【Study】Tips for Acquiring New Skills

This article is written from the perspective of a game programmer, but it applies to web programming, software programming, 2D art, 3D art, design, music composition, language learning, cooking, gardening, and more.
Even if you're not interested in game development, I believe this article will be very helpful.

Thanks to the power of the internet, acquiring new skills has never been easier.
However, not all learning methods are equally effective.

For context, I'm a highly skilled software programmer and adept at creating designs or memes in GIMP and Photoshop, but I'm a beginner in drawing and 3D modeling, which I'm currently learning.
Next, I plan to learn music composition.
Below are tips for acquiring new skills based on my own experiences.

Limit AI Use at First

AI is a powerful tool for learning, but relying on it too heavily early on can hinder growth.
For example, proprietary console SDKs are often inaccessible to AI, and copying AI-generated code without understanding it teaches you nothing.
When I learned Go, I wrote code line by line by hand, tested it, understood its purpose, and then moved on.
This hands-on approach builds real skills.
Start with books, YouTube tutorials, or self-experimentation, and only use AI after struggling with a problem for a few days.
This balance ensures deep learning while leveraging AI's potential.

Don't Set Deadlines

This is a very common pitfall because we want to learn as much as possible, as quickly as possible.
This is completely understandable.
However, the problem is that the focus shifts to time rather than understanding, retention, or skill-building.
It becomes a race against the clock, shutting down your ability to learn new skills.

I understand that workplaces or schools require deadlines, as taking too long costs companies money or you might have an exam in a month.
However, this article assumes self-learning during your free time between work or study, so take as long as you need.
If something you planned to learn in 5 days takes 21 days, that's perfectly fine!
What matters is not how quickly you learn but the progress you make.

Don't Fear Getting Stuck

Getting stuck on a problem is essential to learning.
In Japanese culture, negative experiences are often seen as bad, but humans learn best from negative experiences.
Getting stuck is a negative experience, regardless of duration.

For example, have you ever done something embarrassing without realizing it?
When you found out it was embarrassing, did it feel bad?
Did you repeat it afterward?

Learning a new skill is the same.
When you get stuck on code that doesn't work, allow yourself to be stuck and don't rely on AI, videos, or books for help.
Using API references is fine.
This forces your brain to think critically and builds the courage to experiment.
Give yourself up to 96 hours, and if you still can't solve it, ask AI for an explanation.
You can set it to 48 hours, but in that case, instruct AI to provide only hints, not solutions or code.
This is all about learning skills, not just getting things done!

Don't Fear Failure

Failure is another critical tool for growth.
You'll fail many times before succeeding.
To succeed, you need to fail big.

For example, I built a game engine in Vulkan, but it failed to handle model files, so I started over.
Later, I got a full Vulkan engine working, but I realized NVN was better suited for Nintendo Switch games. My design was too complex to change, so I started over again.
Now, I've built a new game engine with NVN and Nintendo's proprietary framework in mind, and it's a good engine.

The same happened with web development.
Before creating Little Beast (the framework powering 076.moe and technicalsuwako.moe), I developed 60 different frameworks over 5 years, failing 59 times before finally creating something good.
The result is a framework with excellent performance and easy portability, running on any server OS, eliminating the need for static site generators or manually writing HTML for articles, and making maintenance and management easy.
This server runs on OpenBSD but also works on FreeBSD, Devuan, and even Windows 11.

In Japanese culture, perfectionism is deeply ingrained, but in reality, perfectionism is an obstacle to success.
It's okay not to be perfect.
You can't be perfect.
Just focus on doing it right.

Think of Thomas Edison.
He failed thousands of times before inventing the light bulb.
Just like my 59 failed frameworks led to Little Beast, every failure teaches you something new.

Before DOOM was made, its developers created 96 failed projects.
Before Kirby, HAL Laboratory made many failed games.
Before Pokémon, Game Freak made many failed games.
These are all part of the process.
Every failure holds the potential for success.

Start Learning Immediately

If you're thinking about learning something, start now.
Saying "I'll learn it later" often means you won't learn it at all.
Don't say "someday".
There are 7 days in a week, and "someday" isn't one of them.
Plans change over time, so what you planned to do tomorrow often doesn't happen.

Don't Hesitate, Just Try

Learning a new skill means diving into uncertainty.
Questions like "Can I learn this"? or "What if I can't"? will come up.
It's understandable to feel hesitant about trying something unknown.
The key is to let go of hesitation, muster courage, and give it a try.
You might find it fun.

A recent example is learning to draw concept art.
A few years ago, I bought a drawing tablet but didn't use it and lost the pen.
Next, I bought a liquid tablet, thinking I could draw directly on the screen, but it overheated, the screen was too dim, my neck hurt, and I stopped using it.
A few weeks ago, I bought a new drawing tablet but procrastinated learning for the first week.
Then I finally started, learned to draw digital art for the first time, and realized I was enjoying it.
Now I draw every day and improve the more I practice.

This is the tablet I'm using:

Summary

Here are my tips:

Limit AI use at first

Don't set deadlines

Don't fear getting stuck

Don't fear failure

Start learning immediately

Don't hesitate, just try

These tips are based on 30 years of learning new things, which have shaped who I am today.
I was born 34 years ago, but it's never too late to learn new things and become a better person.

Pick one skill you've been wanting to learn and start today.
Whether it's coding, drawing, cooking, or anything else, apply these tips, embrace the struggle, and share your progress with others.
What new skill will you master next?

That's all