From: Kbwms AT aol DOT com Message-ID: Date: Tue, 13 Oct 1998 13:20:44 EDT To: eliz AT is DOT elta DOT co DOT il, rudd AT cyberoptics DOT com Cc: djgpp-workers AT delorie DOT com Mime-Version: 1.0 Subject: Re: libc math function upgrade work Content-type: text/plain; charset=US-ASCII Content-transfer-encoding: 7bit X-Mailer: AOL 3.0 16-bit for Windows sub 38 Reply-To: djgpp-workers AT delorie DOT com Dear Eli Zaretskii, On 10-13-98 at 10:47:42 EST you wrote: > > > How come these are so close in run time, while the rest are roughly > twice as fast as libm.a? Could you make sure the times are right, > even as relative ones? I suggest to run them on an unloaded system. > Rudd's analysis was correct: >> I would guess that these times are dominated by something other than >> function computations, since they seem so similar. My claim that some of his functions are twice as fast as those in libm.a has yet to be verified by credible evidence. Here are the execution times in an unloaded environment: Execution Times Name of Test(s) Rudd libm.a ---------------------- ------- ------- acosh(x) vs. xacosh(x): 31.648 31.978 asin(x) vs. xasin(x) & acos(x) vs. xacos(x): 78.022 78.132 asinh(x) vs. xasinh(x): 25.110 25.165 atan(x) vs. xatan(x) & atan2(x,y) vs. xatan2(x,y): 49.560 49.725 atanh(x) vs. xatanh(x): 13.297 13.297 exp(x) vs. xexp(x): 8.462 8.462 log1p(x) vs xlog(1+x): 4.560 4.560 log(x) vs xlog(x) & log10(x) vs xlog10(x) 14.560 14.670 pow(x,p) vs. xpow(x,p): 31.813 31.868 sin(x) vs. xsin(x) & cos(x) vs. xcos(x): 27.033 27.308 sinh(x) vs. xsinh(x) & cosh(x) vs. xcosh(x): 25.549 25.604 sqrt(x) vs. xsqrt(x): 17.473 17.418 tan(x) vs. xtan(x) & cot(x) vs. xcot(x): 38.626 38.516 tanh(x) vs. xtanh(x): 13.352 13.352 K.B. Williams