Returning script ignored. Bug?

[Edi]

As was noted before, if there was a gem overflow issue, that might have caused problems.

[ano]

Edi

As was noted before, if there was a gem overflow issue, that might have caused problems.

I didn't know that... What is meant by overflow, btw? How many extra games are counted as overflow?
vfb

When you tried the exact same build, did you just give your unit 28 D gems? Or did he gather them in battle (killing more than 30 units)?

Gathered with sickle a turn ago, of course . I understood that such golem would be a tasty dish for an enemy and therefore scripted returning.

[Xietor]

Try it with an 8 hp mage. If the 8 hp mage returns every time with no pd, then my point is proven.

[ano]

Did you mean using 8hp mage instead of golem or instead of duel caster?
And where did you find that magic number 8? Not a secret?

[Xietor]

I am playing pythium in Fallacy. my mages have 8 hps. They have never failed to successfully cast returning this game for me, except once.

When i put in a large pd and the attacking force was small.

[thejeff]

It's still possible it's just an override, but largely based on the debug log, I think it's some kind of gem bug.
Isn't there a "Can use gems" flag of some kind? That's the AI deciding it's not worth wasting gems.
A test might be to try a weak (8hp if you wish) mage with and without the 28 death gems.

[Psycho]

I've had the same problem, except that I had a seraph with a sickle and was attacked by shadow seers casting magic duel. This one is completely deterministic. I tried it in SP a dozen of times and the seraph never cast returning (always personal luck), when attacked by a group of 3-4 shadow seers. The same goes for golem, but not for a lowly guru mage, who did cast returning. I thought that it had something to do with the fact that shadow seers are stealthy, but maybe that's not true. I also tried it with gurus instead of shadow seers on the attacker side and both seraph and golem cast returning in this case, so I don't think it's related just to hit points. I tried it even with a scout empowered to 2S and again no returning from seraph or golem. I didn't look into the logs though.

My solution in MP was to always script phoenix pyre as the first spell for seraph, but my golems mostly became obsolete.

Anyway, I agree that this spell should never be overridden with another spell if it is scripted.

[Xietor]

Well look at it from the Ai's perspective.

Why in the world would a serpah run from anything? Of course the Ai is not attuned to look at spells that instant kill.

[Psycho]

Well, it's a nice way to get back into that domed capitol.

[ano]

Xietor
When you tell mage to retreat he always retreats be it Seraph or Theurg and there's no AI perspective in this case: he retreats because you told him to. And absolutely the same behavior is expected from Returning spell because it makes no sense if it is unpredictable. I still think it is very wrong that returning can be overridden by anything else at all.
You may ask a hundred of players if they expect returning be different from retreat in the sense of scripting and I'm absolutely sure that more than 70% will say they should be treated in the same way.