|Anonymous | Login | Signup for a new account||2019-08-23 09:27 CEST|
|Main | My View | View Issues | Change Log | Wiki | Tavern | News|
|Viewing Issue Simple Details|
|ID||Category||Severity||Reproducibility||Date Submitted||Last Update|
|0010134||[DCSS] Bug Report||minor||always||2015-12-04 08:45||2016-11-24 02:11|
|Status||closed||Product Branch||0.18 ancient branch|
|Summary||0010134: Pakellas' Device Surge does not report improved chance for hex wands to work|
|Description||When using Pakellas' Device Surge, the chance shown for a hex wand to work is displayed as unchanged from not using Device Surge. This may lead the player to believe that the ability does not work on hex wands.|
|Tags||No tags attached.|
|According to 0010177 the bug might be that the success rate isn't improved (not just a display issue).|
So I did some testing on this, using 0.18-a0-822-gb52b7b4.
I was using a 10.0 evocations confuse wand against a shock serpent (mr 40). This displays as 50% chance to succeed. According to learndb this is 40 power for the wand (as expected if success rate is 50%).
The confuse spell at 40 spellpower succeeds 50% of the time also as we expect, as a check. With 1 enhancer, you get 59% success displayed; with 2 you get 67%; with 3 you get 73%. I did not test the spell success rates, but I assume they are correct.
Results of testing device surge and no-device-surge confuse wand against the shock serpent:
No enhancer: 22 successes in 40 attempts (55%, <1/2 of a standard deviation away from 50%).
1 enhancer: 16/41 for 39% success (roughly 2 standard deviations away from 59% success)
2 enhancers: 28/47 for 60% (~2/3 SD away from 67%)
3 enhancers: 14/30 for 47% (2.1 SD away from 73% success)
All device surge results combined: 58/118, for 49% success.
(Using 1 SD = sqrt(successes) as with other stochastic processes e.g. nuclear decay.)
So either I was extremely unlucky or there is something wrong with device surge, at least on hex wands (I cannot easily test how well it works for other devices). All three levels of device surge underperform expectation, and with any amount of enhancers we still end up with a net 50% success rate.
I think it is not working.
|Fixed in 0.18-a0-939-g1e913c8.|
|Probably fixed? It's a moot point currently, so closing.|
|2015-12-04 08:45||prozacelf||New Issue|
|2015-12-04 16:59||Amnesiac||File Added: Amnesiac.cs|
|2015-12-04 17:00||Amnesiac||Note Added: 0029921|
|2015-12-04 17:02||Amnesiac||Note Edited: 0029921|
|2015-12-04 17:58||Medar||File Deleted: Amnesiac.cs|
|2015-12-04 17:58||Medar||Note Deleted: 0029921|
|2015-12-14 19:44||Medar||Relationship added||has duplicate 0010177|
|2015-12-14 19:45||Medar||Note Added: 0029971|
|2015-12-16 11:27||crate||Note Added: 0029992|
|2015-12-28 02:33||sgrunt||Note Added: 0030090|
|2015-12-28 02:33||sgrunt||Assigned To||=> sgrunt|
|2015-12-28 02:33||sgrunt||Status||new => resolved|
|2015-12-28 02:33||sgrunt||Resolution||open => done|
|2015-12-28 02:33||sgrunt||Fixed in Branch||=> 0.18 development branch|
|2016-11-24 02:11||prozacelf||Note Added: 0031295|
|2016-11-24 02:11||prozacelf||Status||resolved => closed|
|Mantis 1.1.8[^] Copyright © 2000 - 2009 Mantis Group|