Devlog 45

Maybe I should take my words back from the previous devlog. I said AI art is not limited to making portraits. Well, that's true, there are many AI artworks that are not portraits, but it is mainly designed to make portraits, at least for now. It's also very hard to make exactly what you have in mind. It's also hard to fix the mistakes of the AI. The AI is always so stubborn that it refuses to fix something as obvious as patches on a wall, or that the road doesn't start from the door. 

So, I may just have to go back to using Daz3d only for now. Sorry for the delays in this update, but I really wanted to explore about AI art, since I kept getting restricted in making what I have in mind while using Daz3d. This AI image of a farmhouse and a barn is how I imagined the house of the MC to be, but I couldn't make it in Daz3d because there is only 1 environment asset that has the exterior of a farmhouse, and my computer couldn't even render the scene because of how badly optimised it is. Now I could probably find a picture of a real farmhouse with a barn that is better than this, but the point is, I always imagine a scene in a particular way, and then I make do with whatever assets I can find that best match my imagination, but I can't always do that. So, I still think AI art has a lot of potential, and it can already do many things that Daz3d can't, but it is not quite there in terms of the control an artist has over it.

Now, I probably picked a bad time to learn about AI art because I couldn't simultaneously make renders for the update while learning about AI art. But, I looked at the assets available, and I couldn't really make an interesting ending with them. That's why I wanted to see if AI art could help me, otherwise I just have to write the endings that I can make with the current assets. That has always been the case with Daz3d. You look at what assets there are, and then you think about what kind of a story you can make with it. But I didn't plan that far ahead with the assets. I just had a rough outline of a story that hopefully I could find assets along the way to portray it. Right now, I couldn't, so I'll have to change things up a bit.

I will continue to follow the development of AI art, because I am very certain it is superior to Daz3d. It already is in many ways. For example, very early on, I talked about how I like using effects such as light rays, fog, or reflections but adding these in Daz3d increases the render times by like a thousand times or more. I tried to add these effects in GIMP for some renders, but it's too time-consuming to add them to every image. On the other hand, adding them in AI art costs nothing extra.

I don't know how much more I can make for this game with the limited assets of Daz3d. There is a reason why most games made with Daz3d are the romantic/dating type in a school/home/work setting. Occasionally there are sci-fi or fantasy settings, but even then those are not packed with action. I took a daring step in making lactation and pregnancy and other things I can't talk about here as the main themes of this game, but such assets are probably the rarest among Daz3d assets. The other route, the demon route, was also meant to be heavily focused on action and special effects, but those are also very lacking in Daz3d.

Don't worry, I will finish this game. I won't leave you hanging, but I'm also sorry for the quality of the game, especially the endings. It started off pretty good in my opinion because I spent a lot of time searching for all the relevant assets before I started making the game. I did not know that there aren't many relevant assets left after I have found the ones I used.

I don't know what's next after I have finished this game, as I am not really interested in making romantic games. I have high hopes for AI, not just in art, but in other areas as well. I'm not just a writer or an artist. I tried making a 3d video game in Unity and Unreal Engine before. I made my own 3D models, animations, codes, music. Admittedly none of them were very good, just like my stories and art. But AI is making progress in all these areas. So who knows? Maybe later on the AIs in these areas will be easy enough for me to make a full 3D game. I already feel limited by the format of visual novel. I'm not content with just making the same thing over many years. I want more. Maybe you won't stay with me through this whole journey, but I think you will see me again because I am always pushing to the limit.