Talk:Sif Muna

From CrawlWiki
Revision as of 13:01, 29 November 2016 by Gammafunk (talk | contribs) (random2avg)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

This page has never had a version tag applied to it, and it's one of the pages I've been least confident about updating (I'm a much bigger fan of Vehumet's style, TBH). I have no idea how out-of-date and inaccurate this page actually is. Can anyone who honestly feels confident about Sif Muna's behavior update this page for 0.14 or 0.15? If you can fill in a History section as well, that'd be even better. Code diving for specific mechanics behavior welcome. --MoogleDan (talk) 21:07, 20 August 2014 (CEST)

As far as I can tell Sif hasn`t been changed much: the skills (channel, amnesia and miscast protect) and spellbook-gifts are still there for 0.15 as well as not accepting the ignorant. I can`t say anything about the specific numbers, but I didn`t notice any changes in the average amount of MP channeled and when Sif starts to gift books. There wasn't any mention of Sif in the changelogs, so I think Sif is unchanged. --BlueCrake (talk) 13:26, 21 August 2014 (CEST)
Actually, do a CTRL+F on the change log. She's in there. It says that she no longer inflicts amnesia. I'm not sure if that was suggesting that she no longer allows you to forget spells at will (which would be a weird nerf to add), or that she no longer blasts your memorized spells in her wrath (the phrasing makes this sound more likely, in which case we're doubly screwed because we never even had that listed in her wraths in the first place). I think she's probably mostly the same nerdy bookish library goddess we've always known and loved, but I want specifics, dammit. What's the point of a wiki that can't provide needlessly in-depth calculations??!? --MoogleDan (talk) 13:50, 21 August 2014 (CEST)
As always, CC, fantastic work! --MoogleDan (talk) 15:55, 21 August 2014 (CEST)

Channelling success rate

The channelling success rate this page indicates seemed strange to me, so I went to see the code for myself. I found out the failure chance of this ability is :

failure = 40 - (you.piety / 20) - you.skill(SK_INVOCATIONS, 2);

However, I can't find the definition of the "skill" method so I don't know what the "2" means and it bothers me. If anyone knows about this I would be very grateful. (btw this is in -- Fingolfin (talk) 11:53, 21 December 2015 (CET)
It's the scale parameter, if it's higher than 1 then progress towards the next skill level matters too (1.5 invo results in 3 here). -- Medar (talk) 12:40, 22 December 2015 (CET)
Thanks for the answer ! However I'm not sure that I get it, does it mean the formula is failure = 40 - piety/20 - ceil(Invocations skill) ? -- Fingolfin (talk) 21:50, 22 December 2015 (CET)
The return value is skill (including progress toward next level) multiplied by the scale parameter (as an integer). So it should work out to: failure = 40 - piety/20 - floor(Invocations skill * 2.0) -- Medar (talk) 10:10, 23 December 2015 (CET)
Thanks a lot! I'll update it. -- Fingolfin (talk) 15:04, 24 December 2015 (CET)


How exactly does it work in the formula here given that no max roll or number of rolls are listed? If it's just one roll of 1d (2/3*Invocations) it seems like we should just say that instead, because the function name just causes confusion (and most readers don't keep a copy of the source to grep through). If I'm wrong in how it works, we should expand on how it's being used/how it works, because it's causing me confusion! Fambida (talk) 23:02, 28 November 2016 (CET)

I've updated the formula to add the number of rolls (two), but the max roll is already given and what you've listed: 2/3* Invocations (subject to random-rounding to make it an integer). The actual formula does indeed use random2avg() to average over two rolls, so it wouldn't be accurate to simply use 1dN notation. Perhaps random2avg() needs a wiki page, since it's a commonly used RNG function in DCSS, or the wiki could use standardized language to describe averaging die instead of using the function name. --Gammafunk (talk) 12:01, 29 November 2016 (CET)