The Mana World Forums

Specifically, the family of sprites belonging to poltergeists, wisps, and spectres normally have a few wandering sparkles surrounding them. When they take hits, though, it can get to be a lot more than a few of them. Let's observe:


Here we see a wisp in one of its natural habitats (as natural as that map can be considered, anyway). The Atmospheric Particle Index(tm) gives a fairly sane 42, just before I'm about to sneak up on it...


Now the wisp is on its last legs, or whatever it uses to stand on anyway. It's picked up quite a collection of fireflies or some such thing, not that they're helping it much, except by giving it the faint glimmer of hope that the result might cause so much lag as to cause me to leave. Clearly that glimmer doesn't shine nearly as bright as I do. Note that the second nearby wisp heard its distress call immediately and has been attacking me the whole time, yet its aura looks completely normal, demonstrating that it's the act of taking hits--not dishing them out--which triggers the particle accumulation.


Just one more hit after the above snapshot caused the wisp to disappear. It does properly clean up all the stray particles over the course of a few seconds, and now the particle count is back to normal (considering that the second wisp is in the middle of its attack animation). So it's not a permanent memory leak kind of thing, but it can potentially cause slowdown on older machines until the monsters are disposed of.

Hi radiant,

Thanks for the very precise description and humour put in it.

What is your particle number settings?

For older machines, I'd say that this should be kept to 'low'.

Anyway, you're maybe right, and there are maybe too much at once for the 'normal' effect.

Best regards.

It seems like the particle effect of the monster is respawned when the monster is hit.

I have no idea what could cause this effect, but AFAIK the code sections in questions got some considerable changes by other people since I wrote them.

Ya I had some issues with monster particles as well. They seemed to have the same 'emitter stacking' problem when the mob was just moving. Later versions of the client, as of nightly build 20110426, don't have this particle stacking problem (both the one you described and the one I had).

I haven't checked any of the latest builds because of the particle y-axis positioning problem (too high up!).