heh.. the saga continues..
black edges around objects caused by smoothed normals should be fixed next release, at least for diffuse mats. Phong is a bit trickier tho

Very interesting comparison Wzn, I think the cause of the dark area in the maxwell render is the infamous 'sunlight thru glass' issue.

Sweet.. very nicely done!

This is a definite possibility.
I don't really know anything about IES, so it depends on the complexity of the format.
Also this is reasonably specialised, only architects will use this presumably.

I'm using my own thread lib which is cross platform: implemented with win api + pthread for *nix

ah, I think this is a bug with sky + bidir tracing, no colour shows up.
Try it without bidir.

noted, thx

Quite an interesting difference in the amount of colour bleeding; apart from that, the two renders are quite similar (altho the indigo one has more noise)

edit: perhaps this is due to the reflectivity of the indigo walls being higher?

awesome!!!!!!!!

Very nice render!

linky?

LOL.. that is one of the funniest things i have read for a while.
' After several hours rendering . . .'
lol
Their path tracer implementation totally sucks balls.. it's not even converging to a correct result!

Looks promising.. I'd like to see a nice highly baked version of this render.
Try setting fresnel_scale on the floor's phong material to something like 0.4 if the floor is too shiny for your tastes.

hehe awesome.. I love it!

nice!
The metal looks good.. personally I would increase the phong exponent on the plastic.
Interesting to see there's some banding visible in the background... Did you do any levelling or anything on the png?