Hi all, I think that this is a serious problem with the FloatMgr (SDK 3.5 that comes from the NewFloatMgr in SDK 3.1), IT CAN'T SUBSTRACT !! The result for something like this
Uint16 accuracy=0; FlpCompDouble x, y; x.d = 0.999999; y.d = 1; x.fd = FlpCorrectedSub(x.fd, y.fd, Accuracy); will give a result of -9.999999998067111E-7 ...and we all know that the answer is -0.0000010000000000 I search for the FlpCorrectedSub in the last Reference Manual (print 6/00), in the KnowldgeBase, and in the mails to this forum, and found no reference to the function, why _d_sub doesn't work? (if it gives the same answer that FlpCorrectedSub !!) Why SampleCalc use it? What is the Accuracy? And what is _fp_round (I changed it before the function call, to flpToNearest, flpTowardZero, flpUpward, and flpDownward and always the result was the same). Maybe Palm needs to put better FloatMgr in next OS? (a Re-Re-NewFloatMgr?) Any comments? -- For information on using the ACCESS Developer Forums, or to unsubscribe, please see http://www.access-company.com/developers/forums/
