HOME | DD
Published: 2012-05-07 18:43:18 +0000 UTC; Views: 2852; Favourites: 53; Downloads: 96
Redirect to original
Description
Another test render with a new variation pack. Still has a lot of work to go though.Related content
Comments: 25
cyberxaos [2012-05-08 12:48:21 +0000 UTC]
Beautiful result! If it's slow it's worth the wait
👍: 0 ⏩: 1
dark-beam [2012-05-08 09:20:42 +0000 UTC]
Superfantastic creation!
Btw, here you find a book of math functions in C - seems to be good and uses approximations
[link]
👍: 0 ⏩: 1
MichaelFaber In reply to dark-beam [2012-05-09 01:23:37 +0000 UTC]
It's a good thing math hasn't changed since the 80's
👍: 0 ⏩: 1
deepbluerenegade [2012-05-07 20:18:37 +0000 UTC]
This makes my nerdy heart go pitter-patter. It is interesting how it looks like a printed circuit board.
Are the renders still extremely slow?
Molly : )
👍: 0 ⏩: 1
MichaelFaber In reply to deepbluerenegade [2012-05-07 22:02:53 +0000 UTC]
Yes, it is slow. The code has an inverse hyperbolic cosine function which is slowing things way down. I'm not sure that there is much way around it.
👍: 0 ⏩: 1
deepbluerenegade In reply to MichaelFaber [2012-05-08 03:48:38 +0000 UTC]
Could you figure out what the Taylor series expansion (about zero) is, maybe the 3rd or 4th order and use that as an approximation?
It is good enough for engineering grad school : D
Molly : )
👍: 0 ⏩: 1
MichaelFaber In reply to deepbluerenegade [2012-05-09 01:20:49 +0000 UTC]
The approximation wasn't close enough. It wasn't much faster anyway though because it still had a log function. I did some other optimizations though and i'm 37% faster
👍: 0 ⏩: 1
deepbluerenegade In reply to MichaelFaber [2012-05-09 15:56:13 +0000 UTC]
Interesting. I'm glad to hear you are making progress
Molly : )
👍: 0 ⏩: 0




















