Matrix Division on the 28C/S (and later)

[Thomas Okken]

(03-02-2020 12:26 PM)Werner Wrote:  Hi Thomas.
Do you want me to give you examples of DOT products, SYS solves etc that would show the effects of the increased accuracy?
Werner

Hi Werner,

Yes, the DOT products in particular. The references you cited look detailed enough that I should be able to write the code from them, but since I'm not too familiar with the numerical issues involved, I could use some help with test cases. (Of course I'd also be happy to provide a test build for you to play with once I get to that point, if you like.)

I haven't really thought about the SIMQ refinement technique, since it's not built-in 42S functionality, but it does sound like something I should look into as well, and maybe put up on my 42S programs page.

Thanks!

Thomas