If Intel were that smart then pray tell me why did they go with Netburst for so long knowing that it was slower than the P3 at the same speed?
Why not ramp the P3 with some fine tuning ... er the M??
http://www.emulators.com/docs/pentium_1.htm
Analyzing the results - why the Pentium 4 fails to deliver
MISTAKE #1 - Small L1 data cache -
MISTAKE #2 - No L3 cache -
MISTAKE #4 - Trace cache throughput too low
MISTAKE #5 - Wrong distribution of execution units
MISTAKE #6 - Shifts and rotates are slow
MISTAKE # 7 - Fixed the partial register stall with a worse solution
MISTAKE #8 - Instructions take more clock cycles to complete
MISTAKE #9 - Thermal issues at higher speed
Make no mistake ... engineers do not run this world ... or they would have butchered that cow and rejigged the P3 before it ever saw the light of day.
The executive at Intel overlooked all of the advice they were being given by a nuimber of engineers because the lead team on the P4 project were obviously very good at pulling the wool over their eyes.
Any company is capable of being mislead.
People run the world ... people make mistakes.
Thankfully they learned and the new core2 Penryn are next.
Hopefully the Nehalem project will be something equally as good.
I have a soft spot for the underdog (AMD) but hey ... never said I was an engineer either.