Re:DCompute: GPGPU with Native D for OpenCL and CUDA
by willvarfar on Monday July 17, 2017 @02:49PM
(#14791295)
(On the Mill CPU the retire stations snoop/snarf writes and spot aliasing as it happens at runtime. This is a big part of our auto-vectorisation)
[ Parent
| on HN ]
Re:DCompute: GPGPU with Native D for OpenCL and CUDA
by nicwilson on Monday July 17, 2017 @08:08PM
(#14793359)
Cool, I'm looking forward to seeing more news about the mill.
Re:DCompute: GPGPU with Native D for OpenCL and CUDA
by willvarfar on Tuesday July 18, 2017 @11:58AM
(#14797546)
https://news.ycombinator.com/item?id=14797350
probably won't get on the front page of HN :|
Re:DCompute: GPGPU with Native D for OpenCL and CUDA
by CyberDildonics on Tuesday July 18, 2017 @04:36PM
(#14799847)
Interesting, are there benchmarks I can look at?