5 ms·
with regards to compounding rounding, is this not why we have significant digits? I haven't done physics in ten years so my memory is hazy on them, but I rememb
by tyre 7y ago
with regards to compounding rounding, is this not why we have significant digits? I haven't done physics in ten years so my memory is hazy on them, but I remember them being some control on what level of precision we can remain confident with.
(Side note: significant digits were a royal PITA)
- pmiller2 7y agoYes, but certain calculations are inherently numerically unstable, meaning that rounding errors can quickly dominate the calculation.
- tyre 7y agowhat does "numerically unstable" mean in this context?
- lonelappde 7y agothat rounding errors can quickly dominate the calculation.
- zentiggr 7y agoIf the number of significant digits required to maintain precision takes you past Planck length, you've reached an absolute physical limitation on your measurement ability. If the process being measured is dependent on changes of sub-Planck length magnitude, you have an inherently chaotic system, and that makes it irreversible in practice.
- tyre 7y agowow that's crazy, thank you for the explanation