Factorio Forums

eradicator wrote: @BlakeMW:
"Too few turbines" does make it easier to always have them fully running. And i was speculating that maybe if the input pipes are always full they only flow "in one direction" and not in both. Also other machines have certain optimations if they run "at full speed" so maybe turbines do too? But without actual measurements that's just wishful thinking on my side .

As I said turbines consume steam each tick regardless of whether they are 1% or 99%: you can easily test this by using a slow game speed, disconnecting the steam source, and watching it tick down. AFAIK even if there is no steam in the engine or if it is disconnected it still consumes just as many cycles because the "inoperable" state is so unusual it was deemed not worth optimizing for.

So far as flow direction; liquids only ever flow in one direction but it can rapidly alternate. I don't believe there is any optimization whatsoever to try and identify "steady state" pipes. At least it is easy to test: just place zillions of pipes on an otherwise empty map, whether you have them do useful work (like connect a pump to an operational boiler powering a beacon), leave them all empty, or let them all be full (removing the beacon) there is no real difference in the UPS. The UPS cost of "plumbing" related stuff is constant regardless of usage.

BlakeMW wrote:The UPS cost of "plumbing" related stuff is constant regardless of usage.

Hm. Interesting. I'll take your word for it. How about pumps/one directional tanks, i guess they don't offer any benefit then either?
I.e. if one builts a straight line of 100 pumps, vs a straight line of 200 pipes.

eradicator wrote:

BlakeMW wrote:The UPS cost of "plumbing" related stuff is constant regardless of usage.

Hm. Interesting. I'll take your word for it. How about pumps/one directional tanks, i guess they don't offer any benefit then either?
I.e. if one builts a straight line of 100 pumps, vs a straight line of 200 pipes.

IIRC fluid calculations work on relative fluid levels and pressure between two entities. A "pipe" built with pumps has the maximum possible flow rate, but for UPS calculations you'd want to reduce total number of entities. For that you'd be better off using underground pipes since they only count as 2 entities for their entire length.

xfir01 wrote:

eradicator wrote:

BlakeMW wrote:The UPS cost of "plumbing" related stuff is constant regardless of usage.

Hm. Interesting. I'll take your word for it. How about pumps/one directional tanks, i guess they don't offer any benefit then either?
I.e. if one builts a straight line of 100 pumps, vs a straight line of 200 pipes.

IIRC fluid calculations work on relative fluid levels and pressure between two entities. A "pipe" built with pumps has the maximum possible flow rate, but for UPS calculations you'd want to reduce total number of entities. For that you'd be better off using underground pipes since they only count as 2 entities for their entire length.

I know about entity count. But the point of the question was UPS of "pump" vs "pipe" (for modding purposes that are not affected by underground pipes).

I imagine a pump consumes more UPS than pipes as it is an active power-consuming entity that calculates its power usage according to how much fluid is moved each tick (and vice-verca, it also has to take into account power availability to see if it can move fluid), this is probably more expensive than pipes checking their neighbor's state to decide which way to move fluid though I don't see why it'd be unreasonably more expensive, like I'm sure it's fine to use pumps where it makes sense, but you're probably not going to gain UPS by using pumps cunningly vs a simpler setup.

BlakeMW wrote: So far as flow direction; liquids only ever flow in one direction but it can rapidly alternate.

There have been many reports of fluids flowing better in one direction than the other. For heatpipes too. Where direction is depending on build order.

Not sure if that's still the case.

What I mean is that each tick fluid flows in either one direction or the other, relative to each pipe. I'm pretty sure there's nothing happening at a larger scale. Yes, flow does propagate in some directions better than others but that is to be expected with this model (because each pipe takes its own turn moving fluid, so if the pipes take turns in the right order fluid initially zooms along, but if they take turns in exactly the wrong order it basically only advances 1 pipe per tick), though beyond the initial propagation of fluid it's not a big deal because of the "flow acceleration" thing that moves more liquid in the direction liquid moved the previous tick.