Adding to Cart…
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2024 Daz Productions Inc. All Rights Reserved.You currently have no notifications.
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2024 Daz Productions Inc. All Rights Reserved.
Comments
Thanks all for the great info in this thread - I've been lurking and decided to post my last attempt - for one of the PC Club contests.
I've found in general using the delivered iRay skin shaders and not using a spec map on the base layer and bringing down the glossy gives a good base then adding more localized spec with the Top coat. It will take a while.... but it looks like everyone is making great progress.
More tweaking on my general ideas. Got the gloss on the face that I was missing, tuned the rest of the skin.
No specular maps. They are too inconsistent and too limiting. A better mix of Glossy Layer and Top Coat gets a better effect, I think.
I came across a post by Jimbow that suggests using a gamma of 1 in the tone mapping, so I tried a few tests. These are all ISO 100, f/16, lit with an HDRI and an emissive plane to the left of the figure, and using V6 Belle skin with no tweaks.
1 - gamma 2.2, EV 11, and defaults for Burn Highlights (0.25), Crush Blacks (0.2), and Saturation (1.0)
2 - gamma 1, EV 10, Burn 0.05, Crush 0, Saturation 0.9
3 - gamma 2.2, EV 12, Burn 1.0, Crush 0.5, Saturation 1.2
4 - gamma 1, EV 11, Burn 0.25, Crush 0 (default 0.2 looked too dark to me), Saturation 1.0
I tried my best to match the color ranges in 1 and 2, and in 3 and 4, but to me it looks like the ones in each pair with gamma 2.2 are a bit washed out compared to gamma 1.
This one is very good. Congratulations!
The problem with Specular is: there are specular maps and there are Specular Maps, so I would suggest that if you found a good Specular map, stay with them. It makes difference at all.
There are no good reason to change gamma from 2.2 in tone mapping. It is a very long history about why use gamma correction and I am not as proeficient in English language to write about the matter in this post, but may be would like to take a look at http://renderman.pixar.com/view/LinearWorkflow and http://www.vfxwizard.com/tutorials/gamma-correction-for-linear-workflow.html to get the idea.
I hope this help.
8eos8. I know this is not your intent, tho some may take this as that it what everyone should use for tone mapping settings, when your just looking at where the skin textures are at, lol. Blindly throwing contrasts Hue and gamma dials, I tend to be a little more reserved then that, lol. I tend to ask, where is the scale? The common ground, etc.
(Iray settings are a tad different) http://www.daz3d.com/forums/viewreply/697733/
(Metallicity 0.00, diffuse roughness 1.00, Glossy Layered Weight 0.00)
Yes, that is a side effect of Gamma, it lowers the saturation. because everything is either pushed towards white (greater then 1), or it is all shoved towards black (less then 1). Mid tones end up being mixed with white to make them brighter, or simply darkened. 128r,0g,0b may not be as drastic gamma adjusted up to 192r,0g,0b and at the same time 255r, 128g, 128b gamma adjusted up to 255r, 192g, 192b is a drastically different color. (I need to add colors to that cube, don't I, lol)
I will for the time being, assume that the makers of 3delight set the default gamma at 1.0 for a very good reason, and Nvidia set there Iray engine at 2.2 for an equally good reason. So the question is, that gamma offset thing in the surface tab for maps? Is it set for 3delight, or Iray by default?
(EDIT, I think "gamma2.2-f16-ev11_.jpg" looks best)
bohemian3, evilded777, Impressive renders. :-) I spent the day, and need to spend some more time, working on outfits, lol.
3delight mid-tone Lineup, in the works. Some of the outfits cover to much to see what the skin tones are, and I've been asking myself how much is perception, vs what is common practice in media. What is real skin tones, vs what is exaggerated to get our attention.
Is there an Asian skin map that is compatible with either Genesis 2 Male base or Michael 5?
I want a good starting point for G2M, but I'm not really interested in getting stuff that requires specialized UV maps.
May I ask you where you see 3Delight setting the default gamma at 1.0?
Should be an automatic way to do this, as Reality creates a face mask from some texture to protect the eyebrows and lips from the SSS effect. Would be amazing if Studio gets a parameter to make something similar automatically.
May I ask you where you see 3Delight setting the default gamma at 1.0?By clicking the gear, and clicking "Reset", it puts the value back to Default. Daz3d or 3delight, Whomever. It's the 'Default' setting for 3delight in Daz Studio, And it works.
Yes, I intend to do some G2M lineups, I just haven't gotten there yet. Been a tad buisy with Iray.
http://www.daz3d.com/forums/viewreply/774402/
Oh, the 1.0 value is not effectively working as Gamma Correction is Off. In fact until some version 4.x.something this parameter did not exists. I am pretty sure at that time Gamma Correction was always set on and with a 2.2 value even so the parameters were hide. In some release the switch for Gamma Correction and Gamma Value was implemented in the render settings, so at that point you get the option to set it away from the default. It is beyond my knowledge why DAZ don't set the default to Gamma Correction to ON and the value to 2.2, but that is the usual way how they should be set.
Oh, the 1.0 value is not effectively working as Gamma Correction is Off. In fact until some version 4.x.something this parameter did not exists. I am pretty sure at that time Gamma Correction was always set on and with a 2.2 value even so the parameters were hide. In some release the switch for Gamma Correction and Gamma Value was implemented in the render settings, so at that point you get the option to set it away from the default. It is beyond my knowledge why DAZ don't set the default to Gamma Correction to ON and the value to 2.2, but that is the usual way how they should be set.I beg to differ, this is with GC off, and Gamma 1.00 in the render tab. That color scale looks true enough to me.
(EDIT)
Me and Szark (prixat, etc)have already gone rounds on this. They use GC on, and I use GC off in 3delight with Daz Studio. Iray is a different beast, the default is GC 2.2
Have people found SSS all that noticeable in combination with translucency?
Seems, off-hand, that translucency does most of what SSS had previously been 'asked' of...
WIP. Something like this?
No, actually, it doesn't.
I have no idea why the default is set at 1.0 for 3Delight renders. Its incorrect.
Gamma should always be set to 2.2.
This argument goes back a long, long, long time.
Why gamma of 1 is wrong
Oh, the 1.0 value is not effectively working as Gamma Correction is Off. In fact until some version 4.x.something this parameter did not exists. I am pretty sure at that time Gamma Correction was always set on and with a 2.2 value even so the parameters were hide. In some release the switch for Gamma Correction and Gamma Value was implemented in the render settings, so at that point you get the option to set it away from the default. It is beyond my knowledge why DAZ don't set the default to Gamma Correction to ON and the value to 2.2, but that is the usual way how they should be set.I beg to differ, this is with GC off, and Gamma 1.00 in the render tab. That color scale looks true enough to me.
(EDIT)
Me and Szark (prixat, etc)have already gone rounds on this. They use GC on, and I use GC off in 3delight with Daz Studio. Iray is a different beast, the default is GC 2.2
DS used to use gamma 1. Then we gained the ability to set the render gamma, but not to set the input gamma for textures in the standard shaders (so that in effect most colour maps, which had an effective gamma of 2.2, were double corrected). Finally we have both the ability to set render gamma and to set the input gamma for maps - in 3delight turning gamma correction on enables that, as far as I am aware the gamma value of a map is always used in Iray. Gamma sets the mapping used on creating the rendered image, for 3Delight gamma correction on tells the renderer to anti-correct any maps marked as other than gamma 1.
And here in lies a major issue. IF we are not able to decide what is good consistent photo-nutral lighting for our test renders, and such settings for Gamma on or off are unable to produce the same exact color scale in a non-post-processed render with the same exact color-scale map. Then why are we even looking at skin tones then?
Not intending to sound completely confrontational, tho the point is, consistent settings to base our work off of is important.
(Iray settings are a tad different) http://www.daz3d.com/forums/viewreply/697733/
(Metallicity 0.00, diffuse roughness 1.00, Glossy Layered Weight 0.00)
The map is free. You see what it looks like on your screen in that and this thread, that is what it should look like coming out of the render engine.
I don't understand what you are attempting to achieve with this. Perhaps you could explain?
And I pose this question: what color space is this image in? Because that makes the a huge difference in how you set things up to render and how its going to look in the end.
One can argue, from an artistic perspective, however one likes. But from from a scientific perspective, there is a correct way to do things and an incorrect way to do things.
http://en.wikipedia.org/wiki/SRGB
Thus, it is not a bad scale for calibrating light levels in a test room. White should be white, mid gray should be mid gray, etc.
Poking around, I'm finding that of the basic skins I have, only Bree Genesis and Genesis 2 actually has a decent translucency map.
The other skins either don't have SSS or translucency or whatever at all, or they take simpler options. A few of them just slap the diffuse map into translucency. A few others use a grayscale map on a base translucency color (particularly the newer ones making use of OmUberSurface and so on)
This includes:
Genesis: Gabi, Keira, Medolian, Tamesis, Tori, Benjamin, James for M5, Michael 5
G2F: Belle, Ej Rafaga, FWSA Stacia, Victoria 6 Elite (??)
Oh hey, I found another one... Eisa for Vic6. It's a slight effect, but it's there.
Frankly, the most useful thing I would like to see in Iray is simply a bunch of translucency maps.
So... Zarcon, that image has a gamma of 2.2.
Render with it appropriately: with Gamma Correction ON.
That's the point of Gamma Correction: telling the render engine that your images are already gamma corrected or what their gamma is.
With GC on, that image will be treated as if it had its gamma set at 2.2, which it IS.
snipped for clarity
Without a complete understanding of the two different render engines, one is never going to be able to produce something that looks identical. Point of fact, since they are completely different render engines, they are NEVER going to be exact.
How are you measuring?
Please bear in mind, I am a dabbler, a hobbyist; I'm not trying to get into an argument either. I am all for open discussion, testing, sharing knowledge and experience. I just don't think what you are trying to achieve is actually possible. The scientists might be able to do it, but the artists don't really care and the hobbyists tend to fall somewhere between the two.
If you want to work in two different mediums, you need to learn how to do so. Painting with oils is not the same as painting with water colors, and that is what we have here: two different mediums.
Fred Winkler seems to produce masks for all his texture sets (each one I look at has one; Ash, Oliver, Sebastian, etc)... which is not the same as a SSS/Translucency map (its really a SSS map).
I am finding that I much prefer the color produced by adding the diffuse texture to the translucency color. In my experiments with both masks and SSS maps, I highly prefer the result using the diffuse maps.
I'll have to give Dimension Theory's SSS maps a try... I have not done that yet. If you really are in search of those... check out the Interjection products, they have multiple SSS maps for lots of different UVs.
Jayden with Logan texture and Real Short Hair, my skin materials.
Not sure how I feel about this... its quite saturated, maybe a tad too glossy. Artistically, I like it.
I think an unbiased render engine does not need a translucence map to show the effect. The translucence effect is very dependent from the lightning. I posted an image showing translucence working very well without any special map in Victoria 6. To me it seems that is working fine. Don't you think?
Yep, a little bit glossy, but I like it also. Very nice!
It seems to me that some parts of the body might be richer in blood vessels (like the ears and nose) than, say, broad parts elsewhere.
Although I do agree that the effect is pretty slight, and I HAVE used translucence with single colors without much problem. So, point taken.
A test with SSS maps (Interjection maps by Dimension Theory). See captions on the images. I rather like the results with the SSS map... I am surprised I do not see the difference I expect to see between using the diffuse map in the Trans Color and not using it, but I think that might be more dependent on the individual texture. I expect a closeup of the face would show the tinting of the eyebrows.
So, I think I've figured out my Iray hair settings. I'm off to eat now, but if anyone wants I'll post my settings when I get the chance.