I realized that this is a general 3d question, but I posted it in the poser forum because it seems to deal most with 3d human figure work by non-professionals/hobbyists.  I'm curious; in what time frame do you think the computing hardware and software will be powerful enough that the average 3d hobbyist (ie someone who spends maybe 10 hours a week on it, but doesn't do it for a living) will be able to create renders where the 3d figure looks completely human and not computer-generated?  Note that I don't mean that you couldn't tell it was computer-generated if you really looked, but that the casual observer wouldn't be able to tell the difference.

I've seen examples already that are hard to distinguish, but in most cases those are made by professionals, with very specifically tuned lighting, skin settings, poses, etc.   Rather, I'm referring to the point where someone who's invested less then $5000 to $10000 (including the cost of their computer) will be able to quickly and easily make renders (as is possible with programs such as Poser) that they wouldn't have been able to before.  Also, what advances do you think will be made that will allow this?  Personally, I think that we're going to see skin itself become a seperate, high-polygon mesh, which will better allow for effects such as wrinkles.

Never. Just like any other artistic disipline, it takes skill and craftsmanship.

Ask how long it takes before you could oil paint a picture to look perfectly real.

Ask how long it would take to sculpt so your sculpture looks perfectly real.

Ask how long it would take to draw, and have it look like a real black and white photo.

It rerally has nothing to do with hardware or software.. I saw poser images 3 years ago that looked almost like photos to me.

Perfect renders are a combination of all those things you mentioned pros to.. specific lighting, textures, ect.

Heck, look at all the plant products for carrera 7.. I have a VERY hard time distinguishing them from real life photos!

Its not the tools.. it's the artist behind the tools...

Its actually pretty close now , with suitable textures and lighting , which does take some skill.

...Just as soon as we scan these two in.

The casual observer is already capable of being fooled.

For a non-casual observer, there are still several things that it is very hard to get to looking correct, the give-a-aways are usually hair and eyes, followed by things in the environment (reflections/reflectivness tends to get over done).

You're being awfully generous ;) 

$5k - $10k will easily get you an i7 system + one of MAX/Maya/XSI + a pro to do the render for you :p

"When will truly human figures be available to the common 3d hobbyist??"

When slavery is reintroduced.
DPH

Actually, Max and Maya work on the principle of LOW polygon mesh/skin coupled with HIGH res texture.  The rule of thumb in these high end apps is 'Don't model what can be done by texture!'
DPH

*When slavery is reintroduced.

Would slavery solve unemployment?

Actually, Biblical Slavery, as opposed to that of the Greeks and Romans, or even of the Antebellum South, was indentured service, with a finite specified period of time.  A person who fell onto hard times could thus always sell himself into service, with the knowledge that sooner or later he'd again be free.
DPH

Actually, Biblical Slavery, as opposed to that of the Greeks and Romans, or even of the Antebellum South, was indentured service, with a finite specified period of time.  A person who fell onto hard times could thus always sell himself into service, with the knowledge that sooner or later he'd again be free.
DPH

7 year jubilee! 

No, slavery was regulated by the Sabbatical year, IIRC.  The Jubilee, or Yovel (as it's pronounced in Hebrew) is the year following 7 Sabbatical years, i.e. 50 years.  On the Yovel year, all land was to be revert back to its original owners.  Thus, every house hold would reinherit its homestead, after having been sold off during hard times.
DPH

For stills, the tools exist to do it now at the hobbyist level; it's skill and training (and the 4% inspiration that Edison talked about). Animation still needs a few technologies to filter down to the masses, like dynamic skin, a muscle system of some form, to get all the wrinkles and bulges as something organic moves. But even if -that- becomes accessible, there is still the skill set to use it that has to be mastered; if you look at the pro level setups, there are usually all sorts of controls that are outside 'common sense' parameters, that enable a trained person to do things you simply couldn't do otherwise in a reasonable amount of time. And even then, you get into the computing power needed to do all the simulation calcs (lessee....first the muscles, then the skin over the muscles, then the clothes, if any, over the body, then the hair over all of that....). Multicore computers seem like the answer, but the software industry =still= is beating its head on the wall about multithreading. It's happening, but nowhere near as fast or as smoothly as some imagine. And just because an app -uses- multiple threads doesn't mean it uses them =effieciently=....which is a must in graphic programs, due to things like memory limits and time to completion of a specific process.

It may happen soon....or in a few years....or maybe never. There are industry politics at play in this as well, and the idea of the proles getting uppity is......distressing.....to some.

***"It rerally has nothing to do with hardware or software..

Its not the tools.. it's the artist behind the tools.."***

Sorry but that is the most misleading statement  ever made in the world of CG
honestly.
yes technical proficiency and even artistic "vision" is important

But there is a reason  why the Hair &Cloth system  in Maxon Cinema 4D
Will get you better results the the one in poser.

there is a reason why the subsurface scattering  Ability in Modo301
is superior to poser
on the other hand other IS a reason why the Fluid & hardbody Dynamic engine
in my $2700 License of REALFLOW 4  is better than the Dynamics IN Cinema4D

there a reason Why Vray is better than "Firefly
there is a reason Why Vue infinite was Used for some scenes in "Pirates of carribean" instead of Bryce.

yes we all use what we can afford and should try to maximize the potential of what we have.

But at some point the ability of the toolset/program will always play a factor.

Now as to the OP's desire for a "truly real"  human in CG.... please google
uncanny valley.

Cheers

Quote - Actually, Biblical Slavery, as opposed to that of the Greeks and Romans, or even of the Antebellum South, was indentured service, with a finite specified period of time.  A person who fell onto hard times could thus always sell himself into service, with the knowledge that sooner or later he'd again be free.
DPH

Ummmm, we have that now in Ameriva. It's called "JOB" and the release is called "RETIREMENT"...  Currently with the economic chaos going on over here and throughout the world, many "owners" are forced to release their "slaves" into the general population without any future provisions for the slaves or the slaves' families but they hoard the good stuffs for themselves.

:tt2: :lol:

UofOstudent

I know you want a anser like 5 years.

but we can not say that because we do not no wen Poser or D/S will get desent hair,render engines ect ect.

dynamic hair in poser might help maybe.

textured hair is kool for games but not the best realist renders.

a lot of nice renders in Vue.

not video but for stills you can get a lot out of a 2D app tweaking renders.

app's like C4D do use low polymeshes but then thay are subdivided and then thay get a micro displacemtn map.so in the final render the charater would have a polycount of a million or two.

$ 10,000 for hardware software + years to learn it all and get good at it all.

we do not do all this for hoby stills but for pro videos.

it would be a bit of a over kill for a render to show off.

You can not compare Poser ,zBrush ,C4D thay work nothing a like.

but you can get a killer renders out of Poser, D/S & Vue.

just practise at render settings with defrent render engines and lights are very importaint.

tweaking with a 2D app always doable. 

Quote - [

Ummmm, we have that now in Ameriva. It's called "JOB" and the release is called "RETIREMENT"...  Currently with the economic chaos going on over here and throughout the world, many "owners" are forced to release their "slaves" into the general population without any future provisions for the slaves or the slaves' families but they hoard the good stuffs for themselves.

Actually its called "Debt" and the wage slave  is actually working for the true masters
...The blood sucking predator called "Bankers"

See here for the truth about our "monetary" system:
**
http://www.vimeo.com/3384195**

Quote - > Quote - [

Ummmm, we have that now in Ameriva. It's called "JOB" and the release is called "RETIREMENT"...  Currently with the economic chaos going on over here and throughout the world, many "owners" are forced to release their "slaves" into the general population without any future provisions for the slaves or the slaves' families but they hoard the good stuffs for themselves.

Actually its called "Debt" and the wage slave  is actually working for the true masters
...The blood sucking predator called "Bankers"

See here for the truth about our "monetary" system:
**
http://www.vimeo.com/3384195**

Thanks for clearing that up for me... I was wondering about what to call it and you hit it on the nose squarely!

:lol: :lol: :lol:

If we're talking about a "create reality" button someone could press right out of the box, that's going to take awhile.

But, "realism" is available right now if you are willing to bridge the gap yourself between technology and desired output by using skill and knowledge.  That "gap" is going to be there for a very long time.  The levels of skill and knowledge required to cross it may shrink as technology becomes more powerful.  That could be within a few years or a decade, it depends on a lot of factors.

As someone said above, the ability to produce stills is already here.  In fact, the ability to produce an artificial human that looks like a real one has been around for a lot longer than some might thing.  Michelangelo's "David" is a good example even though he is somewhat idealized.  Several "Realism" painters successfully breached that gap in their own medium. Animation is troublesome because so much information is exchanged in the interplays of light.  It's difficult to quantify it all.  That will improve with technology.

Painting, sculpture, even literature and theatre.. those were the mediums of the day to produce that "alternate" reality so many seem to crave.  Today, it's computer generated graphics.  What will it be tomorrow?  The medium is just as important or, in some cases more important depending upon how it is perceived, as the finished product.  Will we still be dealing with two dimensional displays of three dimensional environments in the next five years?  Maybe not.

In short, the answer is this: When the level of technology reaches the point where it is easily able to produce the desired output while depending only upon a minimum amount of skill and knowledge supplied by the user necessary to bridge the remaining gap ... then that is when it will happen.  :)

Check out what they are doing at Stanford.  http://graphics.stanford.edu/projects/mich/

An 18th century "gentleman" (hobbyist) photographer might have asked the same question as he struggled to emulate the pros using glass plates and  toxic chemicals.

"In short, the answer is this: When the level of technology reaches the point where it is easily able to produce the desired output while depending only upon a minimum amount of skill and knowledge supplied by the user necessary to bridge the remaining gap ... then that is when it will happen. :)" 

Absolutely true and well put.

This question comes up in one form or another every so often. I agree with Dale B that the essential underlying technology/knowledge already exists. The physics of light, the surface properties of materials, (human skin to perhaps a lesser degree), etc. are known. Some of the techniques are known only to a few but with all due respect to artistry, I don't think reasonable realism is confined to the realm of dark incantations and sheep entrails. I'm not talking about emotion or composition or color sense, which are part of the realism equation but the hard physics.

Perhaps in my ignorance, I see no reason why software cannot analyze a scene and given the included geometry, selected materials and light placement render it realistically. It might take a month to render but that's mainly hardware and hardware is easy relatively speaking - the software is hard. The software doesn't exist because, IMO, the companies don't see a market beyond theose who need/want to spend hours tweaking.

I liken Poser and DS to the 1st 35mm SLRs, a far cry from glass plates but still a long way from point and shoot in terms of getting good technical (as opposed to artistic) results. I think the potential market is huge but for the moment, selling $5,000 Nikons to shutterbugs rules. My prediction, 5 years, but I'm sure I said that 5 years ago :-)

Funny, it was precisely the photography analogy that occurred to me.

When it comes to think of e.g. skin looking like skin etc, I don't think the issue is artistry (in the traditional sense) so much as technical capability and the know-how to use it. The artistry side has to do with the power to observe and interpret. You can have a perfect render in terms of materials, and it still won't look real. Why? Well, let's say we're talking about a portrait. Something isn't quite authentic in the precise way the hands are held, or the fingers are bent, or the forehead is raised... etc etc. It takes a lot more of an artist to read things on that level.

Indeed, that's why some cartoonists can have characters that in significant ways look more "real" than CG work that is technically more "accurate."

Indeed, that's why some cartoonists can have characters that in significant ways look more "real" than CG work that is technically more "accurate."

Google, " Uncanny Valley "

 "Uncanny Valley": VERY roughly - the more human a figure (robot) becomes, the more positively humans will respond to it, until it reaches an "uncanny" state of being almost but not quite human and people's response drops sharply (the valley in the response curve). After that as it becomes even more human people's responses begin to become more positive again.

Which explains why I'd rather date Vicky than Paris Hilton. OK, Vicky with the NGM pack and one of those - ahem, add on thingies they sell at 'Rorica.  

Interesting hypothesis patorak. Do you think it applies to 3D as well as robots? And if so, does that mean that Daz is due for a really bad slump somewhere between V5 and V99?

Oops, props to the horned one for the original uncanny reference.

Actually, guys, I'm far more optimistic, it seems. I look at what passed for "reality" in Poser five years ago versus what we have now, and the difference is like night and day, thanks in large part to better textures and more sensitive modeling. Just the other day, I saw an image for a product, and I swear the image was almost photorealistic, to a degree I hadnt seen in a very, very long time.

And yes, I do think it has a lot to do with what you do with the tools. We had some pretty amazing photoreal stuff floating around here about four or five years ago from a former vendor (whose name I forget, sorry): she had done work using Poser for the History Channel, and the results were astounding. So yes, I think we may see the day when it gets to the point where you really, really have to look to know for sure.

I´m a hobbyist too, but I´m playing with Poser and other Software since Poser4 and 2002, nearly daily. With Poser7 and different adds of scripts, top class materials and textures and the know how what I really can do with Poser, makes most images very photoreal.
Perhaps a week ago someone asked me if my images really would be renders only or if I did use real persons as characters. So I think the images are very photoreal.
But my render settings are always very high and not standard or preset only.

I think most beginners with Poser make the mistake to use the probs, hairs or figures as they are. But to make an image real, you also have to look that hairs or clothes are falling with the pose, and this you have to do by yourself.

"...We had some pretty amazing photoreal stuff floating around here about four or five years ago from a former vendor"

You probably mean Catherina. Yes, her textures and what she can do with them are amazing. Her gallery here is under mec4d. Take a look at "Buddie." So yes, the capability exists today - easy - No. Clone Cath's brain and put it into an AI driven render engine then maybe.

Well I think the OP has a valid point; of course an artist can do photorealistic stuff with a pencil without the need of expensive computers and software, but we are talking about hobbyists here.

I mean, going back to Poser 3, could someone new to 3D redering get the same results as with Poser 7? I doubt it. Could a computer back in the day handle a V4 or a G2 figure?