Hey, lately I was little more enthusiastic about my answer on Blender’s Subreddit than any sane person should be and since I am pretty sure the guy asking didn’t read it, I’ve decided to make a bold move and publish it here. Just to ensure the time spent on it wasn’t completely wasted.
The question was pretty simple, maybe even the most basic one after “How in the name of Ton Roosendaal am I supposed to remove the default cube?”
And it’s also pretty difficult to answer. I am sure even you, professionals and hardcore 3D art veterans, are asking yourself the same question all the time.
"How to make my renders not to look like garbage?"
I am far from being expert, but I’ve found this topic pretty interesting in past and I study it a bit. And since I wasn’t able to work on my own-self, couldn’t sleep and was forced to use only mobile, I had plenty of time to write…
So after 27 hours, I had a virtual paper with too many virtual words Reddit refused to send at once. My very own piece of pretending I know something, and I eagerly wait to be ripped apart by you, people you actually know your stuff. Of course, only if anyone will have time and patience to read it. But the part of me hopes this can start a discussion that can correct mistakes and further deepen my knowledge.
(As I said, it was answer to question so pretend you are asking it)
Interesting renders are results of many things combined, and modelling, ironically, is almost the least important thing (at least if you’re not rendering with the intention to present your topology, interesting texture painting, normal map etc. - mesh needs to be clean in detailed view). Bad topology or unclean mesh can cause shading issues but if your final goal is just render, you can easily hide these flaws in shadows or hide them from the view of camera.
Even visual art tells the story
Firstly, before anything else, you need to know what you want to tell. Think about the scene, the story behind it.
Should it be mysterious? Scary? Should it look realistic? Cartoonish? Funny and crazy? Idyllic? Naïve? Dreamy? Liminal? And if liminal, should it have more nostalgic feel, scary feel or feel of emptiness?
Even with abstract art you can answer this question so make it a priority every time, even before you start modelling.
Answer this question and then go and study images expressing the intended feel. Try to discover what all images have in common. For each image find what area, object or part of the object works as the leading element - the central point, “main character” everything else is spinning around.
Eg. mystery - man standing in a fog, full body, visible but with the face covered in thick black shadow with minimal fading. What’s his position on the picture? For example: Horizontally in centre, vertically in upper-centre… depending on the image it might mean he is either in the distance or there is something else in front of him (bottom of the image = closest to camera) and main character must be clearly somehow connected to it.
The position of the objects on scene is part of the thinking about a composition, and the composition is in visual storytelling essential since it can easily change the whole meaning.
Be careful, leading element doesn’t always have to be the object with the strongest lighting nor the sharpest one (horror - is the main horror element building tension the family, standing in the spotlight and in the focus of camera, or the scattered and blurred eerie body standing behind them?)
Basically, if it’s not the single object just for the showcase, without any intention to provoke emotions or drama (simple bust or statue for example), try to think about the rendered scene as if it was the story.
It is. Even images are the form of storytelling, visual one.
I’ll use one of my own projects as an example from my practice to show you how my mind worked. hope it makes things clear.
I was once creating the cute, stylised gnome but didn’t want to go for obvious look, so I’d decided I wanted him to be a gangster. In the next hour I was writing his whole background, made him English-like upper class brat who didn’t want a boring life of businessman. He should have inherited the factory, but he went full-criminal, racecourse gangster, instead. But he never had a chance to live and understand the life of poor, those whose ranks he’d decided to join. So, he partly despised his origin, partly felt superior to poor ones. So, he was sitting on the fence and wanted to be both but also neither one of them.
This excessive story was really over the top (I just had so much fun writing it and kind of forgot my Gnomster should have been cute. And I presented this with my image, so it helped others to understand what they were looking at and what to focus on.)
It doesn’t matter if it’s one sentence or detailed story, the output is the same: this time it helped me to create other assets to fill the scene, find main colours and have a general idea where the camera needs to be put.
He should have been British gangster so immediately, films from Guy Richie and the show Peaky Blinders came into my mind. So he got gentleman’s flatcap and golf sweater, and classic English manor-like furniture - art deco and empire style. Every gnome has to wear a beard but this guy was twisted gentleman, he wouldn’t be wearing long face-hair like some common miner or Snow-white slave, so I gave him chic trimmed full beard instead.
But more importantly, I knew thanks to the story, that I need to go for dim, smoky atmosphere to keep the aesthetics of 1920s. Use lights and gradient shading to give the otherwise fancy room the feel of non-aired pub. And the scene needed to be at night.
But dark, shallow and gloomy, almost black and white style of Peaky Blinders was too much, he was cute gnome after all, not PTSD gypsy genius from trailer. So I chose earthy brown, green and blue as the prevailed colours (brown and green for the furniture and walls, light brown for floor as the scene needed light colour for contrast and I wanted lines of the furniture to be clearly visible. Then I used dark blue for night outside and also as a tint for other colours. To make it omnipresent and spread over the whole scene).
Of course, I haven’t set up lights yet so the scene was dark as a grave, but I could see in material preview that colours match and they support each other well.
Let there be light
Then lights. Lighting is maybe the most important aspect of the rendering. Good or bad lighting can save/destroy any scene, good textures or composition, nothing really matters if you fail to set up light correctly.
Standard in visual industry is Three-point lighting - three separate positions of lights, google it - basically you have one main light shining directly to the main object and two support ones to control shading and shadow.
Standard doesn’t mean that it’s the law, feel free to experiment but for your own sake, not before you know what you want to achieve. And not before you studied 3PL at least a bit and understand what is doing with/to the scene and why.
You have to practice so why not to experiment a bit and have fun with it - different intensity, colours and color-mixing, radius, angle or distance from object, different types of lights Blender offers (area, sun, point, spot). Every one of them acts differently so play with it as much as you want and need and definitely find some documentation explaining lighting behaviour for each of them, use differently shaped meshes and switch between open space and closed one with the walls as barriers - try both, sharply edged boundaries like walls in room as well as gradually edged ones (bend plane - just google it).
Anyway, switch between all spaces often and check how shadow behaves in different environments, try to notice in which situation it’s sharp and dark and in which it smoothly fades away, how angle of surface/mesh/light changes the atmosphere of the scene… and so on. You get a picture.
Then return back to images you studied before and this time check lights. Are they 3PL or not? What light is dominant and from what direction it shines? What and where are the support lights if distinguishable? Where do all lights meet each other and blend? And most importantly, how is the lighting helping to achieve the main feel the scene has?
One professional photographer doing 3D art as a hobby told me maybe the most important advice I’ve got in this topic (unfortunately can’t find our conversation so cannot credit it):
3D artists are working with artificial lights, and even naturally looking lighting, like from HDRI background is in fact artificial. 3D modelling software computes it like that so as study source, the software generated light is insufficient (if you don’t know what hdri is, google it, one of the best lighting shortcuts if you need natural lights. Often it solves the light problem easily in few clicks, at least when the advanced, more delicate lighting is not necessary for the scene).
Anyway, he said to me that he often sees digital 3D art, even the one made by professional 3D artists, doesn’t truly use full potential of lighting - the intensive and absorbing feel of depth or intentional flatness good lights can achieve. Even brilliant 3d art pieces allegedly often suffer from the lack of the knowledge (I can’t say why since I am not a photographer not to say professional. And as a mere student only discovering all what this magic wardrobe can offer, I am not skilled enough to recognise it myself. But I have no reason to question this statement. He’s not only professional photographer but 3D artist as well so he has no reason to look down on 3d art. Otherwise, I would expect it since these kind of trifling disputes are common everywhere in every field.)
So to the point…he suggested me to learn about lights from the “art of reality” rather than from 3D art. Closely inspect photos to get a knowledge useful for static images and film scenes/professional videos for animations. And use visual theory sources for photographs and cameramen for learning theory, not only digital art ones.
He was right at least in this; I’ve learned a lot from them, so I recommend you to do the same. (As I said I wouldn’t even dare to question the quality of lighting of professional 3d artists, but it definitely doesn’t hurt to broaden the scope of study. Especially with Blender - open-source app available to anyone = internet is overflown with free tutorials and questionable advises. Mostly they are old or bad, done by people who learned the stuff they teach just a day before and cannot truly understand the magic behind it.)
I suggest you to learn 7 general categories of lights in digital art, or at least to acknowledge they exist. Then you’ll understand that all that blah blah from above about bouncing and scattering have very practical reason and use. So, if you want, look for point and directional light, spotlight, emission, ambient, diffusion and specular light.
Basically, every group represents the way the light can behave in 3D, how it affects scattering and how to combine this groups to get the wanted result.
**Colours and materials **
Also, I strongly recommend checking colour palettes, how the basic colours can be mixed and what colours are created by various mixtures. Also, try to check different colour combinations and decide if they fit together or not. Experiments are welcomed but don’t let yourself to go too much “weee!” Sometimes the kitsch is exactly what you want but let your colours be ugly and unfitting intentionally, not as a side effect of the lack of taste and knowledge. Visual art doesn’t have to be beautiful or pleasing but it’s too obvious when the author is the equivalent of crazy and colour-blind owner of clown circus.
One more thing, a drop of physics, little bit advanced and I am not going this way for the sake of both of us but one thing about light is essential and every visual artist should keep it in mind all the time.
Your perception of light is influenced by the ray of the light source.
When the light in the form of straight line - ray - touches the object, it scatters and part of it bounces towards you. Another part is absorbed by the object while something passes through. So, what you see is not actually the object per se but the reflection of scattered illumination.
How much light is absorbed, bounced or goes through the object depends on the type of material - in Material Properties, you can, well, set up properties of material. Material nodes can easily get out of hands so for now, stick with Principled BSDF (or BDSM for some with peculiar tastes) and learn the most important settings on the node (subsurface, metallic, roughness, transmission and IOR - for glass).
Then play with the different settings and observe how the ray is affected by it. You’ll see how strongly the light depends on material.
Emission (settings in Principled BSDF) is interesting fella – this guy simulates light. Object emits the light, and so becomes itself the light source. In reality it can be used for example for scifi buttons or all those dings and beeps flickering erratically to make the mesh more high-tech.
Apart from the Gnomster in the centrepiece and lamp, there were the chair he was sitting on and the table with the glass of whisky. I wanted drink to be clearly visible so I shamelessly added another light, strongly shining point, into the liquid and changed its colour to honey-like. The strongly-shining distorted light, passing and scattering through the liquid and glass, gave it, I dare to say, delicious look.
Images are everywhere just waiting for discovery
But don’t let them knock you off
This is basically all you need to understand to be able to step up and let yourself be fully enlightened by light. If you wanna get this under your skin, you need to build up a routine and start to think about lighting not only few minutes before rendering but whenever possible. Let yourself stop for a while, look around and try to concentrate at the world around you, track paths of light, all shadows… and even better, remember and learn to distinguish the behaviour of at least some materials. Soon you’ll start to notice that the effects of light, its distortion, shadows, colouring… can be seen everywhere. Because it happens everywhere. And after a while you will see it automatically. Like a drummer who feels the rhythm all of the time and in everything.
Next one is going to be little abstract. It’s time to finally understand what’s up with these emotions. Try to thing about what you feel while looking at some scenes or things. Does it makes you happy, sad, nostalgic, calm, angry? Is is inducing, disgusting…?
What about old and abandoned swing in the former kindergarten park? Singing birds? Gamekeeper trying to shoot singing birds?
Turn your concentration, energy and observations skills into yourself and use them to find what do you feel and why. Try to describe what exactly it might provoke it. Colour combination? Shape? Particular object that reminded you something else? Did it happen because it symbolised something that makes you emotional, or was it something you connect with past? Important person? Once important person?
Whatever it is, always try to answer these questions.
And pay attention to abstract igniters, not to solid objects or living beings (these only in relation to colours and everything mentioned before.)
Yes, spiders or snakes cause fear in many, luggage full of dirty and musty laundry son brought home from the dormitory causes upset stomach and anxiety… but this is the issue of modelling or references.
Your main object in the scene is the product, ask yourself what do you need to do to sell it?
Just think of this hypothetical situation: big marketing company offers you amazingly paid job, but you need to pass the test: to model, compose and texture the scene in such a sexy, luxurious and tempting way it would sell even this dirty and musty laundry son brought home from the dormitory.
Pro league. If you want to know how to get your ideas, messages, stories – art - to others, firstly you need to understand them. You need to actively use your empathy to understand and imagine people’s emotions. Find what makes them feel the way they feel and why. Observe things around you and try to find out what kind of emotions it can evoke – but not in you, in the other people.
Of course, this level of empathy cannot be achieved without knowing the people you focus on. So, feel free if you want to establish one or two leisure time clubs for like-minded people and socially experimenting on them.
But it might be safer to try it with your family, friends or anyone who wouldn’t get you arrest for stalking, wouldn’t injure you for the suspicion you are police informant or wouldn’t think this is your weird and somehow creepy way to start an affair with them (unless you are).
But most importantly, all practices focused on emotions must be done without emotions. Of course, it’s impossible examining own emotional reactions to various stimuli and feel nothing but it’s important to keep your distance and don’t let it go over the head.
You don’t want to be depressed cry-baby every time the “dark shadow meets the forest green, and the cold autumn sun suddenly shines through the dark forest and touches the last ochre autumn leaf with its dying summer fingers…” nah, let this crap for dreamy poets.
You are the student, someone outside the epicentre and with clear mind and analytical view.
But don’t get me wrong, when you actually start working on scene, emotions are welcomed. The art is the domain of crazies and over-sensitives. It serves as an emotional and strongly subjective way to express yourself. But to be able to get your mood expressed, you need to know what to look for, which colour combination you can use, what light you need and what is the reason behind it…
And what is the story this sly emotion is hiding behind.
Now it gets more, much more meta.
(It was after circa 15 hours of mobile-writing so my mind was feeble and slid to philosophy - I was thinking to remove it and keep my honour but then I’ve decided to leave it to keep authenticity.)
The “images and photos”, I’ve told you to study to understand storytelling, lights, colours and so on at the beginning, are actually everywhere around you. Because people, even with all their creativity, can never be truly original. Everything, no matter if it is a photo, digital art, film, drawings or even anime, fact or fiction, all what is created or captured by people is basically just a copy of the world around them. Our imagination is limited by the reality we live in for generations.
However, even though it’s the real world and our inability to punch above our own weight which limit us greatly, humans don’t have their hands tied at all. Because this “real world” is actually objective as well. People are emotional creatures, they can’t see the world objectively no matter how hard they try, therefore even the reality is actually just another of ours subjective fantasies. And fantasy we can dynamically shaped.
“Limits of our creativity are strictly defined and enclosed by the reality we all live in. The same reality every one of us actively reshapes with our individual perception of it.”
If there aren’t two people in the world sharing the same vision of the reality, nothing is really set in stone. Each impulse, thought, experience, fear, prejudice, inspiration… all is constantly changing your own perception as well as your personal limits, so they never feel truly limiting. And even better, you can use them to your advantage – the core of these limits is shared with the rest of the world (or at least by majority). So it should never be too difficult to communicate your vision.
Example: if you imagine tree, it can have various shapes, various types, colours… But even when you decide to create pink-red-orange tree, it will still be easily communicated with other people. Do you want it upside down? Or make a big bullet and call it tree?
Since you know how the tree looks like, probably you have your own reason to change its colour or make it inverted. But you know how tree looks like so default tree in your mind is still the same. And probably, if your intention is not to troll or be silly for a sake of it, you’ve left some hints in the rest of the picture for others to understand. Therefore, very likely the bullet is a metaphor for something logically connected to the meaning and story of the picture. It is not the first thing you come up with when someone says tree.
To describe it without the actual image is really difficult if not impossible but I have them in broken laptop that can’t be even booted. But maybe something clicks, and you’ll understand my intention behind it. Or just skip it.
But back to little gnome. Lights can be (also) divided into two categories – one with real source and one without it. This scene had two actual light sources - lamp and the moon outside. I needed to keep the smoky-pub-office mood intact, so lamp light needed to be dimmed and served only as the support. Logically, only the moon outside of the window left to be the source of main light. I changed the position of the gnome to get him under the window and used blue-toned sun lamp to create strongly lit moon god rays (you know the drill, google it).
Light blue moonlight and dim ochre-yellowish light of the lamp made very nice contrast. And the (scattered light) colour blend had short, distinctive, sharp and aggressive gradient fade.
With a little imagination, it looked like the pivot point of the clash between the “pure and natural” moon outside and “luxurious but dirty and dark” lamp inside.
Then I needed to lighten up the gnome’s face since this part wasn’t well lit. His cap prevented the light from the window to pass through. To solve this, I added the support light, this one without the real source, and then changed its angle. Now the new light subtly shined frontally from the ground up to gnome’s face.
The camera was on the bottom, pointed up to the gnome and giving him the aura of strength and power. But at the same time, I chose to capture him from his left side, with camera significantly distanced from him and the rest of the assets to enhance the feel of loneliness that should have indicated he doesn’t belong to either side and therefore is in fact alone. My primary intention was to communicate with viewer that my gnome is an outcast on both sides but too strong, proud and arrogant to be down because of it.
Window with the main light source was opposite the camera.
Now I could have captured the whole scene.
God rays created natural but conspicuously bordered area of light. Moreover, I kept all scene edges naturally dark and carefully checked that nothing, no ray of light or anything else, interfered with them. This way I achieved the illusion of enclosed scene and focused all the viewer’s attention to lightened area.
And then I needed to set up the FOV.
The end of Gnome scene-babbling
FOV - playing with distance
Field of view or better say angle of view is very important parameter in all 3D graphics, photography and filmmaking.
This is definitely something you should check and make yourself familiar with. But in short, the higher is the figure of millimetres the narrower is the angle.
135mm=12 degrees, so you see only small area but the object is enlarged. In photography, it’s used mainly to capture detailed shoots of small animals or detailed photos of objects in the distance.
On the other hand:
16mm=84 degrees. In this case the angle is larger so is usually used for landscape photos.
Natural angle is similar to the range of humans’ eyes, and its length is ranging from 50-80mm.
As usual, art is art and likes to play with unusual options and looks to come with something unique so people often play with FOV to find a new, interesting way to present their work.
For Gnomster scene, I used slightly lower length (higher angle) to support the feel of loneliness. By this, the view was little distorted and the gnome looked farther away.
Where to place models?
Even the place you choose for objects play a vital role. If not done right and with clear intention in mind, placing can easily damage or destroy your whole scene. Just grabbing the model and place it without thinking to the centre of view is really bad idea. It will be boring, unpleasant for eyes and viewers brains will just ignore it.
While everything in the nature calls for symmetry, it’s actually asymmetry that catches our attention. Even though symmetrical face should be the most beautiful, our brain can find it boring and insignificant. The most attractive people are symmetrical but with some asymmetrical flaws that causes a distortion of uniformity. Our eyes notice this asymmetric anomaly and force brain to pause and focus, think about it and deal with it. This flaw may even annoy us at first, but it provokes emotion and in some cases ultimately leads to an attraction.
This works not only for human faces but for images as well, and that’s when the composition arrives on the scene. Simply explained, composition is the organisation and planned arrangement of objects in the harmonious but also catchy way.
Composition - often ignored but important
Make your render remarkable
Many artists completely ignore composition and think it’s unimportant. Nonsense. This and good lighting are main reasons someone actually stops, looks at your render again and maybe decides to hire you or buy your asset.
Ironically, artists often overlook this discipline but marketers make it their priority. Good composition with other senses stimuli supporting each other simply sells.
Composition is another huge topic I recommend you to study and keep in mind. Symmetry and asymmetry were described a bit but there are so many principles - perspective, depth, leading lines… There are so much to consider that I let it to you to discover it by your own means.
But the easiest starting point to get into composition is to learn following basic techniques: golden ratio or its simplification rule of thirds, then rule of odds and rule of space.
There are much more to know and learn about but this novel is more than long enough. But you should also check different camera angles and distances and learn what of it can be used to provoke various feelings.
Eg. Remember the part of Gnome workflow where I’ve mentioned camera placing:
On the bottom, directing up towards leading character to make him look powerful and mighty.
But with the different composition and and settings, this angle can also make the scene look threatening. All depends only on the story you want to tell.
Bottom-up angle was often used for war propaganda posters. Either for allied soldiers standing firm and on the guard to protect the country or for enemy’s soldiers depicted as a monsters similar to King Kong destroying cities or eating babies.
And some sources to start with:
Arthur Tasquin - This page is one of the best sources for visual theory that I’ve found so far. Tasquin brilliantly explains many topics I’ve mentioned and many more I haven’t. This source is not overly deep and comprehensive learning source for all related and relevant, it is the intro quickly explaining aspects that play a role in scene building. If you actually invest your time and effort into it, I am pretty sure you will have good knowledge base to continue studying. And even if you won’t continue, this should be enough to learn the basics not only for 3d image renders but also for animation, photography, film-making and basically any visual art you can imagine with exception of games where their unique settings need unique solutions:
https://arthurtasquin.com/blog/visualjourney1
Studio Binder - these guys teach film theory and are really great. They offer amazing, professional content for free but if you’re not going to focus on 3d graphics for film industry and not planning to do animations or game cinematics / cutscenes, this might not be what you are looking for. But even there you can find great amount of important information useful even for static renders. Even film scenes need to deal with composition, lighting, depth of fields. This link is intro to camera angles:
[https://youtu.be/wLfZL9PZI9k?si=TJAGeqGH7-bNa74v](https://Studio Binder)
