Question about logic triggering weapons computers

[Johnathan Dahmnation]

Based on info in another thread I was experimenting with potentially offloading power production to docked ships since apparently a ship can receive surplus power from it's docked entities. The idea was simple, using a series of powerdrain beams the ship leaches from the docked power producing ships. With a bit of help from logic i could maybe even automate the process to allow for one man operation.

Only issue, I couldn't get the powerdrain beams to actually hit the docked ships. The seem to shoot right through them. I tried having the power drain on the main ship as well as a turret. Neither seemed to work in getting the beams to hit the docked ship, but were definitely firing.

So I'm wondering, does this actually work in theory, and is there something in the implementation I'm doing wrong? Does in not work when the computer is slaved to a logic system rather than the core?

 

[NeonSturm]

You need power supply beams in your reactor.

You get about 45% efficiency for block count: 150 power tanks (vs lag spikes), about 6k reactors (to reach 1kk regen), 3k power supply beams (each one works for 2.5 .. 3 reactors, but instead of every second they may fire only every 1.5 seconds due to logic or 2 seconds with huge lag spikes)

 

[Johnathan Dahmnation]

Ok, so I had the method backwards. Got it. Though @ 6k reactors I wonder if it's worth doing as compared to the space of brute forcing below a certain desired power.

 

[NeonSturm]

As I sad: you have 45% efficiency.
6k reactors to get 1'000'000 energy per second, 3k for supply beams, 150 mass for tanks (they rapidly exponentially increase capacity at this point, thus why 130 are too few).

 

[Johnathan Dahmnation]

Please keep in mind you are several formulas ahead of me. I only know of the basic principle here. I'm missing most if not all of the underlying math.

 

[Crashmaster]

Basically you get high e/sec efficiency out of the 6k reactors but suffer from the additional mass of the 3k power supply beam blocks, logic and power tanks. Therefore around 45-60% efficiency (depending on reactor design/ number of power tanks used) compared to 1 mil e/sec out of 6k reactors efficiently built into the ship itself.
At the 25 e/sec that you get from reactors once over the soft cap it takes 40000 blocks to make 1 mil e/sec so a docked reactor made with these block stats would be roughly 400% efficient in comparison to that. Roughly.

 

[Johnathan Dahmnation]

Ok, got something backwards there explaining my confusion. "Reactor" referring to the individual blocks, not the docked ships being created for power generation. Not sure how I came to the opposite conclusion. Pardon my derp. Though that brings another question as to whether going for smaller generator ships with higher e/reactor would yield better results. Or does the scaling of the power tanks and supply beams favor larger configs?

 

[Crashmaster]

Power gen (previously referred to as reactor) blocks give a bonus that increases the longer the line of power gen blocks is that you are adding them to. So the longer the line of power gen blocks the better. Therefore the larger the docked power generating ship is the longer the line you can build in it will possibly be.
Longer lines = more bonus per block = greater average e/sec per block = less total mass of power gen blocks to get to the 1 mil soft cap = more efficient e/sec per mass = less thrust needed = less power consumed by thrusters

 

[Johnathan Dahmnation]

Feel free to correct me if I'm wrong, but I thought bonus gains diminished as you approached 1mill bonus power, making smaller arrangements give more e/block.

Since the dimensions scale linearly in a line now, or any other shape where each block expands the dimensions by 1, if this is true wouldn't each block give slightly less than the one before?

 

[Crashmaster]

Well, you are correct, It seems 400-500k e/sec per reactor may be the most efficient mass-wise.
One line reactor;
@ 300k e/sec one additional block adds 1560 e/sec
400k - 1640 e/sec
500k - 1630 e/sec
600k - 1550 e/sec
700k - 1375 e/sec
800k - 1100 e/sec
900k - 723 e/sec
1 mil - 210 e/sec

So two 500k reactor ships take 856 power gen blocks and one 1 mil reactor ship takes 983 power gen blocks.

But that is for lone long single lines of power gens. Split into 8 equal lines it takes 2060 power gen blocks to make 1 mil e/sec with an additional block at 1 mil e/sec adding 175 e/sec. It get real complicated when you try to weave multiple lines of various lengths and shapes into an efficiently sized and shaped dockable ship that can fit inside the main ship.

Forgetting the maths as is getting over-complicated, the main reason I used docked reactors that push to 1 mil e/sec is to minimize the number of docked entities on the main ship due to lag and some server rules though efficiency is not as good as I had thought.

I have certainly realized that I do not have as complete an understanding of efficiency optimization of docked reactor ships as I had thought. But I do have a headache now.

 

[NeonSturm]

With the old system, I've built a 123x7y7z 1kk reactor (would yield less now), but it was really hard to dock and created heavy lags if you accidentally undock or the dock gets killed.

You also have to count in the enhancer blocks, but enhancers (40-100 mass) and power tanks (about 140-150 mass) are not much compared to 3000 blocks for power (<5% for 1kk) and about 8% for 500k

Maybe you want to try some reactors with 35'000 power per second to reduce loss from lags but neither require power tanks -> I have not tested these and it somewhat depends on your server's default energy capacity for cores, thus I can not give information about these,

 

[Johnathan Dahmnation]

I haven't changed the core capacity from the defaults, so 50k. Seems I'm got some ideas now. Though one thing I noticed, and I'm almost certain I'm making some sort of mistake here, is that for some reason adjacent power supply beams seem to not stack effects. 1 alone consumes 60 power same as 10 in a row as I'm testing. Is that supposed to happen?

 

[NeonSturm]

I guess not.
/data/config/blockBehaviourConfig.xml

        <PowerSupply>
                <BasicValues>
                        <SupplyPerHit>40</SupplyPerHit>
                        <PowerConsumption>60</PowerConsumption>
                        <Distance>200</Distance>                        <!-- timeBetweenHits = 1 / (unitSize^pow)*mult -->
                        <SpeedPowPerUnit>1</SpeedPowPerUnit>          <!-- this non linear scaling up of beam ROF the larger an array is will impact balance in a nasty way, can we make it a set value -->
                        <SpeedMultPerUnit>7</SpeedMultPerUnit>
                        <CoolDown>1</CoolDown> <!-- Time it takes to fire beam again from the start-time it first activated -->
                        <BurstTime>1</BurstTime> <!-- Time the beam will fire -->
                        <InitialTicks>0</InitialTicks> <!-- Ticks to do at the initial contact of beam with a block -->
                </BasicValues>

Each hit should supply 40 power for 60 drain at 200 distance (dunno if drain is consumed in your tests though it produces no supply)

SpeedMultPerUnit should multiply 40 and 60 both by (7*blocks in your array)

Did you really select all blocks? Mouse over your cpu, press C to select, mouse over your power-supply, press shift+V to select all.

 

[Johnathan Dahmnation]

I definitely had them all selected. When I deleted the connecting ones the power increased. 5 in a line consumes 60e, deleting #'s 2 and 4 and doing nothing else and it jumps to 180e (now 3 separate elements).

Edit: Solution discovered, I need a target to shoot at. If aiming at empty space a group only consumes 60 energy regardless of size.

P.S. Man I am derp today, apologies for noobing this us so much and thanks to both of you for the responses

 

[NeonSturm]

Hey! I did not know either that it drains 60e while shooting into the void.

I guess I would have filled a bug report and you should too (if I understand it correctly). It should either consume noting or 7*60 and suply 7*40 for each block in an array each second - that is what players expect
Do you really need to split that array? or just set the right block being an output?

 

[Johnathan Dahmnation]

The array works fine with a target, even combined. It's just firing into the void that confused me. Not sure if it's just me but the generator I'm working on still reports 60 consumption when aiming at nothing. drains 1mill per shot when aimed at an unconnected core. At first I had no target ship, but once I did I saw the real numbers.

 

[Keptick]

So, did I misunderstand or is there a 1/3 energy loss when transferring power?

 

[Johnathan Dahmnation]

Yes, there is, though once you exhaust the 1 mill power bonus, or come close enough to it, the loss involved still allows more power to be transferred than brute forcing power of the same mass.

 

[Crashmaster]

I'm not so sure about that energy loss. My system of 20 reactors producing and transferring approx. 1.1 mil e/sec plus the 1 mill or so from the ship's power gens will show stable power while the power consumed reads in the 21 mil range.

 

[Johnathan Dahmnation]

Interesting, I was thinking the supply and consumption scaled proportionally. More than happy to be wrong on that.