[R>Programmer] Share Artificial Intelligence Operating System development

[EjemploKuro]

Artificial Intelligence Operating System development

Project:
development of cluster-oriented operating system that natively supports and execution of artificial intelligence able to manage the cluster and native parallel processes. O.S. will count with three main engines:

1. cognitive motor
A.I. for the development of programs that require executing logical and abstract tasks based on artistic scripts, for example in games real-time dialogue generation in response to the user, automatic events generation based on historical context, creation and execution of personality for NPCs.

2. physics engine
engine for the development of virtual spaces that simulates a real environment managed by A.I. generating and developing the environment based on the artistic scripts.

3. graphics engine
engine to convert the abstract elements of the A.I. in graphic elements that can be displayed on screen.

Looking for:
investors and developers or curious people interested in the project.
low-level programmers (assembler lenguaje ADM64 and ARM64).

contact:
Discord unreal_os

Additional comments:
English is not my native language.
This project hopes to be a step towards full-dive games. The idea of the project is that the O.S. allows to support a full A.I. within a server cluster and this A.I. is engine that facilitates the development of complex environments and games that require autonomous NPCs, events and developments based on the artistic scripts, summarizing living virtual worlds.

 

[Saint Blackmoor]

I applaud you for your enthusiasm. How much do you know about AI, and what have you done so far?
Your comment is a perfect example of how complex AI is. Can you break down your comment into more relatable concepts?

 

[EjemploKuro]

I applaud you for your enthusiasm. How much do you know about AI, and what have you done so far?
Your comment is a perfect example of how complex AI is. Can you break down your comment into more relatable concepts?

hi thanks for asking, I like civilization games and NPCs games like https://f95zone.to/threads/school-game-v0-942-sloths-command.48601/ and https://f95zone.to/threads/life-is-paradise-v1-0-sugar-star.24045/, where the NPCs act beyond a single function and become part of the game world, the player not only enjoys the main story of the game but also all the sub-stories generated by the NPCs, such as the player who is exploring for a certain mission when he unexpectedly encounters an "unexpected event" and can ignore or intervene to receive different types of responses depending on the player's relationship with the NPCs.,as the player who is more interested in the stories and relationships between the NPCs than his own or how the treatment of the game towards the player changes according to the actions of this.

These mechanics allow the game to continue indefinitely but its development presents some serious problems such as:

1. choice depletion, current game engines do not have enough facilities for building these games, which leads developers to build action modules but the combination of these leads to innumerable errors and performance failures for this reason, although these games appear to present a great variety of choices, reality converges in few results.

2. randomness difficulty, these games base their decisions on random choice for example in "rain world" but uncertainty of randomness puts the player in circumstances that are impossible to resolve and generates illogical or lacking common sense events.

3. linearity, these games are born from the idea of elements autonomy, but for the above reasons, the actions and events occur only with the player, therefore all possible sub-stories is limited. there is a reference to the monotony of the NPCs in the "NPC rape" games and the doujin "NPC Kan MOD" can appreciate what a "full NPC rape" game would be like.

the use of A.I. to solve these and other problems allows a more feasible and profitable development. this project does not use the current model of A.I. because these current software lack "intelligence", current A.I. are correctly database analysis programs, For example the known A.I. that can draw actually can't draw, they only associate sequences of pixels based on tags from a large number of drawings in databases they are completely incapable of drawing something that is not on the internet. and therefore are unable to be used in the developmentof these games.

the A.I. in this project is born from a certain thesis under the concept of full A.I. a software that is capable of thinking, theories and algorithms associated with this A.I. are currently private to investors and developers of this project and will be published in the developer documentation after the release of the O.S.

 

[Saint Blackmoor]

Thanks for the explanation. I understand it a little better now. Do you think there are parts of your concept that can be implemented in current game engines? Let's use Renpy as an example. Character creation in the game based on input from the player at the beginning. Similar to the prompt system AI uses now.

Even though the concept of AI has been around for a while, the big advances are very recent. The quickness and its widespread use have brought up all sorts of concerns. In one sense, the Skynet scenario that everyone thinks about when they think AI could already be happening, just at a slower pace than anyone thought it would.
I personally don't think that way. In the past, I'm sure people felt electricity and the telephone were some kind of "Magic." We just need time to understand it better.

Your thoughts on the new "game engine" is pretty amazing.

 

[peterppp]

how is this anything but an unrealistic fantasy? a few people from an adult site is gonna beat the big ai companies in making a full ai? if you had anything to offer in the development of full ai, you would work for the ai companies, not coming here looking to make smut

 

[EjemploKuro]

Thanks for the explanation. I understand it a little better now. Do you think there are parts of your concept that can be implemented in current game engines? Let's use Renpy as an example. Character creation in the game based on input from the player at the beginning. Similar to the prompt system AI uses now.

Even though the concept of AI has been around for a while, the big advances are very recent. The quickness and its widespread use have brought up all sorts of concerns. In one sense, the Skynet scenario that everyone thinks about when they think AI could already be happening, just at a slower pace than anyone thought it would.
I personally don't think that way. In the past, I'm sure people felt electricity and the telephone were some kind of "Magic." We just need time to understand it better.

Your thoughts on the new "game engine" is pretty amazing.

I don't think they can be implemented efficiently since the development method is different, for example if we compare the game with a tree currently the development focuses on the creation of the tree, with full A.I. tree is not developed but the seed, game developers create seed and the full A.I. ensures that it grows into a tree, but as a prototype, cross development would be interesting where the full A.I. on server defines real-time dialogues and choice to character in a renpy game.

 

[EjemploKuro]

how is this anything but an unrealistic fantasy? a few people from an adult site is gonna beat the big ai companies in making a full ai? if you had anything to offer in the development of full ai, you would work for the ai companies, not coming here looking to make smut

unfortunately an full A.I. is not pleasant for A.I. companies because they have already invested and worked on an Artificial neural networks (ANNs) model so they are not going to invest in a new model and i'm a programmer not a trader this is one of the few ways i know. I prefer to work with people who have a personal interest in the project.

 

[peterppp]

unfortunately an full A.I. is not pleasant for A.I. companies because they have already invested and worked on an Artificial neural networks (ANNs) model so they are not going to invest in a new model and i'm a programmer not a trader this is one of the few ways i know. I prefer to work with people who have a personal interest in the project.

no way they would miss out on full ai if they could make one. they are surely working on it as we speak but it could take many years before they get there. sorry, but i can't take you seriously hoping the birth of full ai will come from this

 

[natsumeku]

i think it's super over scale project. i mean this is:

• an operating system
• a software development
• a game engine development
• an A.I solution

i don't know what are you trying to solve. but i don't recommend to reinvent the wheel.

 

[Mike the Red]

I agree that the scope of this is too great.

In my mind, what you're trying to create is a server-type program that can interface with a database (or several) and a game engine.

If you plan to use existing components for databases and write an interface for existing game engines, rather than designing them yourself, I think this project might be possible, but still extremely daunting.

That said, components two and three of your project should be split off into a follow up project. While using assembly will definitely give you the best performance, the process of writing in it is so onerous that this project will still not be completed for a decade. Stick to something like C++ for the bulk of your development and implement small bits of manually coded assembly only if you absolutely must optimize a specific subprocess.

 

[EjemploKuro]

i think it's super over scale project. i mean this is:

• an operating system
• a software development
• a game engine development
• an A.I solution

i don't know what are you trying to solve. but i don't recommend to reinvent the wheel.

sorry for the misunderstanding, currently this project deals exclusively with the O.S.

 

[EjemploKuro]

I agree that the scope of this is too great.

In my mind, what you're trying to create is a server-type program that can interface with a database (or several) and a game engine.

If you plan to use existing components for databases and write an interface for existing game engines, rather than designing them yourself, I think this project might be possible, but still extremely daunting.

That said, components two and three of your project should be split off into a follow up project. While using assembly will definitely give you the best performance, the process of writing in it is so onerous that this project will still not be completed for a decade. Stick to something like C++ for the bulk of your development and implement small bits of manually coded assembly only if you absolutely must optimize a specific subprocess.

Currently, assembler is used for kernel development, then we will continue working with c and c++, leaving assembler for post compiled

 

[Mike the Red]

What exactly is the benefit of writing your own kernel versus using an existing one for this project? That seems like a ton of effort and, unless you have an unusual hardware setup, I don't see why that effort is necessary.

 

[EjemploKuro]

What exactly is the benefit of writing your own kernel versus using an existing one for this project? That seems like a ton of effort and, unless you have an unusual hardware setup, I don't see why that effort is necessary.

to meet the objectives of the O.S. Extensive optimization of scalar calculations, native support of parallel processing, and fewest abstraction layers are required.

 

[Mike the Red]

Thanks for the response. I guess that I don't know enough about kernel development to recognize the need to redo all that.

 

[EjemploKuro]

Thanks for the response. I guess that I don't know enough about kernel development to recognize the need to redo all that.

it's a simple matter the A.I. that O.S. has to endure, theoretically has a huge consumption of computing power, so it is obligatory to reduce hardware load. we are grateful for all your questions