Floating Point

• Interactive IEEE-754 - Interactive IEEE-754
• What Every Computer Scientist Should Know About Floating-point Arithmetic - What Every Computer Scientist Should Know About Floating-point Arithmetic
• The floating point guide - The floating point guide: NO comparison, rounding, propagation
• If you really have to compare floats... some thoughts - If you really have to compare floats... some thoughts. Plus many other interesting links about this hell/stuff!!!
• Do not use neither floats nor doubles for time - Do not use neither floats nor doubles for time, even if here says you could use double
• IEEE Arithmetic - DOC from Oracle about IEEE-754. Really well explained!!
• Subnormal/denormal and performance - Performance problems with subnormal/denormal numbers
• Using FTZ or DAZ; avoiding performance problems. - Using FTZ or DAZ; avoiding performance problems. Problem: you do not have anymore IEEE-754!!
• Avoiding denormal values - Again using FTZ or DAZ; just available with SSE instructions!!! And no more IEEE-754 again.
• Be aware!! Some platforms just support FTZ - There are platforms that support flush-to-zero only, and there are many platforms where flush-to-zero is the default