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
I was actually looking for some principles and guidance, but what the hey - I'll bite and get the ball rolling :)
I've numbered the images, in case they get mixed up in the download.
1. Is the original render, deliberately overdone to give me some something to work with
2. Is 1, with some adjustment to brightness, hue and saturation
3. Is using the render passes to see what improvement I could make, with very little understanding of the principles and eyeballing it. Moved layers up and down, trying different blending modes and generally messing about.
Don't know whether there is any improvement - there is certainly better detail (see the seats in the rowboat in the last image) and the lighting is better suited to the time of day.
Since doing that, did some research to get some understandable starting point, without having to take a college course - yes, there are such things just on this one subject - or having to buy Jeremy Birn's book, which in my country costs about three times the cost of Carrara!
Anyway, found this - see the very bottom of the page - which will at least give me some sort of understanding:)
http://www.samwirch.com/blog/ambient-occlusion-and-render-passes-maya-2011
Roy, what passes did you render? Also, what software are you using to composite the image and passes?
The full menu, to see what does what :)
Editing done in GIMP, which does read, but not write, PSD.
While I have you attention and seeing as you've beaten me to acquiring C8.1, I'll ask a question that I've done several times, without any response. Has Daz fixed the problem with shadow catcher being visible when using GI?
I know you are averse to GI, but just for me - please :)
Hi Roygee,
I've only ever used 8.5 so I'm not sure what the issue with the shadow catcher being visible is.
Here's a quick test - the scene has an HDRI in the background, a single light, a sphere primitive and a plane with the material set to shadow catcher. Full Raytracing, Sky Light and Indirect Light are all turned on with default settings and I don't see the edges of the shadow catcher plane. Is this what you were looking for?
Mark
I think that was the issue. The shadow catcher would catch the GI. To be honest, I never really tried full GI with the shadow catcher, as I usually try to match my lighting to a photo or video.
I think that was one of my challenges....figure out what the passes are in Carrara, and how to generate a final image from combining the various passes.
I think that with a little thought it should be pretty easy to figure it out. For example, you know you need shadows, and specular, and diffuse, etc....just start out with the most simple passes (no reflections or GI or stuff like that) and figure out how to combine them to get the rendered image.
I'm still trying to understand if its possible to composite 8-bit per channel info successfully (no visible artifacts). OpenEXR allows 16-bits per channel to be transferred from 3d rendering to compositing applications like AE.
Maybe AE and other apps like it can take 8 bits per channel and sample it up (interpolate) to 16-bits per channel so it is a non-issue?
I have done this with still images and Photoshop by applying gaussian blur after changing the image from 8 bits per channel to 16 bits per channel.
looks like it has been fixed - the first pic shows 8.1 with shadow catcher set to "catch GI" and the second with that setting disabled.
Thanks for showing :)
Roy, I re-read your post and I'm not really sure what you're asking. I *think* you're specifically interested in info on the various Carrara render passes, and what you can do with them.
Like I said, that's a huge topic. And I think I lightly covered a bit of that when I discussed the object index, depth, shadow, reflection, etc. passes. And the important concept is this: look at your image, decide what you want to modify, and then figure out how to do it. Are your shadows in need of a little blur cuz they're too sharp? Are your reflections in need of a little blur cuz they're too sharp, and you don't want to spend all of that render time on the internal blurry reflections ? Do you need to focus the viewers' attention on a central character and want to apply some depth of field or fog to the background? Do you need to apply a color correction for any of the reasons I mentioned?
I think the natural tendency here is to focus on the tool and how to manipulate it. But I'd suggest focusing on your image first, how you can modify or improve it, and only then do you start to look at your tool and its capabilities. Guaranteed you can find a way to do it, but first you need to know what "IT" is...
I think I gave a bunch of examples of stuff you might want to change, so I'd suggest starting off with one of those things and explore it a bit. Or if you can think of something else, then post it as a question and we can help you figure how to do it.
I'd suggest starting off simple. Forget the GI and all the fancy stuff. It's too easy to get off on a confusing and complex side issue. Just take each pass, figure out what it looks like and why, and use your imagination to see how you might change things with it.
Also, with regard to the link you posted...
I'd be careful trying to apply what works in Maya or anything else to the passes in Carrara. Chances are they are different. I mentioned the challenge of re-constructing a final render image using the component passes. That process is different in different apps, as are the blending modes you'll need to apply.
Personally, I haven't taken the time to study the Carrara render passes in depth, since I tend to use other software when I'm going to do any serious 3D compositing. Which is one reason I suggested a challenge. I'm not convinced that the Carrara render passes are all working in a useful way. I wouldn't be surprised if some of them are broken, since I seriously doubt many users spend much beta testing time checking them. And some sure look broken, like I mentioned
And Roy, regarding how you might modify your image....
As I've recommended a million times, I'd first search the net for photos to give you some ideas. If you've ever seen the adobe buildings in New Mexico, especially during a gorgeous orange sunset, you know how breathtaking the colors can be. And as overused as the "blue/orange" tinting scheme that I mentioned might be, an adobe sunset is a perfect opportunity to apply something like that. A deep blue sky with a bright orange sunset reflecting off the adobe buildings.
Here's a sample image I found after a 30 second search for "adobe sunset".
Now you have a decent starting point for modifying your image. Even if you don't decide to "copy" the photo, it's a nice learning experience.
As you look at the image, try to decide what it is about the image that affects you. Is it the colors? Is it the fact that you can almost "feel" the texture of the soft and slightly rough adobe walls? Is it the balance of blue and orange that feels good? Is it the composition of the image? Is it the angles and perspective that are interesting, rather than just a straight shot? Do the birds add anything to the image? Maybe you like the clear, deep blue sky, and find that the clouds in your image might be taking away some of the impact.
You can go on and on analyzing what you like or dislike about the image. And all of that gives you ideas for your own image. You don't have to use the ideas, but it can't hurt to think about them.
Carrara tends to like Linear Dodge (Add) and Multiply if you choose the Correct Layer blend mode. Layer masks are particularly useful. I don't use Gimp, so I don't know the interface or if it uses layers.
Carrara hair doesn't show up in a depth pass, but if does show up in a Volume Primitive pass. You get three layers, a color layer, a black and white layer and an empty black layer.
To add hair to the grey depth pass, I ditch the color layer of the hair pass, and use the black and white hair pass multiplied with the depth pass. I adjust the fill until the hair part looks to be a similar shade of grey as the head. Where the two layers overlap will look black. Fortunately in many cases, I can simply draw a rough line around it with the selection tool and either use the paint can tool to paint it the same shade as the rest of the head/hair, or use the fill tool. I usually feather the edges of the selection in case I'm a bit off in eyeballing the color or there are gradations in the area surrounding my selection. Once the hair shape is added to the depth pass, you can use it for any number of things, such as what Joe was talking about earlier. It's also a very good way to add a DOF.
In my screen caps below, you can see the color layer and the black and white layer of the Volume Primitive Pass. I must have forgot to take a screen shot when I multiplied the black and white pass with the depth pass, but you can see the final result.
Now, I wanted to use caustics with the earrings so I rendered a slightly lower quality, and slightly smaller resolution image as a test. I rendered a caustics pass to see what I could do with that. As it turned out, I felt the effect was to strong, so for my final render, I turned it off.
After the render was complete, I opened the test render and copied the caustics layers and pasted them into the final render. I had to upscale the size, and added a slight blur to hide the slight jaggies from up sizing the render passes. I had two layers generated from that pass. The color layer and a grayscale layer. The color was set to Linear Dodge (Add) and the gray was set to Multiply. You can see I get a nicer, less extreme look.
What I was looking for are the general principles, such as where in the stack to place the various layers and the blending modes to use. Carrara may not have the same quality, or use the same naming conventions as Maya, but the guide on how to do the compositing in PS gives me a good point to start.
EP, thanks for that very detailed explanation. From what I can make out, GIMP works exactly the same as PS, can use PS brushes and there is even a skin to make it look like PS. With plugins, you get literally hundreds of filters. Your explanation is of great value to me in getting a starting point for my own explorations into this fascinating subject.
Great render, btw :)
Oh, I found that by combining my pics 2 and 3 above, I can get the richness of the colour of the sunset on the walls of the cottages, etc., in 1 and 2, get rid of the overdone clouds and keep the fine detail in 3.
Before I hit the sack....
You can use a depth pass for lots of stuff. Sometimes Carrara saves the pass as a separate file and sometimes it is in the same file as the final render. For PS, it is placed in the Channels tab.
My workflow is to copy the background image and paste a couple layers of it. The reason being, if I do something stupid or that I don't like, and for whatever reason can't undo the problem, then I at least have the original version untouched.
So, in this case, I wanted to add a bit more atmosphere to the scene, and introduce a DOF. It's supposed to be a misty morning and I wanted to sell that a bit more.
What I did was to make a couple copies of the background. I created a Layer Mask in the top layer, then I selected the Channels tab, selected the Distance (depth pass) layer and copied it. I pasted it in the Layers tab as a new layer (in case I needed it later- I didn't), and I option-clicked the mask square in the Layer Mask layer, and pasted the depth pass into it. I then inverted the image, otherwise, the layer I'm going to use for misty air would be stronger in the foreground and not in the background.
After I pasted and inverted the depth pass in the Layer Mask, I created a layer below it and set my brush colors to a bright orange/white, and bright blue white, then used the cloud filter to generate a cloudy pattern. It was to strong, so I adjusted the fill.
The next post I will cover the DOF.
Actually I need to add a couple more pictures for the atmosphere effect.
As I mentioned a couple posts back, the effect was a bit strong so I lowered the fill on the cloud layer to around 55%.
Once I had the look I wanted in the atmosphere, I merged the visible layers. I copied and pasted the merged layer so that I had an extra layer. Again, in case I screwed something up. I selected the top layer and chose the Lens Blur filter. It had automagically had the distance pass (depth) selected as the source. I then used the cross hairs to select the area that would be in focus. In my example, I set the blur diameter stupidly large to illustrate the effect. I ended up settling on a radius of 8 pixels.
Not a problem. Sometimes I think that people that do this for years tend to get in a habit of describing what they do in a sort of shorthand. I catch myself doing that at work occasionally when I train someone.
Thanks for the compliment. That's a procedural shader for the skin by the way.
I should also add that the skin is procedural in my last render. I have been converting the skin to work on Genesis 2. The eyelashes are an image map unfortunately, and the eyebrows are dynamic hair. When I have it done, I'll stick it up on ShareCG, just like the one I did for V4.
Absolutely true.
But on the other hand (and it's something I've never quite understood), is this...folks who have very little background in a subject can ask what, to those experienced people, is an extremely ambiguous and open ended question, but their peers with little background can somehow understand what they're asking and focus in on the answer right away.
It's like "hey, I've got this thingy and I need to make it better, how do I do it?"....and guaranteed, someone will step up and say, "oh, yeah, I know what you mean. Just download this plugin and push this button and it will do exactly what you need".
And the answer is generally "Wow, exactly what I need. Thanks." And I'm generally sitting there scratching my head thinking "huh? Is that what he was talking about??"
I *thought* I had described in great length with great clarity some of the ways in which you can do this stuff, but as usual I seem to miss the mark. I think it's usually the step by step detail with pictures that somehow is what people want or need. I keep thinking people can take general concepts and convert them to that detail on their own.
Frustrating...especially after spending so much of my time trying to provide good information. Oh well, such is life.
A picture is worth a thousand words.
And a picture with an arrow drawn on saying "click this" ... priceless
;)
I'm of agreement, especially for anyone not proficient in English, too much literature is tiring in the long run.
Excellent thread all the same, the link is in my favorites of Firefox !
Thank you Joe, like EP and also Roy, which raises interesting questions !
Absolutely true.
But on the other hand (and it's something I've never quite understood), is this...folks who have very little background in a subject can ask what, to those experienced people, is an extremely ambiguous and open ended question, but their peers with little background can somehow understand what they're asking and focus in on the answer right away.
It's like "hey, I've got this thingy and I need to make it better, how do I do it?"....and guaranteed, someone will step up and say, "oh, yeah, I know what you mean. Just download this plugin and push this button and it will do exactly what you need".
And the answer is generally "Wow, exactly what I need. Thanks." And I'm generally sitting there scratching my head thinking "huh? Is that what he was talking about??"
I *thought* I had described in great length with great clarity some of the ways in which you can do this stuff, but as usual I seem to miss the mark. I think it's usually the step by step detail with pictures that somehow is what people want or need. I keep thinking people can take general concepts and convert them to that detail on their own.
Frustrating...especially after spending so much of my time trying to provide good information. Oh well, such is life.
Don't get frustrated.
You did a fine job explaining compositing, what it can be useful for and the concepts behind it. You even touched on the some of the general mechanics of it. The thing about concepts is that they can be somewhat academic until you actually get in and see the step-by-step application in a project. I think that's why video tutorials are so popular, especially when dealing with subjects that have their own complex language. Intellectually you may grasp what is being described, but maybe the jargon is a little hazy, or maybe you don't know where to find something or how to take point A and transform it into point C.
Many folks here also prefer written tutorials that they can go back and refer to, but it helps if there is an illustration or description of not just the depth pass and the results of what you can do with it, but some of the intermediate steps as well, you also see those steps in video tutorials. They are the steps and things that experienced folks do automatically without thinking about it because it's second nature. It's like asking a fish to describe how to swim- they just do it. That's what I meant about a certain shorthand when writing or describing something that's become second nature.
There are also different ways that people learn and perceive the world. There are those that are very analytical about the learning process, and then there are those that tend to "feel" their way through the process, and then there are people in between. I think it has to do with how the brain is wired. I tend to lean towards the less analytical side of things.
For instance, I'll follow a written or video tutorial, and if there are included project files, I tend to ignore those and use my own. My reason is that there is something I want to learn because I bumped up against the edge of my knowledge trying to do a certain thing, so naturally, I want to learn how to do it with my project, not someone else's because I have no vested interest. I may refer back to the project files if I feel I missed a step or failed to grasp a key bit of information. For instance, I followed a tutorial for rigging and weight painting a character way back in Carrara 5, because I had an idea for a video with an animated T-Rex. The tutorial used a human figure, but I wasn't interested in that, so I followed along with the tutorial, but rigged the T-Rex instead of using the project files.
Sorry for rambling on about it.
Thanks for starting this thread about compositing. No matter how folks learn, it is valuable information and insight.
Ask and ye' shall receive! ;-)
And more receiving! Oh yeah, yer receiving dem all right- whether youse like it or not! :coolgrin:
I use simple filters, but as Joe mentioned further up-stream, you can use this method and different render passes to isolate different elements in your scene. For instance, I could just as easily have composited a photograph in the back ground and adjusted the layer mask image to blend it in.
I think what's frustrating is that we've entered the age of what I call "spoon feeding". People *expect* to sit back and have someone do all the work to provide detailed, step by step tutorials and pictures and videos and arrows pointing to show you exactly what to do.
My problem is that I grew up in an age where none of that was available (pre-internet), so the prevailing attitude was that you do the investigation yourself. And the side benefit is that you learn better, with a deeper understanding, and it tends to stick with you if you put in the effort. I just hope the skill and desire of taking concepts and figuring out how to implement them isn't becoming a lost art. I keep feeling that the internet age is nothing more than a "I'll sit back while you entertain me" age.
If someone tells you "push this button, then this one...", then yeah, it might do something you want, but you don't necessarily understand what you're doing and why. You just know "how". But the emphasis here has always been more on the "how", and operating software, as opposed to an in-depth understanding of "what" and "why".
Anyway, just the frustrated ramblings of a clueless old guy I suppose... :) :)
Evil,
I just did a brief glance at your images and the procedure, so I may be off on this, but wouldn't it be easier to select individual objects in your scene using the Object Index pass? That way you can, say, separate the character from the background and apply depth or other effects to only the background. Just select the desired objects in the object index channel and those will apply automatically to the corresponding objects in the rendered image. Then just cut it out as a separate layer that you can work on, or invert the selection and cut the background out.
Sorry, no photos or videos for that.... :) :)
It probably would be easier, but I'm self taught with PS, and never claimed to be an expert. I haven't tried every pass, and didn't know really what the benefit was of using some of them until I read some of your posts. Rest assured, I'll get to trying some of the other passes as I get time.
I'm interested in compositing 3D elements into video background plates, such as a 3D element that is supposed to be in the distance looking as if it really was in the distance, with atmosphere in between the camera and the 3D element. I'm sure some of the passes you have mentioned can simplify the process greatly.
Okay, in terms of the basic layer components and passes in Carrara...
I did a very simple scene, no reflection or GI or anything fancy. And in order to generate the final image from the component passes, here's what I got. Carrara gives you layers in the final image (.psd in this case) as shown in the attached image. Basically, all you need is the shadow, diffuse, and specular passes in order to reconstruct the final rendered image.
Just bring up the image in PS, and the layers are already in the necessary blend modes, so you can start learning which modes to use just by clicking thru the layers.
And the answer is, for those who don't want to take the little effort to try my challenge.... :) :)
You use a Multiply blend mode for the shadows, and Linear Dodge for the specular and diffuse.
Really, like I said, all you need to do is, after you select the basic passes in Carrara, load the rendered image into PS and it tells you what to use for blend modes. No need for detailed tutorials.
And now that you have the basics, you can start adding elements, like reflections, etc., to see how those passes are handled.
BTW, it took me all of 10 minutes to do this simple exercise.