Daz 3D is part of 
Connect
DAZ Productions, Inc.
224 S 200 W, Suite #250
Salt Lake City, UT 84101
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2024 Daz Productions Inc. All Rights Reserved.Daz 3D is part of
Connect
DAZ Productions, Inc.
224 S 200 W, Suite #250
Salt Lake City, UT 84101
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2024 Daz Productions Inc. All Rights Reserved.
Comments
I'm liking these a lot!
Philw, the settings in Carrara portrait are different. What's your opinion on the subject ?
Personally, after different renders, I tend to agree with Eyos.
Philw, the settings in Carrara portrait are different. What's your opinion on the subject ?
Personally, after different renders, I tend to agree with Eyos.
Glad you are finding the "glow lighting" useful. Regarding the other settings, I'd generally opt for Object Accuracy set to 0.5, as when set to higher values, it can sometimes miss very thin strands such as hair or tree branches, so you get a "sometimes there, sometimes not" look along the length of the strand, which is not good. The filter sharpness is more a personal taste. I have to admit that I tend to have it either at the default (75%) or boost it up to 100%, but I can see that setting it lower will give a softer and perhaps more realistic look to an image. Again image size will play a part, with smaller images requiring higher settings. I'll have to play with this one some more!
I think it depends what you're aiming at. If you want to go photo realistic, you have to consider what kind of photographer and camera you're emulating.
Emulating a high end camera manipulated by a professional photographer in a very controlled environment (ie a studio) will need a crisp render.
OTOH, rendering a picture taken by an amateur with a phone camera in the street will look far better if not crisp.
Another consideration is depth of field. All cameras have a depth of field, even if it is very deep and therefore not very noticeable. Some of the blur in an image may be due to the object being out of the focus plane. A little DoF usually adds to the perceived realism. I just wish it rendered dynamic hair properly!!
Thank you Philw. I've tried it and it proved very efficient. From now, I'll use it for interior scene
That's the kind of test I've been doing
#2 is with a shape light (rectangle), #1 with the glow channel on a single plane, rendered with all default parameters and global illumination.
Except the fact that the shape light picture is more luminous, the difference is subtle, but there with more realism on the glow channel picture. It's easy to set up, physically correct, so it's a winner :-)
It's funny I've missed it. Maybe it's because I'm reluctant to use GI because of render times. I used to use light boxes for interior scenes which have given me good results associated with anything glow lights.
Yes the glow channel one definitely looks better IMHO. I'm tending to use these more and more for artificial lights (ie. not sunlight or skylight).
Ambient lighting - this little devil - was 20%. I've set it now to zero and I'm happy to see that it seems better than before.
(Usually I set it to zero, but its easy to forget all about it).
I'm doing some more experiments, and maybe I'll take the sharpness up a bit.
I don't use Photoshop, but rather Paint-shop Pro, do you know of something equivalent for PSP?I am still very new to PaintShop Pro, but I am really liking it. For animations, you can use Project Dogwaffle Pro: Howler, where you can even key frame in various changes to the effects, or isolate where the come from, say with the camera moving. To assist with a moving camera, it has several tool for motion detection and such as well. I really like Howler. It comes with several photographic filters and a plethora of effects that work in animations!
That's the kind of test I've been doing
#2 is with a shape light (rectangle), #1 with the glow channel on a single plane, rendered with all default parameters and global illumination.
Except the fact that the shape light picture is more luminous, the difference is subtle, but there with more realism on the glow channel picture. It's easy to set up, physically correct, so it's a winner :-)
It's funny I've missed it. Maybe it's because I'm reluctant to use GI because of render times. I used to use light boxes for interior scenes which have given me good results associated with anything glow lights.
The thing that's truly missing from the glow channel light version is any illumination, whatsoever, on the subject in the middle. Personally, I feel that the shape light image wins. It has the bursting glare from over exposure - as one would expect from such a dramatic change in light. Either way you chose... it's still a very nice technique - and I can see where the glow channel version could be the choice to use, simply from ease of use and predictability.
Thanks Dartanbeck. I never tried Howler before, I'll take a look at it.
I have a bit of a thread about Project Dogwaffle and here are Dogwaffle Video Tutorials - Playlists Very entertaining and informative!
Just to help give you an idea of what it's all about.
That's the kind of test I've been doing
#2 is with a shape light (rectangle), #1 with the glow channel on a single plane, rendered with all default parameters and global illumination.
Except the fact that the shape light picture is more luminous, the difference is subtle, but there with more realism on the glow channel picture. It's easy to set up, physically correct, so it's a winner :-)
It's funny I've missed it. Maybe it's because I'm reluctant to use GI because of render times. I used to use light boxes for interior scenes which have given me good results associated with anything glow lights.
The thing that's truly missing from the glow channel light version is any illumination, whatsoever, on the subject in the middle. Personally, I feel that the shape light image wins. It has the bursting glare from over exposure - as one would expect from such a dramatic change in light. Either way you chose... it's still a very nice technique - and I can see where the glow channel version could be the choice to use, simply from ease of use and predictability.
The subject in the middle IS illuminated - otherwise it would be black!
The subject in the middle IS illuminated - otherwise it would be black!
I think it's the missing specular effect that he means, because of the specular effect of the shape light you can better see the depth of the cone at the top, while with the glow only it looks a little more flat. In every other way glow light is superior and more realistic though, but it does have the limitation that it doesn't seem to have much highlight/shininess effect. This could be countered by adding some reflectivity to the shader of the cone, but depending on the settings for blurry reflections that could hit the render times (probably not much in a scene as simple as this though).
I kind of like the highlight/shininess specular cheat though, and want to keep it, even though I believe in using the glowing mesh light as the best lighting option. My favorite solution to this is to use the glow channel of the shader of the 'light' object in tandem with adding an Anything Glows light to the scene which is also attached to the same item. Anything Glows can be set for distance squared falloff, so it can match perfectly, so simply lowering the objects shader glow intensity a little and adding in a little intensity to the anything glows light will give you nice specular effect and still preserve the lighting of the glowing-shader-object-in-GI approach.
One word of caution though, I've found that you should use the Anything Glows light in conjunction with real vertex objects in the scene - a simple primitive plane will give odd effects when paired with Anything Glows light, though I don't know why, but I still have learned to avoid using a primitive plane with Anything Glows..
Interesting approach to combine the two, I'll have to give it a try.
On the subject of glow and such... Check This Out! It's a tutorial from DCG using just one of the many features of Shader Ops 2. A Carrara friend bought that for me, along with Shader Ops 1 and Enhance C and has just launched another super fun learning curve. This one does a great job of showing off just a small inkling of what new functions you get with Enhance C. Amongst my complete collection of Inagoni texture room enhancements... my shader possibilities look pretty different and I can see that I'm going to enjoy learning this stuff.
The funny part is that I was always impressed with how the texture room behaves when it's bare-bones stock out of the box. Considering that I'm feeling rather addicted to collecting these fun mods to the kit, it's safe to say that I'll have all of the shader helper plugins made for Carrara by the end of the year, or nearly thereafter.
The ability to truly expand upon SSS in that first link is just super cool!
Oh and I plan to collect all of Fenric's great new shader utilities as well. Just fun, fun stuff. I guess it really kind of relates directly to this thread, in a way, as all of the amazing plugins available for Carrara now is really an astonishing amount of expanding Carrara's abilities to help you to get your renders the best that they can be. We truly are fortunate to have such passionate plugin creators for Carrara.
Dang it Dart, the pics are too small for me to make out exactly and the first link doesn't seem to go anywhere/is broken. :( I remember DimensionTheory mentioning once that DCG had a plugin that would allow to use a texture map in the SSS channel somehow, not sure which plugin it would be though.
And yeah, I'm totally getting all of Fenric's plugins (the ones that I don't already have) just waiting on his PA sale day :)
I was going to say that the first link is broken but Jon got there first.
I have seen somewhere a technique to use maps with SSS - use a layer list with two almost identical shaders, one with and one without SSS, and then use the map to blend between them. Clever, eh? (I can't take credit for it though).
Sorry about that!
I had inserted some strange anomaly within the "url=", Try it now?
I am fairly certain that it was Shader Ops 2 that Dimension Theory meant, because that's what this tutorial is about, and shows how you can add a gradient to the SSS. My goodness there's a lot here to have fun using! But don't let that, simple fact frighten anyone away! Treat them like the rest of the multitudes of Carrara shader possibilities and discover them and their potential as you need them.
So when you need a new shader, and you're just not sure how to go about it, start going through the process adding what makes sense. If you're just not finding what you need, go to the top level of the shader and change it to Shader Ops or Shader Ops 2... now work your way though and amaze at all of the new glee!
EDIT:
But while you're in the usual, multi-shader mode, you'll have so many more available options with Enhance C, and/or Velouté 2, Primivol, Baker, and Deeper that you get with the Advance Pack, along with it's Replica and SWAP tools.
I understand that Primivol has the power to turn any mesh into a volumetric shader. I really look forward to messing with this one.
I love all of these this shader plugin is they are very powerful and lots of fun to use.
Regards
I just wanted to share this latest image for no better reason than I liked how it turned out! Lighting is totally HDRI (a free one from here http://www.hdrmill.com/Freebies.htm) with the exception of a low powered light only on the hair to add a highlight on the window side. I particularly liked the reflections in the eyes, which is a true reflection of her surroundings. The hair is Dynamic Skye Hair.
And of course it follows the gamma=2.2, no ambient and full indirect lighting guidelines.
Excellent render Phil!
That's an incredible render, Phil! If it wasn't for the fact she's sporting Carrara hair I would have sworn that was an unbiased render.
If you have a good hdri and patience, you can get similar results to an unbiased renderer in Carrara, DT did it once with an interior scene. It looked like something out of Fryrender/Arion or Octane.
I remember, the Kitchen Table scene, near the window. It was incredible, but he cited using blurred reflections and if I recall it took more than 20 hours to render.
I remember, the Kitchen Table scene, near the window. It was incredible, but he cited using blurred reflections and if I recall it took more than 20 hours to render.
Blurred reflections are something that I have previously avoided due to the impact on render times, but with my latest computer I have started experimenting again with them and - while it clearly increases render times - it sometimes is not too bad, and it does add a lot of realism. I think maybe because people are not used to seeing that in a Carrara or other "traditional" render. The thing to really avoid is having two surfaces (or one concave curved surface) that can "see" each other and both with blurred reflections - then render times go up exponentially!
A close-up view of the character (MRL Raven) in my earlier post. This uses a low level of reflection on the skin which I think helps the realism.
NVM
What's NVM?
Phil--
very awesome render.
I think its interesting how folks are always searching for "other" renderers when often with a little work and tweaking one can get surprisingly amazing results right here in Carrara.
smiles.
Rich