FROM THE COMMUNITY COLLECTION
One Object
The removal of the water-ball just saved me 3 days of render time. I reduced bump on the torus knot considerably and slightly increased it on the ground. This is just to confirm Rashad's observation (and my own theories). The same HDRI was used, obviously. I made it myself from two series of four pictures, two f-stops apart each, of a mirror ball set up in my living room one evening.
Loading unique views6,673 unverified legacy hits0.0 archived rating2006-11-12 added
Comments 9
A conversation around the artworkSign in to join the conversation.
Horo, very nice render! I've always been fascinated by objects like this....makes a great subject....very enigmatic. Would like to see it on a flat, or glass, or metallic surface with AO, etc....bring it into DS to use the Reality plugin? Animate the lighting? Materials? BTW, re: HDRI apps.....I tried several but settled on Dynamic Photo HDR.
Programs used for HDRI Sorry richter, I haven't answered your question yet. I use several: mostly Picturenaut, HDRShop and Artizen, sometimes also Photomatix and HDRIE.
Render Time It saved me 3 days - the first one needed 3 days and 6 hours. So the answer is 6 hours.
This looks even more real then the previous, Horo. In your descr I read the answer to my question asked in "Two Objects" about the creation of the .hdr. Here you're saying you're hesitant - how much does it take to render this time? If you ask me, with the kind of tests/result seen here continuously improving I'll keep it going. Nice one too, Horo!
To be honest, Rashad, I've just rendered a part at half size and regular quality with AA and compared the results. I plan to do some more experiments to bring back Specularity in IBL lit scenes. The unplublished test shows high frequency bump to be a promising way to go. These tests just eat a lot of render time even on a dual core machine and this makes me a bit hesistant.
increased bump The increased bump should definetly help. I am curious though about the quality of those bumps. In the 10,000 ranges it is difficult to appreciate the full effect at normal AA. If you go that route you will probably need to increase the render quality. I remember a piece submitted recently by LitoNico with a noisy floor element and some geometric shape he had modelled. Do you notice quality issues with the high frequency bump?
To increase the frequency of the ?grain? dramatically proved to be a good suggestion - something to keep in mind.
IBL Settings Yes, you've asked that on the previous picture already and I didn't answer. I will now because there is something for you on the ftp anyway: Quality 25, Intensity 56, HDRI-Effect 81.
This is an interesting experiment. I wonder what happens if you boost the frequency of the mat up into the 10,000 range so that the bump is much smaller than the ray resolution? Also, I was wondering what settings you had employed for the IBL?