Some Work in progress...

[orsobruno]

Hi! this is my first post in this forum, so i would like to thank everyone who take part at the process of dovelopment of this amazing software!

Right now i'm testing Corona starting from some old v-ray scenes, converting materials by using the converter created by DeadClown (wich is amazing!! Thank you very much about that!!), and than by adjusting the properties of the converted material...but i'm trying to make as much better material as i could, so i don't have to spend more time later on other scenes with the same material!

This is some work in progress of some jobs i'm on it right now, so they are not finished and i can't public the whole project until the client say it's over :)

But i'm now using corona as my main renderer, so i'll try to keep this post up to date as much as i could!

I would also like to know if i can take part at the development part, of the program,  by having access to the daily builds...it would be nice for me to help the developers testing out the new stuff!
Mainly becouse i see on facebook that distributed render is available to daily builds testers...and i would really like to speed up rendering time by using other machines...i don't know if ti would be possible, but, in any case, i would really like to thank you all for this amazing software...keep it up!!

ps...any suggestions about everythings (light, materials, render settings...) it's more than well accepted!!

(sorry for my english...)

[orsobruno]

Here is another image made today...as always any criticism and comment are welcome!

Hope you like it!

[yagi]

the 2nd image with sofa is amazing.... but the 25hr render time is not for such a system configuration. was that intentional? did u only use hdri too?

[orsobruno]

Hi yagi! Thank you for your kind words...Unfortunately it was not intentional, i'm quite new to this rendering engine, and i don't know very well how to optimize render settings to make it work faster, i left pretty much everytinghs untouched except changing the secondary solver to hd chache and rise up the amount of the max records to somethings like 5000 (i read on the forum that rising it up could made something good, but honestly, i don't remember what it could do!)
If i remember correctly i also rise up the light sample multiplier to somethings like "8" and the texture resolution to 1, and i think the path tracing sample to 40 and the max sample intensity to 10 (for the same reason of the max records.........)
For the illumination it was made by using an hdri in the corona enviroment rollout, overriding the max one... (the texture was load by using a vrayHDRI...)
Also the DOF was rendered by corona!

i don't know if i act correct, but probably i made somethings wrong, becouse 25hr is a long time for production, do you think i made some "settings mistake"?
i would really appreciate some help :)

anyway, thank you again for your interest in that image!

[romullus]

Having in mind huge resolution of your image, rendertime is not so insanely high. If you'd render fullHD resolution, rendertime would be ~ 3 hours. It's not that bad.
You could've leave hd cache max records at default - it's very rarely needs to be changed. As far as i know, it adds nothing to quality nor to render time (in most cases). Since you raised PT and light samples, you leaved less sampes to antaliasing. Maybe it's dof that has most of noise at the end of rendering? Also, did you put portal in window's opening? Curtain seems to block a lot of light from outside, maybe it have not ideal materia settings? Can you show it?
Btw, i don't see a reason in your scene, why should you lowered MSI from default. If anything, i'd leave it at default or even try to rise it to 40-100, although it doesn't affect rendertime.

[orsobruno]

Hi romullus!! thank you for your answer!

Ok from now on i will left the hd cache value to the defoult setting :)

About the noise question i could say that after 16hr the render was almost done (not at that quality, but acceptable!), but the dof was still noisy, so i decide to let i calculate during the night time!

Sure i can share the curtain material! right now i don't have this image on this pc, but i'm pretty sure it is a default coronaMTL with translucency enable with a value of 0.28/0.35, a map in the bump slot with a very low value, wich is barely visible due to the dof (i could avoid it :) )...i don't remeber if i change the reflection value, but i don't think so!

About the portal at the windows i put a plane of 1 polygon x 1 polygon with the portalMTL applied to it, and the face of the polygon oriented in the direction of the scene!

Regarding the MSI value today i tested some value and, yes the value that you suggest it works very well!! i never rise up above 40 but now i will try 100! thank you very much!
I realized that if i change this parameter some very bright withe point in the render, will disappear, and the image still render in the same time, is this only a lucky case, or this is how it should work normally?

Thank you again for your answer! it will become very useful!! thank you again :)

i think that now the only thing that could speed up the rendering time is the distributed rendering!

[romullus]

About the noise question i could say that after 16hr the render was almost done (not at that quality, but acceptable!), but the dof was still noisy, so i decide to let i calculate during the night time!

In that case you should lower PT and LSM a bit. Very often default values works pretty well.

Sure i can share the curtain material! right now i don't have this image on this pc, but i'm pretty sure it is a default coronaMTL with translucency enable with a value of 0.28/0.35, a map in the bump slot with a very low value, wich is barely visible due to the dof (i could avoid it :) )...i don't remeber if i change the reflection value, but i don't think so!

As long as you didn't touch refraction, rest of parametters shouldn't cause any problems.

Regarding the MSI value today i tested some value and, yes the value that you suggest it works very well!! i never rise up above 40 but now i will try 100! thank you very much!
I realized that if i change this parameter some very bright withe point in the render, will disappear, and the image still render in the same time, is this only a lucky case, or this is how it should work normally?

The higher MSI is set, the higher is risk of getting very bright dots (fireflies) in rendering. Lower values are safer, higher values gives better GI.

[orsobruno]

The higher MSI is set, the higher is risk of getting very bright dots (fireflies) in rendering. Lower values are safer, higher values gives better GI.

Thank you very much romullus, that's the information i was in searching of :D that's very helpful to me! thank you again!