[2.1.9] pipe circuit signal not showing 0 on empty pipe

[Maeximus]

What did you do?

connect signal to fuel pipe which is leading to thrusters

What happened?

pipe is getting empty due to speed control of spacecraft, but signal is still 1 while gui shows 0.00 for the piece of pipe where the circuit is connected to

grafik.png (245.26 KiB) Viewed 426 times

What did you expect to happen instead? It might be obvious to you, but do it anyway!

signal should switch to 0 at a sensible point

Does it happen always, once, or sometimes?

always, I also had another case where both pipe segment and pipeline showed 0.00 but signal still on 1

[Hurkyl]

The pipe isn't empty is it? It should have a trace, but nonzero, amount of fluid in it. A small enough number that it rounds to 0.00, but not actually zero.

Try using a pump to actually remove the fluid from the pipe segment.

I have no opinion on if and when the circuit signal should round down rather than round up.

[Harkonnen]

That's when showing +0 or even +0.00 would come in handy.

[gridstop]

It's critically important that the signal not drop to 0 until the pipe segment is completely, truly empty. This used to be the behavior pre-2.0 and it was awful, the new fluid system fixed it by rounding up fluids below 1 to 1. Without it there's no way to detect what is in a fluid network even though it's still locked to a fluid type. Systems that depend on switching fluids intentionally can get blocked. It's a little more recoverable now because of how pump input filters work, but it would still be really bad to reintroduce this.

[Rseding91]

Thanks for the report however this is working as intended. Since circuit networks treat "0" as no-signal it must be 1 until the fluid is truly gone. For the purpose of fluids circuit networks treat > 0 && <= 1 as 1.

[Maeximus]

It's critically important that the signal not drop to 0 until the pipe segment is completely, truly empty. This used to be the behavior pre-2.0 and it was awful, the new fluid system fixed it by rounding up fluids below 1 to 1. Without it there's no way to detect what is in a fluid network even though it's still locked to a fluid type. Systems that depend on switching fluids intentionally can get blocked. It's a little more recoverable now because of how pump input filters work, but it would still be really bad to reintroduce this.

I totally get this, thank you for throwing that in!

[Maeximus]

Thanks for the report however this is working as intended. Since circuit networks treat "0" as no-signal it must be 1 until the fluid is truly gone. For the purpose of fluids circuit networks treat > 0 && <= 1 as 1.

Thank you for the information!
I'm not sure if I now should open another bug report with "Thrusters are not emptying pipe segment completely" or if this will be fixed with the next version of the fluid network...