Quote - personally I saw it as a fake soon as I glanced at it. there's a dead give away that glares.

look at her groin. wrong...! the skin is perfect.. not a mark, blemish or imperfection (if you are that perfect there please call Hugh Hephner at Playboy and get into the magazine.. it'll save him a fortune in airbrushing overheads!) ... and the specular looks like plastic there.

it's not bad work I grant you... but no one else saw that error?

Yes, I noticed errors and continue to pick up on some the more I look at it, but not at first glance.  The first time I saw it, my eyes didn't immediately relay a message to my brain telling me this was CG, unlike 99.99% of the realisitc renders I see every day.  It took me more than a few moments to realize this was not a real photograph, and I give the artist credit at being the first one to render a human figure, in a still image, that actually had me fooled for at least a few moments.  I've seen many architectural and automotive renders that were nearly indistinguishable from the real thing, but almost never a human figure.  Unlike the vast majority of renders, I believe this is one that could easily convince the average person into believing it's real.

Both my wife and brother, who have no CG experience, were totally convinced it was a photograph, and I had a hard time convincing them otherwise .  They never acted that way when I showed them any of my work, so I was a little insulted by that actually.

when i saw it I thought it was a cg figure pasted on the background of a real photo. that chair really gets me. the background looks great though once i was convinced that background was cg i see some of the texture "distress" has scaling issues. I am 65 and that is exactly how smooth my groin is! ;)

In photoshop I did a modification to remove the spotlight effect and even out lighting (old photo-offset techniques of layering inverted images to "fill flash" and some levels adjustment) inorder to see how the image looked. There was a lot of artificial grain apparent but still the results of the experiment are interesting.

Quote - the scaleof the character is way out when you compare her to the door on the left...she must be about 7 feet tall. .

Who knows? Maybe she's studying science at Dwarf's University, why not? LOL

Back to the crits.
I've been using Photoshop for about 10 years to a professional level and despite his work is a good work, I must admit that at first look that image seemed to show a bad usage of blurring, expecially concerning both hair's contour and face skin's itself. It seems to be an head which does not belong to that body, or viceversa. "Photomontage", what you call it.
Some minor mistakes like the toes (as already stated by many of you) but for the still, what doesn't convince me is how both head and body melts together.
As for the 3D, I can't spread anything than praises. Good work.

Though I have kudos for the amount of work involved in this image (just finally saw it, been offline for a while doing a lot of stuff), I can't exactly jump on the bandwagon of "the best CGI I've ever seen.".     Granted I cant "DO" better, but i have seen better.  

Again, great work, but some things on looking at it throw it off, and actually scream "not photograph" right off the bat.  These might be technical overlooked errors.  The chair in particular, i noticed right away.  the bottom is fine, very 3d, but teh back in which her hands rest on seems part of the wall and very flat.  Like that hand was "pasted" on a flat 2d image. I'm sure its not, just looks that way.

I have to agree, the head and body don't seem to match, as if two different images are there, and the CGI'ness really shows up past the waistline down.  The panty part doesnt seem as photo quality as the top of the outfit as well.

Again, maybe the voice of dissent in the crowd, and its sure better than any CGI I could do, but even on first quick glance it neither seemed a straight photo, nor straight CGI to my eyes.

My question is, what the features are that people respond to in in this image? There is always this debate as to whether attempts at "realistic" cg images should be photographic or not but for the time being I think that it makes sense for them to given there are no immersive media yet that I know of, all the images I see are photos, paintings, or drawings. I have never seen any image that approximates human vision even in cinema so it would make sense for cg images to simulate other media rather than natural vission.

What I call the "render-realism" of the no-post-work movement wherein every pore is sampled, textured and displaced reminds me of the hyper realism of Renaisance painting or 1970's representational minimalism and process painting which in art history class was described as a simulation of going over the entire scene with a magnifying glass from a scanning POV rather than replicating the distortion of focus of natural vision in a single point lensed POV.

Would some one isolate the factors in this image that lend it photographic credibility?

I come up with:

1. surface incident (degree of entropy on the textures)
2. the use of spotlight effect
3. the flattened contrast
4. the strong simulated chemical/pixel grain which seems more prounouced in the dark areas and less in the light areas as would occure with photographic media.

I did tests on the image removing certain factors which can not be posted here due to TOS but would any one volunteer to construct an original Poser scene and issolate and compound various factors demonstrated in this image to show what is going on here as they understand it?

Conceptually I have always wondered if the use of sampled textures is more compremising to the cg integrety of an image then aggresive post-work. People who disdain post-work seem to embrace hi-res photo textures as opposed to "hand painted" or procedural textures. I don't know what I think on that issue myself.

I think that the average non-CG-aware person would believe that this was an actual photograph.  Sure, 3D pros with 10+ years experience, or even advanced 3D amateurs might be able to immediately spot the fact that the image is CG.  However: I doubt that any but the most observant of non-CG types (which represents most of the human population) would be able to tell the difference.

Lee did post a "before postwork" and "after postwork" comparison. It added a lot to the realism effect. And... CGTalk thought enough of the image to present him with an award. He did good. ;)

scanmead, I know it is lazy of me to ask but would you give the link for that before/after... I have been trying to read through the threads to find the supporting images but only have found some.

Looking at his other work my question is why these independent modelers can get figures that look so much more realistic than those from the teams at DAZ and EF?

simple to answer..

they are not making a figure that has to be all things to all ppl (see poser figures +morph demands, where a figure is almost expected to be everything from a baby to an old crone and everything inbetween) and..thats about it really.

if you think about it, we put demands on the figures we use that defy logic. where it would be better to use a purpose built figure to be the old woman, (which would give far better results), we try to make a generic figure into that old woman... and expect that figure to also be the man, six armed thing over there, that tree and oh, can I find a morph to make it into a young perfect woman?

they can make a figure that just is.

:) It's the thread that refused to die,,, I should give a "kitty" a pet for that,,,*

Would some one isolate the factors in this image that lend it photographic credibility?
momodot, If you're asking me,,,

*One thing is; Poser does not blend mesh or texture in a render. What I mean is, the render bucket registers: face tex, face tex, Stop. Eye tex, eye tex, stop. Cornea tex, Stop...(I'm not a programmer)

I mean to say,,, there is no averaging between meshes or their textures. It goes Red, Red, Black(stop), White, White, White... There is no transitional Pink. Nature blends these details to our eye in the form of Dof, reflection, scatter, proximity shading(AO), shadows... to name a few. When 2 white boxes are put together, the line we (usually) see is grey, Not black. I see separate objects in equal resolution in most renders.  Poser and other apps need tons of info to get past this.
 
Some of the best parts of that render (to me) are the seamless blend of separate meshes into each other. Also its proper resolution. I mean,,, you cannot count the threads on her corset or strands of her hair at this focal distance. But it's there. Pointilism, Impressionism really played with this. Again, we forget how little we can actually focus on in real life. Renders freeze "real" life with everything in focus. True,, you can choose what to focus on but you loose an important element for realism.

The subject matter is counter-academia and this is what made it shine so much. Being human is important, too.  Skin is still one of the most difficult things to simulate. But once these little obstacles are surmounted, many avenues will open up for film, tv, gaming, hobbies and poserites (separate catagory).

Edit to add: This render was moved to the CG Choice Gallery. 60,000 views in a few days. Gotta love it