Realism like the movies! Help!

A little background - I’m just about to break out of Noobsville, and I’m just about to the city limits :). I’ve got all the basics down, and I know what most things on the menus are and what they do, but I would really appreciate it if you could be as detailed as possible with where things are and what to do with them, just in case!
Okay, here is basically what I’m looking for:
I’m making a fan-made trailer for a sci-fi movie that doesn’t exist, taking my inspiration from a video game. Since it’s already been a game, I would like to make it look even more real, somewhere along the lines and look of the new Star Trek movie. Now, I know I’m not ILM, but is there any way to get close to that look in blender without my computer blowing up? How can I make it look real (or at least really cool) ? Is it all in the texturing, or in the detail of the model? Is texture painting the best way to go, or should I just assign materials and textures to different parts of the ships? (Oh yeah, sorry. It’s the ships that I’m worried the most about) To simplify, here’s my question: What specifically do I need to do to make my renders look really good?

Thank you so much!

Ok, no offense but there really is no way to answer that question. If you want the most photo real renders possible, the only thing you can do is study real photos. I’d suggest sitting down and watching the commentaries for Wall-E and Star Trek, I know one of the Wall-E commentary’s had a very in depth of how they achieved that look. You could also check . Doesn’t tell you how to go about things, but definitely tells you where to start.

Yes, you can get extremely realistic (“movie grade”) renders from Blender. Big Buck Bunny did it a couple years ago now, and Sintel is well on its way to doing it this year. Plenty of folks are achieving very realistic work with Blender.

But, as QS Dragon aptly says, “there is no way to answer your question.” You need to use the Internet to explore, extensively, the topic of realism … in texture, in material, in lighting, indeed in every aspect of image-making.

You very much need to study real photographs. The very first step in creating a realistic CG image is to know how to design a realistic photograph: one that can actually be printed. You need to know about “color management” in Blender 2.5 (and to be using only that version now). You need to know what “gamma” and “tonal range” and “linear workflow” all mean. You need to learn about nodes of all types. You need to know what “color timing” is, and how to do it.

Go look at lots of “shot breakdowns.” Carefully observe how the completed image is constructed in many individual (and individually tweakable) steps. At no point does the finished image come popping out, whole and complete and beautiful like Venus :eek: emerging from a clam-shell on the beach, no matter how many hours you waste on re-renders.

I find that, as I am “sneaking up upon” what I know will be a very fine finished render, it doesn’t really look like much until the very end. There are, I would say, “lots of disassembled parts laid out all over the workshop floor.” But I know how each one will come together in the final composite, and I know what adjustments I can and cannot make along the way. I’m moving systematically from one group of objects to another; from one render-layer to another; moving the overall project closer to what it will finally be. I also know that there will at some point have to be a “stopping point.” This will be the point where I decide that either the project is now “good enough,” or I will risk throwing the whole darned computer out a third-floor window. :wink: You can always tweak one more thing…

All the above commentary and advice is very true and helpful, but I think it pre-supposes more of a knowledge base than you already have.

This is your first post here, so there’s no way to know if you have even attempted any type of rendering yet. I sort of think not, or at least nothing beyond pushing the RENDER button or F12 and seeing what results. I don’t mean this in a derogatory way, just recognizing the possibility that you may not have enough basic experience to move into studying the areas previously described. You seem to be looking for tips on how to skip over a great deal of study and experience with Blender and its capabilities, in search of a Holy Grail formula for instant success in rendering. Ain’t a-gonna happen.

What’s your animation skill base like at present? Beyond any questions of rendering technique, you have to able to animate like a pro to get pro results – there’s no way around that fact. Even if using mocap options, you need a solid base of experience using the animation tools to be able to make any animation work well.

Have you done any study or practice of basic lighting principles? No way you’re going to be able to jump into HDR lighting and other advanced practices if you can’t set up a properly balanced key and fill scene. Do you know how and when to use ambient occlusion? Do you know how best to use the various kinds of Lamp options to achieve your “vision” for an image?

These few considerations are really only the iceberg’s tip.

Don’t try to jump into the deep end of Blender until you’ve fully-conquered the “shallows” – the basic knowledge and experience that comes from starting with the basics and building your skill base up from there. Along the way you might even develop a unique approach to the work that will make it completely your own, instead of just trying to imitate what you see in the movies.

Most of all, be sure to post your work on this forum (and others, if you’re so inclined) for feedback. Honest critiques and comments are an invaluable resource for your growth as an artist (in any medium), and is also the premier source of help with solving those knotty, gnarly probs that always crop up, no matter your skill or experience level.

funny recurrent topic. a 1 post guy comes here and either:

a) asks how to get instant pro results
b) posts a amateurish sketch and pretends to get help to make the next big animation movie

I’m ok with that, sometimes its a start point for talented and ambitious guys, but consider the analogy: a kid comes to a chess forum and says he knows the rules and made a tutorial on how to move the queen. And then asks how can he play like Kasparov, or pretends to have invented a killer opening and nearly demands everybody to join him in a team to play the chess olympiad.

Whats with CG that so many people consider it a cake to master?

People are exposed to high-quality graphics almost every moment of every day of their lives, in these modern times. They see them in television commercials and episodes, in the movies and in games. All of it is au fait accompli, with no indication of the “blood, sweat, and tears” that is the actual process.

It’s really interesting to me, when I look at a really good photograph (or render), just how unobtrusive “the goodness of it” is. I know what to look for, because I’m one of those guys who still loves to mess around with toxic chemicals in dark rooms under funny red light, but the image does not call attention to the characteristics of it which make it appear “truly outstanding.” There is basically nothing obvious in the final image which says to you, “this is why this works.” Instead, it invites you to say, “but of course it must look like this. What’s the big deal?”

And then … you try it … :eek:.

Now, “geek that I am,” my response (once I became aware that the issue existed) was to analyze it. To find the books that take a picture apart piece-by-piece to see what makes it tick. I don’t just go for pretty picture-books of the works of Ansel Adams or O. Winston Link; I sought out the textbooks that they wrote. What did they do in setting up or selecting the shot; what did they do with the camera; what did they do in the darkroom. There are technical reasons why they did these things, and those reasons persist to this day in the world of CG.

BrillianceFilms, “first of all, welcome!! Please post some of your work. It doesn’t matter how simple or how complex it is: just let us see it. A picture is worth …

check this out:

that is a tutorial e-book of doing what you want.

But like said above, don’t jump ahead. you need to figure out the basics first, & if you don’t you’ll just get frustrated & nowhere. i’d suggest starting with this tutorial:

Right from the start. That should help quite a bit!

A forty-seven dollar :ba: tutorial e-book, mind you. But that’s another story for another day, so let’s not let this thread get hijacked.

The place to begin … is at the beginning. Frustrating though it may be, take every modern tutorial you can find and actually do them. Then, having completed one, think of something slightly different and attempt to do it on your own.

“Pardon me, sir. How do you get to Carnegie Hall?”
Practice, man. Practice!”

And? He could take months scrounging around to figure out what he wants or buy a book that teaches him what he wants. Everybody here owns a computer after all. Nothing wrong with spending $$ to learn things.

the devil is the details. Lots of stuff around; no clean or perfectly smooth walls. The cloud texture is your friend.

Sundial, you have a point about not always being able to see what makes or breaks a picture.

As an ex-lighting-tech, I always worked on the idea that my contribution should generally be invisible. It was the greatest compliment when someone once said to me, " I realised it was getting dark, and thought ‘someone should turn a light on in that room’ and then someone came in and turned the light on" They had lost the sense of this being a set, and thought of it as a room with the sun setting outside.

To answer the original question, up to a point; It’s about getting all the details right. What makes something look ‘real’ is that there is nothing in the scene that says otherwise. If you cannot model it realistically, then don’t put it in. PLAN every part of the image/movie. How dark are the shadows, how rough is the metal, where is the light coming from ( light does not just appear from nowhere.) What makes a room look like a room is the reality of what is outside it. Even if you can’t see it, it’s still theoretically there, and your brain can deduce if it doesn’t make sense. I have seen many sets and CG images where the sun is shining in two directions at once, or things don’t throw shadows where they should.

Where is the ‘ship’ coming from, where is it going, why is it flying in a curve. Is that even possible at that speed? How big is the planet, or the sun, and how far. Still images can get away with a lot of fakery, but moving images have to be far cleverer.

Blender will certainly render a totally convincing image, or even movie. It still needs modelling skills, lighting and texturing skills, a good eye for composition, and above all, an intense study of what reality looks like. Study everything you see. Don’t just ‘look’ at it, but look at each part of it. Where does the light fall, are different parts of it discoloured, and if so, why? Does it have dust, fingerprints, grease, etc. on it, and if so, why and where?

Finally, remember that ultimately, it is only what is seen that matters. If you use a photo to create an image, it doesn’t matter, - it’s not cheating. Don’t model details that will never be seen, when a texture would do the same job.

Oh, and don’t try to do everything at once. Draw it on paper, take your time. Throw it away and start again. Have fun learning to do it better. The journey is the fun.



Don’t forget that the username is either something with *studios, *films, *inc and so on.