I'd like to start a discussion about what makes an image photo-real. If you look on-line at those "Can you spot the real from CGI" tests most of them are very clean, brightly lit studio shots. There is not much out door imagery unless you get into film work, set extensions, props and characters. I would not necessarily call these photo-real but rather photo matched. The CGI landscape or prop has been graded, textured and lit to match the film print (and visa versa).
So what is it that makes an image photo-real. To simply dismiss it as "looking like a photo" is missing a point. "Real" photo's are manipulated either in camera or photoshop according to the desires and skill of the artist.
Are there tricks we can employ to fool the eye, to make the viewer "abandon disbelief", to make the viewer want to believe this is "real"?
LOTS of detail, great subtle lighting, accurate scale and models all spring to mind. Real world textures also.
All ideas, comments and suggestions very welcome.
Cheers
Richard
I frankly think it is so highly dependent on context/content, that it's hard to have a useful conversation about it in a general sense. If you want to try to distill it down into one single thing, or an overall *type of work* (or aspect of imagery) that has the *biggest* influence (even if this influence is only, say 15%, with all other elements sharing e.g. 5-10% each), my personal opinion is: grading and coloration. In some ways this is a bit of a cop-out, because there are plenty of scenarios where you cannot take just any (CG) image and adjust it to look real, however if you tend toward a particular, one might say "cinematic", post-processed look, I think you can make an overall improvement in the *average perceived realism* of almost any image. As I say, this is probably only 15% of the problem, where other elements like detail, lighting accuracy, sufficient randomness, etc. could be 5-10% each, so it's not the "biggest" by much, and other elements *combined* surely have a bigger influence. So again I'm not sure how useful it is to try to draw any conclusions from. But I feel pretty confident that you can make an incredibly detailed, realistically random image with fully correct lighting but, if it's not either tuned by the render engine, or in post processing, for a "realistic look", all the assets and render accuracy in the world won't make it look "right".
Lighting/shadow (two sides to the same coin) would, I think, be a close 2nd in terms of overall influence on perceived realism. One could argue the cope of this, i.e. is translucency a lighting component, a surface component, a combination of the two...
All I know for sure is I've seen quite a few images where an incredibly simple, even very sparse scene, but with great, realistic lighting, and suitably "graded", just screamed "real" to me, despite the lack of detail, the lack of randomness, the lack of reality cues besides color and shading. Perhaps this is best exemplified by the ability for some animation to be perceived as highly realistic, or for a "realistic style" to be still possible even in a minimalist artistic expression. But then one could argue that goes right back to the context issue, if you're depicting a scene that in reality is very simple, sparse, i.e. an unusually (but still realistically) uncomplicated scene that exists in the real world (let's say a sterile office building), then perhaps that makes photorealism easier to achieve *in that specific context*. But does that necessarily say anything about achieving realism of, say, a human figure, or a landscape? Perhaps not. Perhaps, as I began by saying, context is key to meaningful conversation.
So, if you're talking specifically in the landscape context...
- Oshyan
ok. so thread closed.
:-X ;)
Really! Huh?? :o
Oh well ... never mind.
Richard
I didn't mean to kill discussion, I guess I just think we should define the bounds of what we're talking about here. If you want to consider the entire spectrum of "what makes an image look real", that seems like an almost impossibly broad topic. If we can focus more on landscapes in particular, I'd say that helps narrow it down quite a bit. In which case I'd say the single biggest contributing factor is probably atmosphere. :D
- Oshyan
Ok, I'll try to kick it off again :)
Obviously I am talking in a landscape context (and before anybody jumps on me I know my "roadside" contest image is not strictly "photo-real") but that's kind of my point. What is it about certain imagery that makes us "believe". Oshyan has covered so good points and yes context is important. Lighting, shadow and detail are of course vital. Would the landscape CGI out of Promethius, SWATH (cue Chris:) ), or Oz (check the trailer) be considered photo-real. I would say yes to all. They are realistic in that "fit" in the context they are intended for.
The photo's most off us take on holiday are for the most part pretty dull (well mine are), in the old days of 35mil film photography most of the shots taken were pretty worthless, we'd be lucky (or highly skilled) to get 10% decent shots! With digital photography at least we can preview, and delete when necessary, so the good percentage has increased and so by a vast amount has the volume.
When it comes to creating photo real CGI shots do we aim for the rather dull snapshot or try to emulate the great photographers who would try to capture the essence of a scene with drama, contrast and camera views none of which we would expect to experience as part of our daily lives. One of my favourite landscape photographers, Ansel Adams (http://www.anseladams.com/) creates imagery that is highly evocative, beautifully framed and very dramatic but if you visited those places, yes you could recognise them but his shots they don't look like the original, the imagery has become idealised versions/interpretations of reality. If I were to take a photo of Yosemite Valley it would not look like the one that Ansel famously took. But which would be "photo real"? Of course they both would but I would prefer the one with drama to hang on my wall.
So atmosphere is possibly the most important element and I would have to agree.
What aspects of atmosphere help the most? Any tricks of the trade that the pro's use? I'm not necessarily talking in pure Terragen terms here but hopefully any thing tricks/info/knoledge can be applied to TG and other apps.
I would like to hear other's point of view on this.
:)
Cheers
Richard
Just for the record, I was referring specifically to "atmosphere" as a component of the scene construction and rendering process (i.e. the atmosphere and cloud nodes in TG, or more generally, haze, red decay, atmospheric scattering, etc.). I think that adds a ton to realism. Even in real life when I see a super clear day, it looks "unreal". Haze is a huge part of how we judge scale and distance, especially when we're talking about landscape-type distances. Even in small, localized scenes, there is often noticeable haze present, that adds to the feel of a scene, of an image. Without this, I think realism is virtually impossible. Whereas you can have a very, very simple scene, a barren mountain top for example, covered in snow, no plant models needed, very few - if any - rocks, simple texturing, etc. and it can still look highly realistic, as long as there is that realistic sense of "atmosphere". So I really feel it's one of the primary keys in getting a scene to look realistic.
- Oshyan
Quote from: cyphyr on March 06, 2013, 07:56:12 PM
Really! Huh?? :o
Oh well ... never mind.
Richard
For the record: I was making a joke. ;D
I have found the answer to your question, Richard, at least for myself, this seems to be it:
1) The rendered elements need to look like the real thing
2) The lighting needs to be perfectly realistic
3) In the post work, you HAVE to add the imperfections of the camera lens.
The last point is so important, but mostly people don't go there. Why to most renders look fake? And I don't mean TG specifically, I mean generally most CG images look like CG. With ones that don't, the artist has added the imperfections of a camera in post:
- chromatic aberration
- DOF
- vignette
- sharpening
- edge blur
- ....
Cheers
Frank
Check this out guys:
http://3d.about.com/od/Creating-3D-The-CG-Pipeline/tp/8-Tips-To-Increase-Realism-In-Your-Renders.htm
Just what I said and more.
Mmm, very good article Frank! Very relevant.
- Oshyan
Ah yes I had forgot the camera imperfections, chromatic aberration in particular. DOF tends to often be over done making the scene look too small, like a model. For both to be effective higher resolution images need to be rendered in the first place. (as large as possible!)
I do however think there is a huge viewer element. I remember a documentary about "Pathe News" where they showed the sinking of a war ship in WWII. It was just a small model filmed in a tank but the audience left the theatre in shock believing they had seen the real thing. I think I remember hearing similar stories about "Jaws", a film who's SFX are by to days standards laughable (You can see the shark deflating at the end!) The reason I say this is that to some extent *all* the FX guys had to do was exceed audience expectations.
Cheers
Richard
ps: apologies for my "sense of humor-ectomony"
pps: great article
I think it's necessary to add DOF to a render. Every lens, even my eye lens, puts a focus somewhere, so in my image, I will absolutely add a small amount of DOF, ever so slightly, though, but notable on a second look. Perhaps I will take a reference shot here at our street, and see how much DOF is going to be visible with the camera near the ground, but focused on the distance or some element in the mid ground, should I add such an element to the scene.
Guys you are talking about different things and not strictly about realism as we see our world i know.
Like on the 48 FPS Hobbit thread realism isn't DOF or the look we see in this or that film stock.
What you talk in the last posts is more about mimicking the look of some kind of Photo shooting techniques or film stock etc.
When Spielberg filmed "Saving Private Ryan" in the same way as some old WW2 documentaries
it looked realistic as in that documentaries but not necessary realistic as it really should have looked.
I hate DOF and chromatic aberration for example.
When you look at your hand very near your face your hand is in focus
and when you look at the background your eyes quickly focus on that part and you see it clearly too.
But on an image or movie that focused part is static.
You can look on the background (that is out of focus) but that background stays out of focus so much you look.
Is this realistic?
It has its part in storytelling etc. but realistic look is sometimes very subjective.
Giving people the look they expect and not what it should be too probably.
And that is what i hate about DOF and chromatic aberration...
In that aspect photo realism is using the same methods as in that or that camera or in that or that film stock look etc.
If you want to make a image that is shot like with a micro setup then using DOF and chromatic aberration is normal but not necessarily realistic .
Some users here are very good on building landscapes but the lighting -atmosphere they use is dull or painterly.
But some have such a atmosphere combination that their images look like photos so much you can see fractals etc too here or there.
Oshyan puts it better then me.
If you want to make photo-real images with TG2 then you have to take TG2's strengths and weaknesses into consideration.
Then it's matter of trying to make the best use of the strengths and to avoid situations which TG2 can't cope very well with.
In regard to that, in my opinion, TG2's strengths are:
1) atmospheric model, especially the atmosphere node/shader itself.
Picking the 'right' atmosphere density and haze settings can make or break the sense of scale and will determine how saturation of shading and shadow detail in terrain will look like.
2) procedural displacements for adding detail and shading/texturing to your terrain.
This requires no explanation of course, because this is why we all love TG ;)
There are some aspects though which you should be careful with, later more about that in the TG's weaknesses part.
All in all these strengths offer a lot, especially if you want to make barren scenes without vegetation, like deserts and the like.
Also imho, TG2's weaknesses are:
1) object rendering -> objects nearby look too CG. Mostly due to lack of some advanced kind of shading.
A simple example is Silva3D plants' example renders on their site which are rendered with Vray.
If you reproduce such a close-up in TG2, it's always quite inferior.
Mostly textures don't come out as crisp and clean and also (self)shadows aren't that good looking.
You can go through quite some lengths of getting them to look better by hugely increasing AA together with cubic b-spline filter and also upping the resolution. If you don't up the resolution then the self-shadowing is too dominant and you can't get nicely detailed lighting.
Which leads to the second weak point...
2) GI -> to continue with the objects. In the distance GI is uncapable of correctly sampling the vegetation. This leads to too dark vegetation as if the renderer decides to treat the whole object as one shading result, thereby destroying all the detail.
Similar results and issues are with displaced terrain where GI gives unexpected results over distance.
These 2 phenomena I describe can be quite clearly observed in my DeviantArt gallery:
http://tangled-universe.deviantart.com/gallery/
It's difficult to say though whether it's GI which is undersampling or atmosphere or some other process in the renderer which pre-calculates occlusion/surfaces.
There's no documentation about the renderer which gives a detailed explanation how it step by step treats the scene so that you can devise a detailed and especially direct approach on tackling these issues.
So far the only solution is to increase rendersize considerably to ensure that the renderer 'sees' all the geometry of the vegetation and thus applies correct lighting and shading to it.
Increasing GI to 6 along with 512 atmo samples didn't improve it, so my suspicions is that there's some pre-filtering/calculations of surfaces which is limiting.
However, like I said before it's difficult to say without knowing more about the renderer.
Other issues are in-terrain-shadow objects which tend to look plain wrong as soon as you're dealing with either quite displaced surfaces or denser atmospheres than default. Not to speak about a combination of the two. Avoid that at all cost, you'll never get it to look photo-real.
3) noise sampling -> especially in surface shaders the renderer has parameterizations skewed towards high frequency noises which it at AA-stage can't eliminate.
If you design a scene with crop renders of your foreground surface and add subtle brighter speckles to your surface's base then they will look nice.
However, over distance these subtle speckles become dominant and overrule the darker surface base.
So all in all a lot of aspects are involved in considering something photoreal and a lot of that is esthetic and highly specific for each situation and scene, as Oshyan layed out.
However, I think that besides that there are also some technical aspects involved, which I summed up above, which you can keep in mind to stay on the 'photo real road' ;)
That's an interesting comment Kadri,
Funny that photographers do everything in their power to eliminate lens-artifacts like CA and that on the other hands people who have the ability to use absolutely optically perfect lenses want to put in lens-artifacts.
If you simply eliminate these two you end up with zero, so the key is probably not in lens-artifacts and the like, but in other aspects.
Let's not forget that for the purpose of the contest, we're talking about PHOTO real, so it has to look like a photograph as much as possible. (Of a half-way modern camera and lens).
So in that sense, adding the imperfections of a lens is a must.
Quote from: Tangled-Universe on March 07, 2013, 06:56:28 AM
Other issues are in-terrain-shadow objects which tend to look plain wrong as soon as you're dealing with either quite displaced surfaces or denser atmospheres than default. Not to speak about a combination of the two. Avoid that at all cost, you'll never get it to look photo-real.
Can you explain this differently? I don't understand what you mean by "in-terrain-shadow objects".
Thanks
Frank
Quote from: FrankB on March 07, 2013, 07:02:55 AM
Let's not forget that for the purpose of the contest, we're talking about PHOTO real, so it has to look like a photograph as much as possible. (Of a half-way modern camera and lens).
So in that sense, adding the imperfections of a lens is a must.
Yes, that's true.
Quote from: FrankB on March 07, 2013, 07:07:52 AM
Quote from: Tangled-Universe on March 07, 2013, 06:56:28 AM
Other issues are in-terrain-shadow objects which tend to look plain wrong as soon as you're dealing with either quite displaced surfaces or denser atmospheres than default. Not to speak about a combination of the two. Avoid that at all cost, you'll never get it to look photo-real.
Can you explain this differently? I don't understand what you mean by "in-terrain-shadow objects".
Thanks
Frank
I mean objects which are placed in shadowed parts of the terrain.
They are either underlit or overlit or have an odd hue to them.
right. Thanks for explaining!
Quote from: Kadri on March 07, 2013, 06:52:09 AM
...
Is this realistic?
It has its part in storytelling etc. but realistic look is sometimes very subjective.
Giving people the look they expect and not what it should be too probably.
And that is what i hate about DOF and chromatic aberration...
Don't hate it, that's quite a strong emotion for an imperfect lens, which was innocently created this way :)
Richard was asking about what makes renders photo real, with an emphasis on the word "photo".
Of course a photo gives a different impression compared to seeing things with your own eyes. However, because a lot of our day life and memories are supported by photographs, we come to perceive that photographs are the depiction of what's real.
If an image, mostly a render, doesn't look like a photograph, something in me rejects it as "realistic looking".
Now, I hope you participate in the contest, Kadri! That's going to be an interesting challenge for you (like for everyone participating). Try to make it what you think looks real. If you go through creating a WIP thread, you can ask others how much they perceive your results as photo real, and take it further from there. Come on! :-)
Cheers
Frank
Yes this has long been a problem with Terragen, shadows within shadowed areas don't render well. Nor dose detail in shadowed areas (there's a thread about it somewhere). Soft shadows in the sunlight node do help as can multiple sun light sources and high GI settings but obviously this is not a desirable solution; the render times will sky rocket!. If the entire terrain is in shadow the renderer dose seem to be able to compensate to a degree and "find" the lost detail.
I've done a few searches on Google images to look for "Photoreal Landscapes" and there is not that much. Franks Cloud is the third image followed by a bunch of game, and Vue renders and f course "photos of real landscapes".
As you say Franck so much of our lives are recorded in photographic form that we have come to see the photo as a "real" representation of reality.
Hmm this is getting more philosophical that I was expecting ...
There are an awful lot of photographic styles, BW, sepia, high grain, high contrast, monochrome, the current HDRI and "Instagram" fads and I need not say, many more. I haven't checked but I would imagine there are Photoshop lens pre-sets for many cameras so in theory pretty much any render "could" be made photo-real.
I know that's not the point of the contest but don't you think it brings up some interesting discussions about the nature of reality and our perception of that nature. Ultimately the definition of "photoreal" will have to be subjective. As we have already seen we can disagree about this but in the end we will all have our own ideas about what is and is not "photoreal".
Cheers
Richard
now back to those clouds ...
Frank i thought about that "hate" years ago and i think it is because i have myopia .
People who have the same problem will understand me easier probably.
When i looked around me in the first years when i did not know i had to use glasses and everything was blurry-out of focus it was painful trying to see everything clear.
So when i see blurry or out of focus images the same reflex comes in and i see it more as a fault that has to be corrected then as a aspect of a photo.
Regarding the contest...uh...oh... i hate contests too! LOL!
Joke aside, i have not an image in my mind at the moment that i feel to share .
But if i get that inspiration somehow why not :)
Quote from: Kadri on March 07, 2013, 08:30:15 AM
Joke aside, i have not an image in my mind at the moment that i feel to share .
But if i get that inspiration somehow why not :)
You could also start somewhere and see where it leads you. That's how I do it at the moment. It's probably not the most clever approach, but it's fun.
Cheers,
Frank
Unfortunately I didn't have the time to read all the posts of this very interesting thread carefully, but anyway I have to post a comment.
There is not THE way to make an image photoreal. It depends on what you want to achieve. Actually it's photoreal. That means you want to create something that looks like a photograph and not what you would actually see in the real world.
It's a great exercise to take a photo and try to replicate the look as much as possible. And I agree that imperfections of the lens are essential. But of course DOF for example is very often overdone. If you have a panoramic wide shot with no objects close to the lens DOF would destroy the sense of scale and it would look like a miniature.
Too much grain may look artificial too and vignettes are imho an imperfection of very old cameras (?).
Some subtle chromatic aberration is my personal favourite. It can change the look of an image dramatically (sorry, Kadri ;).
Anyway I think reference images are the key to photoreal CG images. Taking the time to tweak an image in TG until it's really close to the real thing is really worth it.
Chromatic Abberation is also a nice way to very subtly blur your image.
As I was reading this thread I also thought was Hannes thought (
QuoteActually it's photoreal. That means you want to create something that looks like a photograph and not what you would actually see in the real world.
). I didn't realize the contest was really about simulating a photo, but if it's stated that way, it should be done. 'Realistic' is not the same as photo-realistic indeed. If you look at a landscape, your eyes wonder over the details, and you only see clear what you focus on, the rest is a blur, although you perceive it as a total (sharp) landscape in your brains.
And I also like the comments about all the types of photography... what photo-reality are we talking about? Instant camera, Ansel Adams' camera... philosophic indeed.
Quote from: Dune on March 07, 2013, 09:52:09 AM
As I was reading this thread I also thought was Hannes thought (QuoteActually it's photoreal. That means you want to create something that looks like a photograph and not what you would actually see in the real world.
). I didn't realize the contest was really about simulating a photo, but if it's stated that way, it should be done. 'Realistic' is not the same as photo-realistic indeed. If you look at a landscape, your eyes wonder over the details, and you only see clear what you focus on, the rest is a blur, although you perceive it as a total (sharp) landscape in your brains.
And I also like the comments about all the types of photography... what photo-reality are we talking about? Instant camera, Ansel Adams' camera... philosophic indeed.
Let's not over engineer this! Just a normal photo camera.
But...but Frank "normal photo camera" ? ;D
I think that you can't create a photoreal image if you don't have a realistic render as a base.
You can push it in extreme directions to make it photoreal but that would become obvious and would also conflict with contest rules.
So I think we shouldn't be influenced too much by this "photoreal discussion" as this is rather end-stage stuff in my opinion.
A purple carrot still looks like a purple carrot in B&W, so you have to get everything straight before adding in the photoreal look.
Quote from: Kadri on March 07, 2013, 10:13:27 AM
But...but Frank "normal photo camera" ? ;D
I think you understand.
Quote from: Tangled-Universe on March 07, 2013, 10:47:10 AM
...
A purple carrot still looks like a purple carrot in B&W, so you have to get everything straight before adding in the photoreal look.
Ah that's cleared that up. I shall sleep easy now. lol
Richard
ps: did you know carrots used to be purple (http://www.todayifoundout.com/index.php/2010/04/carrots-used-to-be-purple-before-the-17th-century/)!
Quote from: FrankB on March 07, 2013, 11:16:53 AM
Quote from: Kadri on March 07, 2013, 10:13:27 AM
But...but Frank "normal photo camera" ? ;D
I think you understand.
Yes and i think you too Frank :)
http://photoreal3d.wordpress.com/category/photorealism-in-3d-cgi-imagery/#
Quote from: Dune on March 07, 2013, 09:52:09 AM
... I didn't realize the contest was really about simulating a photo,
Neither did I. So now it seems it is more a rendering style contest than a landscape contest. Since that's the case I'll be going for something like these advertising postcards for my roadside diner:
[attach=1]
Hi Greg and others,
No it isn't such a contest. This has little to do with render "style".
It's a landscape contest with the road(side) theme and with the aim to produce the most photo realistic look.
I have deliberately written photo realistic seperately because the renders need to look realistic in the beginning before you can even start thinking about the photo part of photo realistic.
Guys, you are totally rat-holing into a secondary and subjective aspect of the contest.
It's interesting to debate what makes a render look realistic, and we have mutually discovered that it all starts with making the render elements and lighting look realistic. There is no need to postwork anything per se, if your render comes out looking realistic in your eyes, however in a lot of cases a slight postwork can increase the perception of "real".
It's up to you to decide. We have only started to share know-how about techniques to do so.
I would say: focus on getting the most realistic looking render in the first place. Everything else is secondary or tertiary.
There is no such thing that the contest is "more a rendering style contest than a landscape contest".
Quote from: Kadri on March 07, 2013, 12:48:55 PM
Quote from: FrankB on March 07, 2013, 11:16:53 AM
Quote from: Kadri on March 07, 2013, 10:13:27 AM
But...but Frank "normal photo camera" ? ;D
I think you understand.
Yes and i think you too Frank :)
Not sure you really understand. Otherwise you would not have made that statement back to me. I wouldn't have thought that this subject would require any more clarification, but here we go:
Maybe you understand it this way: You go into the store of an electronics retailer of your choice, in year 2013, go to the camera section, and buy any camera from there of your choice, and a general purpose lens with it. The pictures you made with this camera would give you photo real images. Alternatively you make pictures with your iphone or whatever.
Keep it simple! :-)
Cheers
Frank
Quote from: cyphyr on March 06, 2013, 09:44:38 PM
When it comes to creating photo real CGI shots do we aim for the rather dull snapshot or try to emulate the great photographers who would try to capture the essence of a scene with drama, contrast and camera views none of which we would expect to experience as part of our daily lives. ... which would be "photo real"?
I agree they both would, but in terms of getting a viewer to suspend disbelief I'd go for the dull snapshot, not only because it is more common, (and just as difficult to achieve) but also because when I see a dramatic image, after the shock of the first glance fades, I look closer to try to see what the photographer did to make it so powerful. If I'm trying to fool the eye I don't want extra analysis done by the viewer. I want him to see a photograph, believe it, and move on, so the shock comes to him only after he learns that it wasn't real at all.
Quote from: FrankB on March 07, 2013, 02:42:15 PM
... focus on getting the most realistic looking render in the first place. ...
There is no such thing that the contest is "more a rendering style contest than a landscape contest".
Then I am confused because earlier you said
Quote from: FrankB on March 07, 2013, 07:02:55 AM
Let's not forget that for the purpose of the contest, we're talking about PHOTO real, so it has to look like a photograph as much as possible. (Of a half-way modern camera and lens).
So in that sense, adding the imperfections of a lens is a must.
I said that, and I mean it, but the first and foremost thing that the contest aims for is a realistic look of the things rendered, including the lighting.
(Look at the 3 bullet points I also wrote to that in the beginning of the thread, to get your quote into some context please.)
That alone makes a render more or less real, depending on the quality of the execution. Already with the rendering, you will determine how much photo real the image looks.
In my personal opinion, I would not stop there, and continue to work the render further in post to add the imperfections of a normal, general purpose camera lens.
You do that too, or you don't. Your call.
Frank
Quote from: FrankB on March 07, 2013, 03:05:08 PM
I said that, and I mean it, but the first and foremost thing that the contest aims for is a realistic look of the things rendered, including the lighting.
(Look at the 3 bullet points I also wrote to that in the beginning of the thread, to get your quote into some context please.)
That alone makes a render more or less real, depending on the quality of the execution. Already with the rendering, you will determine how much photo real the image looks.
In my personal opinion, I would not stop there, and continue to work the render further in post to add the imperfections of a normal, general purpose camera lens.
You do that too, or you don't. Your call.
Frank
In the case of my diner environment the only thing not based in the real world is the camera. The highway is graded properly with displacement maps (so water runs off) according to modern highway code, the signs are scale models built using the U.S. Manual on Uniform Traffic Control Devices (even down to the nuts and bolts holding them on to the steel posts), the diner and other structures are built from measured drawings and photos, the tar seams are placed by tracing an orthophoto, heck, even the junk along the roadside is placed by referring to Google Streetview! As for lighting and atmosphere, I use the U.S. Naval Observatory Sun Position Calculator so that it is accurate for time and place, and refer to local photos of the location for cloud cover and haze, fog, etc.
With all that, I'd say only a handful of the hundreds of renders I've made of the area look like photographs. They do look like the place they represent though. So a closely scaled model environment in itself isn't going to look photoreal on its own.
Since you're judging the contest, not me, it is more important what you consider photoreal, and what you think lens distortion and how much chromatic abberation should be applied, especially since I don't generally postwork my images. What do
you want to see done to our renders, what make
you think a rendered image is a photograph?
Frank do you have to be so much serious? Did you saw the smiley?
I was only implying that with "normal" we could debate endlessly what it means in a humorously way (and had no intent to) and that you understand that...
So yes everybody knows what you mean i hope , but if everything were so easy this thread would have ended in the second post as you said.
I am sure some users will learn something from here.
No smiley this time !
I think you get the smileys in a wrong way.
That is in general a way to show that the user isn't serious and joking.
I am not angry but looking at your post and that you didn't use smileys at all hard to say what you think.
No Greg, you are making things more complicated that they need to be. Let's stop this now, as I can just continue to repeat my previous messages.
For the record, we haven't decided yet if the winners will be determined by a jury, or by the forum users, but in any case I am not the only judge by far, and I believe most people can tell a photo real render from a non photo real one, hence it is not about my perception alone.
It's great if you can achieve that without further work in post.
Regards
Frank
Quote from: Kadri on March 07, 2013, 03:26:35 PM
I am not angry but looking at your post and that you didn't use smileys at all hard to say what you think.
Well look again, I did use one ;)
I thought it was worthwhile clarifying it more, that's all.
Cheers! And I expect you to enter, now that you have already spent so much time talking with me about the contest :)
and an extra one for you: :)
LOL! :)
Quote from: FrankB on March 07, 2013, 03:32:39 PM
...you are making things more complicated that they need to be. Let's stop this now, as I can just continue to repeat my previous messages. ..
It isn't helpful to repeat what you've already said, as that is what confused me. What I would like is clarification. Say it in a different way.
The point I'm making (and I think the point of this thread) is that different people perceive photos differently. Earlier I posted a photo-postcard used to advertise another real diner; that to me expresses the feeling I want to convey with a render for example. It speaks of the time and history of the diner/highway era and conveys the spirit of the time more appropriately than a modern pixel-camera. On the other hand, if I render in black and white, and hand color the render, apply paper texture and lettering I get the effect I want but out of the contest bounds. If I simulate the blown out really saturated film look of the late 1950's, or the paler washed out bleached look of the 1970's it will have a different impact. (Now that I think of it, that might be neat, to make a series of simulated photos from the diner's construction in 1953, to a camera of the 1960's, then have one with cars and film from the 1970's, and so on, a faked photographic history of the place...)
Oh dear, this thread has developed in an unintended direction, my apologies.
My intent was to discuss and share different techniques that we could use to create realistic landscape renders, lighting techniques, visual cues etc and not to get bogged down in the definition of "photo-real". However I admit I have been complicit in just that, trying to work out just what the amorphous concept "photo-real" actually is. Evidently it means different things to different people, and also it appears that a simple answer, "... so it looks like a photo ...", is insufficient simply because as any photographer knows a great deal of their art occurs within the camera it's self and it use. For the purposes of the contest I will continue to develop my image as best I can and if that is not deemed "photo-real", I won't mind, so be it. There are a lot of articles on the net about how to achieve photorealism on a small scale (worth googling for even in our context) but few if any that deal with this on a large scale landscape level so I think there is a worthy discussion to be had on the subject.
gregsandor makes a good point in reply to my post earlier about the difference between a dull snap shot and a dramatic professional shot. We do internally critique the professional one much more than the holiday snap. However making a CG dull holiday snap really would be a challenge! Not a very interesting one I'd have to say but incredibly difficult to achieve. I mentioned Ansel Adams towards the start of the thread and I would much prefer to try to emulate that kind of dramatic style. Much more fun.
Well hopefully we can move on now and actually come up with some helpful TG suggestions.
Cheers
Richard
oh almost forgot ;D
Cypher we need to define what we are talking about first. Otherwise we would all just be looking for that mythical Terragen "Make Art" button.
::) :D
Greg, perhaps for the coming hours/days forget this discussion to touch base with common sense again.
You and I understand perfectly that the postcard you posted doesn't fit the common sense description of photo realistic ;)
Generally speaking everybody shares the same common sense about what's considered photo realistic and what's being looked for in the contest.
There are of course extremes at the boundaries of common sense and that's what this is directing.
It's pretty straight forward and yes there are always questions, which are fine, and speculation on how far you can stretch/pull the rules to see where they break.
That's with every contest, but I think it's both a waste of your time as well as others to discuss this in depth and through significant lengths to see where the aim/goal of the contest breaks by over-analyzing and meticulously picking apart every sentence covering contest aim/goals/rules and their interpretation.
There are a lot of GREAT prizes to win and it's a big opportunity to see and share workflow, knowledge, ideas, inspiration and so on from other participants.
I think it's better to focus and put energy in to that :)
Ok, game on!
Quote from: Kadri on March 07, 2013, 03:52:12 PM
::) :D
We're trying to fix confusion and mis-understanding here as well as getting everyones noses into the same direction here Kadri.
So perhaps for the time being it would perhaps be more appropriate to refrain from kidding around in this specific topic and it will definitely be very much appreciated by us.
Bring back "Button 9"
I really like your idea of exploring the development of photography using the same scene over the last 50 years or so. Not sure if it will qualify but go for it anyway. I think we are talking about how to achieve realism in our renders, getting scale, lighting and detail right and believable. Possibly trying to chase down the definition of "photo-real" is unhelpful and we should just stick to "real". Although thinking about it trying to define "real" has occupied philosophers for millennia and it's unlikely we shall do much better than them in a few weeks. lol :D
cheers
Richard
Replaced my previous comment with an image of a goat wearing a hat. There is no subtext.
:)
[attachimg=1]
Richard
I can give you links which define common sense for you if you like.
I don't feel like posting here anymore, because this is getting very immature.
Quote from: cyphyr on March 07, 2013, 03:55:25 PM
Bring back "Button 9"
I really like your idea of exploring the development of photography using the same scene over the last 50 years or so. Not sure if it will qualify but go for it anyway. I think we are talking about how to achieve realism in our renders, getting scale, lighting and detail right and believable. Possibly trying to chase down the definition of "photo-real" is unhelpful and we should just stick to "real". Although thinking about it trying to define "real" has occupied philosophers for millennia and it's unlikely we shall do much better than them in a few weeks. lol :D
Richard
I've laid out some of the scale aspects of my scene, hit me with yours. I have a "real" environment, now I'm looking to learn how to make the rendered result resemble a photograph.
The purpose of this thread is to
Quote from: cyphyr on March 06, 2013, 06:49:35 PM
So what is it that makes an image photo-real. To simply dismiss it as "looking like a photo" is missing a point. "Real" photo's are manipulated either in camera or photoshop according to the desires and skill of the artist.
Are there tricks we can employ to fool the eye, to make the viewer "abandon disbelief", to make the viewer want to believe this is "real"?
LOTS of detail, great subtle lighting, accurate scale and models all spring to mind. Real world textures also.
...
I don't get why some would prefer to shut down discussion; either add to it or you don't have to participate.
Cyphyr,
As I've collected reference images over the years for this scene I've come across a lot of those old postcards, I imagine they'd have them made up and would place them in local motel lobbies and similar places to attract business. Anyone seeing one in 1957 would instantly recognize them as photographs. I think some were painted or drawn too, but most seem to be photos. So it really depends on what your target audience is. A man of the time would have seen them as "modern" just as we do the things produced today. So if you make a picture in sepia (as you suggested) then folks who grew up in the era when that was "modern" wouldn't think twice about it, and a guy who saw Kodachrome when it was new would respond similarly.
As for techniques in Terragen apart from the modeled scene, we can adjust the camera to let in more or less light, adjust the lens etc. If we're going for a modern photograph I could simply render at 480 pixels and then blow it way up to get it to look like Google Streetview: Crappy quality, but its the real thing! :)
http://maps.google.com/maps?q=clarks+hill+indiana&hl=en&ll=40.252976,-86.699595&spn=0.001646,0.003484&geocode=+&hnear=Clarks+Hill,+Tippecanoe,+Indiana&t=h&z=19&layer=c&cbll=40.252976,-86.699595&panoid=-695YyBTkKnVhipzBxtR2Q&cbp=12,29.69,,0,6.72
Umm I'm not trying to shut down the discussion, anything but, you must be referring to someone else.
:)
Richard
I will amend the rules in the following way:
"Each submission must strive for photorealism as best as possible. "Photorealism" means that the submitted render should look, at close as you can get it, like a photograph made of the scene by any modern camera with a general purpose lens. Submissions that attempt to emulate vintage photo realism, or the look of worn and bleached photographs do not qualify."
I hope that helps.
Frank
Quote from: cyphyr on March 07, 2013, 04:11:48 PM
Umm I'm not trying to shut down the discussion, anything but, you must be referring to someone else.
:)
Richard
I know you're not. I was referring to the ones who were. I'm genuinely interested in this topic, as I am very comfortable recreating real environments in TG, I made my first real-world scale landscape in Terragen v4 or something in the mid-90's. What I want to learn here now is how to use TG2 to create renders that make those environments as real to others who see them as they are to me.
LOL
Actually that photo brings up a very good point, the lighting is almost entirely indirect with a very overcast sky, the shadows are very soft (look under the station wagon).
I've never had much success with overcast scenes, others have.
Is this the actual scene you're trying to replicate?
:)
Richard
On a related note, Greg, let's assume your scene is very accurately modelled like a real world place, that will not suffice to make it a winning render. You will want to have detailed models, non repeating texture tiles, etc, including realistic looking light and shadow, well rendered and shaded vegetation models as if taken by a modern camera and lens, and so on.
Ideally (although that can be hard) a normal, non CG artist, would look at the picture and confuse it with a photograph. Getting a place accurately modelled is a big step of the way, but it's not all.
Regards,
Frank
Quote from: Tangled-Universe on March 07, 2013, 03:52:25 PM
...
... and it will definitely be very much appreciated by us.
Who is "us"?
Martin i wrote some things in a little angry mood but than changed my mind ...
This is the "Terragen 2 Discussion" board and i think some posts here are worth to read for others.
Speaking about photorealism or realism etc. is a nice thing and i would be only sorry about that if it stops.
But the problematic parts about the contest (questions general directly to the contest) would be better to be discussed in his own part on the forum as it seems .
And man smileys can be deathly as it seems. Frak !
Quote from: cyphyr on March 07, 2013, 04:20:03 PM
LOL
Actually that photo brings up a very good point, the lighting is almost entirely indirect with a very overcast sky, the shadows are very soft (look under the station wagon).
I've never had much success with overcast scenes, others have.
Is this the actual scene you're trying to replicate?
:)
Richard
Nah, at this point I'm still gathering ideas and thinking this through. I probably will go for a modern look, but the architecture and even the way the roads are laid out come from an earlier time I want to work that old influence in somehow. Maybe you're onto something with that flat look, perhaps simulating a modern camera but using old-style framing and lighting (overcast sky, etc) would lend a sense of age and speak subconsciously to the viewer.
Simple solution, just run your render through Instagram. Then you get that "magical postprocessing" look that's so popular these days ;D
Participants; better read this again, then decide what to do with your renders before submitting: http://www.planetside.co.uk/forums/index.php/topic,15724.0.html (http://www.planetside.co.uk/forums/index.php/topic,15724.0.html)
Interesting discussion, by the way.
Quote from: Dune on March 08, 2013, 02:38:02 AM
Participants; better read this again, then decide what to do with your renders before submitting:
Yeah, that change was made as a direct slap at my part of the discussion of what constitutes photorealism in rendering in this thread, but the artistic question that Cyphyr asked when starting the thread still stands: "what makes an image photo-real?"
Well, sad to see a bit of crossed wires going on in this thread, but I think we all have similar goals! So let's get back to constructive discussion.
Greg, I think you're trying just a bit too hard to get into the philosophy of this, like trying to define "art". It's a potentially interesting discussion in itself, but probably best for another time.
Basic aspects of landscape realism, quickly defined as per my specific perspective, in rough order of importance to my mind: atmosphere, lighting, practical/geological accuracy, randomness, detail, baseline render quality. Anyone else care to post their own ranking? It does not have to use the same terms as mine, can use more, fewer, totally different ones, whatever. This is just my view, briefly stated. I can argue my perspective if desired, but I'd rather see others post their particular personal summary first.
I do think the discussion about dramatic vs. pedestrian imagery is interesting and potentially relevant. I was going to say that a good approach in scene construction could be to ask yourself "Would I be surprised to see a photograph that shows this?" and if the answer is "Yes", then you have a greater challenge in front of you, if nothing else (i.e. it could still be "photorealistic", but would be harder to make it so). This does ignore the possibility of aiming for dramatic presentation, unusual depictions.
I'll give you a practical example from our user gallery, that I think both highlights the issue here, as well as my points above:
http://planetside.co.uk/images/rapidgallery/slides/f56e3c259ba20200452593a4c6195e88~sauravs-subedi_tgd5.jpg
This is an incredibly simple and, to my eye at least, rather realistic image. The biggest contributors to realism here, for me, are the atmospherics, lighting, and geological accuracy (from DEMs). There is little randomness, not much detail, and the render quality is not even outstanding (or perhaps the image compression is a bit too much, or whatever, hopefully you get my point). This is one particular approach you can take to realism, a somewhat minimalist one, but one that demonstrates the potential power of the "dramatic" presentation. Essentially Saurav used the strengths of the system - atmosphere and lighting - to make up for either TG's own weaknesses in other areas, or the simple challenges of actually building realistic depictions of more detailed elements (e.g. foreground rocks, vegetation, etc.). Of course he was probably also using some reference photos that looked quite like this. The argument can shift in another direction here too, that the choice of subject matter can heavily influence the path you must take to reach realism...
Honestly, it's a very broad topic, but I think I am seeing the pragmatic goal here, and it's not necessarily what I have even been talking much about: the goal, as I understand it, is ultimately to get to some broadly applicable, but specific and useful advice that people can help to improve realism in their renders (improve = make them more realistic)! To which I say, good idea. With that in mind, the above example is perhaps useful if you can consider it as advice to potentially try to avoid or hide elements you are having problems with, but it may also be a rather limiting or defeatist strategy. ;)
So, to be more useful... the question that remains to my mind is, how advanced do we want to start at in our consideration? A good example is the simple but powerful benefit of good lighting choice. It is, I believe, a relatively basic lesson that front-on lighting tends to be very flat, boring, and indeed generally unrealistic-looking. So to my mind that could be lesson #1 (and it even plays into the example I post above, hehe). But is that too basic?
- Oshyan
Continuing the discussion about "realism" (I've dropped the "photo" since it seem so contentious).
Randomness Well the world isn't really is it. There are lots of surfaces that appear random but in truth they follow rules, order etc, dirt tends to fill crevices, crevices tend to form along lines of structural weakness, the list goes on. We use random noise (Power Fractals) to simulate this apparent randomness but by necessity this can only go so far. We need to find a way to simulate the "ordered randomness" we see in nature.If I was working in a normal 3D package (Lightwave, Maya, Max etc) then among other things I could render out occlusion maps which could be re-projected (and inverted black for white) onto the scene as a mask for dirt. Something like this might be possible in TG. Texture the entire scene flat white (grey might be better) and light with Ambient Occlusion only (loose default the blue hue and replace with white). You would then have an image that could be re-projected back on to the scene or used in post as a mask to apply some "ordered randomness".
I'd be interested in other ways of "ordering randomness". ("Two to go easy on the mayo") ::) ;D
Variety One population of the same tree really won't cut it, Three is not much better. There was a thread a while back about adding a global scale texture to a population so there was a variety of textures across a single population. The same is true of "Fake Stones" and "Surface layers", basically the more layers the better. Even if an area is predominately one colour/texture there are subtle variations that may be hard to pick out specifically but their absence detracts from the scene.
Human/Natural interface We can use Terragens powerful fractal nature to create to create pseudo realistic worlds, couple that with using something like World Machine and very realistic terrains can be created (see above for how to surface them in an equally realistic manor). We can also use external programs to create road layouts, buildings, human activity evidence etc. How "realistic" this is is down to the skill of the artist, nothing more. But the interface between the entirely natural fractal world and the entirely human ordered world is tricky. The occlusion map trick or something like it might work although I suspect it may be too subtle. I don't have an answer for this one and I'd be interested in what the board has to say on the subject.
Ok time for some more coffee
Cheers
Richard
PS just saw your post Oshyan (aren't you supposed to be out enjoying your self... ;) ) and I agree with all the points you make
There is no "Button 9," no magical formula or "Make Art" button. Unless the artist has considered some of these things, at least subconsciously, his results will be hit or miss, and mostly misses at that. There are guys who make beautiful paintings and photos and never consciously "know" this kind of stuff but they are rare. Yesterday I ran across an old thread on cgsociety with some guy crowing about running a test of TG2, he said he cranked all the settings up for "godlike" (his words) quality, and in spite of several advices from experienced TG users telling him that it wouldn't produce "godlike" results, he went on. You can give out whatever combinations of quality settings, heck, have Matt code in a preset default for great renders, and still you'll see what you see today in the galleries.
I propose that though it is as technically demanding to produce a pedestrian snapshot as an artistically composed work, the "ordinary" shot is more likely to more often fool the eye. Second, in my experience, even with quality scale models on accurate digital elevation model-based terrains you can occasionally get lucky, but that in itself isn't enough to consistently produce the kind of image we are discussing. Also you must consider the target audience and its expectations of what constitutes a photograph in creating a picture.
Quote from: cyphyr on March 08, 2013, 04:28:32 AM
Randomness Well the world isn't really is it. There are lots of surfaces that appear random but in truth they follow rules, order etc, dirt tends to fill crevices, crevices tend to form along lines of structural weakness, the list goes on. ... render out occlusion maps which could be re-projected (and inverted black for white) onto the scene as a mask for dirt. Something like this might be possible in TG.
You can do exactly this for models, create AO maps (rather something like them) to apply fractal-mixed dust and dirt and grime. It helps "seat" the model in the local earth. The same principle works on terrain features too of course.
[attach=1][attach=2][attach=3]
Quote from: cyphyr on March 08, 2013, 04:28:32 AM
Variety One population of the same tree really won't cut it, Three is not much better. There was a thread a while back about adding a global scale texture to a population so there was a variety of textures across a single population. The same is true of "Fake Stones" and "Surface layers", basically the more layers the better. Even if an area is predominately one colour/texture there are subtle variations that may be hard to pick out specifically but their absence detracts from the scene.
You can texture populations so the color varies by location.
[attach=4][attach=5][attach=6]
Richard, you bring up some very good points and potentially useful ideas. The "Ambient occlusion" approach (i.e. "dirt map", I think) is a particularly interesting one, and I wonder if you could not accomplish something like it with some fancy node work. I believe, for example, that you might be able to create a map of how "rough" the displacement is across the scene, and use this to control texturing. That might be a similar idea. It wouldn't work for objects, nor for the intersection of object and terrain, however.
Greg, "target audience" is mainly what I am arguing against being a particularly important part of this discussion. It feels fairly important to me to narrow down the potentially very broad and diffuse conversation to target the key, most impactful ideas, because otherwise it will be easy to get "lost" in discussion and I think the momentum of the thread quickly diminishes (as I would say has already been seen here at times).
I say that "target audience" is less relevant because we know that at the least the people judging the contest will be in this forum, and I would hope we can agree that we have a *general* sense of people's taste and perspective on "realistic" here, obviously with some possible exceptions (people with unique ideas of realism I guess), people with different taste, etc. But let's keep in mind this is less about "I like that as an artistic piece" (on which I would say we'd have a wider variety of opinions) and more about "I would believe that is a photo of a real place", the latter of which I think many more people (here and elsewhere) would agree on. I think even outside this forum the general perception of "realistic" is not hugely controversial, it usually comes down more to how discerning, picky, and critical a person is, and how much knowledge they have about 3D rendering or at least special effects techniques.
So, to use a practical example, it seems quite obvious to me that your postcard image, with the hand-colored, sketched look, is not what the goal is. Even if it might have been considered realistic or "photographic" at one time (it's arguable if that word even meant the same thing at that time, and then we just get into literal semantics, etymology), it's certainly not now, it is quite clearly more of a "retro" look. Perhaps you brought it up merely as an extreme example (in fact I consider this likely), but just because there can be an extreme example does not mean there is not in fact an already existing - even if not specifically definable - consensus. Even if not, I don't think we need to start at the broadest ends of possible definition in order to get to common understanding, rather, I would be more interested in comparing 2 potentially realistic, "modern photographic style" images, but with different subjects or artistic approaches, and getting some perspectives from various people on whether they think each is realistic. If you want to take the "target audience" approach, perhaps starting with something like the Autodesk "Fake or Foto" tests would be useful:
http://area.autodesk.com/fakeorfoto/
In any case I haven't said "you can't think what you think", only "I don't think a discussion of some aspects is the point of this thread". Maybe I'm wrong. I like the direction Richard began taking above, practical, functional ideas and advice. If you feel a discussion of target audience is critical to that, I won't argue the point further.
- Oshyan
Quote
You can texture populations so the color varies by location.
From the look of the preview screen you have multiple populations each with it's own texture. Or is one population with a texture that varies over it. It's the second option I'm interested in. As memory serves (not very well atm) there was a problem getting the global texture to work with opacity channels whicvh becomes important with complex objects but is less of an isue with simple models like grasses etc.
Oshyan - Blue nodes might do it but I don't think it's just an issue of "roughness". Think of a cliff face meeting a flat horizontal surface, not very rough but definitely the sort of area that would need some sort of dirt map. I do see however that my suggestion of a camera projected inverted AO map is something of a blunt instrument, pretty sure there will be a better and more effective way.
Cheers
Richard
The left hand image (red white and blue corn) is one population, masked (thats why you see the spaces between the fields) with one image to color it. In practice I blend a colormap or orthophoto with the natural leaf textures to produce leaves that have local color variation. I use the method for grasses, crops, and trees.
As for your AO map, if you want to generate one procedurally you can begin with finding sharp angles in terrain, and probably could do it with objects. I'm sure I tried it at some point, but the method I use now has worked best and gives me the best control over the result.
Here's the map I used when I was developing and testing the corn shading I think (its dated 3/28/10, so its been a while).
[attach=1]
And here's the crop mask
[attach=2]
Wow, I missed this thread until today.
I've spent my whole career trying to achieve photo-real imagery. It can be incredibly hard to do because ironically the closer you get to photo-real the more unreal it starts to look right up until you cross over the barrier and achieve the result. This is especially true with CG people. There's even a phrase for it, called Uncanny Valley. That would be a cool titled TG render, -one that's almost photoreal. "Uncanny Valley". I might have to do that.
Anyway, to my eye, the paramount thing is lighting. Everything else is secondary. If the lighting isn't working, nothing else will save it. You can have a plain white box in a white room and if lit properly it'll be photoreal. Things like chromatic aberration, barrel distortion, fringing, DOF and such in my experience are things people do to essentially put lipstick on a pig, assuming they haven't already achieved the photoreal result before adding these secondary filters. I never let artists pile on that stuff until it's agreed we've taken the image to as close as possible to photoreal in the time we have, and only then can these subtle camera lens artifacts be utilized.
Once the lighting is working, I'd say that having good models with the appropriate level of surface detail added is next in the chain. This includes things like beveled edges to catch light, displacements, and enough subdivision to negate faceting, etc. This is something that bumps me sometimes with the Fake Stones. I find at times I can see large facets on them in areas that look unnatural, but because the rest of the image often looks so good, I can overlook it to a degree. It's still there and bugs me though.
Next is good textures and shader work. Textures that don't reveal pixels, don't repeat, are in the correct color space, and aren't unnatural in their range of tone are important. Shaders should have the appropriate diffuse, specular/reflection levels with proper fresnel angles set, bump or normal maps to add further detailing, and something that's lacking in TG at this time which would be a good sub surface scattering option. Certain plants, like succulents for example would be very difficult to do at this time in TG, though most stand alone renderers out there such as Vray support it. This is a more esoteric point though and doesn't hold TG back from doing photoreal. It's more of a want than a need.
After that I'd just say common sense when putting an image together. Make sure things in your scene are layed out and arranged in a natural way. Paying attention to detail and logic are important. This is where artistic flair comes in to play and separates so so renders from truly great ones. I'd say to people struggling to achieve something photoreal that duplicating something you see in a photograph you like would be a great way to approach the problem. Frank B did this with great success on his cumulous cloud render. This approach, using photographic reference and copying it is used all the time in the CG industry and is a great way to force yourself to critically look at things.
Once all this stuff is done and your image is looking like a photograph can one start putting in the lens distortion effects, like flares, DOF, fringing, light wrap, grain, etc. These are all things that degrade the image to varying degrees, which is why artists tend to want to use them right away when their image isn't looking photoreal yet and they start to lose patience or run out of ideas to otherwise take it the image where it needs to be before putting on those final touches. I had an artist once who used to put a heavy diffusion on all his renders because in his mind, that made it photoreal, and it wasn't helping when other artists who were easily persuaded by that kind of hocus pocus would ooohh and aaah when they saw it.
Anyway, my .02 on the issue.
-greg
Hi guys, this is a great topic and I'll try to chip in with a few tips and hope they might be a little useful.
Oshyan's tips on lighting and atmospherics are absolutely spot on, and also highlight a lot of the core strengths of the terragen renderer.
In most other 3d apps, depth and haze have almost always been achieved in post using the zdepth pass, this always frustrated me as it a very linear method of producing depth, and lacks realism as well as the lights relationship to the atmosphere.
The ability to layer atmospherics in terragen can produce far more realistic and natural results, it can even be used to simulate dust and random particulates in the air, so in that sense I often like to employ several isolated cloud layers to create different effects at different depths in my scenes.
It's also good to know the limits of your render engine and discover ways of circumventing the issues, for instance, the lack of shadow and colour detail in gi areas. Before the advent of better light simulation in apps like arnold and vray, we always used carefully placed bounce and fill lights to compensate, and while it can take some time, trial end error to get good results, it's still a very valid technique. I'd say that in most situations it's very tough to get away with simply one light and gi, and it's worth taking the time to experiment with lighting techniques, placement and even switches on the lights, ie using lights that only provide one or two forms of contribution like specularity/ diffuse/ atmospheric contribution. One thing we almost always employ in lighting is the use of our shaded spheres, 50% diffuse, fully reflective, and glossy etc as well as macbeth charts, these can be really helpful to have in your scene when lighting, as they allow you easily cross reference and check against any kind of reference plate, as to weather you are providing the right kind of values to your lights.
In terms of realism, shading is also crucial, and requires much study of your subject, how does the light react to the surface in terms of levels of reflectivity, glossiness and diffuse reaction to light, and how does it break up and scatter? going the extra few yards, say with road textures for example, to use maps that break up the specularity / reflections and diffuse contribution, can produce much more realistic results. Another tip here regarding textures, is that we almost always employ a linear workflow (which I'll avoid going into too much detail about here), terragen automatically converts your textures to linear space, which is great as it helps produce a more accurate response to the additional values of light, but you also have to look out for the colour saturation levels and not just the gamma. A lot of colour is reproduced by the addition of light, and more realism can be achieved by using slightly desaturated tonal values that work in conjunction with your light sources.
In all these situations, while trying to achieve specific realistic results, referencing your subject and type of image can't be underestimated, doesn't matter how experienced you are, without close study of relevant reference, it can be very hit and miss if you're looking for realism.
Richard makes a good point about bringing in maps such as AO to help create natural effects, and it's always worth exploring how we can create any kind of rule that lets us get into a specific part of a surface, I often use zbrush for it's multitude of different masking types, which are great for isolating different types of areas.
In terms of photoreal finishing, and matching plates in film, the end result is always done at the composting stage, where all the lens artifacts and specific attributes regarding the cameras used are matched as closely as possible. The matching of the camera types and lenses is very difficult to achieve in pure rendering terms and when looking for those types of results from your renders it is worth looking into how a lot of those effects are recreated at the post stage. Plenty of info out there.
Hope it helps, if just a little.
Chris
Also, +1 to everything Greg just said :)
Where do you work Chris if you don't mind my asking. You obviously know exactly what you're talking about.
-Greg
Wow, two great and very informative posts, thanks guys. A lot to take in here but this is exactly the kind in info and discussion I was hoping for in this thread.
By the way I had a quick go at creating camera projected AO maps of a TG scene and using them to apply a grunge texture into crevices and hollows. Not great results as yet, maybe working at higher res and detail will help.
Cheers
Richard
Oh, a good point that Chris brought up is the use of gray balls and calibration images. Professionals use these all the time to determine if their scenes are properly balanced. We employ the use of calibration scene setups that consist of Macbeth charts, spheres with 18% gray values added to them so that we know right off the bat if our lighting is even in the ballpark or not. Key to fill ratios are easy to set when you're looking at uniform shading values.
Another technique is to just put an 18% gray value on everything in your scene and then light it. This is more a technique used in stand alone renderers to help with lighting but it can be useful in TG because these surfaces render very quickly and are good proxies for how a camera views the world from a luminance perspective. You can quickly "sketch out" your lighting angles and find problems that might otherwise get hidden in shaders and textures. Once you're happy with the overall look of the lighting, again what I consider the most important aspect of the image, you can start layering in the textures and shaders. If problems start to pop up at that point you now know where to troubleshoot as you've already determined that your base lighting is correct.
-Greg
Great info guys, this is shaping to be a very, very useful thread. :)
- Oshyan
Hey Greg,
No problem, I'm currently at Iloura in Melbourne, lighting on the great gatsby.
Still learning from some great lighting artists :) as I usually work as a generalist.
Back on topic, I'd just like to add that there is so much information on the net regarding lighting that's worth researching, and its pays dividends to try a few new techniques to expand your knowledge of lighting.
Here's a good one that also references someone you all know, Ansel Adams.
http://www.scratchapixel.com/lessons/3d-advanced-lessons/things-to-know-about-the-cg-lighting-pipeline/0-18-stop-and-exposure-explained/
Chris
Chris's link is great example of the linear workflow. A lot of work has gone into facilitating this approach at the various effects facilities around the world and is pretty much standard practice everywhere. Artists, and especially lighters routinely speak in the same terms that DPs and photographers do these days with regards to lighting. It's not at all uncommon to hear someone say "make that two stops brighter", or use the gray ball to calibrate your light intensities to match what was shot on set.
Terragen pretty much has already built it's world internally to take this all into account, but it's still a good idea to understand this stuff if you're striving for photorealism because at the end of the day we're trying to make our images look like photographs (assuming that's your goal of course) so it's handy to understand why photographs come out the way they do.
As I said a lot to take in but one thing occurs to me right away. The techniques you have both described, Macbeth Charts and various grey/reflective balls in particular are for matching virtually created environments (and elements of those environments) to a real world image or film stock. Will this even work in an entirely virtual environment? If I drop a Macbeth Chart into my Terragen render what do I have to compare it to? I *think* I should be able to simply compare the luminance and colour values from the rendered image to the source Macbeth chart and adjust exposure accordingly.
Hmm much to ponder ...
cheers
Richard
ps: Uncanny Valley would be a GREAT title for a contest :)
Yes, if you took a Macbeth Chart and dropped it into your TG scene it should be useful as a calibration tool see what your exposure is, but when doing all CG environments it's all relative. The chart will tell you however if your TG assets and lighting are calibrated properly in relation to real world lighting or if you were cross pollinating TG renders with renders from other programs that went through the same calibration process to combined in a compositing program.
Macbeth Chart & grey/reflective balls
Dose someone have a link to a good resource on reading about these things; what they are, how to use them, both in relation to compositing film and CG, as well as using them in a complete CG environment.
I have seen both many times in threads on line, but never with much discussion of them.
The so called Macbeth Chart sounds like it comes out of photography ("middle grey"), so Im interested in seeing how to use it in CG.
Check the link Chris posted a few posts up, it's a whole series of articals, pretty technical but understandable in the most part.
Richard
A lot of this stuff is mostly useful when you're trying to marry CG rendered stuff into live action plate photography, but that's not say all CG stuff won't benefit from it.
When we go out on set we bring all this stuff along because it helps us when we get back to the facility later with the calibration process. The Integration Department uses this stuff to make sure HDR latlongs are properly graded, and having the gray balls and Macbeth charts in the latlongs allows them to make sure the exposure is right and the grades are neutral.
Once that's done and we know that the on set HDRs are properly color corrected we take them into our CG environments knowing that IF we use these images on dome lights with their intensities set to 1 that it'll be the correct luminance value to start lighting our 18% gray valued CG models and environments. At that point it's just a matter of apply the shaders to the assets, which have already themselves most likely been developed in a calibrated "look development" scene that contains a latlong attached to a dome light with the same settings as our CG scene light. The assumption is that if the asset created in the look development scene was made from the same baseline lighting intensities that you can just plug them into any new CG scene and it'll work, assuming of course you new CG scene lights have also been through the same calibration process. This save a ton of time and allows easy re-use of assets from previous shows without having to redo a lot of work.
Terragen already pretty much enforces this workflow because everyone is developing their scenes and assets with the same lighting model and shaders, so if someone does a good job building and texturing a tree for example and uploads it to the forum it'll be pretty much guaranteed to work in any other TG scene out there unless someone has really done something whacky with their lights or sun settings.
So we're really back to "that art thing" as one of my studio managers used to call it. Terragen's lighting and atmosphere model is pretty good to begin with so when it comes down to it making a good picture is a matter of having a good subject photographed in a visually interesting way. Composition makes or breaks it.
I haven't read through this disucsiion fully so I apologize if someone already said this. From my own experience, another way for a picture to look real is if no one suspects it to be fake. If I make a picture of a planet or of some scene where the viewer knows a camera could never be, then people know it is CGI and tend to notice where it is different unless of course the artist did an extremely excellent job. However, if I made a Terragen picture of say my back yard or of some other place that is real, people would not expect it to be fake and would more in likely believe it even if it had the same flaws as the picture of the planet. Of course, these "flaws" are more in likely picked out by CG artists such as those on this forum. Again this is just from my experience.
Quote from: gregsandor on March 09, 2013, 11:19:41 PM
So we're really back to "that art thing" as one of my studio managers used to call it. Terragen's lighting and atmosphere model is pretty good to begin with so when it comes down to it making a good picture is a matter of having a good subject photographed in a visually interesting way. Composition makes or breaks it.
I haven't read through all this thread except I saw gregsandor's post. I don't think art has anything to do with making something photoreal. Art will make it interesting to look at though. You can be very skillful at making the picture look real but putting art into it is a different matter. Some TG2 images have a lot of art about them but aren't realistic looking. Maybe they just have a good mood. It's when you get both these things right that you get a really great render.
I think you need detail. Real landscapes usually have a huge amount of detail. You need sense of scales i.e lots of different scales of things we know about. For example, If it looks like sand we know what size that is. I think the scale thing lets down a lot of TG2 images that otherwise would be convincingly realistic. Lighting is important but TG2 pretty much handles that without too much tweaking. Atmosphere is also quite important. How thick mist is etc. Then make it look like a photo. How a lens reacts to light, exposure, how photographers frequently post process photographs, those sorts of things. People are used to looking at photography.