Re^3: Portability of floor(log(N))

Thank you for the suggestions.

I think they were completely irrelevant to the problem you're facing.
Silly me didn't think to actually check what the correct calculation of floor( log($num) / log(1000) ) should yield under 'long double' (64-bit) precision.

When I do check, I find that the correct answer is -12.
The error is in your test script ... either that, or in Math::MPFR:

use strict;
use warnings;
use Math::MPFR qw(:mpfr);

# Set precision for extended
# precision 'long double'
Rmpfr_set_default_prec(64); 

my $num = Math::MPFR->new('1e-33');
my $thou = Math::MPFR->new('1000');

# Convert $num & $thou to their
# respective log()
Rmpfr_log($num, $num, MPFR_RNDN);
Rmpfr_log($thou, $thou, MPFR_RNDN);

print $num / $thou;

# prints -1.10000000000000000009e1
[download]

This indicates that when the floating point precision is 64 bits, the correct floor() result is -12.
And, IIUC, the test script should accept '1000e-36' as correct when NV precision is 64 bits (as is the case with those 2 FAIL reports).

Interestingly, with full quad precision of 113 bits, the calculation reverts to returning -11.

Cheers,
Rob

Comment on Re^3: Portability of floor(log(N)) Select or Download Code

Replies are listed 'Best First'.
Re^4: Portability of floor(log(N)) by syphilis (Archbishop) on Mar 14, 2018 at 02:42 UTC
And, IIUC, the test script should accept '1000e-36' as correct when NV precision is 64 bits Alternatively (and since you're already using POSIX) you should get the result you expected, if you replace the log() calls with POSIX::log10(). That way both numerator and divisor should be exact (for these particular inputs), as should also be the result of the division ... irrespective of the precision of the NV: `$ perl -MPOSIX -le 'print POSIX::floor(POSIX::log10(1e-33) / POSIX::lo +g10(1000));' -11` [download] Cheers, Rob	[reply] [d/l]
Re^5: Portability of floor(log(N)) by toolic (Bishop) on Mar 14, 2018 at 17:31 UTC
Using POSIX::log10() looks like the simplest solution. I will attempt to pester the tester with a modified version of FormatEng.pm. Do you think I can simplify it by replacing POSIX::log10(1000) with 3, or will it be better to keep your code as-is? `my $e = floor( POSIX::log10($num) / 3 );` [download]	[reply] [d/l]
Re^6: Portability of floor(log(N)) by pryrt (Abbot) on Mar 14, 2018 at 18:22 UTC
log10(1000) will never change, and is exactly 3, so there's no reason not to use it	[reply]