1

Hope this hasn't already been asked, but... is there any simple way to get high precision floats (something lime 1024 bits precision) working on CUDA without having to code it from scratch? I'd need something very simple, and I need only operator + and *... is this possible?

mskfisher
  • 3,291
  • 4
  • 35
  • 48
Matteo Monti
  • 8,362
  • 19
  • 68
  • 114
  • 2
    http://stackoverflow.com/questions/6769881/emulate-double-using-2-floats/6770329#6770329 – ArchaeaSoftware Nov 03 '11 at 12:00
  • crmlibm's scs_lib uses multi-precision approach using double's if you have double support you could directly use that otherwise you need write similar lib using float. – kanna Nov 30 '12 at 17:22
  • Possible duplicate of [Emulating FP64 with 2 FP32 on a GPU](http://stackoverflow.com/questions/29344800/emulating-fp64-with-2-fp32-on-a-gpu) – phuclv Dec 01 '16 at 03:04

1 Answers1

1

CUDA Compute 1.3 and above cards will do double precision out of the box without you having to implement it yourself. Basically anything after GTX 280 and Quadro 5800. If you need more precision than that then you will have implement it yourself.

talonmies
  • 70,661
  • 34
  • 192
  • 269
jkysam
  • 5,533
  • 1
  • 21
  • 16