Anonymous | Login | 2024-04-26 03:20 CEST |
Main | My View | View Issues | Change Log | Wiki | Tavern | News |
Viewing Issue Simple Details [ Jump to Notes ] [ Wiki ] | [ View Advanced ] [ Issue History ] [ Print ] | ||||||
ID | Category | Severity | Reproducibility | Date Submitted | Last Update | ||
0010333 | [DCSS] Bug Report | minor | always | 2016-03-27 06:12 | 2016-03-27 07:14 | ||
Reporter | crate | View Status | public | ||||
Assigned To | neil | ||||||
Priority | normal | Resolution | done | ||||
Status | closed | Product Branch | 0.18 ancient branch | ||||
Summary | 0010333: Speed brand is a bit broken by integer math | ||||||
Description |
Caused by 2733cf4, if I'm reading this correctly. That commit sets int weapon_min_delay in fight.cc to (min_delay*2/3) if the weapon has speed brand. Since this is integer math, this means min_delay is always an integer. E.g., this sets min_delay to 4 for a longbow of speed. The problem is that speed brand isn't supposed to work this way; weapon_min_delay (the way it is used now) is actually supposed to be noninteger (4.7 for longbow of speed for example). The result is that a longbow of speed does have the 0.4 min delay that the game claims (this claim is supposed to be a bit of a lie, if you read the commit message for 2733cf4), but reaches this delay at skill 22, not skill 20. So speed brand once again actually changes the skill for min delay for many weapons (and is particularly good for weapons with specific non-speed weapon_min_delay), which of course is amusing because avoiding this is one of the reasons the brand was changed in the first place. |
||||||
Additional Information | |||||||
Tags | No tags attached. | ||||||
Attached Files | |||||||
|
Mantis 1.1.8[^] Copyright © 2000 - 2009 Mantis Group |