Make 2025 The Year You Learn Blender
In this article I will convince you that you can, and that you should learn Blender in 2025.
I will continue to provide Blender content and tutorials on this blog and various other platforms. But first, let's cast away some doubts about learning Blender.
- Is it worth learning Blender?
- Is it even possible to learn such a "difficult" skill if you don't have a background in Computer Graphics or other disciplines?
- Do you need an expensive, powerful computer/laptop to learn Blender?
- Before learning Blender 3D, should you start learning 2D drawing/painting instead?
- Finally, is it worth learning any skill at all in the age of "Artificial Intelligence" (AI)?
- Summary
I will answer these questions today.
Is it worth learning Blender?
Short answer is, yes.
Long answer is, yes it is.
Perhaps if you are already aware of the possibilities of 3D creation, you won't need any convincing that learning 3D art and design is so rewarding and fun. When you learn Blender, you can create all sorts of things—from video game assets to animated films, and from 3D printable models to assets for scientific research and development.
You can model your house with actual dimensions and move around digital furniture to test your house layout, without even lifting your finger (well, you only need to lift your finger). You can even use Blender for video and photo editing.
Here is a non-exhaustive list of professions and tasks that can make use of Blender: 3D Modeling, Sculpting, 3D/2D Animator, Visual Effects Artist, Game Designer/Developer, Game Asset Creator, Automobile Designer, Architectural Visualizer, Virtual Reality (VR)/Augmented Reality (AR) Developer, Film and Video production, Physics Simulation, Educator or Trainer, Graphic Designer, Concept Artist, Story Board Artist, Advertising and Marketing, Technical Artist, Programmer, Scientists and Researches from a wide range of studies, Medical Professionals... This list could literally go on for a long time.
If you have an interest in bringing your imagination into life in the digital realm, Blender is both an extremely powerful and cheap - no, excuse me, free - way of doing so.
I heard other 3D software are more powerful than Blender...
What about other 3D software, and those which cost money?
Aren't they simply better than Blender? To that I would answer: No, not really. There are certain powerful features that exist in paid software. But I don't think those features justify the price. These paid software programs used to be expensive. Today they are really expensive because most of them have switched to subscription models. This means that you can't even buy these software programs. You must pay monthly or yearly subscription fees for a program that you won't even be making money off of, or for a program that you won't even be using that often.
Another point to consider is the following: You will never hear people who pay for these programs say that Blender is actually better. They will defend their paid software harshly, and they have a good reason to do so: They paid money for them. Nobody who pays money for something would admit that their purchase is not worth it. Don't listen to people who talk negatively of Blender. Blender is a very capable program. You only need to learn how to make it work for you.
Remember that Blender is just a tool, like any other 3D software. It is pretty much capable of doing anything. At this point in time, it is loaded with a wide range of features. And if a certain feature is missing, you can extend its functionality by programming that feature into it. You might say, I am an average Joe, I can't possibly code a new feature to Blender. Yes you can. And it can be a very rewarding experience. We will talk about that in a different blog post as well.
Finally, here is another strong argument for why I think learning Blender is worth it: The knowledge you will learn in Blender is transferable to other digital creation software. The other way around is also correct. If you already know a 3D modeling software, you can transfer a lot of your knowledge to Blender. This is because, fundamentally, all 3D modeling software involve some level of vertex, edge and surface manipulation. This is called polygon modeling. Another type of modeling is sculpting, which Blender supports as well. If you come from another sculpting software, then you can use your existing knowledge to use Blender effectively as well.
So choosing Blender is a free and fun way of learning 3D digital creation. Blender shines among other programs as the most perfect candidate in case you intend to begin your journey into the 3D world.
In this part, I explained some of my main opinions on why learning Blender is worthwhile. Let's jump into the next topic.
Is it even possible to learn Blender if I don't have a background in [insert a shiny background]?
Short answer is Yes.
All you need to learn Blender is a passion for it and the ambition to push forward. Let me tell you my personal story: I learned English by watching Blender tutorials. My desire to learn Blender was so strong that not knowing the language in which it was taught did not stop me. I learned the language too. I was like, 12 or something. Tutorials and lessons in my language at the time were almost non-existent. Even today it is very difficult to find up-to-date Blender tutorials in my language. So if you already speak English, you will find more tutorials online than you will ever need (even then I am on a mission to create Blender tutorials in English).
All you need to do is find good tutorials that show you how to create similar things to what you dream of creating. Follow these tutorials and apply them. You will learn along the way. There are good tutorials and then there are bad tutorials. The only way to distinguish between them is to just push forward. Along the way you will also develop a good sense of which tutorials are good for you and which tutorials don't work for you.
Personally I learnt a lot from early Blender Guru tutorials. Andrew Price's videos were absolutely amazing. He would just show you how to do something step by step. He would not loose much time explaining the theory behind, because he would not know them himself. This way, in my opinion, he created one of the best tutorials online. Just show me how it is done. Show me step by step, without skipping anything. This way after having finished watching and applying your tutorial, I will have created the exact scene/model/whatever that you showed in the beginning. Such style of tutorials were what clicked for me.
Unfortunately these days Blender Guru and also most tutorials online are more about summarizing how they created this or that, without showing every single thing step by step. In these new concept tutorials, a lot of steps are just "mentioned" or skipped through quickly. Such tutorials are full of good insight. But in my opinion, they are intimidating for the beginners and they do not really help the beginners take off.
Personally, I am planning to create tutorials that speak to the new beginners, as well as seasoned players in the game. I want to skip over the boring parts like, this button does this, this slider does this, and just jump straight into creating a scene/model/whatever. I will show every single click that I do, every thought process behind a creation. I also want to teach the theory behind things. This way, you will be able to follow my tutorials step by step, and after having watched a tutorial and applied everything I did, you will (more or less) get the same exact result. And along the way, this should teach you how to use Blender.
At least for me, this was how I learned Blender. Let's discuss the next topic.
Do you need an expensive, powerful computer/laptop to learn Blender?
No, you don't need a powerful computer to learn blender. Here is my current setup, and the setup with which I am planning to write and record all my tutorials:
- Laptop: ThinkPad X230
- CPU: Intel i7-3520M (4) @ 3.600GHz
- GPU: Intel 3rd Gen Core processor Graphics Controller
- Memory: 8 GB
- Hard-Drive space: 512 GB
- Laptop resolution: 1366x768
- External monitor: 1920x1080
- OS: Archlinux
This laptop is from 2012. Now if you are reading this from your RTX 4090 equipped new desktop computer, you might cringe over my setup. Can I really compete with your setup?
Well, I cannot, and I don't need to. The point is that, if you want to learn Blender, if you have any setup similar to mine or better, you can absolutely learn Blender. Along the way, I will not only teach you Blender, but I will teach you ways to optimize your Blender scenes to such a degree that it will either make RTX 4080 look like an overkill, or it will make such a powerful GPU akin to a render farm in your eyes. Having a strong computer will most of the time save you time during the rendering process. This means that when you finish creating a scene or an animation, you will have to render out your result. Rendering is a compute intensive step and having a strong computer can dramatically reduce render times.
Back in the days, I had a desktop that was more or less as powerful as my current laptop listed above. I remember when I would leave my computer on for the night, or sometimes leave it on for a day or two, to finish out a single render.
Well, today I still have a mediocre setup. However, we live in the age of technological wonders. Today, for example, we have a renderer in Blender called EEVEE, which takes advantage of real time rendering. Back in the day, real-time rendering meant a render quality similar to that of a game like Half-Life (1998). Today, thanks to EEVEE, you can render in real-time quite realistic scenes with dynamic lighting, realistic materials and compositing effects.
It is worth noting here that Blender has 2 main renderers. One is EEVEE, which allows you to render things in real-time. EEVEE works by a bunch of optimizations that allow you to create realistic scenes in real-time. Due to these optimizations, the resulting render will not look as realistic as real life. The other rendering engine, called Cycles, on the other hand, uses real-life physical rules to render your scenes. Cycles is much more compute intensive, meaning that it works slower than EEVEE. It won't allow you to have real-time speeds, but it will give you pretty realistic results. However with the newest hardware like RTX graphics cards, you can also get pretty fast renders in Cycles. I have used RTX 2080 Ti for a long time. In that graphics card, I could get almost real-time rendering speeds with Cycles (of course that also depends on the complexity of your scene). We will only have to wait to see what technological wonders and novelties await for us in the future.
Do not block your progress by fooling yourself into believing that you can't learn Blender unless you have a strong computer setup. If you think like this, the following will happen to you: You will buy a strong computer, thinking that you will be able to run any scene in Blender no matter the complexity. Since you won't learn how to optimize your scene, you will end up creating horribly complex scenes that will slow down even the strongest commercial computers. You will be left wondering whether you should have bought something even better. Meanwhile I will be chilling with my Thinkpad X230.
Next topic.
Before learning Blender 3D, should you start learning 2D drawing/painting instead?
This one will be a shorter section. This is a question that I have come across quite a few times, including myself. Before I explain what I think, please remember that this is solely my personal experience. Everybody might have a different experience on this topic.
I think that learning 2D art is absolutely not necessary if you want to learn 3D digital art. Here is a confession: I really suck at 2D art and I want to learn it myself. Not that I ever though that if I knew 2D art, I would be better at 3D art as well. However, I am slowly realizing the power and efficiency that comes with 2D art, and I am starting to appreciate its merits. Personally, I am starting to think that knowing 2D art could potentially help you in your 3D endeavors as well.
Having said that, if you plan to learn 2D art before 3D, I would rather suggest you to learn both at the same time. Learning 2D and 3D art does not have to be sequential. One does not come after another. And I must add: Apart from the most basic theory, like coloring and lighting, 2D and 3D art are very different from each other. Having learned both will give you super powers in the art world. However not knowing one or the other is not really going to hinder you either.
The reason why I suggest you to learn both at the same time is the following: If you focus on 2D art for a whole year and not even open Blender, you will have lost the opportunity of having tinkered around with Blender for a whole year. In other words: Our brains perform passive learning. Even if you just open Blender and tinker around here and there occasionally, meanwhile dedicating most of your time to 2D art, at the end of a year, you will have learned a tremendous amount of 3D as well. Opening Blender for the first time can be intimidating. But opening it the 10th time will not be as intimidating. So don't wait to have opened it the 10th time.
Finally: Blender has a very strong 2D suite built-in as well! It is called grease pencil. If you are interested in 2D and 3D art, Blender is the right tool to take you the next level by combining 2D art itself into a 3D scene. It is quite mind-blowing, but I must stop here. Otherwise this topic will turn into a long read.
Finally, is it worth learning any skill at all in the age of "Artificial Intelligence" (AI)?
I will try to keep this section as short as possible. In another blog post, I would like to discuss this section more in depth.
This final question must be bothering all of us. AI news are everywhere all the time and it became unavoidable to talk about it. If you haven't heard about it, you must be living under a rock. Actually, if you do live under a rock, don't come out here because honestly it is getting stressful.
AI, or more specifically Generative AI (GenAI) has been dominating our attention in the last 3-4 years by now. One of the most shocking applications of this new technology has bothered artists quite a bit, and for a good reason: Real human artists' hard-work, their art pieces has been stolen from the Internet and used to create these GenAI tools. Perhaps you could say, what is on the internet is free to grab. Surely, but what bothers most people is that these GenAI tools are commercialized, and the people behind these new technologies act as if these artists' hard work was really created so that they could make these GenAI tools. They act as if the real artists' work was nothing more than a stepping stone to create this new piece of technology. Furthermore, they go on to claim that these image generators can now replace human artists. Big claim. Finally, we have a horde of people who can't even draw a stick figure go around and claim that they are now "artists", because they write a prompt and the image generators give them an art piece. These people are indeed an artist as much as a customer who orders a burger is a cook.
All these political and bothersome attitudes from people outside of the art world aside, I would like to discuss technical aspects of this new technology to set your minds at ease.
Because here is what I think:
- Generative AI tools do not work AND will never work as good as they are marketed to be,
- They cannot replace humans,
- And in fact they are going to make human artists' work even more valuable than before.
Finally, I will share my opinion on another crucial point:
- What position should artists take against AI, and against intellectual theft that they are exposed to?
Let's talk about each and every claim of mine one by one. Before I begin, here is my disclaimer: You don't have to trust my words on this topic. But I feel obliged to tell you that I am a Computer Science major. I know quite a bit about machine learning (the field of study behind current day "AI"). I have also worked quite a bit with linguistic applications and transformers (the tech behind LLMs, aka ChatGPT, Claude, Gemini etc...). I also studied western philosophy and pondered on human language, intelligence and consciousness quite a bit.
I am not an expert, and I am surely not trying to sell hope. The following opinions are not my own; I did not come up with them, but I support them.
1. "AI" does not work, and will never work as good as it is marketed
All GenAI tools are based on a field of study in Computer Science called "Machine Learning", which itself is based on Statistics. This is the same technology that has brought us innovations such as Google Translate, visual recognition systems, speech recognition systems, and even self driving cars, albeit non of these technologies are perfect.
These systems "learn" from vast amounts of data, or by getting trained in a simulated environment which generates them the data they need, and they end up becoming useful tools. For example, after they are trained on some data, they can be used to classify further new data that looks like their training data. Or they can generate more data that looks like their training data.
Let's give a practical example on how they work. Imagine that we want to build a machine learning model that can tell apart a cat from a dog. In order to build such a magic model, first we need to get plenty of images of cats and dogs, and manuall label these images as such. Let's say that now we have 100 labeled images for cats and dogs each.
We can now feed these images to a machine learning model. Now it can magically "learn" to associate the word "cat" with cat images, and the word "dog" for dog images.
Great! Now we have an "AI" that can tell cats from dogs. But here is the catch: If I show this "AI" an image of a crocodile, the AI will either guess for a cat or dog! Why? Because it was trained to tell you only that. During the training, we did not show the AI what a crocodile looks like. So it simply cannot answer you correctly.
This is the reason "AI" does not work as intended and never will. Currently, gigantic and very sophisticated AI models like ChatGPT are trained on all the text data that exists on the Internet. This is surely a great achievement. Now ChatGPT can recycle and spit out the information that already existed in the Internet.
However, if you ask ChatGPT a question that does not already exist in the Internet, it will simply fail to give you the answer you are looking for. ChatGPT sounds smart, but it really is a parrot that is spitting out what it already heard.
So these Generative Artificial Intelligence models cannot really come up with something that is "new". They cannot come up with "something new", because they do not have the cognitive faculties to do so. Machine Learning is based on statistics at the very bottom of it, and we are well aware of the strengths and limitations of statistics. A statistical model can generalize, can converge and it can extrapolate. But it cannot get out of the box in which it was created.
The same goes for AI models that generate "art". Such models will only recycle and generate the swath of human Art which was utilized to create the AI. Any new artistic expression remains to be a human endeavour.
2. AI cannot replace humans.
This point is already explained above. But I can further explain why.
We still don't know how humans work. We don't know how human intelligence work. We can't even define it.
When we say that AI is not intelligent, AI advocates often answer that we must first define intelligence. Since we cannot define intelligence, we cannot call AI not intelligent.
However, since we cannot define intelligence, we surely cannot claim that AI has any intelligence at all, or that it will continue to improve from its current state, and somehow become more "intelligent". Honestly I think it did not age well to keep saying "just wait for the next AI model release" or "just imagine what AI can do in 10 years from now".
I do not believe that AI in its current form will exceed humans any time soon. Maybe in some near or far future we will invent a new paradigm that will allow us to create human level machine intelligence. But this is as hypothetical as it gets. Such hypothetical scenarios are not worth considering. We can start talking about them once somebody shows a promising paradigm that would actually help us create human level machine intelligence. Such an invention would, in my opinion, create more good than bad. Unfortunately we are nowhere near such a creation.
I am confident about this. I am confident that the current paradigms (statistical learning) won't exceed or even reach human intelligence, because we have seen this before. The promise of inventing AI has been around before. Since Second World War and the invention of modern computing devices, this AI hype has happened at least 2 times. You can read this Wikipedia page to learn about the history of "AI Winter".
I am confident that AI in its current form won't exceed humans, because the technical foundation of the current cutting edge AI tools is nothing new—it is still based on statistics.
There are indications that human brain also uses statistics to think. But I think that we can all agree that there is more to it in the human brain. All current AI tools are based solely on statistics. I think this just won't be enough to create something that is as good as human, let alone something that is better than humans.
After all, why the hell did we think in the first place that some machine, trained on data generated by humans could ever become better than humans?
3. AI will make work done by humans more valuable.
In my opinion, as the AI hype will die out and when people realize its limitations, it will make human artists even more valuable. When junk art becomes available for everybody, human art will become more in demand. Without a doubt, AI art will also find its community and fan base, but it won't affect human artists—if anything, it may make them more valuable.
I think a large portion of people will continue to consume AI generated art. I truly feel sorry for these people as their eyes and souls will be limited to the local maxima of a soulless machine learning algorithm. They will see what they prompted, what they wished to see. I think Art forces the artists to create itself. It is the Art that manifests. The artist, who is the messenger, delivers Art through faculties of a human brain.
Those who mistaken machine generated "art" for real Art will therefore move away from the unpredictability of human ingenuity and venture into the yet early stages of mechanical art. Just as we observe that those who consume only junk food develop heart and artery problems, we will also observe that those who consume only machine-generated art develop problems in their artistic "veins."
Overtime, I believe this will place Art from human hands into a superior position, as it deserves.
What position should artists take against AI, and against intellectual theft that they are exposed to?
If you are following news, you might have heard about the legal battles between artists and AI companies who stole their digital paintings to train their commercial AI products.
I absolutely stand with artists. Their work is being used without their participation, consent or even payment on their behalf in order to train a machine that generates art similar to the art of human artists. The same thing happens with ChatGPT and other chat-based AI models when they looted the entirety of human generated Internet, including copyrighted content that humans created. Without even giving original creators any credit, these large corporations act as if it is within their right to such things. And then when another company uses their AI models to trained the next AI model for a cheaper price, they cry out that their intellectual property has been stolen. What a shit show.
If it is within your power, and if your copyrighted content has been violated, by all means, take a fight against these looters and take your stance in the correct side of the history.
If not, there are still certain things that you can do to "make it right". First of all, for all your work, even if it is for free, i.e. licensed with Creative Common Licenses, do make it compulsory to give attribution to you. An example of such a license is Attribution 4.0 International. If you license your work under this, anybody can use your work, however they have to give credit to you. Now the thing is that, AI models generate text in such an unreliable way, that they fail to give attribute to their sources. This means that AI models, if they used your work, will fail to give attribute to your work. This could make such AI models illegal to be served to public. I am saying "it could", because whether something is illegal or not depends on what the jurisdiction and specific copyright laws say. The important point here is that these matters are still being hotly debated. So if you believe this is illegal, then take your position and defend your point.
That being said, in order to follow proper rules of conduct, one must always give credit for a work if they were not the original creator, even if the license does not require it. If you use somebody's work without giving them credit, you are essentially committing to plagiarism. I only wish if big corporations had the same level of moral code that we individuals have.
Finally, a US court ruled that AI generated content cannot be copyrighted. This means that no "prompt artist" can really claim that an AI generated art is "theirs". However this still does not stop them to put AI generated art on social platforms and try to sell them. What is happening is really ugly and tasteless, to put it mildly. However, it is happening, and the leechers will continue to leech. As we cannot change their nature, we must adapt ourselves, at least our stance against these new developments to protect our value.
So well yeah. Here it is. I guess there is not much we can do, ha... I will just repeat myself here. There is really nothing to be afraid of when it comes to machines creating art. People who has never looked at an art piece or paid for art are suddenly becoming connoisseurs. If you are an artist, these people would never really become part of your network. So in either way, you will still not loose your real network. In fact, you, your work and your network will only get more distinguished.
Summary
I believe Blender is a great tool to create 2D/3D digital art.
You don't really need a good computer to get started with Blender.
You can learn 2D art and 3D art together.
Finally, I don't believe that we should be worried about machines replacing humans in the field of Art. This will remain to be a human practice. Therefore perhaps there are more reasons to try to be artists today. Today it is more important to learn a new skill. Today it is more important to self improve.
So let's learn Blender in 2025.
Footnotes
- a render farm is a collection of many computers that can render a scene very, very quickly↩