Photoshop is extremely advanced in Color Management: it can can handle color profiles at the image-window level so you can open two windows on the same image, and use a monitor profile on one and a printer profile on the other. All these are compensated for the profile which drives your monitor.

So, when you do not have CM implemented at the system level, an image window using sRGB will produce exactly that on your monitor. This, amongst others, will prevent you from baking gamma compensation into your images. But when you save the image and open it with another paint or viewing program, the colors will differ from those in the Photoshop window.
If you have CM implemented, Photoshop will correct for that, and will still produce sRGB on your screen in that window. Again you might experience some slight differences when using sRGB in Photoshop, and having another (say device specific) profile loaded into your system CM. But when you either assign "monitor gamma" to the Photoshop window, or you assign sRGB to your system CM, all differences will be gone.

By the way, from Vista on sRGB is the default profile loaded into the CM, while sRGB is the Photoshop default for the last so many years now.

How does this effect PP2010? Well, it looks a safe guess to me that most textures were created in Photoshop. And as these have hardly any hardware compensation baked in, there should be no need for a GC feature in the rendering software.

(Un)fortunately, there is JPG.

When the first digital cameras created BMP or so for output, sending it to a monitor (or TV or else) in a CM-less environment made it look ugly thanks to the monitors non-linearity. Correcting the image made the hardware compensation to be baked into the image, and it was time-consuming to do all those corrections. For video reproduction on TV, things were even worse as the player had to perform the corrections while streaming to the screen.

Therefore, JPG (and MPG) were introduced, having the sRGB / non-linearity correction embedded by the device creating the image or video. Now, images and video could be dumped right onto the screen while still looking good. This enabled cheap DVD-players for MPG playback, and prevented most photo-enhancing from baking hardware specifics into the image.

While Photoshop takes the embedded correction out of the JPG before displaying and manipulating it, Poser up till now did not, but PP2010 does via the GC-feature. From this point of view, and from the point that rendering does require linear input, it's not really a feature but a fix of the failure to handle the JPG format properly. But okay, it serves more purposes than this one only and one can argue that linear rendering is  the feature at hand, so no hard feelings.

So, while CM on your system or the use of Photoshop (or both) make the PP2010 GC-feature sort of redundant, saving any texture to JPG immediately brings its relevance back again. Hence, for JPG images in the diffuse/color channels, GC should be used at the usual gamma=2.2 setting. By the way, the same holds for PNG although more variations may occur. For other image formats, GC should be used for textures not made in Photoshop on machines without CM enabled (pre-Vista etc). In all other cases, and for the other channels, GC should be off (or; gamma =1).

People creating images on a CM-aware system had to take into account that other people, viewing those image in an online gallery, did not have CM enabled on their box. So, they had to put the monitor/gamma-correction back in again before publishing. Like I apply a printer profile before sending our my images for print.

Fortunately, just publishing the image in JPG solved the issue thanks to the correction embedded in the JPG format. This is what JPG was all about: the issue gets solved before you're aware of it.

As a side effect, looking at JPG's or other formats having monitor compensation baked in, on a CM-aware system, provides double compensation for the same issue. And the more people use Vista or up to look at JPG's in the Rendo gallery, this side effect is becoming commonplace gradually. Something similar happens, when you allow PP2010 to put the non-linearities back into the rendered result, then you save it as JPG, and look at it on your Vista/Win7 (aka CM-aware) system. That's triple correction.

The side effect enables you to see more detail in the dark areas, than presented by a normal photocamera shot. But a camera is not the human eye. Our eyes are adaptive, pupils open up to get more detail when it's dark. And seeing details in the dark is something that eases our mind, and make us consider an image as "good". This is why no-one complains about it. Your images become even more real than photoreal. But be aware of the triple compensation that lures behind a corrected PP2010 output. Things can become too much.

That's all folks.

Quote - If you prefer un-realism, then Poser 5 is more than anybody needs and we would not need to discuss purchase decisions at all.

This is a really interesting thread and helped swing my decision to side-grade to pro with the extra 50% off at CP - $99 didn't seem too bad a deal to me.

However, as one of those who has had little interest in photorealism in the past, I would like to correct BB on this statement. The chief way of getting non-photorealistic results is through the use of filters/styles such as those in Postworkshop or the old Painter plug-in that I used to use. While these offer some control over the result (and that can be a lot of control when you have the node based style creation in PSW or FF) they are still limited by the initial image and their output is particularly dependent upon the lighting.

This is where 3D users have some advantage over photographers with filters, in that we have greater control over all the variables that produce the final image. Being able to set up lighting to best suit the filter you want to use is a major boon, so any new lighting trick that your software can offer is worth having. A good example is in sketch effects I was using a while back - I was getting much better results using IDL than any other lighting set ups. I dare say that if I spent a little longer setting up other lights I might be able to get results close to it but IDL is much quicker to set up (so far, I've not yet needed anything too sophisticated for the sketches) and then leave to render, so why bother? (It is slower to render but there are always other things to do in the mean time.)

Another anticipated boon will be the 64bit rendering. If you want to over-paint images in Painter or Art Rage, then a bigger canvas makes a whole lot of difference (at least to a ham fisted painter like me!). I hope this will do a better job of bigger renders.

Having said that, you've all awoken an interest in playing with some realism - if only for fun. I did try a couple of test renders with different textures and MAT set ups and wow, what a difference a simple set up makes with GC against the DAZ ones! However, those baked in highlights in many textures look awful with it. I'm going to have to have a good look through all the old stuff in my runtime to find the right texture sets and then work on simplifying the MATs for them. I have a feeling that Semidieu's new script might make short work of the later.