Research:Armour

From Discworld MUD Wiki
Jump to: navigation, search

Chance to hit

I've verified the chance to hit stats are correct for no-focus (same height target). Details, for my attacks on humanoid targets over the last 6 months:

Area Hits Chance
Chest 13290 16.3%
Stomach 10594 13.0%
Left leg 8258 10.1%
Right leg 8184 10.0%
Right arm 8137 10.0%
Left arm 8037 9.9%
Head 5497 6.8%
Back 3047 3.7%
Neck 2763 3.4%
Left hand 2762 3.4%
Face 2741 3.4%
Right foot 2704 3.3%
Left foot 2702 3.3%
Right hand 2693 3.3%

Note that the ratios in the article aren't exactly those given in the mudlib. I've determined the differences from the mudlib as follows:

  • There's now a face area, with weight 1.
  • The weight for the chest has increased from 4 to 5
  • The weight for the stomach has increased from 3 to 4
  • The weights for the arms and legs have increased from 2 to 3.

With these changes, you can see the experimental data is pretty much spot on with the predictions.

It'd be handy if someone can verify the other hit-location chances with different focuses. Things in particular to be aware of:

  • There are quite a few bugs in this area in the mudlib, which look to have since been fixed (for example, an enemy using 'focus lower body' would otherwise, unerringly, hit your right foot every single time...).
  • Therefore, I think it's likely that the rather anomalous algorithm that causes dwarves to fixate a lot on the right foot has also been fixed.
  • There's also a couple more subtle bugs in the mudlib's algorithm for being attacked by giants - observe that they have no chance to hit the right foot, and that their to-hit on the head is a little higher than it ought to be (both of these are caused by an off-by-one bug in the calculation).

So far I've collected a few data points for focus lower body; the stomach seems to get hit a little more often than it ought to and the right arm isn't being hit often enough, but I'll need a lot more data to be sure.

--Chat 18:58, 21 July 2009 (UTC)

Condition

The mudlib seems to have commented out the section on applying armour condition to AC - this would have applied an increasing chance that armour would block nothing based on how poor the condition is. Can a couple of PKs do some experimenting with damaged armour and see whether it actually makes a difference after all?

--Chat 18:58, 21 July 2009 (UTC)