Paralyze vs Stellar Cascades

[Corinthian]

I have personally seen 100 prec spells miss in a storm so there is that.

[JonBrave]

May I ask where you guys get your precision formulae from? Where does it say that 100 Precision spell is anything like 100% accurate??

[Corinthian]

Guesstimates? You never see them miss. Except in storms and presumably under cover of darkness were its actually more like prec 50 due to the halving.

Edit: Except were there are more than one unit in a square. Precision only narrows things down to a square. I once had a supercombatant that attacked a fort that contained a few weak astral mages. The astral mages were scripted to cast Horror Mark x5. But most of them missed because my SC was only size four and shared the square with and indy commander. Every spell hit that square but they all struck the poor indy commander instead of the SC. The SC did not get a single mark before he was finished with his long buff cycle.

[Knai]

Originally Posted by JonBrave

May I ask where you guys get your precision formulae from? Where does it say that 100 Precision spell is anything like 100% accurate??

The manual describes precision as the number of squares a unit can aim away before possibly hitting the wrong square. The presented formula is essentially (precision)+((precision - 10) ≥ 0). With 100 precision, that is 190 squares away - longer than the entire field.

[Corinthian]

As anyone with thugs that have ever failed to bless themselves despite the distance being zero, knows. The formula in the manual is wrong.

[JonBrave]

Originally Posted by Knai

The manual describes precision as the number of squares a unit can aim away before possibly hitting the wrong square. The presented formula is essentially (precision)+((precision - 10) ≥ 0). With 100 precision, that is 190 squares away - longer than the entire field.

Umm, I'm sure I'm being dumb, but my manual p.77 states (Precision / 2) - 2 as range beyond which deviation will occur. Which I make 48 for a 100 Precision. I admit I'm looking under "missile combat", but I thought it was same for spell-casting mages....

and P.S. p.86 tells me mage casting is as per p.77 !

[Knai]

Originally Posted by JonBrave

Umm, I'm sure I'm being dumb, but my manual p.77 states (Precision / 2) - 2 as range beyond which deviation will occur. Which I make 48 for a 100 Precision. I admit I'm looking under "missile combat", but I thought it was same for spell-casting mages....

and P.S. p.86 tells me mage casting is as per p.77 !

Elsewhere it describes it as precision, with precision past 10 counting twice, which simplifies to the formula I gave.

[thejeff]

Originally Posted by Corinthian

As anyone with thugs that have ever failed to bless themselves despite the distance being zero, knows. The formula in the manual is wrong.

That's not a matter of Precision, but of how the AI picks targets and how the AoE spread works.
In short, instead of picking the sacred thug to target, it tries targeting several random units and picks the first one where the AoE spread in its test casting includes the thug. The AoE spread in the actual casting does not necessarily hit the same squares. This is not necessarily a bad idea when you have multiple sacred units to bless, since it picks a target that has a chance of getting more units with one casting: think a non-blessable target surrounded by blessable ones.

The simple evidence for this is that bless will never miss a single thug alone on his side of the field. If he brings an ally or if the enemy reaches him before he casts (fliers!), it can miss.

[Corinthian]

I understands your reasoning, but I was talking about lone thugs. Or in this case, lone Dai Onis. Granted, Dai Onis auto summons wolves at the start of the battle but these are not sacred and should not be targeted. I rather think it has something to do with the fact that Dai Onis have less than 10 points of precision. Witch would mean that prec 10 is really prec = 0 for most purposes.

[JonBrave]

Originally Posted by Knai

Elsewhere it describes it as precision, with precision past 10 counting twice, which simplifies to the formula I gave.

I now understand your formula, thanks.
Given that we are talking about deviation likelihood, since the manual there states that the distance is "half the Precision, minus 2", a 100 precision spell has a range of 93 (plus a bit maybe from the mage). Not the 190 someone said.