News about 0.04.
Well, I once again missed the deadline I announced earlier
I took some time off from work next week to devote all that time to developing the game. Hopefully I'll be done by the end of next week.
Sorry to keep you waiting.
It's going well, but very slowly. The main problem is that I can't adequately estimate how long it will take me to do this or that. For example, if I think I can get something done in an hour, it might actually take me two days... The second problem is perfectionism. I can do something, I don't like the result and I redo it.
I'm still rendering. Each pose has to include several loops of 2-3 images to alternate these loops so that the animation doesn't seem looped. Each image has to be rendered in 4-6 ways: (with skin redness effect on/off) x (with visible and invisible grass) + (if the camera is not well chosen, I separately render the orc to make it semi-transparent with visible and invisible grass). Render with two options visible grass is necessary because if I render without grass, the characters appear to be hovering above the ground. And if I render with visible grass, then it passes through the bodies and stick out. So I render the two versions and then combine them.
So each pose requires about 40 renders, which then need to be mounted. And that's not counting all sorts of micro actions like slaps and other stuff. Oh! I forgot that I'm rendering two different views of the orc's dick. So conventionally 40 can still be multiplied by two, although this is not quite right because the second version of the dick does not require rendering a full image, but only a small area. But despite the fact that the rendering time of the second variant is less, it still takes time to mount and adjust the pose.
In general, the work in progress, gradually nearing completion, but I do not know how much time it will take me more.
I've finally finished rendering all the necessary images and I'm now busy post-processing and assembling those images. So the release is coming soon.
I'd also like to talk about my plans for the future and how to avoid similar delays in game releases in the future.
First of all, as I've written before, after this release I plan to move all of my work into Blender. But now I also plan to give up rendering images by ray tracing. Blender has a pretty good engine that allows you to get quality images without ray tracing. Working with it a little more complicated, but this largely pays off in terms of speed. Additionally, the plus side of Blender is that it is easier to automate. I very often need the same images, but with slight differences: with/without clothes, with different emotions, etc. And I could write a script that would switch these options and render them automatically. (In Daz Studio this is also possible in principle, but as long as it renders images for an hour it's pointless). To put it in perspective I see my work with the Blender in the following way: I prepare some frames (or the whole animation if we're talking about very distant perspective), run the script, and in 5-10 minutes I have the whole batch of images ready.
Secondly, the thing about working on my engine. I think I need one more release to get the engine to a state where I could implement all my ideas on it. Accordingly, after making these changes to the engine, I will limit the development time to one/two weeks per release. This should immediately reduce the time between releases by one and a half/two months.
I will write more about the planned engine improvements and how the story will evolve after the upcoming release.
Monthly progress report.
At this point, I have two images left to process. Each image takes about 2-4 days to complete.
(By image I mean all the rendered variants within one "animation". And by process, I mean post-processing the rendering; associating different areas of the image with the game's state variables; setting the rules by which these variables will change to create the "animation".)
Plans for the next release:
I remember back in early June, when I first started modeling the scene for this update, I thought it would be hard for me to do everything I had in mind. And that I need to do automation (see below), to switch to Blender since Daz Studio is not a modeling program, and that the rendering takes too much time and my attempts to reduce this time only worsen the quality of the picture. Then I had two choices: to do everything that I'm now planning for the 0.05 release or continue to do as I did before. By that time two months had already passed since the previous release and I was afraid that the first option would take too long, and the second... well, how long would it take? - A month, two tops. Lol. (I realize now that there was also a third option - to make the scenes simpler, like in previous releases.)
But the second option ended up taking six months. Hence the conclusion - with the tools that I am using now, it is impossible to make the scenes that I want within a reasonable period of time. I have to change tools. I've already written about switching to Blender in previous posts, now I will write about the engine improvements I plan to make.
What I want to do for the next release: 1) Additional layers. Controlling the transparency of each layer individually will allow me to create different interesting effects and save memory. For example, in this release I made a setting for the transparency of the orc in one scene. There are three fixed options - fully transparent, semi-transparent, and fully opaque. I could have used the fragment blending mode, but it was already busy with the smooth skin reddening effect. If the orc was in a separate layer, I could adjust its transparency smoothly, and the scene itself would take up three times less space in memory. With additional layers, a system of clothing and accessories could be made. And if I could control the transparency of the layer using masks, I could make a system of destroying clothes. 2) "Slow Variables". Now the engine must keep all of the fragments of the image in memory as it can not know in advance which piece it needs in the next moment, and dynamically to load them during rendering is too slow. "Slow variables" are variables that will change rarely. For example - what the character is wearing, or the stage of pregnancy. Obviously, these things will not change often, unlike the character's emotions or poses. Accordingly, after implementing this system, the engine will be able to keep in memory only those fragments that it needs in the near future. 3) Transformations - so that it is possible to reuse the same image in different scenes. Or render several of the same type of background objects. 4) Automatic splitting into areas and fragments. This is one of the kinds of routines I wrote about above. I do it manually now, for this release this work took several months. If implemented, this work would be done at the click of a button.
After implementing all these improvements, I will be able to do all the scenes on my engine the way I intended them. I will still continue to improve it. I still want to add full animation there, I need to come up with something with UI and so on. But all this is not urgent. I.e. after implementing this functionality I will limit the time of work on the engine to 1-2 weeks per release (not 1-2 months as now).
Plot.
This release consists of one event -> the minigame -> one scene. It ends with Orc leaving to rescue Helga from the hands of bandits. Accordingly, in the next release, the main event is "Orc meets Helga". The minigame will be improved. Also plans to significantly improve the win scene in the minigame. I want to make this scene more interactive. I mean, right now Dora reacts to your actions, but it's more of a decorative thing - it doesn't affect anything. But how much I can improve this scene will depend on how quickly I figure out the blender.
Thank God! You're a lifesaver for updating the latest release. I've been waiting for this in anticipation for a really long time!
Love this game and the slight comical theme but it's also the impending doom for our main protagonist haha!
Keep up the good work
