Jeroen Demeyer on Tue, 11 Feb 2014 12:25:46 +0100


[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]

Re: Large errors for eint() 

On 2013-11-09 23:36, Karim Belabas wrote:

1) For eint1, it was an oversight (1-word cancellation just before
returning the result, plus a few other places where we computed with one
word less than intended). This one is now fixed in commit f0ed0cd2

Unfortunately, there has been a regression again since f0ed0cd2.


*Version f0ed0cd*:

gp> \p481
   realprecision = 481 significant digits
gp> eint1(150);
gp> \x
[&=00000000014472a0] REAL(lg=27,CLONE):050000000000001b (+,expo=-224):5fffffffffffff20 a3fb9f7425aed45a 24754067b452661b 1d01ff6cecf64ff4 61c83629f7270d72 3c5957e50e9e17b1 6cb1edee181625cd b374952ad6652771 448fc4eba3db453d 04e545892636178e bcc81712b31f8f9c adfc5ddca826112e a36f40295ca5d917 278e7862af6d5de8 ab01b5a620a4c317 24a9c69d834109f3 a6c04cc983304f76 212f72b221b33f4c 39b4b69a800cbd3f b2049373fb9d679b 0d8f00c28d295100 407499a23aad0ad8 11b145a51b669795 cf72537f8190d50d fda5948870bd0052 95087aaf6133dbd4 


*Version 9973281* (current master):

gp> \p481
   realprecision = 481 significant digits
gp> eint1(150);
gp> \x
[&=0000000002192cc0] REAL(lg=27,CLONE):050000000000001b (+,expo=-224):5fffffffffffff20 a3fb9f7425aed45a 24754067b452661b 1d01ff6cecf64ff4 61c83629f7270d72 3c5957e50e9e17b1 6cb1edee181625cd b374952ad6652771 448fc4eba3db453d 04e545892636178e bcc81712b31f8f9c adfc5ddca826112e a36f40295ca5d917 278e7862af6d5de8 ab01b5a620a4c317 24a9c69d834109f3 a6c04cc983304f76 212f72b221b33f4c 39b4b69a800cbd3f b2049373fb9d679b 0d8f00c28d295100 407499a23aad0ad8 11b145a51b669795 cf724c7000000000 0000000000000000 0000000000000000 

(note that all the lower bits are 0)