Jeremy Meredith

Matt points out there are good libraries for doing this on a GPU (like curand), though of course we need to abstract it so it's not CUDA-specific. I believe curand...

Upon further review, I like (c). Here's what I'm thinking: 1. a new concrete type of array (say, `eavlImplicitArray : public eavlArray`) which has exactly two values: an origin and...

Matt looked at the performance more than I ever did, so I don't have any analysis beyond his. I was mostly just making sure I got correct answers with _some_...

Am I reading those plots wrong, or is dynamic the _worst_ performer? (With static and auto tied, and guided the best.) The vertical axis is runtime, right?

The test in configure.ac should attempt to try linking against lib/ first, and only failover to lib64/ if lib/ doesn't work. I'm wondering if there's something else going wrong. Does...

The EAVL source has been patched to allow the user to pass CUDA_LIBS. For example, run: `./configure CUDA_LIBS="-framework cuda -lcudart"`. Note, you may encounter errors at the end of the...