News for hackers. Stuff that matters.

A comment on A book to learn R and Python in parallel for Data Science. Parent comment

Re:A book to learn R and Python in parallel for Data Science by tylermw on Tuesday April 23, 2019 @11:04AM (#19729312)
> That raytracing example you linked would run at interactive rates with cudf, I really don't see any basis for perf arguments in R's favour, and 'massive data' arguments are laughable here.

I don't see how the "GPU DataFrames" provided in cuDF would enhance a raytracer in any way.

[ Parent | on HN ]
Re:A book to learn R and Python in parallel for Data Science by bicubic on Tuesday April 23, 2019 @11:32AM (#19729602)
You don’t see how a gpu accelerated numeric array would speed up ray tracing?
[ Parent | on HN ]
Re:A book to learn R and Python in parallel for Data Science by tylermw on Tuesday April 23, 2019 @12:04PM (#19729902)
The bottlenecks for raytracing are primarily in scene traversal/intersection testing--which does not benefit from a GPU-accelerated array structure.
[ Parent | on HN ]
Re:A book to learn R and Python in parallel for Data Science by bicubic on Tuesday April 23, 2019 @12:40PM (#19730287)
You are literally as wrong as it is possible to be wrong.

http://hoxxep.github.io/webgl-ray-tracing-demo/

https://devblogs.nvidia.com/accelerated-ray-tracing-cuda/

[ Parent | on HN ]
Re:A book to learn R and Python in parallel for Data Science by tylermw on Tuesday April 23, 2019 @01:06PM (#19730597)
No, I'm not. I'm well aware of CUDA being used to accelerate raytracing. That cannot be accomplished by simply providing a GPU-accelerated data frame structure, as cuDF provides.
[ Parent | on HN ]