Timeline for Product over a range
Current License: CC BY-SA 4.0
15 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jul 27, 2022 at 18:46 | history | edited | Seggan | CC BY-SA 4.0 |
added 17 characters in body
|
| Jul 27, 2022 at 18:46 | comment | added | Seggan | @Deadcode hmm, that must be a bug then. Or I can make it a "feature" | |
| Jul 27, 2022 at 18:39 | comment | added | Deadcode | I just realized, your post says it's using an exclusive range, but the program is actually using an inclusive range when I test it with Fig 0.3.2. | |
| Jul 27, 2022 at 17:51 | history | edited | Seggan | CC BY-SA 4.0 |
deleted 84 characters in body
|
| Jul 26, 2022 at 20:09 | comment | added | Seggan |
@Deadcode as in N1, not N-1. N is negation
|
|
| Jul 26, 2022 at 19:52 | comment | added | Deadcode |
rRN-1N-4 still does not work, though. It returns 0 but should return 24.
|
|
| Jul 26, 2022 at 19:38 | comment | added | Seggan |
@Deadcode oops, thats a bug in my input evaluator. You can append N<number> for the same effect
|
|
| Jul 26, 2022 at 19:10 | comment | added | Deadcode | How do I pass negative numbers to this program? It appears not to work with anything negative. | |
| Jul 26, 2022 at 15:26 | comment | added | Dominic van Essen | Thanks for the link. I wasn't aware of the new rule/consensus. | |
| Jul 26, 2022 at 14:59 | comment | added | Seggan | @DominicvanEssen chat, currently accepted rule | |
| Jul 26, 2022 at 7:24 | comment | added | Dominic van Essen | Why do you think it's 'allowed by the rules'? On CGCC the 'rule' has typically been that languages are defined by their implementation, and that code size is it's actual size when stored. So if you implement base-96 encoding and store the program then it could be Ok (although I suspect that you'll need bits to finally store it on any existing computer). This is what Stax implements, for instance. But not bothering to actually implement it and claiming a 'theoretical' information content seems like a shortcut. | |
| Jul 25, 2022 at 18:11 | comment | added | Seggan |
@DominicvanEssen see this chat conversation, both above and below. TL;DR it is possible to build an architecture that supports any alog_256(n) byte count. Also it's allowed by the rules
|
|
| Jul 25, 2022 at 10:17 | comment | added | Dominic van Essen | I think it would be fairer to score this with the smallest number of bytes/bits that you can actually store the program in (if the 'Fig' interpreter implements base-96 encoding of the program, and is also able to write/read this as binary bits, then this program could be 13 bits, which could even save you 0.17 bits). | |
| Jul 25, 2022 at 10:17 | comment | added | Dominic van Essen |
I am very sceptical about the 1.646-byte claim, which seems impossible as it isn't an integer number of bits. The size of a program for code-golf is the size of the program when it is stored. Other sub-byte-token languages (like Nibbles or seven, for instance) implement ways to actually store the program in the indicated number of bytes/bits, and to do this can even 'cost' extra bits. Simply declaring that - because you've elected to use only 96 characters - the code 'size' is a multiple of log256(96) isn't sufficient.
|
|
| Jul 22, 2022 at 21:27 | history | answered | Seggan | CC BY-SA 4.0 |