Elon Interstellar Problem ( Alarm leaving crew menus ) | Kodera Software | Page 1

barren flower Nov 16, 2022, 11:07 PM

#

This goes back to the steam topic "Elon Interstellar Problem ?" from today (16th nov 22).

Running Elon Interstellar after leaving the "J" crew menus the game plays an alert, which i am not sure what it is about (suspect power management because of the ignition take in on Fusion engines, but then - there is no actual maneuvering going on).
The ship operates fine otherwise.

It's not entirely consistent - i started a few dives trying to recreate and was unable to. Then went back to playing normally and it started happening again deeper in a dive - so duration of the dive might be a contributing factor.

Using a retrofitted K37 HUD and downgraded to the NDCI autopilot. Engines are default (i think) Z-Axial Fusion and Elon Agile Thrusters; MKIV Big Furnace for Aux Power.

The folder was zipped during the dive, so the [only] savegame is just before launching. I also recorded a bit of video replay.

📎 DeltaV_ElonAlarmGlitch.zip

#

Elon Interstellar Problem ( Alarm leaving crew menus )

scenic sundial Nov 17, 2022, 8:24 AM

#

Interesting.
The alarm sounds because your ultracapacitor is dry when you just exit the OMS; however, I have no idea why it is dry yet.

#

I can't really replicate it on my side. I wonder if I'm missing an component

scenic sundial Nov 17, 2022, 8:58 AM

#

tried a number of things, including importing this save and trying to re-create, but I had no luck. Is there something specific you are doing when that triggers?

scenic sundial Nov 17, 2022, 9:21 AM

#

I can see something in the simulation resolution seems to not align as the time speeds up when you exit the OMS menu

#

but I was unable to trigger it on my side

barren flower Nov 17, 2022, 5:20 PM

#

scenic sundial I can see something in the simulation resolution seems to not align as the time ...

Yes, that was my impression as well.

#

the only thing that might be different between my and your attempt is the time - my dives are pretty long

#

just starting a dive and going into the menu does not seem to trigger it

#

The MO is - i start a dive, jump to the furthest location i track, go from there for depth until i find another trackable location at which point i fill up my hold and return

#

the thing starts happening during the flight

#

I do not know enough about the engine and your code naturally, but functions or states that are time sensitive might be the suspects - either something is accumulating an rounding error over time, or somehow the duration of the dive figures into the calculations when the time re-scales after leaving the menu

#

PS.: most of the time (including when just going straight) i fly with the Autopilot on (playing on mouse and keyboard i find it difficult to control wild rotation otherwise).
So maybe some value in the Autopilot is the rogue time sensitive variable.

#

PPS: thinking of it - one thing i do is quickly switching the tabs in the menu - kinda as you see in the replay - i enter the menu, press page down a few times in a quick succession, to get to the pilot page and then leave the menu (this is to check for new radio contacts).
I do that a lot during flight, so again: if there is something that accumulates error every time the menu is accessed...
Once the alarm starts going tho, it is irrelevant what page i look at and how long i stay in the menu.

scenic sundial Nov 17, 2022, 5:50 PM

#

Okay, please check the following thing next time it begins to occur. Go into OMS/mechanic, turn off reaction wheel (there is on/off button next to everything) and check if that makes a difference.

#

I'll try to run a soak test in the meantime

barren flower Nov 17, 2022, 6:03 PM

#

Will do.

barren flower Nov 17, 2022, 6:55 PM

#

On a sidenote: I'll assist best i can and post update, as soon as i dicover anything new; But if you feel the issue is not worth your dev-time, i'm chill about it.
It is mildly interesting if it indeed points to some inconsistency in the simulation. But on the user side it does not seem to create any issues other than the alarm, which is not all that big of a deal.

scenic sundial Nov 17, 2022, 6:58 PM

#

It might be a symptom of something bigger

barren flower Nov 17, 2022, 11:19 PM

#

So i had 2 dives where it did not happen (too short?).
Then got one where i was able to get it fairly consistently.
Not sure how long the dive took (had interruptions pausing the game to menu - might it be a contributing factor?).
About 150-180 kilometers of depth gain.

#

With that i tested a few things.

#

Firstly it keeps happening with the reaction wheel off

#

(I have some more random video if needed, but i don't think much more can be seen there)

#

It also happens flying slower and sitting still (as still as the autopilot will hold after pressing "X")

#

In the Reaction Wheel video it's seen, that even with the Wheel OFF (the ship was turning noticeably more sluggish and jerky) the autopilot is still doing angle corrections using thrusters.
So if it was some kind of angle-saturation issue, it might still be the case.
(I am using the NDCI Autopilot).

#

As far as i can tell it does not happen flying with the Autopilot OFF (you know how proving a negative is much harder, since something not happening, does not mean it never happens...)

#

When i play next, i'm thinking to try recreating it with another ship.
Or rather a different Autopilot?

#

scenic sundial Nov 18, 2022, 7:38 AM

#

Thanks for all that! Now I do have some idea of what might be happening.

I'm going to bet it won't happen if you turn off your ZAP drives?

scenic sundial Nov 18, 2022, 8:29 AM

#

I think that what is happening here that during the time speedup phase of the game just after you exit the OMS menu your autopilot briefly operates in a higher resolution and is issuing a brief pulse to your ZAP drives, which drains your 500MJ ultracapacitor dry.

barren flower Nov 18, 2022, 3:58 PM

#

It was my initial assumption as well, but it does not seem to line up entirely - i mean each of them takes 150MJ, so 300 together - that should not empty the 500MJ ultracap in an instant, on a ship that is net positive on power otherwise.

Edit: you mean the Autopilot turns them on and off repeatedly before giving the Ultracap time to recover?

#

There is another mechanic in there that i do not quite understand - the Elons Fusion Reactor. The dealer overview says "Powerplant 1500 MW" yet the simulation in the Equipment screen shows 500MW for a naked ship;
But then it seems like the power output of the Stellarator ramps up if needed?

scenic sundial Nov 18, 2022, 7:34 PM

#

The Stellarator looks at how much energy your ship is using and it provides a bit more.

#Elon Interstellar Problem ( Alarm leaving crew menus )