Limitations of (photo-realistic) rendering

Become a member of the CGSociety

Connect, Share, and Learn with our Large Growing CG Art Community. It's Free!

THREAD CLOSED
 
Thread Tools Search this Thread Display Modes
  04 April 2013
Originally Posted by feed3r:
What's technically missing then (given the reasonable assumptions, that you can't spent unlimited time and money on a project and regardless of our artistic capabilities and the fact, that realistic is only subjective perception - I know, this will always be an effect, but as I said, I'm especially looking the technical side)?

To be honest I have a hard time thinking of anything for you to write about.
We cannot even suggest hardware or location limitations much longer...
http://home.lagoa.com/
 
  04 April 2013
Originally Posted by feed3r: Thanks for your valuable response, guys. I'm exactly on your page.

In my thesis I'm especially looking at the technical side of today's (in-camera) rendering techniques. That's why it's totally fine to say, that non-post-images are not photorealistic at the time. Yet I'm wondering, what is missing then.

Let's bringt Maxwell Render to the table: They claim to be absolutely realistic, having features, that are usually done in post, such as lens effects, like chromatic abberation, bloom etc.

What's technically missing then (given the reasonable assumptions, that you can't spent unlimited time and money on a project and regardless of our artistic capabilities and the fact, that realistic is only subjective perception - I know, this will always be an effect, but as I said, I'm especially looking the technical side)?


Nothing is technically missing, its weather the client wants to pay for your time and money to do it. You also seem to be forgetting that all the listed features are done in post so you can change them at client will.
 
  04 April 2013
Originally Posted by circusboy: To be honest I have a hard time thinking of anything for you to write about.
We cannot even suggest hardware or location limitations much longer...
http://home.lagoa.com/


Thanks circusboy. Lagoa looks quite promising. Don't worry - actually I've got plenty to write about (rendering/shading-features & techniques that are in use right now) - it's really only about my conclusion, whether I can confirm our (the renderer's) ability to create photo-realistic images or not. And if not, why.. and so on. I'm still struggling, but I guess it turns out that there are no (or only very few) technical limitations, but depending on the complexity one may need to put an unreasonable amount of time (money) into a project to make it photo-realistic.

Last edited by feed3r : 04 April 2013 at 03:58 PM.
 
  04 April 2013
Originally Posted by DucK-: Another vote for food, especially when you're working with substances that blend into each other. We had to work on animated closeups of dumplings in creamy soup a while ago. Displaying dumplings soaking up the soup to various degrees was quite challenging. Was probably due to the fact that we had no accurate enough way of controlling the changes in substance volumetrically. If it hadn't been for some compositing work I doubt the results would have looked as realistic.


Yeah definitely. Especially when I see cg movies, the food always looks like plastic. Also something else I think has a long way to go as far as true realism, which some would argue is a good thing, is gore. For example, if you have a limb that's cut off, besides needing fluids and a shader for blood, you need to show all of the anatomy, where it was cut off at.
 
  04 April 2013
Originally Posted by feed3r: <snip> I'm still struggling, but I guess it turns out that there are no (or only very few) technical limitations, but depending on the complexity one may need to put an unreasonable amount of time (money) into a project to make it photo-realistic.

Or as a artist with a deadline. Stop rendering and start painting to get that last 10%. Also for cases of 'artistic license'. Its not 'accurate' but it looks 'nice' or the client wants that visual feature hell-or-high-water.
 
  05 May 2013
My apologies for resurrecting this thread from last month but there were a couple of things I wanted to add but was too busy to make my brain work properly.

I agree with what a lot of people have said, namely that convincing/appealing food is probably one of the hardest things to get right. I also get what Jaco is saying about shooting and retouching being the better/faster/cheaper option in most cases and this is true, especially when you need to match a specific product.

However, more often that not you're being asked to create something that needs to look 'photo-real' but would also be extremely time consuming (or impossible) to create in the real world. A convincing jar of honey is one thing, a bear made from honey something else.

SSS and the general speed increases in rendering refractive and transmissive materials have helped a lot but convincing liquids or porous objects are still a real challenge.

Originally Posted by ThE_JacO: BTW, Sponges are actually not that hard to do volumetrically. Rather easy in fact.
I think this could be something that could be application specific but currently with the tools I/we use this is still really hard to achieve. You can cheat the look of sponge with SSS but creating a volumetric shader or surface that holds up to close scrutiny is still a huge pain, for me at least
__________________
SKETCHBOOK
MDI
moonjam.com
 
  05 May 2013
Time to learn Houdini then, Alex
__________________
Come, Join the Cult http://www.cultofrig.com - Rigging from First Principles
 
  05 May 2013
Damn.

But will Houdini give me fluid dynamics as good as this..?

*Bursts into tears*

__________________
SKETCHBOOK
MDI
moonjam.com
 
  05 May 2013
Originally Posted by feed3r: Thanks for your valuable response, guys. I'm exactly on your page.

In my thesis I'm especially looking at the technical side of today's (in-camera) rendering techniques. That's why it's totally fine to say, that non-post-images are not photorealistic at the time. Yet I'm wondering, what is missing then.

Let's bringt Maxwell Render to the table: They claim to be absolutely realistic, having features, that are usually done in post, such as lens effects, like chromatic abberation, bloom etc.

What's technically missing then (given the reasonable assumptions, that you can't spent unlimited time and money on a project and regardless of our artistic capabilities and the fact, that realistic is only subjective perception - I know, this will always be an effect, but as I said, I'm especially looking the technical side)?


Every lens and camera ever created will be subject to its own subtle flaws. There will never be a technology to recreate the EXACT way each individual camera is tweaked as it is impossible. Maybe the pure way our eye sees light, but even then can you compensate for each viewers eye discrepancies?
 
  05 May 2013
Imagination is the Only Limitation.

To wit, there is nothing that cannot be done with 4k Photo-Projection and cards. lol.
__________________
"Your most creative work is pre-production, once the film is in production, demands on time force you to produce rather than create."
My ArtStation
 
  05 May 2013
BTW, about food -

http://renderman.pixar.com/products...atatouille.html

Wasn't 100% photoreal but close, and they did manage to make it look mighty appealing.
 
  05 May 2013
such an interesting topic,
actually if you want to look at things in scientific terms and definitions it can get really complicated.
cameras will continue to evolve and as "scrawford" mentioned they all have their distinct flaws and that's only one thing.

while the effects are magnified in refractive materials in my opinion the biggest drawback are the basic "shaders" we take for granted or how a surface will reflect light.

one of the easiest materials to get looking right is clean chrome because it reflects like a mirror but how about a metal with a double layer of different paints one of them partially oxidized or nonconducting and with some dust on it. that will not behave like any simple mathematical abstraction or any lab material.

basically all real world materials are so complicated to calculate and my guess a lot of render engine makers simply pick a computer graphics textbook and implement the same methods again and again because it would be too costly to rewrite new mathematical equations or do genuine broad scientific research for it, we are talking of course about very subtle differences.
__________________
Ebal Studios
 
  05 May 2013
Thread automatically closed

This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.
__________________
CGTalk Policy/Legalities
Note that as CGTalk Members, you agree to the terms and conditions of using this website.
 
Thread Closed share thread



Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off
CGSociety
Society of Digital Artists
www.cgsociety.org

Powered by vBulletin
Copyright 2000 - 2006,
Jelsoft Enterprises Ltd.
Minimize Ads
Forum Jump
Miscellaneous

All times are GMT. The time now is 12:00 PM.


Powered by vBulletin
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.