|  | Home | Libraries | People | FAQ | More | 
Using user-defined floating-point like Boost.Multiprecision has a very high run-time cost.
To give some flavour of this:
Table 16.2. Linpack Benchmark
| floating-point type | speed Mflops | 
|---|---|
| double | 2727 | 
| __float128 | 35 | 
| multiprecision::float128 | 35 | 
| multiprecision::cpp_bin_float_quad | 6 |