Yeah, looks like it should be something like that. Assuming that gets fixed, it's looking very good!

Whichway

The shape of the artifact changes, but not the color.

It could also be great to have a GI cache like in Vue. I mean you run the IDL calculation, and it's stored somewhere. So the next render will use that IDL pass. And you re-calculate it if you changed something in your scene.
How does it sound?

You can see the pattern of reflected green change with and without the triangle.

The glow effect is beautiful - so smooth and renders really fast. Once this problem with rays in the wrong direction is fixed, this is really going to be excellent.

Quote - It could also be great to have a GI cache like in Vue. I mean you run the IDL calculation, and it's stored somewhere. So the next render will use that IDL pass. And you re-calculate it if you changed something in your scene.
How does it sound?

Perfect, if you're only going to move the camera. If you move or pose anything else, the IDL needs updated. If you significantly change the color of anything, the IDL needs updated. But it would help with a portrait where you're just making small adjustments to speculars or something.

did stefan add a checkbox for indirect specular yet? I ask because, with a similar scene in carrara, I get artifacts on the front of the sphere with low ISL/IDL settings.  not the same artifacts as those, however.

My Poser drive just croaked.  I have Poser 7 and 8 on a dedicated drive, with all my runtimes.  I was in the middle of using the XL library python script when Poser 8 suddenly crashed .  So I restarted my computer, because it wanted to use Poser 6, which is on another drive, to open the file.  I think the drive fried right in the middle,. My compy took ten or fifteen minutes to boot.  You know the Windows screen with the little scanning thing?  It started creeeeeeeeeping along.  Finally sped up again and booted.  No Poser 8 drive.  Feh.  Gonna go to Smith Micro and make a bug report.

Quote - did stefan add a checkbox for indirect specular yet? I ask because, with a similar scene in carrara, I get artifacts on the front of the sphere with low ISL/IDL settings.  not the same artifacts as those, however.

There were and are no plans to make Indirect Diffuse Lighting also do specular. It's a simple problem and he's handling it. I shouldn't have been posting with beta software. My apologies to SM.

I got too excited by how much better it is now. So fast, so smooth. It's like a different program.

[EDIT] Just got notice it's fixed already.

Quote - My Poser drive just croaked.  I have Poser 7 and 8 on a dedicated drive, with all my runtimes.  I was in the middle of using the XL library python script when Poser 8 suddenly crashed .  So I restarted my computer, because it wanted to use Poser 6, which is on another drive, to open the file.  I think the drive fried right in the middle,. My compy took ten or fifteen minutes to boot.  You know the Windows screen with the little scanning thing?  It started creeeeeeeeeping along.  Finally sped up again and booted.  No Poser 8 drive.  Feh.  Gonna go to Smith Micro and make a bug report.

You had a hardware failure, and you're going to report a bug to SM? I'm confused.

Quote - I got too excited by how much better it is now. So fast, so smooth. It's like a different program.

[EDIT] Just got notice it's fixed already.

Wheeee! Go Stefan!

Whichway

I was under the impression that software problems can sometimes cause hardware failures.  If I'm wrong, sorry. 

Hardware can cause software problems but not usually the other way round

Quote - I was under the impression that software problems can sometimes cause hardware failures.  If I'm wrong, sorry. 

Badly written device drivers can put hardware into states that can cause damage.

But an application can't do that sort of thing. There's absolutely no way for Poser or any application to cause a hard drive failure. It just asks the operating system to read and write things.

GI is hard to do. but i see that you guys are really trying hard. respect. respect because you work will not be noticed by more then 50% poser users.

i thinkposer 8 will work perfect with SP2 or 3.

Yeah, bring on SR1, I'm tearing my hair out with these damned dirty splotches and weird effects where one object touches another!

Quote - Also Ambient and Translucent not allowed on Materials.

Whichway

What do you replace Translucent with? (I know that you can just get rid of ambient.)

It would take days to explain, but translucent only works as a translucent thing on a single sheet of polygons. I have a huge post on this somewhere at the Node Cult.

For all other situations, Translucent is ambient. It's the same as just adding some color regardless of lighting. Same as connecting to Ambient_Color, Diffuse_Color, Specular_Color, Reflection_Color, Refraction_Color - these are all just adders with no lighting model built in. If you plug something in there that doesn't include a lighting model, then it just glows.

Ooo, ooo, conspiracy theory time!

"SP1" = SePtember 1

Ohboyohboyohboy.

Whichway

Quote - It would take days to explain, but translucent only works as a translucent thing on a single sheet of polygons. I have a huge post on this somewhere at the Node Cult.

For all other situations, Translucent is ambient. It's the same as just adding some color regardless of lighting. Same as connecting to Ambient_Color, Diffuse_Color, Specular_Color, Reflection_Color, Refraction_Color - these are all just adders with no lighting model built in. If you plug something in there that doesn't include a lighting model, then it just glows.

OK. So if I have a material which is supposed to be translucent, I should rely on either transparency or fresnel (if it is transparent enough to make a difference) with the color coming from the diffuse node.

it will be difficult for a surface of 0 thickness to scatter light in a realistic facsimile of translucence IMVHO. to me it's more volumetric fx, like smoke or cloudy water.

Quote - It would take days to explain, but translucent only works as a translucent thing on a single sheet of polygons. I have a huge post on this somewhere at the Node Cult.

Tracking down the translucence post and being distracted by all the interesting side issues, of course. But there seem to be an awful lot of missing images over there, despite the attachment fix news postings. Are they really gone or am I doing something wrong again? E.g., [

http://www.runtimedna.com/forum/showthread.php?t=26869&highlight=translucence](http://www.runtimedna.com/forum/showthread.php?t=26869&highlight=translucence)

Whichway
 

*"it will be difficult for a surface of 0 thickness to scatter light in a realistic facsimile of translucence IMVHO. to me it's more volumetric fx, like smoke or cloudy water."

Wouldn't volumetrics used for translucence be like treating a mesh like an atmospheric container (one atmosphere on the inside, and another atmosphere on the outside)? Would that be possible? We are talking about FireFly.

Some objects are uniformly translucent. Others, like skin, are translucent to only a certain depth (where the characteristics change), and characteristics of the surface of the skin affect the translucence. However, volumetrics and atmospherics have a wide variability of possible attributes.

How might such a thing be implemented in Poser?

LMK

Translucence post is here, btw, but no pictures. A little hard to follow that way.

http://www.runtimedna.com/forum/showthread.php?t=26024&highlight=translucence&page=2

Whichway

Quote - Translucence post is here, btw, but no pictures. A little hard to follow that way.

http://www.runtimedna.com/forum/showthread.php?t=26024&highlight=translucence&page=2

Whichway

Weird, 'cause i can see pictures...

I see the pictures, too.

Whichway, excellent thread archeaology there, my friend. You found it fast.

That thread is one of the most sophisticated "scientific" experiments I've done with Poser.

There are certain individuals who, ahem, have argued that she understand what this channel does.  But if you really read all that I posted there, you'll come to see that unless you are really really thorough in testing all sorts of combinations, you will come to wrong conclusions about what that channel does.

Even I, now 2 years later, have trouble following my demonstration in that thread. You cannot just skim over it. You have to read every line I wrote, and study every object in the picture, and wrap your head around what you should be seeing, versus what you do see.

If that's too much effort (and I gotta say, I can't blame anybody for going, huuuhhh???) I'd just follow the wrap up I gave which was this:

Conclusion - the translucence either acts ordinary or does something that isn't very useful. In addition, there are situations where it doesn't even do its intended behavior for a completely meaningless reason. There should be no distinction between using two identical but separate Diffuse nodes versus using the same node for two channels.

I say stay away from it.

Quote - *"it will be difficult for a surface of 0 thickness to scatter light in a realistic facsimile of translucence IMVHO. to me it's more volumetric fx, like smoke or cloudy water."

Wouldn't volumetrics used for translucence be like treating a mesh like an atmospheric container (one atmosphere on the inside, and another atmosphere on the outside)? Would that be possible? We are talking about FireFly.

Some objects are uniformly translucent. Others, like skin, are translucent to only a certain depth (where the characteristics change), and characteristics of the surface of the skin affect the translucence. However, volumetrics and atmospherics have a wide variability of possible attributes.

How might such a thing be implemented in Poser?

LMK

You guys are over thinking this, and letting the full gamut of meanings this word has overwhelm you.

The job of the Translucence channel is simple: Imagine a piece of paper, lit from behind. Hold your hand behind the paper. You can see your hand's shadow clearly, but you can't actually see your hand or anything else.

That is the intended Translucence effect. Totally different from Transparency. In Transparency you actually can see your hand behind the paper.

This channel had meaning in the early days of Poser, when every polygon was effectively a double sided polygon. But ever since "Normals_Forward" was introduced, it is very odd and nearly meaningless. For the same reason, the Poser Square and Hi Res Square are useless. They are relics from P4/P5 render engines, and don't work right anymore.

*"You guys are over thinking this, and letting the full gamut of meanings this word has overwhelm you."

LOL. I admit, it isn't exactly as if I have learned to use all of the other nodes to their best advantage, and am looking for another node to fiddle with because I am bored.

In Poser 8, how would one shine a light on the backside of a translucent object, and see the shadow on the other side. Like illuminating frosted glass in a hallway, and seeing a silhouette on the glass in a darkened office?

How would you describe light from the front side of a translucent object that penetrates the surface, and then is reflected back out? Is that subsurface scattering?

LMK

Quote - This channel had meaning in the early days of Poser, when every polygon was effectively a double sided polygon. But ever since "Normals_Forward" was introduced, it is very odd and nearly meaningless.  For the same reason, the Poser Square and Hi Res Square are useless. They are relics from P4/P5 render engines, and don't work right anymore.

What?  What exactly are you saying here?

Something on my machine is screwing up IE8's handling of at least some JFIF files. I will suffer through Firefox for the sake of getting the translucency thread read.

Whichway

I konw this is drifting off topic (again), but there is some chatter on the web that the attachment problem with IE8 may be that some of the images are CMKY while others are RGB. Apparently, IE7 would render both, but IE8 won't render CMKY. On this page:

http://www.runtimedna.com/forum/showthread.php?t=26024&highlight=translucence&page=3

the first post show no image, but the second does for me. It would be great if someone could figure out how to check if they are in fact different color spaces and that is what's causing the problem. Thanks.

Whichway

Don't know, if i drag these two pics in Photoshop, it tells me both are RGB.

For ALL renders :
Smooth polygons was              ON
use displacement maps was ON, ( only used on the brows)
Post filter size                                1
No other Lights are used, just the balls.

Check this test render I just finished. No lights, just the glowing box outside the room.

4 balls, wih true colors at Ambient value 10.
3  Normal Boxes, just plain white.
my Aly in the middle.
Render settings
cast shadows          ON
Raytracing                ON
RT Bounces                2
Irradiance caching   50 (again)
Indirect Light            ON
IDL Quality                  50
Pixel Samples            2
Shading rate            0.2

For ALL renders :
Smooth polygons was              ON
use displacement maps was ON, ( only used on the brows)
Post filter size                                1
No other Lights are used, just the balls.

Whaw BB,
Ok, I'll supend my adventure into the IDL lights till SR1.

And you even get an IDL shadow, something I had not been able to.

Whaw, big improvement.
 

Hey BB, that is an improvement! Can't wait for the SR1!! :D

BB:

I note the reflection on the floor. I don't recall if you had that in the last render of this test I saw. It looks very good to my eyes.

LMK

I did not bother using the D3D settings dialog. I just used the normal render settings dialog.

IC=50
ILQ=30

Render time 1 minute 50 seconds.

[EDIT] The forum showing a reduced-size version is messing this up a bit. Click it to see the real image.

Quote - BB:

I note the reflection on the floor. I don't recall if you had that in the last render of this test I saw. It looks very good to my eyes.

LMK

You're right, my original did not have the wood floor. When I rendered the original setup, it was so perfect (and fast!) that I then decided to give it a bit of a harder problem by adding the wood floor. The floor is 100% procedural with lots of wood grain and procedural boards - one of my wood floor shaders from matmatic, which makes it over 3 years old. I didn't tweak it at all, just loaded it. There was a considerable increase in render time with the large reflective surface, but it was still acceptable, IMO. Of course I'd be even happier if reflection could be optimized some more. But I'm pleased with SR1, no doubt. Perhaps we'll see more goodness in SR2. I'm very impressed that we're getting a very good lift in SR1 in only a month.

Quote - Don't bother doing detailed testing anymore. You're fighting problems that no longer exist. SR1 improvements are enormous. SR1 will be available very soon.

Check this test render I just finished. No lights, just the glowing box outside the room.

is the time longer?

Quote - I'm very impressed that we're getting a very good lift in SR1 in only a month.

Let's see now.... posts about Poser 8 from early buyers started here around the 4th or 5th of last month. Could that mean SR1 is out tomorrow.....?

Quote - > Quote - Don't bother doing detailed testing anymore. You're fighting problems that no longer exist. SR1 improvements are enormous. SR1 will be available very soon.

Check this test render I just finished. No lights, just the glowing box outside the room.

is the time longer?

No - way less, because I was able to reduce the quality parameters a lot. The values above 90 we were using before are usually not necessary with SR1.

Quote - > Quote - > Quote - Don't bother doing detailed testing anymore. You're fighting problems that no longer exist. SR1 improvements are enormous. SR1 will be available very soon.

Check this test render I just finished. No lights, just the glowing box outside the room.

is the time longer?

No - way less, because I was able to reduce the quality parameters a lot. The values above 90 we were using before are usually not necessary with SR1.

IC=25
ILQ=10

Render time  26 seconds.

Time 107 seconds.

As I sit here watching my potential Dystopia scene take 12 hours  to get half way through the IDL precalculation phase I quietly wonder

***When???

• Oh, please, soon. Please. Please?
  **

Whichway

Turnabout

Whichway