If a plant is at 0 integrity and malfunctions this will put you at 20% power output - IF it doesn't get a crit that destroys you instantly.
Yes, this is how it used to work, though it's been a long time since I've seen that 20% output figure show up in the module pane.
When we attempted to test for this we got a destruction first so it seems like an exceedingly rare event. Best way to test would be to drop a plant to 0 integrity and then use scramble spectrum on OTHER parts of the hull until a plant malfunction is achieved. Otherwise the risk of crit destruction is too high to test consistently.
I was thinking the same thing, though to be certain there isn't some functional difference between damage and the special effect I'd want to test both.
I'm not entirely sure what the chance of destruction at 0% integrity is based on...with NPCs I usually see them pop sooner from raw volume of hits faster than raw damage, but I suspect it's based on both. Malfunctions might be the same way.
In test conditions you can just look at the modules pane and it tells you what the current output is. So the information that pops up on the centre of the HUD is really irrelevant to what you see actually happen.
Yes, it is, but a lot of observations occur outside of deliberate testing, and having the wrong error messages show up can lead to confusion.
The fist time I saw that 20% output malfunction I wasn't trying to test PP malfunction levels, someone was trying to shoot down my ship. I believe it was actually in the first PvP league when I was kiting a couple of opponents away and lost power to my thrusters (distinct from a thruster malfunction).
It was only in later testing that I was able to verify that 20% output stage, but it seems pretty clearly to be in effect here, as the priority 1 level I had set during that first warning was well below 40%.
May I ask you to tell me your test results, if there is any?
I'll probably try for a quick test of that 85% output level today, otherwise I won't be able to get to it until after new year's.
The 85% integrity threshold for on-hit power plant malfunctions is also not a test result as far as I I know (it can be difficult to prove, easier to disprove maybe).
I'm going to get someone to put eight c1 gimbaled MCs on an anaconda and slowly grind my vettes PP down. Should take several dozen, if not hundreds, of rounds fired per percent of integrity lost, hopefully prompting a malfunction very soon after it's possible to do so...that's the hypothesis anyway.