Re: current best practice" packed/11/4 to char/5

On May 24, 2019, at 12:44 PM, John Yeung <gallium.arsenide@xxxxxxxxx> wrote:

On 5/23/19 5:56 PM, Jon Paris wrote:

%REM will involve division Raul - which is about the slowest thing any computer can do. I'd be horrified if it was faster than %Int. In fact I'd be surprised if it ran at less than 3x the time.

On May 24, 2019, at 9:21 AM, Raul Jager <raul@xxxxxxxxxx> wrote:
True, but converting binary to char uses one division for each digit...

On Fri, May 24, 2019 at 11:32 AM Jon Paris <jon.paris@xxxxxxxxxxxxxx> wrote:

But they occur at the hardware/microcode level.

Are you saying that division of one digit occurs at hardware level and
division of an "entire number" doesn't? I don't follow the direction
of your argument.

What I will say is: comparisons with %INT are tricky, because %INT is
used for two quite disparate purposes. One use is to truncate an
already-numeric value to an integer. The other is to convert a string
to an integer. I imagine that these two operations could potentially
have very different performance characteristics.

John Y.
--
This is the RPG programming on IBM i (RPG400-L) mailing list
To post a message email: RPG400-L@xxxxxxxxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: https://lists.midrange.com/mailman/listinfo/rpg400-l
or email: RPG400-L-request@xxxxxxxxxxxxxxxxxx
Before posting, please take a moment to review the archives
at https://archive.midrange.com/rpg400-l.

Please contact support@xxxxxxxxxxxx for any subscription related questions.

Help support midrange.com by shopping at amazon.com with our affiliate link: https://amazon.midrange.com