Janonard

Makes software and tells how!

[swiftmusic-deactivated20230101]

Stylish shot from the new promotional video. Loving the stage, lighting and performance of the entire video.

Sexy bass guitar and me.

[janonard]

From which video does this awesome shot come from? Can you give me a link?

[janonard]

"Life" Development Update: 6th of July 2018

I have almost finished the concept! Really, it's almost there! Last week, I checked the writing again for the last time, and I think that it should be fine now. Unfortunately, I've figured out that I will not have the time to start the project: I'm planning to start my studies in Informatics later this year, and the pre-courses are starting soon, which leaves me only about a month of time. If I had finished everything in April or May, I could have started the project, got it rolling and then watch it from the backseat for a month or two, but now, it's too late.

Still, I am going to publish the concept and the demo and adequately promote them, but more with the message "Look at my ideas, I think they're nice!" and less with a "Look at my ideas, I need someone to implement them!". When this is done, I might become a bit quieter, or I will post about some toy projects, but nothing too big or important. Maybe, when I have more time, I will pick up the project again!

This is it for this week, now I will finish the very last pieces of the concept, complete the demo tech guides and then finish this project off. As always, you can follow the development at my GitHub repository and my roadmap aka to-do-list, which luckily is pretty empty right now! Until then, see you next week!

[janonard]

"Life" Game Development Update: 29th of June 2018

First of all, sorry for missing last week's update! Multiple things came together: I had several events regarding my A-levels I had to attend, which is why I had less time to work on "Life". Also, my excitement for "Life" has lost a bit of momentum since I had almost nothing to do except writing the concept, but I hope that I can pick it up again.

Anyway, I've almost completed the concept: Since the last update, I've added more descriptions for unfinished topics (problems without a proper solution or enough time to find one), a reflection/conclusion and an introduction. One might think that the concept is finished now, but I still want to read over it several times to adjust the grammar and the tone until it matches what I want to express. It now fills about 25 pages, which might become 26, but it doesn't have to.

As said before, almost everything for the demo is finished, it only misses some development guides I still have to write. Sometimes I really feel like I am a writer and not a software developer! :D

After all of this writing is finished, I will issue a proper release, which will also contain the concept, and then promote it on several forums. My mind isn't made up yet, but guess that ModTheSims, the Unity Forums or several Sub-Reddits might be a good place to start.

This is it for this week! As always, you can follow the development at my GitHub repository and my roadmap aka to-do-list, which luckily is pretty empty right now! Until then, see you next week!

[janonard]

Game Development Update: 15th of June 2018

I've finished the source code documentation! Finally! It's now generated automatically by Travis CI whenever I push something and hosted on the project's Github Pages site! If you are interested in the code, check it out there, I hope it's understanding.

Of course, the documentation isn't the only thing I did. The next task after the documentation was bug research job streamlining and so forth. Quality assurance, you know. I was surprised that some jobs had no cancel paths at all! When you cancel a job while it is running, it can handle this by transitioning to other states and finishing the job as soon as possible, but the showering and the mixing job had absolutely no cancel handling! The job always had to finish, if it was canceled or not. Especially creating cancel paths for the showering job was delicate since I had to resplit the animations into several smaller clips. In the end, this was quickly finished too.

After dealing a bit with the demo itself, I got back to writing: The concept paper is almost finished. All topics I want to include, except for the conclusion and the introduction, are written out and I only have to read over it again and correct spelling and grammar mistakes I made. If everything goes well, it might become ready to be released next week.

Talking about releases, I've also added a new snapshot build which you can download at the repositories releases page. Unfortunately, it does not include the aforementioned streamlining, so it might feel a bit clunky here and there, but contentwise everything is in there now.

My goal for next week clearly is to finish the concept paper. Apart from that, I might also continue writing guides on how the demo works and how to extend it. As always, you can follow the development at my GitHub repository and my roadmap aka to-do-list, which luckily is pretty empty right now! Until then, see you next week!

[janonard]

In an earlier post you mentioned some flaws you noticed in the Sims that you were hoping to improve upon in your current project. How else are you hoping to differentiate your work from the Sims? Are there other issues you are hoping to do better with?

Hi! I'm glad you asked!

Those issues I mentioned were only indicators of fundamental problems of the "Sims" series. If everything goes right, "Life" will have two fundamental things that differentiate it from the currently released "Sims" games: A branching skill tree and a dynamic emotional or reward system.

The first point, the skill tree, is quite obvious since doing and learning things in "Sims" is very linear: You have one skill that can include many different topics and develops from level one to ten. That's it. That doesn't look like real life, does it? Therefore, I want to introduce a skill system with many small skills that include only one logical task and depend on another. This way, you can have a different skill development for every character while having the same amount of content as "the Sims".

The second point is a bit more delicate: The basic idea is that the characters observe and learn which tasks and which characters make them happy and therefore want to repeat it. This want of repetition is expressed via the user interface (covering the desires from "The Sims") or taken into action when acting autonomously (covering the automated behavior from "The Sims"). At least a part of relationships is also included in this system since characters who like each other want to repeat tasks together and characters who dislike each other want to avoid them. This system elegantly covers many different previously separate parts of "The Sims" and gives them more depth and detail.

These two concepts are more or less the point between "The Sims" and "Life", but since they are so fundamental and imply so much, you end up with a very different game.

Apart from that, I want to make the game Open-Source. Not only because of me simply liking Open-Source, but also for practical reasons since I'm almost in the same position as you, only the other way around: I'm a programmer with little artistic experience. The models and animations that are shown in the demo are okay, but I don't really like them and to pursue the concept, I would need help from modelers and animators to make the game possible. Therefore, the game should be Open-Source to make it easier for artists and developers to join the project.

Another advantage of Open-Source is that you have an easier time establishing a modding community since the modders can use the same tools as the core developers. One could even say that the modders are the core developers!

I hope this answer solved at least parts of your question and that I did not go too much into unnecessary details. If something is unclear, just ask me again! Also, there is a complete concept paper in the works that might be finished next week and contains all information about my plans for the game.

[janonard]

Game Development Update: Balancing and Documentation

So, now I'm trying to do a weekly project update for "Life", the life simulation game demo I'm currently working on. Let's see if it takes off! Although this is supposed to be about this week, I'll also include half of the last week since the last update. You can follow the development in the project's repository, in it's roadmap and of course in this blog! So, what happened? Last week's Thursday, I created a bed, added it to the game and integrated it with an energy need, you saw it in the last video. Next, I wanted to tackle the balancing problems with the game and figured that this would be easier if the game had a day-night-cycle. This way, I could, for example, say that a day should be 15 real-time minutes long and a character should sleep for one-third of the day. This means that the sleeping time is five minutes long and in these five minutes, the character can restore the full energy satisfaction from zero to one, which results in a rate of satisfaction increase of 0.003 "satisfaction" per second. You see, with a working day-night-cycle, you can have a small set of reasonable but arbitrary decisions and calculate the required values from them. But first, I needed a day-night-cycle in the game before I could balance it! There were two important points: The the position of the sun and time warps. Both were surprisingly easy to tackle: In Unity, you have a directional light source that resembles the sun and is connected to the skybox. When you rotate the light, you also rotate the skybox and the sun moves with it. Now, you only have to calculate a rotation from the passed time and e voilá! You have a sun going up and down. If you are interested in the exact code, you can see it here. The other "problem" was time warping: When the characters sleep, you don't want to wait for five real-time minutes, you want to accelerate the flow of time. Again, this is easy in Unity, since there is a time scale parameter which is multiplied by the real world time to produce a, well, scaled time. When the time scale is set to two, everything happens twice as fast, you can even set it to zero and everything halts! Adding a control surface for the different speeds and that also shows the in-game time was easy, the only real problem was that, somehow, the inputs are also scaled with the scaled time and you, therefore, couldn't move the camera when the time was frozen. Switching to raw, unscaled input values solved the problem. Now, I could go and balance the game. All the details are found in this spreadsheet, you can see how everything is calculated there. I think I did an acceptable job with enough time that could be filled with proper tasks, but maybe the days are a bit too short. We'll see! :D Balancing was finished last Saturday and my next task was to document the code. I added Doxygen and GitHub Pages support, set Travis CI up to build the documentation automatically and restructured the code. This took me about two days, Sunday and Monday, with a lot of moving back and forth. When this was finished, I started to document the code on a separate branch so Travis did not rebuild everything I made. I'm almost finished with the first run over every element and maybe I'll go over everything again to check that everything aligns well. I must say that almost nothing pleases me as complete, documented and interlinked code and I hope that you can see that I spent much effort in this project! Conceptwise, I didn't do much: I restructured the text a bit and had some thoughts on relationships, but nothing worth writing about here. Following week, I hope to finish the documentation and write guides on how to modify and extend the game demo. When this is done, I only have to finish the concept paper and then I might be ready to release. Wish me best of luck, until next week!

[janonard]

Commenting!

[janonard]

And here is the bed, according to plan, the last prop and need to be added. The animations aren't great, as always ;), but it's okay for now. Now, I "only" have to balance some numbers, document the code and finish the concept.

Talking about the concept, I have decided to exclude the question of the engine choice from the first paper and tackle it when the gameplay planning is completed. Until then, I'll simply continue with planning and development.

[programmerhumour]

Turn out the lights >:o

[janonard]

Ehm, no. When I program, I have the lights on and therefore I need the light theme!

[janonard]

Concept, Concept, Concept!

Here I am again! Last week, I was on vacation with my family, which is why I didn't post anything. Instead, I had quite some time to think about the concept, which I did. I mostly settled on what to include in the concept and what not and also sketched most of the core contents. Now I only have to write them down. Yay!

I've already written about four pages, but I am only at the beginning. Here is the current structure of the paper, just to give you an image of what I still have to write. Please note that this structure may change in the process because when I start to write my ideas down, I instantly start to think of their problems, which I then have to solve and write down again. But for now, this is what I work with:

Gameplay:

General Gameplay (done)

Reward System (done)

How does the reward level affect the character?

Learning System (done)

optional complexity (done)

Careers and Money

being evil

relationships

Which skill systems to include:

Social

Teaching

Sports

Open for discussion

clothing

building/house editor

open world

online features

targeted age, violence, explicit content and ethics

Technical:

engine choice

How to implement the systems

Voice and sounds

licensing

Reflection:

Will it be fun?

Is it possible?

Who and What is needed?

What to do next?

[janonard]

The fridge is done! I think it looks quite good, although my animations, as always, are a bit ugly. Anyways, I'm a bit surprised that I could create this prototype only in two days using the system and tools I created and I think, if a better animator or even another modeller joins the project, alot more is possible.

If you want to be this animator or modeller or something completely elso, you are very welcome to help. I host both the Art and the Game on Github, both under open source licenses.

I have also published a snapshot version you can simply download and run here. Now, the demo is almost finished, only a bed, balancing, some documentation and, of course, the concept is needed.

Until next time!

[janonard]

Look at my fridge, my fridge is amazing! Well, not so much. It's a fridge, everything I need for now!

[janonard]

Concept Update: Emotions

In the last days, I have thought about emotions in "Life", the life simulation game I am working on. In "Sims 4", you have moodlets that stack on top of each other. The mood with the most/strongest moodlets is the current mood of the Sim, which can change animations and enables or disables certain actions. That's it. It is good but does not go deep enough (Just like Sims 4 in general :D)

The system I came up with includes emotions, autonomy, relationships, personalities, and needs: Most of the things a character can do give moodlets which make it feel happy, sad, uneasy, etc. Being happy is like a reward for the character and therefore it remembers what actions made it happy and wants to repeat that. That also goes the other way around: A character remembers when something made it feel bad and wants to avoid it.

These needs/desires/wants (I have no proper name for them yet) are expressed via the user interface and give the player a guideline want to do, but they are also used to determine what to do when the character is acting autonomously: Basically, if a character is autonomous, it always executes the most urgent need/desire/want.

Relationships are also implemented using this system: The character gets rewards/good emotions when doing something with another character with similar interests. Therefore, they both want to repeat what they have done or want to do other things together. This continuous want for social interactions (or absence of social interactions) with a certain character certainly is a relationship, isn't it?

Last but not least, we have personalities: These are a mixture of predefined capabilities that give certain moods for certain activities and experiences a character got over its lifespan. If, for example, a character is good at cooking (because the player gave it this trait), it gets higher rewards for cooking and therefore starts to like it, which in turn forms its personality.

I think that this is quite a decent system since it covers so much, but will also involve a lot of balancing, especially when the game becomes bigger and you have more things to do. But you don't make a game because it's easy, don't ya? Do you like that idea? I'm still open to questions and comments. Until next time!

[commanderbellrke-deactivated202]

ok I have to take a course on coding, and I was wondering what the difference was between python and java was, like is one easier than the other?

[programmingmadness]

I would say Python is easier to learn than Java. The syntax is slimmer and you need to write less “useless” code to get things done (called “boilerplate code”). You can write and test things quickly because Python is also a scripting language - in Java you have to use an elaborate setup to test your code/code snippets, in Python you just write something and exexute to see results. I always recommend Python to beginners even though I started with Java because it operates on a slightly higher level of abstraction which means you don’t have to get into the nitty-gritty of computers to accomplish something simple. Once you’ve learned Python, you can also easily learn Java or other languages since you already know the basics (but that goes for basically every language you learn). So I would definitely take the Python course if I were you.

If you have more specific questions just contact me!

[janonard]

I would also recommend Python because you can do almost anything with it: Python is said to be "batteries included", which means that most of the stuff you need is included in the standard library. For example you can access databases via SQL, use internet connections, both raw and SSL-encrypted, create user interfaces and so on and so on without a need to know anything except the standard library. Python is also platform independent and supports meta programming.

[janonard]

Hello everyone! This is a new concept post for "Life", the life simulation game I am currently a demo of. I am trying to post more of those regulary while I am working on the complete concept paper, so is an exerpt of it:

The root of "Life" is the learning and skilling system. Everything depends on it and feeds back to support it.

Everything a character, the "Sims" of "Life", can do, they have learned from a skill. This includes everything, the basic things like sitting and walking as well as complex things like building rockets. These skills are atomized and therefore only include as few activities as possible. When you look at, for example, cooking rice, this would include the skills of boiling water and using a timer to know when to remove the pot from the cooker, which would be separated.

A character can train activities by doing them, reading books about them or getting taught by someone who is able to teach the skill. The development of the skill is described by the mathematical functionf(x)=1-e^(-(1/a)x), as seen in the diagram below, where x stands for the amount of effort put into the skill and a for the complexity of the skill. The higher the value of f(x) is, the better the quality of the activities supplied by the skill is.

Usually, you have four stages: The first at f(x)=0, where the character can not do the activity, one at 0f(x)<0.5, where the character understood the basic idea of the activity and is barely able to do the activity, one at 0.5f(x)<1, where the character is properly able to do the activity and teach others how to do it and one at f(x)1, where it mastered the activity. Depending on the activity the skill delivers, these stages may have substeps or flow into each other, but this is the basic idea of the skill system.

As implied before, skills depend on each other, which is based on the fact that certain activities require other activities and therefore cannot be done when the characters can not do these other activities. If, on the other hand, someone tells a character the concept of another skill that currently is not inside the range of possibilities of the character, it still can learn this concept and therefore train the skill.

[chiusan]

i need help with unreal engine, so the problem i am having is: the red arrow marks the point where my mouse is located, i pull and far from where i clicked pops up a yellow tile that changes its size according to how much i pull after clicking. to me it seems as if the whole thing is shifted waaaay to the left. i also cant chose single tiles like people in tutoriel videos are doing. i am using unreal engine on a macbook pro.

[janonard]

Could you show me the code you are using? I might have an answer for you, but as far as I know, there are multiple ways the achieve what you are trying to do. The best way I could help you would be if you'd upload your project to something like Github, but a simple screenshot of the blueprints you are using might do it too :)

[janonard]

My reason for "Life"

In my last post, I hinted at my new project: An open-source life simulation game. It's basically supposed to be something like “The Sims”, just a bit better and open-source!

I am a long-time fan of the Sims series. Sims 2 was one of the first games I seriously played, which lead to my laptop's GPU being fried! Since then, I am following the development and I must say that the current iteration, Sims 4, is the best life simulation game currently on the market. It is mostly bug-free (especially in comparison to Sims 3), you have a lot of stuff to do, it looks nice and has incredibly good animations. I really enjoy it, but it isn't flawless:

The first flaw that came to my mind was, that making music is extremely unsocial: You can't play songs together, you can't write songs together, and you even can't teach others the songs you wrote! Also, the progression of skills, in general, is described as a single value between one and ten. It was always like this and I recognize why EA continued that way, but I think that in the times of exhaustive skill trees in role-playing games, a little more complexity should be possible. But making music isn't the only activity that was treated this way. Most of the other skills are also heavily focused on practicing and learning alone, which is a bit sad, since learning in real life so much more social.

After recognizing this issue, some more caught my eye:

EA's rigorous DLC policy lead to a tessellation of content, where some elements don't interact with other elements at all or some depend on content from other DLCs.

There is almost no reason to dislike or hate another sim, except cheating on their boyfriend/girlfriend and just being simply "evil". This boils down to the problem that there are almost no interactions that have a good or understandable reason for the actor and a negative result for the victim. No one steals something just because they are evil, they have a reason for that and this is missing in "The Sims".

Custom Content Creators have to hack themselves into the game because EA never gave them any tools or useful information on how to create content. They just said: "Here is a folder where you can place mods in. Now you have to figure out how to use it!". If they published any valuable information, I could not find it and therefore they didn't promote it well enough.

From my point of view, these problems can be tackled by creating a new, open-source life simulation game. First of all, we can freely design the game in the way we want it to be and don't have to split it into multiple packages so we earn more money from it. Also, there would be no real difference between developer and custom content creator, since every content would be something like "custom content".

I'm currently working on a concept and request-for-comments paper that fleshes out the basic design decisions and a demo to showcase them. It's been in the works for roughly two months and it already includes the job queue, traits, needs, and a basic level with disappearing walls. You can download it at it's Github repository, which also features a roadmap that represents its current status.