(ref.doc)JOOP 12/6

October 99
Volume 12, No. 6

p 61 C++, by Andrew Koenig and Barbara Moo
Performance: Myths, Measurements, and Morals
  "You can have it good, fast, or cheap -- pick any two"
  "I don't have to outrun the bear; I just have to outrun you"
  Jon Bentley has proposed two rules for optimization:
   "1. Don't do it.
    2. (For experts only) Don't do it yet."
  Myths:
   The faster the processor, the less important speed is
   Speed is always important
   Local optimizations are extremely important
   Local optimizations are unimportant
   Hand optimization yields faster programs than using general-purpose
    libraries
   There is little room for improvement in modern libraries
   It is easy to tell where a program spends its time by looking at it
   It is easy to tell where a program spends its time by profiling it
   It is easy to optimize a well-designed program