Search found 156 matches

by graphicsMan
Tue Nov 07, 2017 5:26 pm
Forum: Tools, Demos & Sources
Topic: SIMDy
Replies: 5
Views: 380

Re: SIMDy

NM, re-reading, it is clear that it doesn't generate object code. I think it's a nifty project, and probably a good way to learn stuff, but I think if you spend effort writing optimized C++, I'd be very surprised to see this perform similarly.
by graphicsMan
Tue Nov 07, 2017 5:24 pm
Forum: Tools, Demos & Sources
Topic: SIMDy
Replies: 5
Views: 380

Re: SIMDy

Does it generate object files? IMO, this is pretty neat. It would be cool to write kernels using python, and then link those into C++ code.
by graphicsMan
Mon Oct 16, 2017 9:08 pm
Forum: General Development
Topic: Quick Question
Replies: 12
Views: 1792

Re: Quick Question

I'd suggest running a profiler to see what your hot spots are.
by graphicsMan
Wed Oct 04, 2017 4:01 pm
Forum: General Development
Topic: Quick Question
Replies: 12
Views: 1792

Re: Quick Question

trig functions will be much more expensive than sqrt.
by graphicsMan
Thu Sep 28, 2017 4:56 pm
Forum: General Development
Topic: Quick Question
Replies: 12
Views: 1792

Re: Quick Question

Your ray tracer will need to be extremely fast before optimizing sqrt at this level will make much sense. Of course there are many cases where you can avoid sqrt altogether, and that makes sense, but I wouldn't go out of my way to avoid it. And with SSE on modern processors, the latency is about 10 ...
by graphicsMan
Fri Jun 12, 2015 3:50 pm
Forum: General Development
Topic: Float versus double
Replies: 12
Views: 9492

Re: Float versus double

There has been work in both fixed-point hardware ray tracing and integer ray tracing. See e.g. Johannes Hanika's dissertation and the work by Gribble et al. (Note that just because people have done it doesn't mean it will be easy ;) )
by graphicsMan
Thu Jun 11, 2015 5:30 pm
Forum: General Development
Topic: Float versus double
Replies: 12
Views: 9492

Re: Float versus double

You can work around most precision issues with float, but in some edge cases, you have to be fancy. Using double would probably reduce (or eliminate) your need for those workarounds, but keep in mind that when a problem exists for float, the same problem exists for double, just at a different scale.
by graphicsMan
Fri Jun 05, 2015 8:11 pm
Forum: Links & Papers
Topic: Looking for robust ray offsets
Replies: 2
Views: 4224

Re: Looking for robust ray offsets

BTW, I was talking about Carsten Wächter's thesis: http://vts.uni-ulm.de/docs/2008/6265/vts_6265_8393.pdf
by graphicsMan
Fri Jun 05, 2015 7:07 pm
Forum: Links & Papers
Topic: Looking for robust ray offsets
Replies: 2
Views: 4224

Looking for robust ray offsets

Hi all -

It's been a while, but there was a paper or dissertation that discussed how to calculate robust ray offsets by pushing the low-order bits of the floats for the vector offset. I can't recall which paper or dissertation; can anyone please post a link?
by graphicsMan
Mon Feb 02, 2015 4:23 pm
Forum: Scratchpad
Topic: A joke.
Replies: 3
Views: 4037

Re: A joke.

wah wah wah wah... Yes, I smiled in spite of myself.

Go to advanced search