Weird questions about rendering options
SnowSultan
Posts: 3,798
Got a couple more questions about some slightly obscure rendering options, hopefully someone can give some suggestions.
* I read in an old thread that Max Path Length at -1 allows unlimited light transport, but that setting it to something like 8 will give similiar results but faster. Is this still true or should we always just keep it at -1 for normal renders?
* Using Spectral Rendering appears to have a huge effect on SSS, completely removing it from the skin in the tests I did. I can bring it back somewhat with large changes to the values, but changing the translucency amount doesn't seem to have any real effect once it's more than 0. Does anyone have any experience or recommended settings for using SSS with these rendering modes?
* Is the Exposure value in Tone Mapping identical to the inverse Exposure value that needs to be applied to an EXR in Photoshop to make it visible (for example, if it's 13.00 in Studio, you enter -13.00 in Exposure when converting the render in Photoshop)? I wasn't able to get identical results doing this, and was wondering if there is an exact value that will carry across so the converted EXR looks just like a standard, non-canvas render.
* I read a post where someone (who is actually very skilled) said they used a Pixel Filter Radius of 1.0-1.2 for distant renders and like 1.8 for close-ups. Doesn't raising the value basically reduce sharpness? Why would you want to do this?
Think that's it for now. ;) Thanks in advance for any information.
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
© 2026 Daz Productions Inc. All Rights Reserved.
Comments
Point one.
I have mine set at 10. What you have to watch for is the eyes of some characters as 10 isn't enough to penetrate all the layers so you have to increase it slightly. The same if you do a room of mirrors. Set at 10 and you only get between 8 and 10 reflections before it goes black, increasing it allows more reflections as they disappear into the distance.
Point two.
I have now changed to having Spectral Rendering switched on for all renders but I haven't paid much attention to its effects on skin and SSS as I usually change settings on most surfaces of everything, not just skin, as I never think they look right
Point three.
I'm not sure what you mean about this one. I open .exr files in PS without adjusting anything except it asks if I want to use a Wide Gammut colour space.
Point four.
I have my Pixel Radius set to Mitchel 1.2 and seldom change it. I do know that, through my Photography, the bigger the image size the bigger the setting needs to be to change the sharpness. As I tend to render the same size of image changing the setting doesn't matter. On the occasions I do a larger size render I would increase the setting but only by 0.1 at a time until it looks sharper. My reasoning behind that, whether right or wrong, is that there are more pixels to work with so increasing the radius slightly allows the sharpening to be less restricted in its range of influence.
As I say, right or wrong, this is how I work
1) 8 can be a little low. I tend to go for 12, but sometimes increase it to 16 if I think the scene needs a lot of reflections/refractions, if it's primarily lit from outside a window (which I think uses up two incidences, for the front and back layers of the glass) or if I just want to play it safe.
Admittedly, the setting is a constant compromise - bear in mind that if you set it to 12, and the light path needed to be 13 long, it'll still need to spend the time calculate to bounce 12, and will then scrap it even though it was almost done. So it doesn't necessarily save much time, although it can still make a difference in interior scenes to stop light paths pointlessly bouncing around the back of shelves and then ending up almost pitch black anyway because of how many reflections it's gone through. (As opposed to the useful light paths that quickly reach a lightbulb or window and thus contribute most of the brightness to the scene).
3) I tend to use a value of -13.6 to get canvases in GIMP (as my version of PS is too old to handle HDR files) to match the default tone mapping in DS.
I believe the common advice is that you're actually supposed to render canvases with tone mapping off, but for me at least that just means that the rendered image is almost completely white as well, so I can't spot any errors or preview the quality while it's rendering.
4) Setting the pixel filter too low at close-ups may cause sawtooth borders to appear. So, the closer the geometry to the camera the higher the pixel filter it needs to be. I usually work at 1.50 for portraits and 1.00-1.20 for everything else.
Thank you all for the detailed answers so far, that helps a lot.
When you say "before it goes black" for light bounces, do you mean before the reflections stop being rendered, or do they actually render as black areas?
I'll try -13.6 for canvases, Matt, thank you. Also, good idea about turning tone mapping off; I usually just use Iray preview on the scene in the main window anyway before starting the render, so that would probably work for me.
Thanks Rafmer, very concise and those settings sound good to me too.
It's a matter of when the path stops updating, because the length has hit the limit or because it's "run out of energy", the current cumulative value is then passed back up the chain.