Daz 3D is part of 
Connect
DAZ Productions, Inc.
7533 S Center View Ct #4664
West Jordan, UT 84084
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2024 Daz Productions Inc. All Rights Reserved.
Comments
.
Tutorial describing concepts behind various render passes. Maya based, but the concepts should apply.
http://www.3dtotal.com/index_tutorial_detailed.php?id=1507#.VS3uMzpFCUk
Karibou, a frequenter of the Commons, has a tutorial on using Daz Studio to generate various renders that can be useful in compositing.
https://www.youtube.com/watch?v=dnqj6K3_uc0
Here is another interesting tutorial with some interesting tips that could be useful for compositing.
"Regardless of the correction you wish to make, or the effect you’re trying to achieve, your goal is to isolate the portion of the photo you’d like to correct from the rest of the image."
http://photo.net/learn/digital-photography-workflow/advanced-photoshop-tutorials/advanced-masking/
Okay, so let's see...so far it's:
1. Headwax: Too busy
2. Evil: Too busy
No Evil, I wasn't being sarcastic, I was being serious. I REALLY would like to see some of the "seasoned" folks here be bold enough to step out and try to learn some stuff, even if it's new to them and they might not be comfortable doing it. Take it personally if you want, but it's not intended to be anything but encouraging, and maybe to give some folks a little nudge to get them to try it.
Lynda.com usually charges for tutorials, but they have an excerpt that has an introductory discussion of combining images in Photoshop.
https://www.youtube.com/watch?v=MmwrWCMdYqI
Okay, so let's see...so far it's:
1. Headwax: Too busy
2. Evil: Too busy
No Evil, I wasn't being sarcastic, I was being serious. I REALLY would like to see some of the "seasoned" folks here be bold enough to step out and try to learn some stuff, even if it's new to them and they might not be comfortable doing it. Take it personally if you want, but it's not intended to be anything but encouraging, and maybe to give some folks a little nudge to get them to try it.
I think that when one asks you a question in connection with the compositing, and that you don't even take the trouble to answer, it is that you do that for yourself and not for the “seasoned folks”…
HI Joe :)
the fragment passes are, as you rightly say, a way to "break up" the edges which could be badly anti aliased or to help in blending with other layers in compositing with alpha channel or masks, but I think it's important to remember that All of these render passes are ways to help in "Correcting" the final image in some way, whether that's adding Camera depth of field, or to adjust the shininess of materials, or to blend better with other footage.
I'm not trying to say that post work, or render passes are a bad thing,
but it helps to have a good image to begin with.
We all make mistakes,.. so having the ability to correct those is a good thing
but do I really need to render out all possible render passes with every clip i make,..
It's a bit like asking if you need to use all the different "blending modes" in photoshop on a single image.
most of the time you'll use modes like multiply , overlay or soft light, and then adjust the opacity to bring it back closer to the original
It's equally important to understand that you can create Masks and Selections from your image, to help you select parts of an image and adjust those parts, if needed. but it's also important to make the shaders and lighting in your original image as good as possible in the first place.
For example,. to use your image of the Car and the girl again,. (while not being negative) and realising it's just a test image...
the shaders on the Camera and boom seem really flat and dull compared to the high reflections on the Vehicle and even the girl.
You could try to use render passes to improve that,. but I think it's better to spend the time on the shaders and lighting, so you don't have to
Interestingly, with the new breed of render engines like Octane or luxrender is the ability to see the shaders you're building with the correct lighting in your scene, and the renderers ability to replicate real world materials and lighting.
Octane has Multipass layers too, so you still have that ability to adjust things in post., but you have a much better ability to create more lifelike images to begin with.
Anybody that is aware of the history of this forum and the personalities that you have named, and the way you name them, and has even a slight grasp of reading comprehension knows how the above post was meant.
Really, if you want us to come out and play, ask nicely.
Joe, I assume you missed this post? Maybe the last sentence? The one where I said I would keep everybody posted as to what I am doing? Just because it's not what you want the "dream team" to work on, doesn't mean that I'm not going to share what I'm doing.
that;s a good link thanks :)
Cool, so you have some background/experience with fragment passes? I'm curious, where did you use them and for what? Do you have any examples you can provide? It might be instructive to everyone.
Well, since nobody will step forward to help with this tutorial, I’ll get the spoon out again. :)
The next two passes that are provided by Carrara are called the Material Diffuse and Material Specular passes. And, as their names clearly imply, they provide diffuse and specular info for each material in the scene.
The diffuse and specular passes for the scene I’ve been working with are shown below.
The Material Diffuse pass shows you what you entered in the diffuse (aka, “color”) channel of each shader. And the Material Specular tells you what you entered in the specular (aka, “highlight”) channel of each shader.
So, for example, the girl’s dress had only a red color in the shader’s color channel (see image below). So the dress is shown in the Material Diffuse pass as a solid red. Her skin, on the other hand, was a texture map, so that texture map is shown for her skin. No shadows, nothing but a pure diffuse color/texture.
The same applies for the specular (aka, “highlight”) channel. So if you have a white color as the highlight, then that material is shown as a solid white. And if you used, for example, a value slider to define the highlight, then that value is converted into a 0-255 grayscale value, and the surface is shown with that shade of gray.
Pretty straightforward.
Now, can anyone tell me what you might use those channels for ? Shouldn’t take too much time to respond to that :)
3DAge, it seems like you view compositing and render passes as merely a way to correct an image that should have been better in the first place. In fact you said that multiple times. And I'm sure others feel the same way. I've noticed a similar perception here over the years, the view that compositing is somehow "cheating" or fixing something you did wrong.
While fixing stuff is obviously is one use of compositing, I think that hobbyists who don't have experience with the use of compositing in the real world don't fully understand the benefit and the concept.
I thought I had clearly provided examples that show that compositing is FAR more than just fixing images. But if you've never attempted to blend a live action image or sequence with CG elements, you are unaware of how critically important it is. In fact, virtually every film you've seen on TV or in the theater has been heavily composited. And that's not because they didn't render it right in the first place. It's because compositing is absolutely critical to the goal of the film.
So LIVE ACTION INTEGRATION is an extremely common and important reason to consider compositing.
And as I mentioned before, there are features that are unavailable in your rendering software, or not implemented well, and they must be brought in from another source and blended with your 3D work. Carrara has a long list of features that are, in the view of many, poorly implemented. And that's the case for every 3D application. Some are good at some stuff, some aren't. Some have really good smoke and fire sims, some don't. Some do great looking fluids, some don't.
If you are happy with what I call "cheesy" looking results from your 3D app, then this isn't relevant. But if you want the best for your images, you look to ALL available sources and choose the best of the best.
So QUALITY is a major reason to use compositing.
And as I also mentioned before, for professionals working to deliver high quality under a strict deadline, efficiency is a key goal. You need to do things well, but you need to do them fast.
If you are a hobbyist with no deadline, then this might be fairly irrelevant. If you can live with a render that takes 27 hours, rather than finding some other way to get a good, or even better result in 1 hour, then you don't need to consider compositing.
So EFFICIENCY is another major reason to consider compositing.
There are many other reasons, 3DAge. It's not just about doing it right the first time, or fixing mistakes in an image you should have rendered "right" the first time. I just don't want people to get the wrong idea about an extremely useful and important technique.
And 3DAge, one of the most important reasons to use some form of compositing is to make your viewers FEEL something. To tell a story in a certain way, to add something to your images that can't be added by a 3D render, no matter what renderer you use.
For hobbyists who are only making renders for their own enjoyment, and don't have to consider telling a story or making viewers feel certain emotions, then again, this might be irrelevant. But often it's not about "reality" and how well you can mimic it with your renders, nor is it something that a single render can achieve.
Often you need to modify your image with a particular lighting effect, or combination of renders, or real life photos, or an endless list of components that will help to tell your story or make your viewers feel something or transport them to the world you want to send them to.
Again, if you don't care about that, then it might be irrelevant. But in the real world, the goal is to make your viewers feel something, or enjoy a story, or whatever. And often a straight render just can't come close to achieving that, no matter how well you do it.
I gave a very simple example of how you can convert your render into what appears to be a photo of a miniature, using a compositing technique which simulates a "tilt shift" photograph. That's just one of a trillion examples. You might also do a google search for something like "composited images" to get some ideas of how you can transmit various looks and feelings to your audience.
I just did a very quick search for "composited images" and one of the first images to pop up was a good example of how, no matter what you do in your CG renderer, you just won't realistically achieve a certain look unless you bring various elements together.
http://www.msjphotography.com/wp-content/uploads/2013/09/msjphotography_Creating-Convincing-Composites_01.jpg
Whether you like the image or not is irrelevant. The point is that, if the artist had that image in mind and wanted to generate it, doing so in a 3D renderer, while possible I suppose, would only be mimicking the same steps as you would take in a compositing app (ie, adding separate elements on plates in your scene), and would be far more difficult and inefficient to generate.
Everyone please stay on topic and keep posts from becoming personal or we will have to consider closing this thread.
Please don't let 'em close the thread........:bug:
.
24 minute video on introductory compositing multiple images
http://www.photoshop.com/tutorials/7602
By: Julieanne Kost
ABOUT
Somewhere between the decisive moment and moving pictures lies the world of digital compositing - where multiple images captured at different times layer together to create a whole that is greater than the sum of its parts. In this episode Julieanne Kost will demonstrate how to transform concepts and ideas into images by mastering the tools used in compositing.
screengrab tutorial of one person's postwork work-flow
http://www.silverblades-suitcase.com/tutorials/htm/38.html
also has tutorials on other subjects on the left side
here is the Daz documentation center on the subject
http://docs.daz3d.com/doku.php/artzone/pub/tutorials/production/postprod-misc86
Hmmm, some posts deleted since this morning.
You forgot mine…;-P
Joe, it's very clear that your thread is super interesting, but I think that this forum is intended more for exchanges of ideas and questions/answers than for a “One Man Show”.
You should compile all your work and the post it on CarraraCafé, there it would be available at any time.
If you would make a .pdf with chapters, it would be still better.
One could keep this thread for our discussions on this interesting topic.
Hi Joe
One of the most time consuming parts of making 3D animations is a thing called "Motion Blur"
In 3D, motion blur normally requires that several renders or "samples"are created, between each frame.
these samples are full renders which can be Forward or Backward of the current frame.
this can dramatically increase the time it takes to render, since you're rendering perhaps 5 or 6 extra frames, per frame,
These "Sample frames" are then composited (automatically) in your 3D application to create a final output of a frame with Motion blur.
Using a Video editor such as After effects, then you can use the "Fragment" passes as the mask layer or control layer for a Blur effect, and if you use RSMB pro (ReelSmart Motion Blur) which has a "Vector" option which can use a layer as the tracking source, then you can save a lot of time, since the Motion blur effect isn't rendering multiple samples of the image it's just using a layer to apply a complex directional blur effect to ( Saving time is saving money ) and that's a really useful way to use render passes in post production.
RSMB works almost instantly, it's a plugin I'd highly recommend
I hope that helps :)
(on the previous page)
I'm not sure, but it seems like you're saying that render passes are somehow necessary to make a good image or a believable image.
This image was one of the first images I posted here at Daz3D, just learning Carrara and taking part in the New members area of the Platinum Club.
absolutely no post work on this (not that post work is bad or wrong or cheating),
post work can make a good image even better. it's the process of enhancing what you have to work with.
Is it possible to figure out that it's 3D composited, sure it is, but it's pretty well matched into the background plate, and in an animation that would work even better.
Using a Video editor such as After effects, then you can use the "Fragment" passes as the mask layer or control layer for a Blur effect,
Andy,
I'm kind of surprised you're using a Fragment Coverage pass for a motion blur effect. Typically, you use what's called a "Velocity" pass (I believe that's also how Carrara names it) as the primary input to any motion blur function/plugin.
Perhaps you're referring to what's called a fragment jitter effect, which is a different technique where you basically copy and paste copies of an image or portion (ie, "fragment") of an image slightly offset from the original image to simulate kind of a shaking/earthquake/drunken photographer effect. :) :) :)
That's very different from what we're discussing here, which is a Fragment Coverage pass, which, as I showed, gives kind of an outline image to define the antialiasing edges where objects in the image overlap one another.
Maybe I'm misunderstanding what you're using it for. Can you explain?
Would you like me to also get you a cup of coffee while you wait? :) :) :)
Still loving this thread! I wish I could contribute as I am learning across all these fronts. Sometimes it seems overwhelming when to layer, when to go at one pass. I imagine experience will help.
Here is a video from Flipped Normals on improving a 3d rendering with various types of postwork.
http://flippednormals.com/tutorial/improving-3d-renders-photoshop/
Which is what I've been asking for for the last 3 weeks. An exchange of ideas and discussion of issue, not personalities. But apparently you have to ask "nicely" or people won't participate. Or something like that.
Anyway, in spite of the difficulties, I guess I'll continue. Although my patience is getting real thin right now.
I guess one area I haven't covered yet is what Andy mentioned, the motion blur/velocity pass. Maybe I'll throw something together on that. Sounds like there are some misconceptions.
By the way, still nobody who wants to venture a guess about what you might use a Material Diffuse and Material Specular pass for?
Anyone? Anyone? Bueller? Bueller? :) :) :)