JonChesterfield

148 ms·

2491.
by JonChesterfield 5y ago
> The expansion of function definitions inline, on demand, is a capability that only a LISP IDE has had, afaik. That is a great idea. Take the definition, splice it into the call site, rename local variables to match the caller. Interest
2492.
by JonChesterfield 5y ago
That's a stretch. Fortran used by group with nukes does not imply Fortran used to contol the nukes.
2493.
by JonChesterfield 5y ago
A missing 'good' - the reference implementation is extremely easy to extend with existing C or lua libraries. E.g. embed sqlite.c and bindings to it and all of penlight (a bunch of standard library like utilities written in lua).
2494.
by JonChesterfield 5y ago
Separate laptop.
2495.
by JonChesterfield 5y ago
Quoted before income tax and employee national insurance (50%ish) and after employer national insurance (13%). Most people seem to forget about that extra 13 since it disappears before the 'gross' number on the payslip.
2496.
by JonChesterfield 5y ago
It would need a lazy data constructor and the lisp ones are (usually) strict. If you make cons et al lazy then letrec on data works as it does in other languages. Alternatively, make letrec wrap the values in thunks. The resulting cyclic st
2497.
by JonChesterfield 5y ago
I think this is available now. The waves/wavefronts on a GPU run independently. Communication between them isn't great, independent is better. Given chips from a couple of years ago have ~64 compute units, each running ~32 wavefro
2498.
by JonChesterfield 5y ago
You might like lua as a compromise between shell and python for this stuff. Embed any extra libraries you want in the lua executable and put that one file on $PATH.
2499.
by JonChesterfield 5y ago
The gfx10 line (6800XT et al) probably work out of the box on a recent release. I think some are even officially supported. I test on a 5700XT which I don't think is officially supported. The change to 32 wide wavefronts took a while t