Microsoft has released a fix for a curious flaw that threatened to confound engineers who use the latest version of Excel.
The patch fixes a bug that caused Excel 2007 and Excel Services 2007 to spit out incorrect results when calculating numbers around 65,535 and 65,536. The former number is the highest that can be represented by an unsigned 16-bit binary number, so the bug had the potential to lead hardware and software engineers seriously astray.
In due course, Microsoft Update will automatically push the update to all users of Excel 2007 and Excel Services 2007. The patch will also be included in the first service pack of Office 2007, whenever it becomes available.
Microsoft's David Gainer, who announced the patch here, didn't explain how the bug came to be. (One commenter wondered aloud if it was the result of over-optimized floating-point formatting.) However, he thanked users for their patience. ®