in reply to Re^12: Need more precision. (Updated.)
in thread Need more precision.

If you want to build a 128-bit int yourself, not using native types or existing code, then my approach would be to look at how it's done at a smaller scale, like 16-bit math on an 8-bit uC, e.g. http://www.8052.com/math - the same concepts apply when building a 128-bit number out of two 64-bit numbers.

By the way, if you like fractals then you've probably already seen this, but if not you may enjoy it: https://www.youtube.com/watch?v=foxD6ZQlnlU

Replies are listed 'Best First'.
Re^14: Need more precision. (Updated.)
by BrowserUk (Patriarch) on Jun 10, 2015 at 22:30 UTC
    you've probably already seen this, but if not you may enjoy it:

    I hadn't. Wow. I thought I was obsessed. My current render has been running for just over 125 hours of cpu and should finish up at around the 200 hrs mark sometime tomorrow. Those guys' renders are measured in months. (And put paid to the "it's all the same" notion :)

    I'm naught but an amateur. Thank you for the link.


    With the rise and rise of 'Social' network sites: 'Computers are making people easier to use everyday'
    Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error.
    "Science is about questioning the status quo. Questioning authority". I'm with torvalds on this
    In the absence of evidence, opinion is indistinguishable from prejudice. Agile (and TDD) debunked