Energy use and deficit in CW2

[Karsten75]

You mentioned this early on, but I can't find it and a wider audience might appreciate this info.

Can you please go over how the deficit is calculated in CW2?

Also, I can't seem to get the stable pattern of building units I got in CW1 - mostly because I can't serialize builds, since one oneit is not dependent on a previous unit. So I mostly find myself going through swings of surplus and deficit in energy.  I was wondering if you doubled reactor cost and output if it would help to even out the swings a little bit?

[Aurzel]

no way dont double reactor cost! lol

but i do wonder about what karsten's said myself, it seems the energy fluctuates a lot more

[Karsten75]

If you double both production and cost, it nets out, it just means you don't have to place a reactor every 5 packets...

[Sqaz]

If you double both production and cost, it nets out, it just means you don't have to place a reactor every 5 packets...

Double cost will mean it's harder to get energy in the beginning, cause while those 5 extra packets for the more expensive reactor are still en route you already get energy from your cheap one.
It also makes terrain management less needed, now you need to dig 1 ground per reactor which means a reactor is 8 energy (*2 = 16) , while the expensiver ones will only be 13 energy (so you'd save 3 energy).

[knucracker]

Deficit represents the number of outstanding structures requesting packets.  So if you have a deficit of "5" it means that you have five places that have asked for a packet but are waiting on it.  Structures won't keep asking for packets until their previous order is 'shipped'.  Hence, deficit equates to the number of structures that are 'starved'.  This differs from CW1 where structures kept ordering packets (and the deficit would grow and grow).

Energy demands can be greater in CW2.  This is because things like Tech Domes and Beacons consume energy.  They can cause an irregular drain on your energy infrastructure that can make a shallow energy reserve dip from time to time.  Most of the time, its a transient dip and doesn't really affect much in terms of weapon starvation.  But I could imagine certain game scenarios where it would matter.  But that's what the tech upgrade for increasing the energy reserves is for (along with careful crystal management)

[Karsten75]

It seems *all* the tech domes and beacons request energy at exactly the same point in time. Is this a valid observation? Or do they request packets a certain interval after their build/consume state?

I'm mindful of the fact that beacons and domes have a store of energy, and maybe, somehow, they should be prioritized or sequenced so they don't demand energy packets the moment they consume one.  Not sure exactly what I'm asking for yet, but just talking out loud to see what it generates.

[ontheworld]

little off topic: how did this get named CW1?

[Karsten75]

little off topic: how did this get named CW1?

Are you familiar with the concept of "typographical errors?"  :D

[knucracker]

Domes and beacons consume energy on their own schedules (at an interval that start when they are built).  This interval can be exactly the same for units that are pre-existing on a map.  I have considered slightly randomizing the start of the  interval for tech domes and beacons that are preexisting structures on a map.

For things that are built during a mission, they can get synchronized if you ever end up in a deficit situation for long enough. 

[Karsten75]

I'm probably overthinking this, but I'm thinking that if I'm starved of energy use, slow consumers (domes and beacons) need not request a new packet after they have only consumed one of the packets in their inventories. But I can see the complexities that may arise here. so I guess we'll (and that means you) leavethis as-is.

Also making this less of a problem is that by the time beacons and domes are significant consumers is the fact that by then I have a large reactor farm usually.

Which segues into another issue for which I'll start another thread.

[Fisherck]

While on the subject of this, I was wondering if ore rigs always use energy. And I mean this by that ore rigs are not always "running". They stop working as soon as the ore bar fills up. So, if they keep consuming energy during their "waiting" period, it would not make much sense because the same state could be created with the deactivate button.

[knucracker]

The rigs should only use energy while their drills are running.  Their drills will stop when there is no place to put the ore (your ore reserve is full).

[Fisherck]

Okay, I was not sure. Thanks.

[Karsten75]

I'm sorry, but I *still* don't understand the energy dashboard/indicators.

Look at the attached image. A second or so earlier, I had a -24 "demand?" and now a -16 demand. I'm building 4 reactors and excavating (if that's the word) two gems and two nanochip deposits. So how does that account for a -16 energy consumption?

What's the metric? What is the time range?

[knucracker]

That number is literally the number of packets dispatched over a 30 frame cycle.  I mod the frameCount by 30 and then look at the running total of dispatched Packets since I last looked.  That is the number shown.  Put another way, it's a packet-per-second count, updated every 30 frames.

The build and supply rate for most things is 15 frames (2 per second).  So if you were building 8 things at once, I would expect to see -16.  As show in the picture I would expect to see -12.  I suppose you could have used some energy right before the picture was taken but more likely I'm going to guess that some packets fell just inside  a frameCount % 30 window.

Now I could smooth this number out by just changing it to a 2 second calculation for instance.  I would still report the value in PPS, it would only get updated every 2 seconds though...