User talk:Danaj: Difference between revisions
SqrtNegInf (talk | contribs) (Created page with "==== Wrong value for π ==== Oh the ignominy! Even 22/7 would have been more accurate... Thanks for catching this.") |
SqrtNegInf (talk | contribs) |
||
| (3 intermediate revisions by 2 users not shown) | |||
| Line 1: | Line 1: | ||
==== Wrong value for π ==== |
==== Wrong value for π ==== |
||
Oh the ignominy! Even 22/7 would have been more accurate... Thanks for catching this. |
Oh the ignominy! Even 22/7 would have been more accurate... Thanks for catching this. |
||
: No worries, and it really has no impact on the output. I was just stumped for a second about the magic value. It should probably be done with a 'use constant PI => 3.141593;' at the top. [[User:Danaj|Danaj]] ([[User talk:Danaj|talk]]) 17:28, 5 September 2018 (UTC) |
|||
== Prime conspiracy == |
|||
I am possibly confused about your statement "using an array rather than a hash to store transitions". Is this what you had in mind? |
|||
[https://gist.github.com/SqrtNegInf/f3e9ef2807f225946871b385aeae6826#file-prime_conspiracy-pl Prime_conspiracy.pl] |
|||
I benchmarked, and didn't see a difference. But if you have a faster version, definitely put it on RC... --[[User:SqrtNegInf|SqrtNegInf]] ([[User talk:SqrtNegInf|talk]]) 15:07, 27 October 2018 (UTC) |
|||
: I commented on the gist. Would you recommend replacing the code or adding it as an alternative? |
|||
: [[User:Danaj|Danaj]] ([[User talk:Danaj|talk]]) 18:00, 29 October 2018 (UTC) |
|||
== ntheory question: Handling very large values with 'vecreduce' == |
|||
Using 'ntheory', the exponential factorial is easily defined as <code>vecreduce { $b ** $a }</code> but for n=5 the result returned is 'Inf'. The straight calculation of <code>5**4**3**2</code> correctly returns a number that has 183231 digits. All this can be seen in context at |
|||
[https://github.com/SqrtNegInf/Rosettacode-Perl-Smoke/blob/master/S/Special_factorials Special factorials]. Everything else works fine. Clearly, the fate of the planet does not hang in the balance, but still it would be satisfying to get a clean result out of 'ntheory' here. Lighting the Bat-signal, as I am not clear on how to proceed, and would appreciate any thoughts you may have on the topic. --[[User:SqrtNegInf|SqrtNegInf]] ([[User talk:SqrtNegInf|talk]]) 03:07, 9 April 2021 (UTC) |
|||
Latest revision as of 03:07, 9 April 2021
Wrong value for π
Oh the ignominy! Even 22/7 would have been more accurate... Thanks for catching this.
- No worries, and it really has no impact on the output. I was just stumped for a second about the magic value. It should probably be done with a 'use constant PI => 3.141593;' at the top. Danaj (talk) 17:28, 5 September 2018 (UTC)
Prime conspiracy
I am possibly confused about your statement "using an array rather than a hash to store transitions". Is this what you had in mind? Prime_conspiracy.pl I benchmarked, and didn't see a difference. But if you have a faster version, definitely put it on RC... --SqrtNegInf (talk) 15:07, 27 October 2018 (UTC)
- I commented on the gist. Would you recommend replacing the code or adding it as an alternative?
- Danaj (talk) 18:00, 29 October 2018 (UTC)
ntheory question: Handling very large values with 'vecreduce'
Using 'ntheory', the exponential factorial is easily defined as vecreduce { $b ** $a } but for n=5 the result returned is 'Inf'. The straight calculation of 5**4**3**2 correctly returns a number that has 183231 digits. All this can be seen in context at
Special factorials. Everything else works fine. Clearly, the fate of the planet does not hang in the balance, but still it would be satisfying to get a clean result out of 'ntheory' here. Lighting the Bat-signal, as I am not clear on how to proceed, and would appreciate any thoughts you may have on the topic. --SqrtNegInf (talk) 03:07, 9 April 2021 (UTC)