Chaos Corona Forum
General Category => Gallery => Topic started by: racoonart on 2013-07-03, 23:54:56
-
I've been showing Corona to a friend these days and we came up with this little rendering. His model, my lighting and rendering :)
Hope you like it.
took 30 mins, 110 passes, 3d dof.
-
Its great like always :)
-
Super!
-
bug looks OK , but the ground is amazing ;)
-
Love it !
-
Super!
-
Nice, looks like Dwemer technology from Elder Scrolls. ;)
It's strangely cropped, I would give it a bit more space to the right, it's somehow disturbing but maybe only for me.
-
...it has to be a so dangerous mosquito!!! ;)
Cool image!
-
Well done, really nice Image!
Liebe Grüße aus Wien
-
Nice, looks like Dwemer technology from Elder Scrolls. ;)
First thing that came on my mind too :- ) It was 10 years I think since I played Morrowind...and stopped with games altogether.
Very nice image
-
How did U make this little blur around highlights? Is this from corona or post-pro??
-
Amazing! I cant believe this only took 30 mins to render. Care to share your hardware mate?
Cheers!
-
Thanks a lot for all your comments! :)
Nice, looks like Dwemer technology from Elder Scrolls. ;)
It's strangely cropped, I would give it a bit more space to the right, it's somehow disturbing but maybe only for me.
Never had that in mind (although I've played elder scolls games :D)
Hm, yes, all in all it may not be very well placed but I didn't plan to do a "real" image when I started to play around with it ;)
How did U make this little blur around highlights? Is this from corona or post-pro??
It's post, in eyeon fusion :) Some stacked glows which add up to a very nice and soft highlight.
Amazing! I cant believe this only took 30 mins to render. Care to share your hardware mate?
I rendered this on a dual xeon E5530 @ 2.4 Ghz. Not a very fast machine today. New i7s tend to be a lot faster ;) I tried to optimize the settings a bit, this helped a little in some areas.
-
Deadclown, have you thought of compiling your current knowledge of Corona into some short blog post topic ? I would be very interested to read that.
Something like "I found X to Y sample in PT to work best in XY scenario" etc.
The current know-how seems to be extremely scattered and chaotic across the forum.
-
To be honest, I don't know what I should write in this article...
My whole optimizing is more or less balancing pt- and light samples and passes. There is not much more I could do in Corona. So I try to find reasonable values for pt and light and passes.
For example:
Normally I tend to believe that 100 passes should be enough (non-adaptive) anti-aliasing (this would mean 1-10 min-max dmc image sampling in Vray or more precisely 10 subdivs fixed rate). Which means that in 100 passes light and pt passes should be clean.
So I test how fast light and pt noise is dissipating and how they are related in specific scene, sometimes I need to crank up pt samples to 64 while leaving light sampling as it is.
In other very difficult lighting situations it's the other way around. HdCache can do most of the GI problem areas so I don't need to invest a lot of samples in pt - but light sampling through a lot of little windows is difficult which means I easily end up using 4 pt and 32 light samples...
That's all the magic ;) And a little bit of scene optimizing of course, checking bump values in material and refractions modes and so on.
I really would not write about "I found X to Y sample in PT to work best in XY scenario" cause I believe that each scene is best handled individually and should be optimized the way I wrote above ;).
-
Yes, I think it would be very helpful to create a new base of Corona knowledge. Or at least a faq. Much has changed since a4. Someone else also thought of this (at least I think so ;) ).
http://forum.corona-renderer.com/index.php/topic,1161.0/topicseen.html
-
I think the problem is that corona is still changing a lot and we should be very careful about what we want to be published "forever" in the internet ;) We may start a thread with tips and tricks with EXPLICITLY mentioning the current used version (and confirmation of keymaster would be great for each posting). I guess a lot of things may again change when we start to use adaptive solvers.
Personally I would recommend to actively read the forums and simply test things out. The key to success is to understand the functions and the tech behind it.
-
Nothing is ever set is stone on internet, that is the last thing I would worry about, confusing people when it is not longer actual.
The thread is very good idea. The one you linked to Maru is bit chaotic, and just demanding answers. It should be something more organized. But maybe even such viral start could be good.
I disagree with notion of reading forum and actively testing things out just being the best solution to be it all, everyone does this to certain degree, but it is nevertheless wasteful use of time, because no one benefits from outside knowledge, which is something every industry is built upon. If everyone decided to "find out for yourself", no one actually gets anywhere. And testing out, doesn't necessarily correlates to understanding, merely just rough idea of it, that could be corrected if compared to other people's experiences. Even the most logical written tech, doesn't guarantee understanding after prolonged use. The value of technical know-how is bit overpriced in these times.
Hmm hmm, I think I'll start with some making-of once I finish my project.
-
Nothing is ever set is stone on internet,
Completely agree on this, and even more than that -sharing your level of knowledge is absolutely wonderful in such cases when we are just in the beginning of stepping on an undiscovered land. For instances sometimes I need to crank up pt samples to 64 while leaving light sampling as it is.
it would be very kind to know in what exactly type of scenes or light scenarios. It is strongly recommended to have a FAQ, completed by the most experienced artists in Corona, off course understanding that the most active and knowledgeable members are also humans and most of time they are busy.
Hmm hmm, I think I'll start with some making-of once I finish my project.
Thats a damn good idea!