Massive fatigue spells

[Twan]

If using more than one gem was really intended (KO is the the guy who designed the mechanic, when Bruce Geryk in the manual just described how he understood the game was supposed to work, so I'm not sure the manual has to be > to KO view on the subject)

Anyway, a good formula for me would be something like :
(carried number of gems) / X rounded down (X being a fixed number between 5 and 10)

It would allow the player to be sure that mages having less than X gems don't use them just to reduce fatigue, but also to make them use several gems by giving them a lot.

A good value for X would be 9 because 8 gems is the maximum gem cost for a battle spell (so with X=9, giving 8 gems to a mage and using a 0 gem spell in round one, you can be sure he will still have 8 gems to cast even master enslave in round 2 instead of using one to reduce fatigue on first spell).

Then if using a formula is considered too complex, I'd simply say one per spell.

[thejeff]

Max, if you've still got that test you just did around, what were the fatigue levels after each cast? Maybe there's some logic in that. Also does anything show up in the debug log about gem use?

As for a better mechanic, I'd like to see it tied to fatigue. Use a gem if it'll save you a certain amount of fatigue. Maybe 20 points?
So for Shadow Blast:
D2 caster 1+0 gems: 100 fat
D2 caster 1+1 gems: 50 fat
Uses 1 gem, saves 50 fat

D3 caster 1+0 gems 50 fat
D3 caster 1+1 gems 33 fat
D3 caster 1+2 gems 25 fat
Uses no gems. would only save 17 fat.

Max's D6 caster would only use 20 fatigue without gems, using 3 only saves him 8 fatigue. Using 6 saves him another 3. That's silly.

I wonder if the AI is figuring in the extra area and damage? No noticeable fatigue savings, but if it does a lot more damage it might consider it worth it.


I'd also prefer if they wouldn't use gems if, even using gems, fatigue would still end up over 200.

[Loren]

How about a different approach:

When faced with gems and scripted spells the AI should look at what combination of assigning gems to spells results in the lowest final fatigue. That's how the gems should be allocated. Under no conditions should gems be used so as to preclude casting all scripted spells unless it's impossible to cast them all.

[MaxWilson]

Sorry, thejeff, I scrapped the test after I was done with it.

I'd be happy if it just capped at +1 extra gem per cast. It's hard for me to conceive of a situation where you'd want anything else to happen. In order to use +x extra gems, you need to be x levels over the minimum path requirements anyway, so using +x gems will merely half your fatigue (it's 1/(2*x) fatigue instead of 1/x). That means it will be a rare spell where spendign the extra gems actually makes the difference between staying conscious and hitting 200 fatigue.

Let's be concrete about this. Say you want to do <Master Enslave, Vortex of Returning>. Master Enslave is S8/800. You need to be casting at +9 (S18) to stay conscious (88 fatigue + encumbrance) after casting it. If gem usage were capped at +1 (9 pearls total), you could do this combo with an S17 caster. Not likely. With the current system, you'd need to be S13 (spend 13 pearls). This is not exactly a mainline scenario. I think the method KO describes (can only use +1 gem total) makes a lot more sense and will almost always be more economical. For the scenario I describe you'd be better off bringing along another caster so the S13 can do <Master Enslave> and the S4 can do <Vortex of Returning> on the same round, anyway.

-Max