So that we may learn (and laugh) at other people's mistakes, could some examples be provided from real life where an idea was tested incorrectly and the actual implementation failed expectations miserably? Such stories of multi-billion dollar loss are sure to make us students feel better about ourselves (if not paranoid about our testing procedures).
jazzbass
@lament There is the Intel iAPX 432 processor, launched in 1981. It was Intel's first 32-bit processor, and even though its design was more complex than the average processor of the era, it was slower than most of the processors that were available at the time of its release by at least 10%. For this reason the processor did not do very well commercially.
One of the biggest problems was that the processors cache was too small. It needed large associative caches to run efficiently, but the chips had no room left for that. Another problem was it's complex instruction set. The instructions were not aligned, had variable length and took a long time to decode. Another problem was the lack of local user data registers, and the protections that the processor enforced for data accessing. These protections slowed memory access (effectively) as a second memory location often had to be read to verify the access rights of the current process. Also, the Ada compiler that was used for the processor was very poorly optimized, generating slow and expensive programs.
So that we may learn (and laugh) at other people's mistakes, could some examples be provided from real life where an idea was tested incorrectly and the actual implementation failed expectations miserably? Such stories of multi-billion dollar loss are sure to make us students feel better about ourselves (if not paranoid about our testing procedures).
@lament There is the Intel iAPX 432 processor, launched in 1981. It was Intel's first 32-bit processor, and even though its design was more complex than the average processor of the era, it was slower than most of the processors that were available at the time of its release by at least 10%. For this reason the processor did not do very well commercially.
One of the biggest problems was that the processors cache was too small. It needed large associative caches to run efficiently, but the chips had no room left for that. Another problem was it's complex instruction set. The instructions were not aligned, had variable length and took a long time to decode. Another problem was the lack of local user data registers, and the protections that the processor enforced for data accessing. These protections slowed memory access (effectively) as a second memory location often had to be read to verify the access rights of the current process. Also, the Ada compiler that was used for the processor was very poorly optimized, generating slow and expensive programs.
Disclaimer: This is mostly a summary of the following pages: Wikipedia, Final project, Intel iAPX 432.