Recently, I have noticed an increase in questions about the accuracy of Perl's floating point results, with all the typical answers about it being the same as C, C++, etc. ( It always brings a smile :-).
But this time, I goggled to find if anything has changed in the non-Perl world. I was very surprised to find that IBM has been shipping a DFP hardware module since 2007 as part of the p-series power6, power7, power8 boxes. Further that many standards have been implemented to support hardware DPF.
Further, since GNU's introduction of gcc version 4.3 hardware and software DPF has been supported. So I looked into the IBM implementation, and found that real results are available. please see the following IBM wiki ( Please Note: It comes up with an 'not-found' error and after several seconds, shows the document. )
Has anyone done any work in this area specific to Perl?
Regards...Ed
"Well done is better than well said." - Benjamin Franklin
In reply to Decimal Floating Point (DFP) and does Perl needs DFP? by flexvault
| For: | Use: | ||
| & | & | ||
| < | < | ||
| > | > | ||
| [ | [ | ||
| ] | ] |