-
Couldn't load subscription status.
- Fork 953
Closed
Description
It would be nice to have a small overview of pros and cons of using double or float for the end of first book section 3.1. The difference became relevant as I was implementing the second book in cuda.
double: more numerical precisiondouble: twice the memory use compared tofloatfloat: less numerical precision.float: cuts down the memory use by half compared todouble
When I tested the final scene of the second book, I got out of ressources error from cuda, but everything worked fine with float.
Though obvious it is, it might be important to mention that.
Delphinidae84