“In 1994, Intel stated that the cause of the bug was that five entries were omitted from the table due to an error in a script. However, my analysis shows that 16 entries were omitted due to a mathematical mistake in the definition of the lookup table.”
Intel found the FDIV bug in its own testing in May 1994, declared it “doesn’t even qualify as an errata,” and quietly fixed the circuitry. Professor Nicely found it in October and got brushed off by tech support. The recall cost $475 million. Thirty years later a die photo shows Intel’s official explanation was wrong too.