Build applications that can scale for the future with optimized code designed for Intel® Xeon® and compatible processors.
Announcements
FPGA community forums and blogs on community.intel.com are migrating to the new Altera Community and are read-only. For urgent support needs during this transition, please visit the FPGA Design Resources page or contact an Altera Authorized Distributor.
Success! Subscription added.
Success! Subscription removed.
Sorry, you must verify to complete this action. Please click the verification link in your email. You may re-send via your
profile.
Read the three-part article "The Perils of Real Numbers", starting here. The correct bits are being used, but you are missing the difference between binary fractions and decimal fractions.
I have a problem with the conversion from single to double precision. With the following statements and results: real a real*8 b
a=3. shows in Watch: 3.000000 b=3.d0 shows 3.00000000000000
and
b=a shows 3.00000000000000 sofar so good
BUT
a=3.1 shows 3.100000
b=a shows 3.09999990463257 and b=2.d0*b shows 6.199999980926514
so the wrong bits are kept and used.
I got the same (!) results with the CVF compiler a reason to try the Intel compiler.
Question: what am I doing wrong or mis-understanding?
0.1 is an infinately repeating binary fraction and therefor cannot be precisely represented in neither real*4 nor real*8. Therefore 3.1 cannot be precisely represented.
b=a shows 3.09999990463257 and b=2.d0*b shows 6.199999980926514
