Stephen Brooks 2002-11-06 22:00:42 | I have a nice toggle-switch on my compiler that switches "Pentium Pro Instructions" on. The only trouble is, when I switch them on, the program starts evaluating 226890<0.1 to "TRUE". In fact it seems that pretty-much everything is smaller than 0.1. Hmm. This also has the side-effect that all the lines I plot come out as points. "As every 11-year-old kid knows, if you concentrate enough Van-der-Graff generators and expensive special effects in one place, you create a spiral space-time whirly thing, AND an interesting plotline" |
Stephen Brooks 2002-11-06 22:05:28 | It appears that replacing 0.1 by (float)0.1 does the job. The value of 0.1 is evidently much higher on a Pentium Pro processor unless you round it to single precision. I think my compiler is retarded. It's come back at me with "Error: Expression too complicated" before now. "As every 11-year-old kid knows, if you concentrate enough Van-der-Graff generators and expensive special effects in one place, you create a spiral space-time whirly thing, AND an interesting plotline" |
Stephen Brooks 2002-11-06 22:11:27 | I have come to the conclusion that it would be proper and beneficial to humanity and the scientific effort in general if I left that compiler switch well enough alone. "As every 11-year-old kid knows, if you concentrate enough Van-der-Graff generators and expensive special effects in one place, you create a spiral space-time whirly thing, AND an interesting plotline" |