Thanks to whoever created this subforum for me!

For those who follow the

y-cruncher project, you'll know there's quite a bit of overlap between large prime numbers and crunching constants to gazillions of digits. (mostly from the need for large arithmetic) So I figured this is a reasonable place to put a public discussion forum for it.

For starters, I'm officially releasing version 0.7.7 on Saturday. Been testing this thing since November and I think it's good to go. If you want to try them out a couple days early, here they are:

The defining feature for v0.7.7 is of course the custom formulas. So in addition to all the built-in constants (which haven't really changed much in the 10 years of the project's existence), you can now compute things that can be expressed as a finite combination of:

**Basic Arithmetic:** addition, subtraction, multiplication, division, integer power
**Elementary Functions:** square root, arithmetic-geometric mean (AGM), logarithms, ArcCoth() of integer
**Special Functions:** ArcSinlemn() of rational, generalized hypergeometric series of rationals
**Constants:** Golden Ratio, e, Pi, Zeta(3), Catalan's Constant, Lemniscate, Euler-Mascheroni Constant

Bug reports can go either here or on

GitHub. I expect a lot of them due to the sheer scale of this feature and the virtually unlimited domain of inputs.

There's nothing new on the hardware/optimization side. I've since shifted my priorities away from low-level optimization to focus on code cleanup and other features/experiments that have stalled over the years (this custom formula feature being the one of them). So there's nothing on the pipeline for optimization for the near future except for maybe tunings for Zen 2 and Ice Lake.