Hagimei
commented
2 years ago xtkoba I find that you closing this issue as ( INVALID ) does an extreme injustice to the whole of ( QB64 ) community. This is an extreme bug that relates to math in which the entire core of (QB64) is based on. I am a MATHEMATICIAN and my position as a Military Tactician relies heavily on math and projections... Open the software in question or contact me direct ( W3L0N1@Aol.com ).. This is a critical err in all releases after 1.2 to 2.02 that must be addressed..... I DEMAND YOUR ATTENTION, not your excusable whim to dismiss this address as not warranted... I can walk you through this process because the software that I am working on is for work.
Grimler91
Problem description
ALL VERSIONS 1.5 to 2.02 releases critical bug found in processing information _integer64 and _Unsigned _Integer64 Err discovered that _Integer64's MATH is not accurate no matter the length in digits that are environmentally controlled. Using _Integer64 V.S. Double.. The math looses data and becomes corrupt.. V 1.2 is the only accurate _integer64 algorithm that functions correctly. 0000000020 2824096036 5167042394 7251286016 <- Using Double 0000000020 2824089600 7599867904 7251286016 <- Using _Integer64 and _Unsigned Integer64 Each set of numbers should match using either function should match since this is using basic math to exceed 20,282,409,603,651,670,423,947,251,286,016 USAGE: Dim shared X ( 2, 1000) as _Integer64 The only Release that is accurate is ( V1.2 ) and all other releases contains the same error using _Integer64
What steps will reproduce the bug?
And basic math formulas that exceeds 8 digits while multiplying
What is the expected behavior?
Expected behaviors are to compare both Double and _Integer64 or _Unsigned _Integer64 Numbers are continuously incorrect
System information
termux-info:
Laptop HP ENVY x360 16Gig Ram, 8Core 16Thread 4tb ssd drive ( Ryzen 7 4000 Series ) HP