List Commands Library for 50g

[DavidM]

(03-17-2018 09:38 PM)pier4r Wrote:  Hmm. Installed 1.1.3b4, but instead of LSHUF, SPLIT is called. ...

Sorry about that. The order of the commands in the library changed shortly after that 1.1.0d was released. I had hoped the limited distribution of that older release would mean that few people would be impacted. There was some discussion in the thread about the potential for this to happen... many messages back I'm afraid.

Library command references in a UserRPL program aren't actually stored by name, but rather by number. They are converted to text just before you edit the program, then are compiled back into a number when you are done editing. Changing the library meant that the older references were now pointing to different commands.

You can "fix" this by re-installing the old library, convert the programs to text with →STR (and then saving wherever you like), then reinstall the new library and convert the text versions of the programs back into compiled RPL by issuing STR→ commands on them.

The good news is that this shouldn't happen again, as the command numbers shouldn't change moving forward. That's one of the benefits of the custom menu.

As for the test, I'm wondering if you would find different results with a list of 5 items instead of 4. See this thread for a potential issue that could be reducing the effectiveness of shuffles on a list of 4 items (specifically). A list of 4 elements seems to hit a harmonic with the built-in PRNG, at least given the way I'm currently using the seed results. If you don't mind performing your test with 5 instead of 4 elements, I'd be very curious as to the results.