Page 13 of 50
Re: Nvidia Fermi Octane Competition
Posted: Sat Mar 27, 2010 8:00 am
by Chris
Here`s my contest entry.
Image title: The alien project.
inspiration: I had in mind to make it look like a screenshot from some kind of 80s sci-fi movie. I think i have managed to do that.
Resolution: 800x1000.
Original resolution: 1600x2000.
Rendertime: 10 hours & 50 minutes.
Photoshop post: Removed a few fireflies, added bloom effect & resized image.
additional info: 3230 samples/px - 0.28 megasamples/sec
1435572 triangles
wip:
http://www.refractivesoftware.com/forum ... ?f=6&t=843
Ps: I will probaly rerender this when i got my hands on beta 2

Re: Nvidia Fermi Octane Competition
Posted: Sat Mar 27, 2010 9:46 am
by Krisonrik
Wow, 10hr and 50min? That's crazy. And your mesh isn't that heavy either. I wonder what took so long? Or are you adding in all the post work in as render time as well?
Re: Nvidia Fermi Octane Competition
Posted: Sat Mar 27, 2010 10:20 am
by radiance
Hey,
10 hours is very long indeed.
It's probably due to the fact that there is light entering into the scenes through small holes,
with our current kernel it's very innefficient.
we need portals and MLT for this, which will be added soon.
Radiance
Re: Nvidia Fermi Octane Competition
Posted: Sat Mar 27, 2010 4:45 pm
by Chris
Yes, this is a lowlight scenario.Pathtracing is far from the best solution for this kind of lightning condition.
I could have stopped it at 5 hours but then i would have to use noise reduction, and i wanted to avoid that. I have a 5 hour render with noise reduction if you want to compare.
With MLT this scene could probaly be pretty clean after a couple of hours.
The post pro only took 3-5 mins so i don`t count that in

Re: Nvidia Fermi Octane Competition
Posted: Sat Mar 27, 2010 5:33 pm
by Chris
what is the lowest limit for entries in this competition for not canceling it?
Re: Nvidia Fermi Octane Competition
Posted: Sat Mar 27, 2010 5:42 pm
by radiance
Chris wrote:what is the lowest limit for entries in this competition for not canceling it?
can you restate that ? i don't understand.
Radiance
Re: Nvidia Fermi Octane Competition
Posted: Sat Mar 27, 2010 6:04 pm
by Chris
enricocerica wrote:If we don't get enough participants we may decide to cancel the competition and to postpone it to a later date, in such a case, the submitted images may be reuse in a future challenge.
What is the lowest number of entries needed before deciding not to cancle the competition?
Re: Nvidia Fermi Octane Competition
Posted: Sat Mar 27, 2010 8:00 pm
by radiance
we haven't thought of that.
we will probably extend the competition so people have plenty of time to enter submissions with the upcoming beta2, so i'm sure we'll get enough submissions.
Radiance
Re: Nvidia Fermi Octane Competition
Posted: Sat Mar 27, 2010 11:04 pm
by andrian
I am waiting for it (beta 2), coz I model something that I cant render right now ( beta 1) it has huge poly counts.
Re: Nvidia Fermi Octane Competition
Posted: Sun Mar 28, 2010 1:12 am
by Krisonrik
When can we expect beta 2? If I knew the deadline can extend to wait for beta 2, my method of approach would be so very different.
