Ben Barsdell

I only had a quick look at this so far but can you try compiling with _M_AMD64 or __x86_64 defined.

Sorry actually for Windows it requires defining `_WIN64` or `__LP64__` as noted [here](https://devtalk.nvidia.com/default/topic/1070522/optix/raygen-sbt-data-access-results-to-illegal-memory-access/post/5425269/#5425269). E.g., pass `-D_WIN64` and/or `-D__LP64__` as compiler options to Jitify (NVRTC). If this works I'll look into...

Hey Ewan, That would be a very cool application to try Bifrost out on; it's exactly the kind of thing it aims to tackle. Your GPU cluster and software plans...

I don't remember much about it but it's probably related to this comment: https://github.com/ledatelescope/bifrost/blob/master/src/udp_capture.cpp#L56 It sounds like maybe gcc5 wasn't vectorizing the loads but gcc7 is trying to.

Interesting test, took me a bit too get my head around it :) To confirm, it's failing on some trials but not others? Does it still happen if you make...

Been thinking about this. It's a symptom of something that will be a problem for the whole library, which is that specifying array strides in units of bytes means things...

I considered having a flag like that; it could help, but it may also make things complicated. If we use nbit strides at all, we will have to handle conversion...

Thanks for finding these bugs. Couldn't you use http://docs.astropy.org/en/stable/api/astropy.coordinates.Angle.html#astropy.coordinates.Angle.dms Astropy does seem like a heavyweight solution for this. How/where do you need to interact with the angle value in the...

This is excellent feedback, thanks Casey! I think several of the issues you mention relate to the fundamental design of Bifrost vs. other approaches to solving similar problems. While the...

Bifrost asynchronicity is based around CPU threads each having their own CUDA stream. All GPU work in a CPU thread must be synchronous with respect to that thread, so it...