaves a sour taste in ones mouth when these AI imaging databases use artists work without them even knowing. However, I don't see how their (Midjourney's) legal troubles would affect properties that used their images, commercially or otherwise. I'm not a legal scholar by any means so I'm sure someone will correct me if I'm wrong, but all I foresee would be Midjourney having to shell out a large sum of money or forced to close down, and don't see how or why it would trickle down to people who used their services.

I'm not sure why it's an issue that would cause someone not to buy the game though? Adam has obviously already paid someone to do this for him (by the sounds of it, he hasn't done it himself) and already has the results. Your money will not be going towards Midjourney, and it's really hard to decipher how you feel you'll be "supporting" them by buying this game.

[Peria]

13 minutes ago, Mammoth said:

The issue is less about licensing (as I understand it, you can use Midjourney results commercially as long as the person who "created" them was a paying member) and more about ethics anyway. As someone who does graphic work, it leaves a sour taste in ones mouth when these AI imaging databases use artists work without them even knowing. However, I don't see how their (Midjourney's) legal troubles would affect properties that used their images, commercially or otherwise. I'm not a legal scholar by any means so I'm sure someone will correct me if I'm wrong, but all I foresee would be Midjourney having to shell out a large sum of money or forced to close down, and don't see how or why it would trickle down to people who used their services.

I'm not sure why it's an issue that would cause someone not to buy the game though? Adam has obviously already paid someone to do this for him (by the sounds of it, he hasn't done it himself) and already has the results. Your money will not be going towards Midjourney, and it's really hard to decipher how you feel you'll be "supporting" them by buying this game.

 

These are closest to my sentiments. I genuinely think it isn't such a bad idea, and it is way more accessible and easier for people to learn who lack the artistic ability to use other rendering software, which means the game can continue to evolve and thrive without 4 or 5 talented artists to keep it afloat. It's a niche within a niche within a niche that is TEW re-rendering. So I'm all for it. My only question now is what will the new background be? White? Or just no background at all?

[Markw]

8 minutes ago, EdJames said:

 

On the left is Adam Matravers as a 30-year-old man in TEW 2010.
On the right is Adam Matravers as a 40-year-old man in TEW 2020.

This is what you're currently playing with 🤷

Below is a random wrestler image I found online. It took me 30 seconds in Photoshop to change his gimmick. ANYBODY can do this with a copy of Photoshop and generative fill. About ten people on the forums can re-render an Adam Matravers and have the technical skill in Daz3D, let alone the paid assets that you need to accumulate and needing the original character model. 

 

I'm not sure keeping things as they are is the easier option whatsoever.

Someone with a £220-odd a year copy of photoshop could very easily add sunglasses and a jacket to the Adam Matravers render, we've had alt threads with people doing just that for as long as I can remember. It's hardly a fair comparison to say we can't age Daz renders (which the creator could easily do if they desired by the way) and then rather than age the character you've found with AI just stick glasses and a jacket on them and say 'look it's so much easier'. Age that man 10, 20 years if that's the point you want to make, because all you've done there is make the same kind of alt in photoshop that people have been making for Daz renders for decades.

Daz is free to download, and has a number of reasonable free and cheap assets if you're willing to look around and wait for the right sales. I just think, on a personal level, it's really sad that the decision's been made to move away from the already built up community of people who spend money and their free time to make renders for this game that they're clearly very passionate about. That's something that's pretty rare for a sim game of this size to have, and trading it for the same AI art style that everyone else is using in their games at the moment just feels like an own goal to me. For me it'll just make the Cornellverse feel souless, but hopefully there'll be enough committed Daz renderers to keep that style alive.

Very excited about the features that have been released and sure the game is going to be excellent, certainly wouldn't stop me buying it.

[Paxlux]

In my opinion AI was used for the wrong thing in TEW XI. Instead of using AI for image generation, it would've been cool to use AI text generation to help AI-controlled companies generate somewhat coherent storylines and shows. It would no doubt result in some wacky stuff, but would probably be better than what was happening with AI-controlled companies in 2020.

Edited July 3 by Paxlux

[foolinc]

7 minutes ago, EdJames said:

Could do it, but haven't. And therein lies the issue.

To your point..."look, much easier".

   

What happened with the 3rd picture? Why is it a different person's face with completely different lighting? 

[Markw]

1 minute ago, EdJames said:

Could do it, but haven't. And therein lies the issue.

And who's to say that someone's going to mass age every character with AI every iteration of the game? I really don't see that switching to AI actually solves the challenge.

There's still lots of people rerendering characters to age them. No offence intended, but I really think most aged Daz rerenders have much more in common visually with the original character render, than that third AI character has with the first one (and on top of that the skin tone/lighting changes from body to head on the third one are quite jarring). Moving to AI is not going to solve the issue that you're highlighting.

[Astil]

9 minutes ago, EdJames said:

Could do it, but haven't. And therein lies the issue.

To your point..."look, much easier".

   

The first two kind look like he has lil horns coming out of his temples hehehe

[Pteroid]

9 minutes ago, EdJames said:

Could do it, but haven't. And therein lies the issue.

To your point..."look, much easier".

   

Like a medieval alchemist of old you’ve accomplished transmutation. You turned John Seesaw into Billy Gum.

And apparently he smelled a bad fart midway through the process.

[TheWho87]

Specific heights and weights is the dumbest change, especally when all you needed to do was to seperate them (like the other booking sim), this just adds unnessessary work for mod makers when that information isn't out there, the grouped weight categories were perfectly fine, the issue was having to bump wrestlers into different categories due to their heights.

[EdJames]

10 minutes ago, Markw said:

And who's to say that someone's going to mass age every character with AI every iteration of the game? I really don't see that switching to AI actually solves the challenge.

There's still lots of people rerendering characters to age them. No offence intended, but I really think most aged Daz rerenders have much more in common visually with the original character render, than that third AI character has with the first one (and on top of that the skin tone/lighting changes from body to head on the third one are quite jarring). Moving to AI is not going to solve the issue that you're highlighting.

They're not going to and never have - that was the exact point I originally made.

(Agree on the render, needed a relight)

[Markw]

1 minute ago, EdJames said:

They're not going to and never have - that was the exact point I originally made.

(Agree on the render, needed a relight)

Yeah but my point is, who's to say someone is going to do it for AI? The switch doesn't solve or address that issue at all.

[Idolized]

9 minutes ago, TheWho87 said:

Specific heights and weights is the dumbest change, especally when all you needed to do was to seperate them (like the other booking sim), this just adds unnessessary work for mod makers when that information isn't out there, the grouped weight categories were perfectly fine, the issue was having to bump wrestlers into different categories due to their heights.

Currently in TEW 2020 the size/weight breakdowns are as follows:

• Very Small (Below 140)
• Small (140 - 180 pounds)
• Lightweight (180 - 230 pounds)
• Middleweight (230 - 260 pounds)
• Light Heavyweight (260 - 290 pounds)
• Heavyweight (290 - 320 pounds)
• Big Heavyweight (320 - 380 pounds)
• Super Heavyweight (Above 380 pounds)
• Giant (Special Cases)

If you look at the Mexican National Titles for (an extreme) example, you will see how the new true weights will come in handy

• Mexican National Featherweight Championship  (126 - 139 lbs)
• Mexican National Lightweight Championship  (139 - 150 lbs)
• Mexican National Cruiserweight Championship (170 - 180 lbs)
• Mexican National Welterweight Championship (170 - 190 lbs)
• Mexican National Middleweight Championship (181-192)
• Mexican National Light Heavyweight Championship (203 - 214 lbs)
• Mexican National Heavyweight Championship (231 lbs and above)

Under the old system, The Lightweight and Featherweight titles overlapped with “Very Small” (but easily fixed by bumping the Lightweight up to small). However, properly setting the Cruiserweight, Middleweight and Light Heavyweight titles is a lot trickier without the AI selecting workers to challenge/hold the belts that shouldn’t be.

In NJPW, the Junior Heavyweight Championship’s cutoff is 220, but as we all know - guys like Styles, Osprey, Nakamura and others that were under or on the fringe of that cutoff ended up as Heavyweight division stars and left the Jr title behind for good. In 2020, you would have to either set AJ, Osprey and Nakamura as Middleweights to properly get them to challenge for the IWGP Heavyweight Title, or allow middleweights to compete for it and end up with someone like Liger who was super over, but never a heavyweight constantly challenging and holding the belt on AI shows. Now, Setting a slight overlap on the weights of the belts should solve that issue (I would assume)

[EdJames]

Just now, Markw said:

Yeah but my point is, who's to say someone is going to do it for AI? The switch doesn't solve or address that issue at all.

I never said they were. His point was that the move to AI prevents it. I said "it's never happened anyway, so why are you bothered?" in a round about way.

All of that being said, let's imagine a new TEW comes out in 2028...I cannot for one second believe that AI image generation won't have come along far enough that you'll be able to age up existing pictures with absolute accuracy, or that you won't be able to change those same people's expressions, outfits at the click of a button (and like en masse). 

I suspect by then we'll be able to send all 3,000 images into a batch process and age them all four years in a matter of seconds. 

I'm not saying AI solves the issues, but I also don't personally see it as a worse position than being reliant on six or seven community rendering experts, all with their own unique style, to feed the entire community. Just my opinion.

 

[Chikbot]

Personally. I think the 2d/3d hybrid renders had a soul to them. I understand that some AI products allow for commercial use but ethically I'm not a fan of AI beyond just having fun with it in a non-commercialized basis. 

[EdJames]

3 minutes ago, Chikbot said:

Personally. I think the 2d/3d hybrid renders had a soul to them. I understand that some AI products allow for commercial use but ethically I'm not a fan of AI beyond just having fun with it in a non-commercialized basis. 

The option is still there to use them. We now have the option to use the original style OR an AI style. 
The community can feed into both channels and everybody's preferences can be met.