poser to DS, C85, final render settings?

Mistara

in PP12, when i'm ready for the final render
set to sync,
Irradiance = 100
min shade rate = 0
pixel samples = 20
raytrace bounce between 10 and 20.

could anyone tell me how to make an equivalent final render settings in DS and/or Carrara 85?

i see the pixel samples x, and pixel samples y in DS render settings.

Thanks. :)

icprncss

What are you rendering that you need to have IR at 100 and that many bounces?

Mistara

overcompensating :) doesn't take as long as a luxrender :lol:

i liked how this came out with the 20 bounces in ps, was enamored for a while.

you haz to use the python script for render settings to go over 12 bounces.

icprncss

I know how to boost the bounces, it just a bit of overkill for the scene. If you want to use Lux Render and Poser, you might want to take a look at Reality 3 for Poser. I think it might still be on sale over at Rendo. It was also 40% off at RDNA.

Or you could look at Luxus for Cararra. You could set up your scene in Poser, take it into Cararra and then use the Luxus plugin to get you into Lux Render.

DustRider

In Carrara it would depend on if you are using indirect lighting or not. But in general to improve render quality in Carrara (for the Photorealistic Renderer) you just increase the values for AntiAliasing, Object Accuracy, and Shadow Accuracy above the default settings shown in the attached image. Checking "Full Raytracing" will also improve the image, but slow the render down.

In general, when using Indirect Light, you just need to increase the Light Quality and Accuracy settings a bit, and turn on (check) Light Through Transparency if your scene uses trans maps. Checking Improved Edges can help if you have a lot of fine details in the image.

For most renders, you don't need to set everything to the max or best settings, the differences are pretty subtle between the best and the value right under it. But, settings can vary greatly depending on what you have in the scene.

Mistara

dustrider said:

In Carrara it would depend on if you are using indirect lighting or not. But in general to improve render quality in Carrara (for the Photorealistic Renderer) you just increase the values for AntiAliasing, Object Accuracy, and Shadow Accuracy above the default settings shown in the attached image. Checking "Full Raytracing" will also improve the image, but slow the render down.

In general, when using Indirect Light, you just need to increase the Light Quality and Accuracy settings a bit, and turn on (check) Light Through Transparency if your scene uses trans maps. Checking Improved Edges can help if you have a lot of fine details in the image.

For most renders, you don't need to set everything to the max or best settings, the differences are pretty subtle between the best and the value right under it. But, settings can vary greatly depending on what you have in the scene.

thanks. :)

Mistara

from what i'm reading it sounds likes the finer DS settings are part of the lights properties. no info on ds render settings yet.

there was some lights info in a mini tutorial with the- D.I.Y. Portrait Lights for DS4

Kerya

For DS render settings read the infos here:
http://www.daz3d.com/forums/discussion/16085/

Edited: about Raytrace Bounces
http://www.keindesign.de/stefan/poser/renderer.html

This one's only important when you use raytracing nodes in materials (reflect and refract). It defines how often a ray can hit an object. E.g. with a value of 0 you have no raytracing at all, 2 will allow reflections, 3 is the minimun to see through an object (the ray has to go through a front- and backside polygon, then hit a polygon behind the transparent object) and with 4 bounces you can see a reflection in an object behind a transparent object.

DAZ_Spooky

Except for very specific circumstances, there is no need to set raytrace bounces above 8. 4 is usually enough. (In any raytraced render engine.) All anything more does is increase render time without providing anything more,

Minimum Shading rate in DS below .5 is usually overkill, for Poser below .1 is usually overkill. (And usually increases your render time for no discernible difference.)

Mistara

DAZ_Spooky said:

Except for very specific circumstances, there is no need to set raytrace bounces above 8. 4 is usually enough. (In any raytraced render engine.) All anything more does is increase render time without providing anything more,

Minimum Shading rate in DS below .5 is usually overkill, for Poser below .1 is usually overkill. (And usually increases your render time for no discernible difference.)

the higher bounces makes the render look 'richer' to my eyes. don't know what other adjective to use.

i'm using hdmi to my monitor, it's a lot brighter than vga. maybe that makes the difference?

after experimenting with luxrender render times :lol:
i'd like to try the best possible render/lighting settings in DS46, CA85P, and PP12 before complaining they inferior to luxrender.

was doing min shade tests with the Old West Mercantile. the stuff on the shelves is all imagemap.

found one way to lessen render time is to put space between widgets with reflective surface, so they don't reflect each other so much.

frank0314

You only need bounce if you are using reflect in the scene. 3-4 is about the max you need to use. If you have mirrors in the scene then I'd say use 4, if its just glass then 3

Mistara

it's interesting, eh?

trying to setup an experiment (hopefully with expert help :) ) to prove / disprove if difference exists between higher bounces.

CypherFOX

Greetings,

DAZ_Spooky said:

Except for very specific circumstances, there is no need to set raytrace bounces above 8. 4 is usually enough. (In any raytraced render engine.) All anything more does is increase render time without providing anything more,

Minimum Shading rate in DS below .5 is usually overkill, for Poser below .1 is usually overkill. (And usually increases your render time for no discernible difference.)

Huh! I find this interesting; I usually render at SR of 0.4 while putting an image together until the final render and then 0.1 for the 'real thing'. It mainly affects the noise in dark or shaded areas.

Here's the difference between 0.1 and 0.4 in a simple bus stop render. In the reasonably lit areas, there's some difference, but I'll admit it's not enough to note. In the darker areas, especially under the seats though, it's very significant.

I can imagine that renders that don't have dark areas, or that make heavy use of backgrounds or DoF would probably find the SR less relevant, though.

-- Morgan

DAZ_Spooky

Cypherfox said:

Greetings,

DAZ_Spooky said:

Except for very specific circumstances, there is no need to set raytrace bounces above 8. 4 is usually enough. (In any raytraced render engine.) All anything more does is increase render time without providing anything more,

Minimum Shading rate in DS below .5 is usually overkill, for Poser below .1 is usually overkill. (And usually increases your render time for no discernible difference.)

Huh! I find this interesting; I usually render at SR of 0.4 while putting an image together until the final render and then 0.1 for the 'real thing'. It mainly affects the noise in dark or shaded areas.

Here's the difference between 0.1 and 0.4 in a simple bus stop render. In the reasonably lit areas, there's some difference, but I'll admit it's not enough to note. In the darker areas, especially under the seats though, it's very significant.

I can imagine that renders that don't have dark areas, or that make heavy use of backgrounds or DoF would probably find the SR less relevant, though.

-- Morgan
I am not sure what lighting you are using. This is Uberenvironment 2 set as AO. Same settings on all three renders, though Minimum Shading rate of .75, .5 and .1 Usually the differences you show can be cleaned up with the lights, without changing the Shading rate. (At a significantly faster render time.)

Mistara

technical term for shade noise = shmootz :lol:

CypherFOX

Greetings,
Interesting; I was using an Advanced Ambient Light. The images are darker than yours, which may be why the 'shmootz' shows up more, especially in that back corner and under the seats. I suppose other variables include the samples the light is set for (both my Ambient and your UE2)...

That said, with those images and that lighting, it definitely doesn't seem to matter what SR you set it to, they're very similar. I guess I just need to spend more time understanding lighting in DS, because it's always mattered for the images I've put together. :-/

-- Morgan

DAZ_Spooky

Cypherfox said:

p.s. What was the number of samples on your UE2? The default for me is horribly shmootzy.

The default is meant to be. The 4XHi Setting is 128 (Default for the AOA light). I usually leave it there or set it to 256 depending on the render.

DAZ_Spooky

One minor point, the lighting, and shaders have much more effect on render quality than the render settings, especially on the higher end of render settings.

CypherFOX

Greetings,

DAZ_Spooky said:

One minor point, the lighting, and shaders have much more effect on render quality than the render settings, especially on the higher end of render settings.

Right, I was just editing my post to include this:

Just keeping track for my own use, bumping from 0.75 to 0.4 took an image from 1m10 to 1m41. (AAL SS32 AOSR4, HM:Prim). Fwiw, in the same image, turning off HD details took it from 4m exactly to 1m10.

Using UE2 w/ Shading Rate of 4 and Render Settings SR of 0.4, image takes 1m16 to render, comparable shmootz. Render Settings SR of 0.75, it takes 1m2.

I see.

Okay, so the argument being made is that the Render Settings SR can be higher (0.5 or even 0.75) if the lighting you're using has a Shading Rate of 4 (or lower), because the lighting's Shading Rate gives the same quality level without as much render time impact as lowering the Render Settings value. You can de-speckle (de-shmootz) using either setting, but you get more 'bang' for your cycles from changing the one in the light.

Okay, that makes sense. If I've got that right.

-- Morgan

DAZ_Spooky

Cypherfox said:

Greetings,

DAZ_Spooky said:

One minor point, the lighting, and shaders have much more effect on render quality than the render settings, especially on the higher end of render settings.

Right, I was just editing my post to include this:

Just keeping track for my own use, bumping from 0.75 to 0.4 took an image from 1m10 to 1m41. (AAL SS32 AOSR4, HM:Prim). Fwiw, in the same image, turning off HD details took it from 4m exactly to 1m10.

Using UE2 w/ Shading Rate of 4 and Render Settings SR of 0.4, image takes 1m16 to render, comparable shmootz. Render Settings SR of 0.75, it takes 1m2.

I see.

Okay, so the argument being made is that the Render Settings SR can be higher (0.5 or even 0.75) if the lighting you're using has a Shading Rate of 4 (or lower), because the lighting's Shading Rate gives the same quality level without as much render time impact as lowering the Render Settings value. You can de-speckle (de-shmootz) using either setting, but you get more 'bang' for your cycles from changing the one in the light.

Okay, that makes sense. If I've got that right.

-- Morgan
Essentially yes.

For both the AOA advanced ambient and UE2, the one that really makes a difference is the number of Occlusion samples. So mess with that one first, though I don't recommend going much above 256 as you hit the point of diminishing returns very quickly. (128 or 256 is usually enough, depending on the render.)

DAZ_Spooky

DAZ_Spooky said:

One minor point, the lighting, and shaders have much more effect on render quality than the render settings, especially on the higher end of render settings.

Just to be clear, the lighting and shaders are the basis of everything in CG.
Your composition can be great, your morph or even sculpting can be great, your textures can be as photorealistic as if you were standing in front of your source with a 38 megapixel camera with the correct lenses taking closeups, but if you have bad lighting and/or bad shaders, regardless of your render settings or what render engine you use, your Render will still look bad.

mjc1016

DAZ_Spooky said:

but if you have bad lighting and/or bad shaders, regardless of your render settings or what render engine you use, your Render will still look bad.

Or improperly configured fantastic shaders...more is not always better.

Mistara

Cypherfox said:

Greetings,

DAZ_Spooky said:

One minor point, the lighting, and shaders have much more effect on render quality than the render settings, especially on the higher end of render settings.

Right, I was just editing my post to include this:

Just keeping track for my own use, bumping from 0.75 to 0.4 took an image from 1m10 to 1m41. (AAL SS32 AOSR4, HM:Prim). Fwiw, in the same image, turning off HD details took it from 4m exactly to 1m10.

Using UE2 w/ Shading Rate of 4 and Render Settings SR of 0.4, image takes 1m16 to render, comparable shmootz. Render Settings SR of 0.75, it takes 1m2.

I see.

Okay, so the argument being made is that the Render Settings SR can be higher (0.5 or even 0.75) if the lighting you're using has a Shading Rate of 4 (or lower), because the lighting's Shading Rate gives the same quality level without as much render time impact as lowering the Render Settings value. You can de-speckle (de-shmootz) using either setting, but you get more 'bang' for your cycles from changing the one in the light.

Okay, that makes sense. If I've got that right.

-- Morgan

yayy for de-shmootz.

one of my render banes is shmootz on the characters cheeks.

DAZ_Spooky

Misty Whisky said:

yayy for de-shmootz.

one of my render banes is shmootz on the characters cheeks.

The "Schmootz" is usually artifacts from not getting enough samples for the light or the shadows. When it is lights it looks like paintballs striking the object.

[digital] Lighting and Rendering by Jeremy Birn is the book you want to really understand this stuff.

Mistara

i actually have that book. ++ points for me. :lol:

back from the days i thought could learn maya5. only thing i learned in m5 was how to super tesselate -the tesselated square for a posercloth sim. i could put it in freebies if anyone wants it. just haz to find the energy to fire up a 32bit compy.

(err, or not - minus points, can't find it.)

Mistara

what draft settings do you use in DS to test if light setup is too bright or not?