190h05m and still rendering

Started by TheBadger, February 28, 2012, 09:53:25 PM

Previous topic - Next topic

TheBadger

QuoteThe patch size is something I remember being suggested to you to make your walls work.
I didn't say it's not good. I only included it in the list of possible reasons why this takes so long.

I know. I was thanking you for helping me solve the problems with my walls because you said they looked good. It means a lot to me that you spend so much time helping people, It really is a good thing.
This is the only forum on the internet I post on. I don't find other forums on this subject are as knowledgeable and friendly. Its one of the biggest selling points of this software, and goes a long way to making up for the lack of in-depth tutorials. But I know you guys are working on that too 8)
It has been eaten.

Kadri

#16

The walls with the ivy's looks nice TheBadger. Curious about the final render :)

TheBadger

Hi Kadri ;D
lol, I haven't forgot about you and your help, Im really looking forward to you seeing the finished work. I'll be glad if you like it, being that your work helped me to understand how to to it.
It has been eaten.

Kadri


Everybody does help each other here  :)
I feel like a beginner in my on going *#&½ still not finished animation.
It will barely useful i hope . Assuming naively that it will be OK and hoping the best is not the best way to do this kind of work .
What i learned ones more is that i should had payed more attention to posts here about animations and posted some questions right at the beginning.
I like this place TheBadger and because of your last post i had to write all these unnecessary things  ;D

TheBadger

Its all about the work, and the people who make it.
It has been eaten.

Dune

Incredible piece, Badger. So you probably saved your walled terrain and imported that in IvyGen? Your walls turn out very tight (nice, I mean). And the wall base is probably a repetitive mask? I wondered if it wouldn't have been easier (a bit late to mention this if I see your render time  ;)) to make an wall object and repeat that as a population. With 2-3 sets of ivy grown on that one wall set, it's easy to give the impression of different growth by mixing them with some blending PF. Just wondering, though.

inkydigit

agree with Ulco...
looking excellent so far!

TheBadger

Quote from: Dune on March 01, 2012, 03:18:52 AM
...So you probably saved your walled terrain and imported that in IvyGen?... And the wall base is probably a repetitive mask? I wondered if it wouldn't have been easier... to make an wall object and repeat that as a population. With 2-3 sets of ivy grown on that one wall set, it's easy to give the impression of different growth by mixing them with some blending PF. Just wondering, though.

Hi Dune,
Yes, you are right. But, I wanted a global environment from which I could make a near un-limited amount of shots from. I also wanted to stay entirely with-in Terragen2 for the terrain for a lot of reasons. I will have to do a lot of work outside of TG2 to finish but at least the main body of the project is in one program. This is just one shot that will be on screen for 5-6 seconds, but its a very important establishing shot.
If you remember, you helped me get started on this project, and helped me trouble shoot more than a few times, so I'm happy that you like it. I don't want you guys to think your waisting your time on me.
As for me, I have been looking at this project for three months and I still like it, that tells me something good.


@Inky, Thank you.
It has been eaten.

FrankB

everything has already been said, but I would like to say this is going to be a great scene!

Is the maze procedural?

TheBadger

#24
Quote from: FrankB on March 01, 2012, 01:59:53 PM
everything has already been said, but I would like to say this is going to be a great scene!

Is the maze procedural?

Hi Frank,
I am not where I should be with my vocabulary, where the 3D world is concerned. A shame, that makes it hard for me to ask my questions intelligently and efficiently. Also to answer back :-[

I believe that I can correctly say that the terrain hills are "procedural", but the terrain details ie., the walls are imaged mapped for shape and size, as well as detail. Additional colors were added via power fractals.
I do not know if image mapping can be regarded as procedural?

I hope you will like how your clouds work in the final image, Frank. They added a lot of personality to the image, and did wonders with the light.
It has been eaten.

FrankB

Thanks for the clarification. So in your case the maze isn't procedural, which means it isn't based on any mathematic formula. Doesn't matter, I was just curiuos.
Looking forward to the final!

Cheers,
Frank

Oshyan

I'd love to see this if it's finished, it looks quite interesting.

A lot of good advice and feedback has been given here, but I want to weigh in on a few things.

As Martin said, the detail setting makes a big difference for relatively small (numerical) changes. When your detail scale is from 0-1, a change of even .1 (e.g. from 0.5 to 0.6) is large relative to the total range of the setting.

I hadn't noticed GI surface details being enabled in your original settings post or I definitely would have mentioned that. As again Martin said, it is seldom worth the render time hit and should generally be avoided. Basically I see it as a last resort when I am otherwise unable to get the detail I want with other combinations of GI settings. In other words, investigate all other options before using it, and definitely try renders - or at least crops - without it first, using settings that are otherwise "final". Only enable it if you feel something is "missing" in the GI (small-scale shadow detail), and even then you should do several small crops to make sure it makes a difference and a positive one.

Seeing the size of your object, I suspect that actually is also contributing to render time. If I understand correctly, you have a *several GB* OBJ file? That's pretty crazy, hehe. Do-able, but definitely taxing to the system.

The patch size no doubt also has an effect. I'd be curious just how much, and how much increasing it would affect your scene results visually. But with the high render time you're probably not interested in experimenting. ;)

Finally, regarding raytrace atmosphere, I think it's worth saying that Martin is right in that it is not always beneficial, however I must disagree with his blanket statement that when using high(er) AA for vegetation, it automatically makes raytraced atmosphere undesirable. As I am just now finishing rendering of the Garden of Eternity animation, I had the need to do a lot of testing and render optimizations and my tests were unambiguous and detailed: raytrace atmosphere was faster for equivalent or better quality in all areas of this scene. This is using multiple volumetric cloud layers, one in particular that creates low-level mist, and is a source of a lot of noise.

Where I think Martin's approach may be making his own tests skew the other direction is the use of "Max Samples". I did try this with the Garden animation but, while it may be suitable for stills, I don't think it works as well for animation. Ultimately I used AA8 with 1/4 first samples and a 0.02 noise threshold. With this, using raytraced atmosphere was faster *and* less noisy, by far. I suspect if I was using Max Samples the results would be different however.

I also want to mention something I discovered while tuning the Garden scene render settings. *Increasing* the number of samples for a cloud layer actually *decreased* render time (significantly) when using raytraced atmosphere! This seemed counterintuitive to me at first, but I realized it's probably because of the adaptive AA settings. There is a relatively low noise threshold (0.02) and relatively high max samples. The AA sampling algorithm is therefore throwing more samples at the problem, up to its max level, in an attempt to reduce the noise coming from the cloud layer. Attacking the problem directly by increasing sampling of the cloud layer alone focuses the sampling where it matters, on the scene element that is creating the most noise. Thus you get reduced noise *and* reduced render time. In fact, at 32 samples, the quality is 2.8 for that layer! Yet it's faster than 0.5 quality. Something to keep in mind.

All that being said Martin is definitely right that RT atmosphere is not right for all situations. Hopefully we can come up with more concise guidelines on all this as it can be rather confusing and time-consuming to test lots of settings to get it right.

I do hope you'll post the results of this lengthy render!

- Oshyan

Dune

Interesting observations, Oshyan, especially the ones concerning AA and cloud quality. Something to keep in mind indeed. And....
QuoteAs I am just now finishing rendering of the Garden of Eternity animation
:D :D :D

Tangled-Universe

#28
Ghehe Oshyan, you keep on trying it he

http://forums.planetside.co.uk/index.php?topic=13679.15

I think for everyone here it's interesting to read this topic.

My tests aren't skewed, they are real life examples of what one will use.
Regardless of whether I use full sampling or default 1/4th, since I stopped use RTA everything renders faster.
The arguments and proof are in the link above ;)

Now I'm going to be an ass, but if you still feel it's not true what I'm stating and what my tests have shown then I think the ways you use it is either in a too advanced way for many many users, but even more you should feel the urge to properly and finally document this all ;)

TheBadger

Oshyan,

Hello,
Thank you for such a in depth post on this matter. I think between you a T-U, you guys may just be able to get my brain to understand this stuff. If you can teach me you can teach anyone.

Ok, so the render in question was quit at 320+ hours. There is just a small area in the bottom left corner that did not finish. I am currently rendering a version with all of T-U's settings + I lowered soft shadows to default settings .
The curent render is almost finished, at this moment it is at 25h22m28s. I expect it will be done when I wake up.

I had intended to not post anything of this project until it was completely finished and as close to perfect as I could ever get it, including creating, animating, compositing my characters. But clearly, there is a great potential for me to really learn something from you guys if I post the WIPs. So I will. A little sad, I was looking forward to "the big reveal" ;D

I also wanted to render the image with RTO off. And now based on your last post TRA on, with the others off. SHould I do all that first? Or do you think you can tell me what the results would be just from the 1st and 2nd renders? Also, I have a lot of questions and statements to pose based on your last post, but I'll wait until I post the renders so it will be easier to answer.

One clarification though. I do not have a single object that comes in at 2 GB, I have 250+ individual IVYs, that =more than 2 GB + several populations of thousands of other plants. I'm not sure if thats what you meant or if it makes a difference, but yes, The file on my hard drive that the ivy are kept in is 2GB+.

Please let me know if you want me to do the other test renders I mentioned, or if I should post the first two only.

Thanks a bunch for yours, and everyones interest in this project, it really helps to motivate!




It has been eaten.