HTML Dune – Breeding a Dynasty [v0.1.1] [lmno]

[lmno]

Any news about this?

Progress is slow, but we're getting there bit by bit

 

[Gasbag127]

So I'm currently listening to the Dune audiobook at work and am just really getting Dune hyped again. I watched both movies probably 3 times each.

I just want to say I'm definitely hyped for this project, probably more than anything on this site.

I know you have a day job and progress is slow. Good luck with it. You have done a good job so far.

 

[Atlan]

 

[reynold.biplane]

So there hasn't been any progress for a long time. I think we owe you guys an explanation.
Let's call this...

Devblog #01

This game has been under development for about a year now and yet we have only released a single version, no updates. Obviously there are problems behind the scenes. Let me fill you in on what's troubling us and how we intend to deal with the situation and become productive again.

Up and down the motivation spiral.
I gotta confess, left to my own devices I'm a bad procrastinator. That's exactly how things were when I was still writing this story alone without any input or supervision. As soon as things become a little difficult and the ideas aren't pouring freely out of my head or I have a difficult creative decision to make, I'd rather be doing something else. And since I'm already sitting on the computer, the ultimate distraction machine...
This changed drastically once I teamed up with lmno. Productivity skyrocketed (relatively speaking). Having somebody to hold you accountable, seeing their progress, receiving instant feedback on your own work and seeing that implemented immediately, troubleshooting together... It was a constant back and forth between us in those first few months. We were giving each other energy. It was like a motivation spiral. We created that first version pretty quickly. Unfortunately the spiral works in both directions...
So what happened then? The honeymoon phase was over? Well, not exactly. Shortly after releasing v0.1, lmno got a new job. Suddenly lmno had a lot less time and energy available for this project. At the same time, part of his responsibility for the project - AI art - got a lot harder. More about that in a second. Anyway, that killed his motiviation, which dragged down my motivation, which in turn dragged down his motivation further etc. until we are at a near standstill.

AI is kinda shit, actually
Don't get me wrong, it is remarkable what AI can do nowadays. There are some real masterpieces out there. And it's not like the graphics in our game are bad. You can do a lot with AI, if you are patient and willing to discard some results and finetune your prompts, and finally if you are doing some manual editing. And we did do that. It wasn't always easy. The problem isn't just that AI can be really wonky and has trouble with certain body parts like hands, or the consistency issue. We were able to deal with all that, for the most part.
The real issue is that AI can do certain things really good, and others not at all. AI has no creativity whatsoever. No real ability to come up with new things. What it sees, it can replicate. So things that exist in large numbers on the internet, AI can replicate in high quality with few errors. Pinups. Yes. Absolutely. Brilliant. But as soon as you want non-standard poses or god forbid have more than one person in the picture... there are a few poses with multiple people that AI can do apparently, but your options are severely limited. You are looking down the barrel of some serious editing. Actual skill required.
We worked our way around these issues in chapter 1. We finagled some things, cut some corners, used direct references, did a lot of editing and multiple passes on some pictures. Like Jessica's handjob. That took lmno several days until it was good enough. And it's still one of the weakest pictures in the game. Still a lot less work than actual drawing, but we were approaching the same ball park on that one.
And now in chapter 2, things are even more problematic. The new artstyle made work more easy, but it didn't fix the fundamental issues. There are some scenes that we just can't do pictures for. Basically, development has ground to a halt because of that. Maybe if we had more powerful computers available for much faster generation... but we don't. Maybe it's partially a user issue. Perhaps somebody else could do a better job with the same tools. But we haven't found that person yet. We have asked a few people, but no dice.

So what now?
Hard decisions. The only thing that's for sure is that we can't continue the same way. So we've been thinking that it might be best to just continue developing the story and the game code and neglect graphics for the time being. We can still use the AI to make backgrounds and character sprites. But scenes would be text only. That way at least there is progress and maybe AI will improve significantly over time and then we can add illustrations later.
Not really that satisfying, IMO. Text-only HTML games are a dime a dozen. Dune deserves better. But better something than nothing, which is what we currently have.
There isn't that much work left to be done for an update, if we leave out art. I really only need to write one more scene for there to be enough content for a worthwhile update and then we need a couple more backgrounds. Easy.

What? Text only?!
I gotta save my honour a bit here and mention that I wasn't entirely inactive the last few months. First I tried my hand at image editing to take some of the load off of lmno. But that wasn't enough to get the art production back running. We still couldn't fix the glaring issues I mentioned earlier. Then I straight up pivoted into practicing drawing. And I definitely made progress in that area. Here's my rendition of the Reverend Mother:

Spoiler: some pictures

You don't have permission to view the spoiler content. Log in or register now.

Not bad, huh? But is it good enough? This style definitely clashes with the existing backgrounds, at the very least. More importantly, it took me ages to make since I'm still super inexperienced. I reckon it would take me at least half a year, if not much longer to become suitably proficient, depending on how much time I decide to invest. And on top of that I'm already the writer, which is pretty time-consuming in itself. Even if I were to work on this project full-time, I don't think it would be feasible.

So what other possibilities are there? It would quite frankly be unreasonable to expect that there might be anyone out there willing to create art for free in the quantitites that we require. Unless it's AI generated, maybe.
There have been people in this thread indicating a willingness to part with some of their money. Subscribestar is an option to raise funds to hire an artist (but not Patreon, fuck Patreon).
But please understand that this is a daunting proposition. Once we receive money, even if it's just enough to pay an artist, we really have to deliver. Then it's an obligation and the pressure is on.

So where does this leave us? I dunno, but I'm gonna put down the drawing pad for a while and get back to the keyboard and write some more.

 

[Arnav Dasari]

So what other possibilities are there?

I don't remember if I proposed that before in this thread, but may be you guys could create the characters and the images you need in some 3D game poser - like Honey Select or Virt-A-Mate. You could even use these images as a base for further AI enhancement. For example a game called 'Sofia's Dark Fantasies' uses exactly this technique.

 

[name]

So there hasn't been any progress for a long time. I think we owe you guys an explanation.
Let's call this...

Devblog #01

This game has been under development for about a year now and yet we have only released a single version, no updates. Obviously there are problems behind the scenes. Let me fill you in on what's troubling us and how we intend to deal with the situation and become productive again.

Up and down the motivation spiral.
I gotta confess, left to my own devices I'm a bad procrastinator. That's exactly how things were when I was still writing this story alone without any input or supervision. As soon as things become a little difficult and the ideas aren't pouring freely out of my head or I have a difficult creative decision to make, I'd rather be doing something else. And since I'm already sitting on the computer, the ultimate distraction machine...
This changed drastically once I teamed up with lmno. Productivity skyrocketed (relatively speaking). Having somebody to hold you accountable, seeing their progress, receiving instant feedback on your own work and seeing that implemented immediately, troubleshooting together... It was a constant back and forth between us in those first few months. We were giving each other energy. It was like a motivation spiral. We created that first version pretty quickly. Unfortunately the spiral works in both directions...
So what happened then? The honeymoon phase was over? Well, not exactly. Shortly after releasing v0.1, lmno got a new job. Suddenly lmno had a lot less time and energy available for this project. At the same time, part of his responsibility for the project - AI art - got a lot harder. More about that in a second. Anyway, that killed his motiviation, which dragged down my motivation, which in turn dragged down his motivation further etc. until we are at a near standstill.

AI is kinda shit, actually
Don't get me wrong, it is remarkable what AI can do nowadays. There are some real masterpieces out there. And it's not like the graphics in our game are bad. You can do a lot with AI, if you are patient and willing to discard some results and finetune your prompts, and finally if you are doing some manual editing. And we did do that. It wasn't always easy. The problem isn't just that AI can be really wonky and has trouble with certain body parts like hands, or the consistency issue. We were able to deal with all that, for the most part.
The real issue is that AI can do certain things really good, and others not at all. AI has no creativity whatsoever. No real ability to come up with new things. What it sees, it can replicate. So things that exist in large numbers on the internet, AI can replicate in high quality with few errors. Pinups. Yes. Absolutely. Brilliant. But as soon as you want non-standard poses or god forbid have more than one person in the picture... there are a few poses with multiple people that AI can do apparently, but your options are severely limited. You are looking down the barrel of some serious editing. Actual skill required.
We worked our way around these issues in chapter 1. We finagled some things, cut some corners, used direct references, did a lot of editing and multiple passes on some pictures. Like Jessica's handjob. That took lmno several days until it was good enough. And it's still one of the weakest pictures in the game. Still a lot less work than actual drawing, but we were approaching the same ball park on that one.
And now in chapter 2, things are even more problematic. The new artstyle made work more easy, but it didn't fix the fundamental issues. There are some scenes that we just can't do pictures for. Basically, development has ground to a halt because of that. Maybe if we had more powerful computers available for much faster generation... but we don't. Maybe it's partially a user issue. Perhaps somebody else could do a better job with the same tools. But we haven't found that person yet. We have asked a few people, but no dice.

So what now?
Hard decisions. The only thing that's for sure is that we can't continue the same way. So we've been thinking that it might be best to just continue developing the story and the game code and neglect graphics for the time being. We can still use the AI to make backgrounds and character sprites. But scenes would be text only. That way at least there is progress and maybe AI will improve significantly over time and then we can add illustrations later.
Not really that satisfying, IMO. Text-only HTML games are a dime a dozen. Dune deserves better. But better something than nothing, which is what we currently have.
There isn't that much work left to be done for an update, if we leave out art. I really only need to write one more scene for there to be enough content for a worthwhile update and then we need a couple more backgrounds. Easy.

What? Text only?!
I gotta save my honour a bit here and mention that I wasn't entirely inactive the last few months. First I tried my hand at image editing to take some of the load off of lmno. But that wasn't enough to get the art production back running. We still couldn't fix the glaring issues I mentioned earlier. Then I straight up pivoted into practicing drawing. And I definitely made progress in that area. Here's my rendition of the Reverend Mother:

Spoiler: some pictures

You don't have permission to view the spoiler content. Log in or register now.

Not bad, huh? But is it good enough? This style definitely clashes with the existing backgrounds, at the very least. More importantly, it took me ages to make since I'm still super inexperienced. I reckon it would take me at least half a year, if not much longer to become suitably proficient, depending on how much time I decide to invest. And on top of that I'm already the writer, which is pretty time-consuming in itself. Even if I were to work on this project full-time, I don't think it would be feasible.

So what other possibilities are there? It would quite frankly be unreasonable to expect that there might be anyone out there willing to create art for free in the quantitites that we require. Unless it's AI generated, maybe.
There have been people in this thread indicating a willingness to part with some of their money. Subscribestar is an option to raise funds to hire an artist (but not Patreon, fuck Patreon).
But please understand that this is a daunting proposition. Once we receive money, even if it's just enough to pay an artist, we really have to deliver. Then it's an obligation and the pressure is on.

So where does this leave us? I dunno, but I'm gonna put down the drawing pad for a while and get back to the keyboard and write some more.

i always assumed that the AI images were mainly there to set the scene like the character portraits, the background, etc. my hughest hopes were that the portraits would change depending on the scene (Jessica in a royal dress at the palace, in lingerie at night, in a bikini at the catchbasin, etc.) you are tagged as text based after all and that's the standard most of them bring.

i was legitimately surprised that you guys were going to illustrate the sex scenes. i think, at least for now, it's better to use the text to describe the acts and have to portraits reflect the appropriate level of nakedness/spunk cover, you could even get creative and make enticing solo shots like facials and the like, but if actual scenes involving two or more people is that detrimental to your productiveness i say leave it aside for now and let the text do the heavy lifting.

Imno probably has a lot on his plate as is, programming wasn't even brought up but it's definitely a sink in its own way too, so i say just move forward, and go back and add actual sex pictures in future updates when you find the time. remember these are "version 0.01"s and the like, you don't need to make a finished product right away.

anyway, that's just my two cents. in productions and especially collaborations you need to make compromises, and it's fine to do so. remember just because there are fewer pictures right doesn't mean the game is lesser for it, there are plenty of interactive game elements you can add to the experience

 

[KingsRaiden]

What? Text only?!

Like the days of Adrift, RAGS, and TADS where you might get an image now and then to help imagine the text scenes. Only thing with this is you REALLY have to know how to write erotic fiction, there was some really bad Adult Interactive Fiction (AIF) games back then and there was little sympathy for it. When all you are relying on are your words, they better make the complete picture or people will be rightly critical as you haven't brought the erotic to the story.

 

[reynold.biplane]

I don't remember if I proposed that before in this thread, but may be you guys could create the characters and the images you need in some 3D game poser - like Honey Select or Virt-A-Mate. You could even use these images as a base for further AI enhancement. For example a game called 'Sofia's Dark Fantasies' uses exactly this technique.

Hm, I have no experience with any of that. Honey Select doesn't look right. Haven't seen enough stuff made with Virt-A-Mate yet.
Your example pictures do look good, but somehow my instinctive reaction to using 3D posers is still negative...

i always assumed that the AI images were mainly there to set the scene like the character portraits, the background, etc. my hughest hopes were that the portraits would change depending on the scene (Jessica in a royal dress at the palace, in lingerie at night, in a bikini at the catchbasin, etc.) you are tagged as text based after all and that's the standard most of them bring.

i was legitimately surprised that you guys were going to illustrate the sex scenes. i think, at least for now, it's better to use the text to describe the acts and have to portraits reflect the appropriate level of nakedness/spunk cover, you could even get creative and make enticing solo shots like facials and the like, but if actual scenes involving two or more people is that detrimental to your productiveness i say leave it aside for now and let the text do the heavy lifting.

Imno probably has a lot on his plate as is, programming wasn't even brought up but it's definitely a sink in its own way too, so i say just move forward, and go back and add actual sex pictures in future updates when you find the time. remember these are "version 0.01"s and the like, you don't need to make a finished product right away.

anyway, that's just my two cents. in productions and especially collaborations you need to make compromises, and it's fine to do so. remember just because there are fewer pictures right doesn't mean the game is lesser for it, there are plenty of interactive game elements you can add to the experience

Maybe the 'text-based' tag was misapplied here. It was just supposed to reflect the fact that the action is mostly conveyed through text and that there's a lot of reading involved in absolute terms, not a lack of pictures.

Changes of wardrobe were indeed a planned feature (and could still potentially happen), although not in the way you desribe.

What? Text only?!

Like the days of Adrift, RAGS, and TADS where you might get an image now and then to help imagine the text scenes. Only thing with this is you REALLY have to know how to write erotic fiction, there was some really bad Adult Interactive Fiction (AIF) games back then and there was little sympathy for it. When all you are relying on are your words, they better make the complete picture or people will be rightly critical as you haven't brought the erotic to the story.

I have full confidence in my writing ability. It's just that games with accompanying visuals have a much broader appeal without any downsides.

 

[Arnav Dasari]

Honey Select doesn't look right.

The good thing about Honey Select (1 or 2) is that the character editor is quite versatile, as well as the poser ("Studio Neo").

Getting Dune-appropriate hairstyles and clothes might be a problem though. Virt-A-Mate's editor is even more flexible and detailed, however the UI is a nightmare and hairs/clothes is even more limited in vanilla - you'd have to search for everything at the

You must be registered to see the links

.

 

[name]

Hm, I have no experience with any of that. Honey Select doesn't look right. Haven't seen enough stuff made with Virt-A-Mate yet.
Your example pictures do look good, but somehow my instinctive reaction to using 3D posers is still negative...


Maybe the 'text-based' tag was misapplied here. It was just supposed to reflect the fact that the action is mostly conveyed through text and that there's a lot of reading involved in absolute terms, not a lack of pictures.

Changes of wardrobe were indeed a planned feature (and could still potentially happen), although not in the way you desribe.


I have full confidence in my writing ability. It's just that games with accompanying visuals have a much broader appeal without any downsides.

if you've ever played classic Goblinboy games that's basically my expectation of every text based tagged game, where most of the engagement is through reading as you said. i never expected there to be no pictures but i didn't expect there to be entire multiperson scenes with them, especially knowing AI