Liquid pump pumping only 2500 g/s

Took forever to find the spot you were talking about. Looking like the branch flip flop for equal distribution still reserves a spot for absent packets on valid paths. Just move the output over on top of the pipe so it'll add it without a branch.
 

I cut all branches it still does only 2,5. I swapped the pumps, the moment lower pump got connected to that line it started pumping 2.5.

There are only two things left that can be affecting it - 2 liquid bridges on the line or that half of the line is made of insulated pipes. There though is nothing suggesting they have a different throughput and I seen then having 5kg inside.

I am taking the bridges off next.

Well after taking the bridges out and clearing the system before turning on, it still does not reaches 10 kg/s but pipes now include 5kg and 7,5 kg.

Now I am stumped. Spots of polluted water should not be causing it, it should be just pumped. Perhaps effect of fluid dynamic from geysers?

Would appear to just be ticking 4 times a second to reach 10kg. I'll try dropping it to 8kg and see if 2000 appears.
 

There is indeed an implicit tick rate of 4 times a second on the simulator.
The consumption radius is also not responsible for it. Does the same at old 1 tile radius.

 

Edited March 21, 2017 by Risu

and I just noticed that those numbers are affected by the speed of the game... hitting 7,5 - 10 on speed 1

Edited March 21, 2017 by Vilda

Either simulation is not making the ticks in time, or something is inverted again and lowers the amounts on higher speed instead of making them bigger

7 minutes ago, Vilda said:

and I just noticed that those numbers are affected by the speed of the game... hitting 7,5 - 10 on speed 1

Could be on to something there. I don't see the game speed being relayed to the simulator.
 

Found the 4.

this.simDT += Time.deltaTime;
if (this.simElapsedTime < 0.25f)
{
	this.simElapsedTime += Time.deltaTime;
	return null;
}
this.simElapsedTime -= 0.25f;

 

So as I understand it, if the simulation slows down enough or the game speed is high, that it does not check often enough, it skips some ticks?

 

7 minutes ago, Vilda said:

So as I understand it, if the simulation slows down enough or the game speed is high, that it does not check often enough, it skips some ticks?

 

Maybe. The simDT gets passed to the conduits so they get the full time scale, but all the messages being retrieved from the simulator are just being limited by that simElapsedTime block.
 

If you run the game on dev speed you will get a lot of empty blob and 2,5 blob

Because the conduits wait for the simulator to be ready, yet the simulator can only run once per game tick if a quarter of a second has passed, the conduits can build up a large delta time while waiting.

Once the simulator has elapsed at least a quarter of a second it can start running every tick until the elapsed time runs out,
but the conduits are still running far ahead. Can't speed up the 0.25 second either because it's apparently hardcoded over in the simulator since it always send a fourth of the consumption rate every tick.
 

I feel like im wrong when i say because pipes can only take 2500 g/s

29 minutes ago, Begeesy said:

I feel like im wrong when i say because pipes can only take 2500 g/s

Pipes can take (at least) 10 kg/s

8 minutes ago, Vilda said:

Pipes can take (at least) 10 kg/s

At most. Though I wish this was tunable from the class config. It's a hardcoded max in the flow process.
 

I was not sure of maximum load hence that quantification.