From: SourceForge.net <noreply@so...>  20120921 05:47:55

Bugs item #3539220, was opened at 20120630 18:06 Message generated for change (Settings changed) made by rtoy You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=104933&aid=3539220&group_id=4933 Please note that this message will contain a full copy of the comment thread, including the initial issue submission, for this request, not just the latest update. Category: Lisp Core  Floating point Group: None >Status: Closed Resolution: Wont Fix Priority: 5 Private: No Submitted By: MukundNaik (mukundnaik) Assigned to: Nobody/Anonymous (nobody) Summary: Romberg Integral yields zero value Initial Comment: build_info("5.27.0","20120508 11:27:57","i686pcmingw32","GNU Common Lisp (GCL)","GCL 2.6.8") m1(x):=(1cos(%pi*x))*(1cos(%pi*x*2/3))*(1cos(%pi*x*2/5))*(1cos(%pi*x*2/7))/16; wxplot2d([m1(x)], [x,32,36], [y,0,0.6], [gnuplot_preamble, "set grid;"])$ romberg(m1(x), x, 32, 36); yields Zero. The plot above shows that it is not zero everywhere in the region 3236. In fact, romberg(m1(x), x, 32, 33) = 0.051305498551289 romberg(m1(x), x, 33, 34)=0.016230140453864 romberg(m1(x), x, 34, 35)=0.0014305334087985 romberg(m1(x), x, 35, 36)=2.3489482722215246*10^4.  Comment By: Raymond Toy (rtoy) Date: 20120815 09:27 Message: 1. Update summary to reflect this is a romberg issue, not symbolic integration. 2. Change category to floatingpoint 3. Changed visibility to nonprivate. Try changing some of the variables that control romberg integration. Changing rombergmin to 1 gives romberg(m1(x),x,32,36) > 0.0692, which is very close to the value of the symbolic integral.  You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=104933&aid=3539220&group_id=4933 