I hear ya, I just mean that it is easy enough to do GC and saturation adjustments to your render in post if you're going there anyway.

i dunno.  i've found the better my render is to begin with, the better it is after post.  or rather, the closer it is to what i want.  i spend a lot of time painting hair, but it's always better if i have a good base hair to begin with.  as a result i've spent a lot of time on my hair shader.  and the only GC correction i can find in Photoshop isn't as good as just using GC materials, but maybe i don't know the right place to look. 

basically, that's not what i've found.  i'm not saying that i know enough to be definitive, but i've definitely found doing as much as i can in the render to give me results i like better.  even for more illustrated looks.

I'm with you on that. I suppose I'm just being contrary with myself in this case. I actually do not know how to do GC in Photoshop and get the same results I get with shader math, or the lens.

 Coming in extremely late for this.. BagginsBill.. You've been my HERO for a long time, and this just proves why!

I love this, especially the B/W one.. I've been wondering for a LONG time if there was a way of actually making Black & white images in Poser. Not just desaturating them afterwards in Photoshop, but rendering them that way. You've shown me the way!

Ah I KNEW coming back to Poser after a hiatus would be a good idea.. New and cool things are bound to crop up!

Gawd I've missed you guys! 

Quote - > Quote - > Quote - I have a question about lenses... is it possible to simulate a toon look with them so I can just use the lens to make a toon scene?

Not really.

Okay thanks for the reply...  I was hoping there was another way to do toon renders.

I'm wondering (actually thinking aloud here).. what if you combined this lens thing with the Z-toon technique? - the one PhilC came up with years ago, scaling the Z-axis to almost-zero thereby giving a toon effect even in Poser 4?

What if? Give it a try.

Actually I can make a lens that does toon shading. I just can't make it draw outlines the way a human would.

While it should be possible to get the lens to do quantising of colour, there's no way it can be made to do the outlining as that requires an edge blend setup on the object because that's the only way the renderer can find the edge.

Quote - As far as Poser is concerned, seeing the background through refraction is not at all the same thing as directly seeing the background. So those pixels are no longer considered transparent.

If you're going to composite in post, then you don't need the lens at all, do you. I mean, the point of the lens was to avoid postwork.  If you intend to do postwork, particularly compositing, then why bother with the lens? To do a proper job of compositing, you'd want to adjust levels and tone the render against the background. My preference would be to avoid postwork altogether, which means I always render with my background in place in 3D, on a one-sided square or on an environment sphere. If I were of a mind to, I could also render in a picture frame, or signature, or other things.

If you have Poser Pro, then using PPro GC is superior to the lens. When you enable PPro GC, it will also fix your incoming material. Which means you won't get washout, which means you won't need the saturation boost. I suggested using the lens for GC when you do not have Poser Pro and you have to GC in post.

Ah, I was under the impression that the gamma correction results were the same with either method, but doing the saturation boost in Poser with the Lens was better than doing it in Photoshop, but even that isn't necessary. I guess I've been needlessly increasing my render times for a while now. Good to know.

BB -- I'm having some issues with using the Gamma Correct lens (one with the hsv node) that's been putting me off my render-fu. (I'm also at work right now....so images will have to wait...sorry)

I end up with my characters having glowing eyes like they're in some crapy B vampire movie...which would be great if I was rendering crapy vampires....however, I'm not. It's like there's something on the ambient channel when there isn't and it's completely off. (unless I've missed something else on one of the other eye materials on the figures I've used...completely possible) There's no hint of the eye texture, just a washed out glowy look - even when it's more of a closeup shot.

I don't have a lot of options for doing it within P6....unless I do it in post, but I'm rather lazy and the most I want to do after the render is done is stick my signature on it and make thumbnails...

Well, of course I'm guessing since I don't know what you have going.

If you are using Ambient Occlusion on the eyes, it is typical that the AO darkens them considerably. In a normal render, you'd probably boost the Diffuse_Value to compensate, trying to keep the overall brightness, while showing some of the AO effect to give the eye some depth.

When I say "boost" I mean simply that the current value is too high. With GC, even darkened eyes will not be dark anymore. You'd want to un-boost the diffuse reflection in such cases.

If you're not using AO, this could still be the case. It's just a matter of what half-assed workaround was being applied to make the eyes appear to be the correct brightness when viewed on a computer monitor. Such choices turn into over-compensation once the GC lens is applied.

I will experiment some more when I get home and post some pictures (I just made myself post from work because I keep forgetting in the middle of my frustrations with Poser recently....)

It's probably some material setting I haven't even thought of or some little thing I've missed, just that after rendering, futzing, rendering and repeating such for hours I throw the towel in and go read a book. If I render without the GC lens, you can see the eye textures, but then the rest of the scene is rather "blah"...and with it, everything else looks fine, I just have odd glowy eyes.

But will be back later with images so I can hopefully get whatever is going on sorted out and fixed.

The skin tone on the corrected side is great - except for the eyes having gone totally funky and the oddness with the hair prop (and a couple others where there's no hair to cover the forehead show the same thing...forgot about those earlier).

I have one simple color IBL at 20% and a white infinate at 93% (little high I know, but I forgot to turn it down when I rendered) No AO because I generally forget about it...99.9% of the time.

And this is usually where I give up and find a book to read...because I have no idea where to start on what to change as none of my past experiments have seemed to work all that well. Eyes don't like me, hairs don't like me.../sigh

It looks like the iris is ok, but the sclera is messed up.

At first I was thinking there might be a problem with the EyeSurface material, but that covers the whole eye, so I don't think that's it.

I'm curious what the shaders are on the eyewhite. I'm wondering if it somehow mis-uses reflection, and that whatever it's trying to do is being prevented because you only have maybe 1 ray-trace bounce, and that got used up by the lens.

In general the lens is going to mean that any ray-tracing bouncing effect, such as reflection or refraction, is going to need one more bounce than it used to.

Can you show me the node setup for the sclera?

Here's how the vendor set the sclera up....and it makes absolutely no sense to me /blush

Interesting. This person has been reading my postings on how to manage AO strength in a material. There is also a Bias node, something else I posted about. It allows for more direct control of the AO shadow evolution. It is, however, acting as a pass-through at the moment, with the Bias being .5.

Can you show me what is in the Image_Map_4 that goes to transparency - just open image map node previews so i can see the patterns in them. Otherwise that's a total mystery to me as to what is in there. I need to see the image. If it is all white it doesn't matter.

Also, I'd like to see the EyeSurface material, please.

I might ask for cornea and iris later. If it's handy to show those while you're at it, do it.

There's something terribly clever going on here. I'll know better when i see what's in that transparency map.

I'm noticing that there is a bright red circle in the color map. I'm wondering if there is some intentional leaking of that via the use of transparency. In which case, this is another example of a technique for faking SSS that is going haywire. The amount of red is effectively already gamma corrected. By adding the GC lens, we're doubling up on that.

Try disconnecting the Image_Map_4 from all the transparency channels, and set all 3 of them to 0.

Sorry I didn't get back sooner...will do some more fiddling and post the rest of the materials tomorrow evening. (got distracted with other things earlier)

And what happens when, on the sclera, you disconnect nodes from all transparency channels and set those to 0?

By the way, the transmap is white only where the sclera is. Which means it is doing nothing. You can delete it.

You can delete the AO stuff from the iris. It is neither useful nor realistic. Light reaching the iris is collected and bent by the cornea, which makes it pretty much evenly lit no matter what.

If you don't want AO on the sclera, disconnect those too.

I didn't have time to test adjustments for a render...just woke up early enough that I figured I'd get these posted before leaving for work.

So will be a bit, but will see what happens when I get home this evening.

Thanks for helping on this, was really starting to drive me crazy.

Need to fiddle with the transparency settings on the hair prop as well and figure out what's going on there...and why it's only a few of them when I use the lens. =P

bagginsbill can you please show me how to get the efect in your first post, it looks fantastic, and I try and try but can't get not even close that good.
Thank you!

Thank you very much!!!! :-)

i love 'distortion'.....!!!!

-s/.

Because the camera looks straight through the middle of the lens, the distortion there is minimal. As you look towards the edges, it increases because more bending happens there.

Therefore, adjusting the position and size of the lens with respect to the camera will alter the effect. If the lens is closer, the edges are at a more shallow angle, and will distort more.

I have a question for bagginsbill. is there any way to make the lens blur the image? ie the render will come out blurred, and is it possible to control the blur amount????

Hello all,

i hope my question is not off topic, but i tried to understand Gammacorrection all day long and searched the net, but i don`t get what it does... (i even follwed the mathematics in wikipedia but args :D)

I have Poser Pro.
I have a single white light intensity 100% with victoria 4 base.
I render without gammacorrection ticked in rendersettings: the image is a little intensive but ok.
I turn gammacorrection on with value 2.2 -> V4 looks desaturated and a little greenish, more like a goblin and somehow "flatter" .

What does the gammacorrection and why to gammacorrect when it messes the output?

Thanks for reading ;)

Saying that gamma correction messes the output is a little like saying using a hearing aid when you are not hearing impaired messes the sound up. That's correct, using a hearing aid messes sound up IF YOU DON'T NEED A HEARING AID.

But your monitor is not able to render linear colors correctly using linear color values. Furthermore, complaining that a 100% light looks bad when gamma corrected is like saying listening to a stereo that is already loud and also using a hearing aid is too loud.

The only reason you're using a 100% light is because without gamma correction, only brightly lit things look close to correct.

There is a diagram explaining the phenomenon in this thread:

http://www.renderosity.com/mod/forumpro/showthread.php?message_id=3388435

If you're using gamma correction you don't need lights at 100%. Not even close.

didn't you say you were using an infinite at 270% in addition to your IBL  in your outdoor images?

Quote - didn't you say you were using an infinite at 270% in addition to your IBL  in your outdoor images?

Yes, what of it. That was because I was trying to simulate what is called "over exposing" on a camera, by about 1 and 1/2 stops. (Roughly a factor of 2.8 more than a normal exposure)

I was not doing it to force my render into a range of illumination that looked "less wrong". Rather I was doing that because I was simulating a render in broad daylight, which means directional light is about 7 to 8 times brighter than the ambient light. Since my ambient light was on average at a level of about 40%, that lead to a "sun" intensity of 270%, 40% * 7.

This not art - it is physics. I'm just using the information I know about how light works, particularly in photography. When rendering to simulate a photograph, you must create the same conditions. When rendering for arbitrary art reasons, you can do otherwise, but I was doing aviation photorealism on a bright, sunny day.

And, just so you know, if I did not have gamma correction turned on, that would have totally blown out my render. Not only does GC brighten dark things, it also darkens bright things. Bet you didn't know that.

Quote - I have a question for bagginsbill. is there any way to make the lens blur the image? ie the render will come out blurred, and is it possible to control the blur amount????

I did blur earlier in this thread, but it was not a general blur, rather a blur like is produced by a less-than-perfect lens. That means almost no blur in the center, more in the corners.

Did you want something else?

The reason i ask about the blur effect was that i had an idea...
DOF in poser is incredibly slow, as those who have tried it will know, and doing depth masks and moving them to photoshop etc is annoyingly time consuming, so i thought......
what about making several inward facing spheres each with the lens material on them but with a blur effect on each one.... then depending on where objects are positioned (ie between spheres) depends on there blurred effect. and hopefully a faster and cheaper version of DOF....
i did try altering the refract node (plugged into the alternate diffuse)softness value and quality but this has two problems.... 1st is it increases the time of the render which wasn't the point and 2nd is the refract not on the closest sphere to the camera negates the effects of all the spheres further out from the camera.... so is there a fast and more stable way of doing this and is this a good idea????
I hope when poser 8 comes along they will make a fast realtime DOF....

Hello it´s me again ;)
first, thanks for your advice,
I spend some time understanding the diagrams... it stayed with trying to ;)

What i did so far... (and please dont laugh at me, i simply dont know better)
I "tried" to calibrate my monitor to a gamma of 2.2 using greytone testimages on the net, hoping we can now speak from equal prequisites.

i rendered props and voila, on high as well as on low intensities they are, at least as far as i can judge, more realistic with gamma correction.

Now it comes to humans ;)
on some intensities (10%-20%) gc doesnt look froggy, but better. (but its too dark to show much anything.
below 10% intensity here eyes look glowing... erm.
above 20% she looks green and it looks far worse then rendered without gc.

now my question... if i want to make a render with just an infinite light, nothing else, why cant i do a bright image with gamma?

Quote -
Yes, what of it. That was because I was trying to simulate what is called "over exposing" on a camera, by about 1 and 1/2 stops. (Roughly a factor of 2.8 more than a normal exposure)

I was not doing it to force my render into a range of illumination that looked "less wrong". Rather I was doing that because I was simulating a render in broad daylight, which means directional light is about 7 to 8 times brighter than the ambient light. Since my ambient light was on average at a level of about 40%, that lead to a "sun" intensity of 270%, 40% * 7.

This not art - it is physics. I'm just using the information I know about how light works, particularly in photography. When rendering to simulate a photograph, you must create the same conditions. When rendering for arbitrary art reasons, you can do otherwise, but I was doing aviation photorealism on a bright, sunny day.

quote from previous post:

Quote - If you're using gamma correction you don't need lights at 100%. Not even close.

so this is the problem people have when they ask you about gamma correction and lights: they say it's not working properly, because the Poser lights at default are designed to work without gamma correction.  meaning the brightly lit scenes that are most popular here are acheived with one light at about 100% or more at lower intensity.   and then you say they don't need lights at 100% or even close to that.  so when they try it, they don't get at all the results they want. 

the point is that for the bright lighting most people seem to want to acheive, you had to boost your light to more than double 100%.  that's important information for a lot of users, and your statement sounded as if you were saying you don't ever need lights at 100% or even close.  not, you don't need your lights at 100% to achieve realism, which is what you seem to have meant.

i think you're expecting people to not follow you as closely as they need to to be able to move forward.  or to do too much advanced work on their own.  and to think renders that don't look like they want and possibly don't look right to them at in general do look right.  me, i make my lights what i need (or at least as close as i can), but most people aren't willing to spend as much time on lights as i am.

Quote -   And, just so you know, if I did not have gamma correction turned on, that would have totally blown out my render. Not only does GC brighten dark things, it also darkens bright things. Bet you didn't know that.

yes, i did. from months ago, when i was first playing with it.  the non-linear response makes handling the lights a huge PITA at either end of the extremes of lighting, and generally makes it harder to make lights with a  significant (to me) differential.  from my totally unstructured, random test renders in the process of making images, from what i could tell, when a light hit something squarely so it was fully illuminated by that light, it was really hard to get darker than a certain point, requiring almost absurdly tiny differences to approach black.  and it's almost equally hard to get very bright. my first real GC image had one bright light from above and one dim light from the side, and i never could get the side light dark enough for my satisfaction.  iirc, i went below 1%.  conversely, i think the brighter light was over 100%.  i'd really have to double check, and i was using spots with falloff, but the dim light was further away.   i haven't been able to blow a picture out yet, either.  not that i've tried, but there are a few overexposed (i've only known photographers to use the term to mean a photo with a blown out area, so that's how i use it) portraits in my references i'd like to take  swing at, and i know i'll be working with absolutely enormous numbers.  

when i have to type in  either decimal point changes or changes by the tens or even more for every adjustment to work as dark or bright as i might like, i wish the controls worked differently. basically, controls with a linear scale that don't have a linear response are hard to use.  which is why people complain.  imho, it's worth the trouble, but it isn't intuitive or (in some cases) easy to get the response you want.

oh, and it's not just that without GC, Poser looks better with strong ambient lighting.  personally, it was my low-light images that were strongest and most photoreal way back in Poser Pro Pack days.  it's just what people want.  look at fashion magazine covers.  you'll see lots of white and plain light backgrounds and bright lights.  it's a pinup/commercial image look, and it's by far the most popular style of lighting here.

Hi CD. I've been pondering your post for a few days.

With regard to the claim I made about "you don't ever need 100%" I meant "to get your figure lit indoors." Of course if you're doing a scene which is supposed to model the reality of a situation where the dominant light source is 10 times brighter than the ambient light, AND you're simulating a camera exposure value that is adjusted to the ambient light, then of course you'd have a light source that is way over 100%, as I did.

Meanwhile, I've been studying your issue about how hard it is to get dark with GC.

Guess what. I think you're right. I think Poser Pro is f'ed up, and the approach I took to mimic the Pro GC is also f'ed up.

Essentially, the point is this:

GC(2.2) does NOT equal sRGB. OH MY GOD.

It has taken me a few days to work out why and when this really matters. In a nutshell, when you're dealing with very dark things, GC(2.2) is too bright if you're going to display it on an sRGB monitor.

All the stuff I read about GC as the way to achieve accurate colors on a monitor is wrong, and I've been foolishly believing it instead of reading the specs and doing my own experiments.

I've now come to the conclusion that GC(2.2) and sRGB have a lot in common, and for most lighting situations they are enough the same that you can't really tell the difference. But in low light, you can tell the difference, and quite easily.

Kudos to you, you have a f'ing eagle eye. :) I didn't see it until I went looking for it, deeply.

I'll post more on this, soon.

Each row uses a different color space correction scheme.

Top row is linear - no correction.
Middle row is GC(2.2).
Bottom is sRGB, in shaders, built today after I studied the sRGB specification very closely.

The first column is with 128% infinite light in front and a 64% infinite light in back. These lights are rotated 20 degrees around y, and the back one is 20 degrees up in x.

As you go left to right, each column has half the light of the previous column.

So they are 128%, 64%, 32%, 16%, 8%, 4%, 2%, 1%, .5%.

The top row drops off to essentially black way too fast. That's why we want some color space conversion.

For months I've promoted GC(2.2) as the solution. That's the middle row. But look at the last few columns. They're too bright! And also, the second and third are slightly too dark, believe it or not.

The third row is new - you've never seen this before because I've never built it before, and probably nobody else would ever put all the nodes together that are required. Look at the last column - almost black - I think that's how it is supposed to be.

the difference is really small.

I made a few more in various colors.

here it is mroe visible.
 but do we need to now change all shaders? i changed now all my shaders to GC 2,2 :) 

Well I don't know if we need to change or not. In these renders I can tell the difference, but in practice how often does it come up that you're trying to render colors in the RGB 1 to 20 range accurately? To hear cobaltdream, it's all the time. :)

I'm doing some test renders using the sRGB value for skin instead of the GC 2.2.

wow!  yes!  that shows exactly what i was experiencing.  there are times i want a response like the 1 and .5  at the end of your 3rd scale, and i just pretty much couldn't get there.  for those last 3 columns, i don't find what you're showing a small difference at all.  they're exactly the range i'd like to work in if i'm doing a low ambient lighting scene.  that's much better.  woohoo!  i'm so impressed!

edited to add: no, not all the time.  just about 1/2 is what i'd like.  i mean, basically what you're talking about is anything that's nighttime and not in the city.  and personally, i can see a significant difference starting in the fifth column.  which is pretty common for anything not brightly lit.  if you want, i can show some of the works i've favorited that use about  that level of lighting. 

Quote -
With regard to the claim I made about "you don't ever need 100%" I meant "to get your figure lit indoors." Of course if you're doing a scene which is supposed to model the reality of a situation where the dominant light source is 10 times brighter than the ambient light, AND you're simulating a camera exposure value that is adjusted to the ambient light, then of course you'd have a light source that is way over 100%, as I did.

right.  my point wasn't that you were wrong at all.  my point is that you're communicating in a way that makes you repeat yourself and others have problematic results.  you're coming at it technically, knowing the reality and how to match it.  most people are just trying to make pretty pictures of pretty girls.  most Poser users are pinup artists, and pinups are 90% bright, uniform lighting.  so when they say, "the lighting isn't any good."  what they mean is, "the lighting doesn't look like what i want or expect given what i was doing before." 

given that people are still using light sets and don't seem (in general) to be using your IBL generator, the relationship between the environment and lighting isn't as concrete in your works.  i mean, most people like a figure against a sunset to be fully and brightly lit from the front.  totally breaks realism for me, but i've been told several times that it's better that way.  same way i've been told burned in highlights in hair are better. 

if you want to tell people how to have "good" lights, i'd estimate that well over half of the time that translates to bright as daylight in a white room lighting.