[DQSD-Users] Math error

2010-03-23 Thread Tom Corcoran
I found what looks like a Math error: 1276.595744680851/47 = 27.16161158895428 1276.6/5 = 255.32 tom. -- Download Intel#174; Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find

Re: [DQSD-Users] Math error

2010-03-23 Thread Glenn Carr
Are those not correct answers? On Tue, Mar 23, 2010 at 4:02 PM, Tom Corcoran tom.corco...@moonbade.comwrote: I found what looks like a Math error: 1276.595744680851/47 = 27.16161158895428 1276.6/5 = 255.32 tom.

Re: [DQSD-Users] Math error

2010-03-23 Thread Tom Corcoran
Apologies guys, it made complete sense at the time...clearly I need to lay of the cr**k pie, damn. On 23/03/2010 21:28, Kim Gräsman wrote: Hi Tom, On Tue, Mar 23, 2010 at 22:02, Tom Corcorantom.corco...@moonbade.com wrote: I found what looks like a Math error: 1276.595744680851/47 =

Re: [DQSD-Users] Math error

2010-03-23 Thread Glenn Carr
Haha... it happens. On Tue, Mar 23, 2010 at 4:47 PM, Tom Corcoran tom.corco...@moonbade.comwrote: Apologies guys, it made complete sense at the time...clearly I need to lay of the cr**k pie, damn. On 23/03/2010 21:28, Kim Gräsman wrote: Hi Tom, On Tue, Mar 23, 2010 at 22:02, Tom

Re: [DQSD-Users] Math error

2010-03-23 Thread Kim Gräsman
Heh, I had a whole answer prepared about IEEE floating-point accuracy [1], but I thought the numbers looked OK :) Cheers, - Kim [1] http://en.wikipedia.org/wiki/Floating_point#Accuracy_problems On Tue, Mar 23, 2010 at 22:47, Tom Corcoran tom.corco...@moonbade.com wrote: Apologies guys, it made