Re: floating-point performance

2010-03-17 Thread john gilmore
/stipulate it if I judged it relevant. (It is not. I did not mention COBOL in my post, and floating-point performance has never been an urgent issue in COBOL, where it is little used, less used than it should be.) His view that there are packages invocable from C++ that yield better performance

Re: floating-point performance

2010-03-16 Thread john gilmore
For n = 50(10)100 I inverted an identity matrix, a unit upper triangular matrix, and a unit lower triangular matrix using first HFP and then BFP. The results, stated as index numbers with BFP=100, are summarized below. 50 60 70 80 90 100n 101 101 100 103 10199I 101

Re: floating-point performance

2010-03-16 Thread David Crayford
john gilmore wrote: For n = 50(10)100 I inverted an identity matrix, a unit upper triangular matrix, and a unit lower triangular matrix using first HFP and then BFP. The results, stated as index numbers with BFP=100, are summarized below. 50 60 70 80 90 100n 101 101 100

Floating point performance

2010-03-15 Thread Miklos Szigetvari
Hi If soemone has some performance comparsion, on z9 or z10 , between HEX and binary (IEEE) floating point operations -- Mit freundlichen Grüßen / Kind Regards, Miklos SZIGETVARI Research and Development ISIS Information Systems GmbH Alter Wienerweg 12, A-2344 Maria Enzersdorf, Austria T: