Ye Ol' Reactor debate

[Diazo]

Except that would kill me early game.

I play single-HW start and have a defensive play style. I would run out of energy so fast it would not be funny.

The unlimited energy option has to be there, just expensive.

Hence my comment about keep the multiple Converter at a planet production penalties. You are not hard capped but it is still to your benefit to spread your Converters out.

D.

[keith.lamothe]

I have another idea for encouraging spread-out for later, but for now you'll presumably always have the collector on each planet, which the AI can then hit.  If you have enough converters next to your home command station (or wherever) to run everything with minimal collector help then your econ is probably deep in the trash can.

[Wingflier]

In that case I hope the Converters are extremely inefficient.  Not only do they have to make up the resources you aren't spending before (57 M+C per planet) but also be sufficiently draining to prevent you from creating a huge energy cushion on a heavily defended planet.

[Diazo]

I am making some assumptions here so we'll see what the numbers are, but early game I'm always building multiple of each reactor type at my HW and first few colony planets.

I'm expecting to be reaching for the Converters early in the game and quite often, not because I'm not using Collectors, but because of my slower more defensive play style which is energy intensive.

As it stands, it's going to be "Out of energy? Build a Collector on my HW", there is no decision there.

With the penalty for multiple Collectors at a planet, I'm now encouraged to spread them out, but not forced to like a hard cap would.

D.

[keith.lamothe]

Ok, I've got it ready to go, but wanted to do a little mathematical due-diligence to make sure I'm not changing the econ game by an unnecessarily large margin.  Currently the numbers I'm using are:

Collector:
1 per planet
150,000 e output
no ongoing m+c cost (which is staying that way)

Converter:
no per-planet cap
75,000 e output
100m+100c per second cost (I don't want to go higher than that to avoid getting too coarse, but the e output can change)

To put it in perspective, here's the efficiency-optimal build in the old system if you're trying to squeeze the most power out of a single planet for the least m+c:

Type	Count	BaseOutput	EfficRatio	Output	M+C	e/(m+c)	TotalOutput	TotalCost
EII	1	40000	0.6	40000	30	1333.33	40000	30
EI	1	5000	0.8	5000	4	1250	45000	34
EI	2	5000	0.8	4000	4	1000	49000	38
EIII	1	80000	0.2	80000	80	1000	129000	118
EI	3	5000	0.8	3200	4	800	132200	122
EII	2	40000	0.6	24000	30	800	156200	152	(Collector)
EI	4	5000	0.8	2560	4	640	158760	156
EI	5	5000	0.8	2048	4	512	160808	160
EII	3	40000	0.6	14400	30	480	175208	190
EI	6	5000	0.8	1638	4	409.5	176846	194
EI	7	5000	0.8	1310	4	327.5	178156	198
EII	4	40000	0.6	8640	30	288	186796	228
EI	8	5000	0.8	1048	4	262	187844	232
EI	9	5000	0.8	838	4	209.5	188682	236
EIII	2	80000	0.2	16000	80	200	204682	316
EII	5	40000	0.6	5184	30	172.8	209866	346
EI	10	5000	0.8	671	4	167.75	210537	350
EI	11	5000	0.8	536	4	134	211073	354
EII	6	40000	0.6	4000	30	133.33	215073	384
EII	7	40000	0.6	4000	30	133.33	219073	414
EII	8	40000	0.6	4000	30	133.33	223073	444
EII	9	40000	0.6	4000	30	133.33	227073	474	(+1 Converter)

In other words, previously you could get 225,000 e from a single planet for (474m+c)/sec, and further increases would be at the rather low efficiency of 133.33 e/(m+c).

With the new model numbers I'm using now, you could get 225,000 e from a single planet for (200m+c)/sec, and further increases would be at the (also rather low) efficiency of 375 e/(m+c).

So I'm thinking of changing the converter to 50,000 for 100m+100c, so that the comparison would be:

Old: 200,000 e from a single planet for (316m+c)/sec.
New: 200,000 e from a single planet for (200m+c)/sec.

In other words, a clean 50% boost (at that stage, it still goes on to being nearly twice as efficient in the infinite-stacking part of the graph), but not an over-100% one.

50% may even be too much, but the other side of the coin is that you can no longer mitigate your energy cost by shutting stuff off (though I did take the 8k/each energy costs off the home settlement stuff because that would just be a kick in the pants with no low-power and no scrapping).  So I'm thinking that the 50% boost is a good level of compensation for the amount of wiggle-room that's being removed (though the reason it's being removed is that it's more fun to shoot stuff than to wiggle around energy costs).

Thoughts?  Any strenuous objections?

[Eternaly_Lost]

Thoughts?  Any strenuous objections?

Multi-Homeworld? Do we get collectors = homeworld numbers per planet? Homeworlds/2?

My current 4 homeworld game, I am using about 18 million energy to power all my stuff.

[keith.lamothe]

Multi-Homeworld? Do we get collectors = homeworld numbers per planet? Homeworlds/2?

My current 4 homeworld game, I am using about 18 million energy to power all my stuff.

Sorry, forgot to mention: collector output is multiplied by the owner's homeworld count.  Converter output is not impacted, but there's no cap on them so it doesn't matter.

So in a 4-HW game a collector outputs 600,000 energy, which means it would take 30 collectors to power all that stuff.  Knowing you, you may well already have 30 planets   Or do you have less?

[Diazo]

Hmmm. Going to turn this around a bit. Using 50,000 output for converter and only a single HW start before capturing any systems as early game is my main point of concern.

Stage1:
New: Collector 150,000 energy, 0 m+c, free energy!
Old: EIx3, EIIx2, EIIIx1, 156,200 energy, 152 m+c, 1028 energy/resource

Stage 2:
New: Collector + Converter x1, 200,000 energy, 200 m+c, 1000 energy/resource
Old: EIx9, EIIx4, EIIIx2, 204,682 energy, 316 m+c, 648 energy/resource

Stage 3:
New: Collector + Converter x2, 250,000 energy, 400 m+c, 625 energy/resource
Old: EIx11, EIIx15, EIIIx2, 251,073 energy, 654 m+c, 384 energy/resource

Stage 4:
New: Collector + Converter x3, 300,000 energy, 600 m+c, 500 energy/resource
Old: EIx11, EIIx28, EIIIx2, 303,073 energy, 1044 m+c, 290 energy/resource

So regardless of what happens, this does not break the game in my opinion. In fact, it makes it a bit easier as under any circumstances you have more energy per resource then before.

I would still argue for something to avoid stacking all the Converters you need in your home system, but you said you've got a mechanic in mind for that so I'm willing to wait and see.

D.

edit: Multi-HW multiplies energy output of the collector? You already get more collectors by virtue of having more HW, that seems like overkill to me. It also does not break the game mind you so I'm not sure how big a deal it is.

[keith.lamothe]

I would still argue for something to avoid stacking all the Collectors you need in your home system, but you said you've got a mechanic in mind for that so I'm willing to wait and see.

Yea, this definitely isn't where I want the leave the energy model forever, just trying to resolve the more fundamental problems of a mindnumbing-micro-heavy model (for the folks who actually pay attention to energy efficiency) so we have something to move forward from.

edit: Multi-HW multiplies energy output of the collector? You already get more collectors by virtue of having more HW, that seems like overkill to me. It also does not break the game mind you so I'm not sure how big a deal it is.

Yea, I'm debating making it multiply by 1.5^(homeworld_count-1) instead.  I was wondering how many planets Eternaly_Lost had in that scenario to figure out how many converters that would mean for him.

Bear in mind that starting with extra territory (and thus extra collectors) is a great early game advantage but really doesn't scale up well mid/late for energy specificallly : sure you start with (say, for 4HW) 4x the energy and 4x the ship caps, but by the time you've taken 8 planets that's down to 1.5x the energy and 4x the ship caps.

[Diazo]

I would still argue for something to avoid stacking all the Collectors you need in your home system, but you said you've got a mechanic in mind for that so I'm willing to wait and see.

Yea, this definitely isn't where I want the leave the energy model forever, just trying to resolve the more fundamental problems of a mindnumbing-micro-heavy model (for the folks who actually pay attention to energy efficiency) so we have something to move forward from.

Heh, that wasn't me. Depending on how a specific game went, the first thing I usually built in my first colony system was 5 reactors so I could turn some of the ones in my home system off.

This thread is making me realize how differently I handled energy then most other people.

edit: Multi-HW multiplies energy output of the collector? You already get more collectors by virtue of having more HW, that seems like overkill to me. It also does not break the game mind you so I'm not sure how big a deal it is.

Yea, I'm debating making it multiply by 1.5^(homeworld_count-1) instead.  I was wondering how many planets Eternaly_Lost had in that scenario to figure out how many converters that would mean for him.

Bear in mind that starting with extra territory (and thus extra collectors) is a great early game advantage but really doesn't scale up well mid/late for energy specificallly : sure you start with (say, for 4HW) 4x the energy and 4x the ship caps, but by the time you've taken 8 planets that's down to 1.5x the energy and 4x the ship caps.

That is true. What about also boosting Converter output on multi-HW starts in exchange for not as big a boost on Collectors?

The general feel I'm getting is (barring high energy units such as fortresses) Collectors only should power the mobile fleet fully built and a few turrets, but if you want to build turrets up to cap as well you'll be reaching for the Converters.

I think at this point we need to mess around with it. The numbers you have posted don't break the game in my opinion, if anything they are a slight buff for the players.

I'll also pay attention to energy when I do, I'd gotten into the habit that about the 10min mark in my games I plunked down a CTRL- Reactor II to build 5 of them. Sloppy yes, but I needed the energy.

D.

[keith.lamothe]

Ok, the stuff's in the svn and the release notes have been updated: http://arcengames.com/mediawiki/index.php?title=AI_War_-_Current_Post-5.000_Beta#Energy_System_Rework

Planning to push the release itself out today at some point, there's a thing or two I want to look at first.

Thanks for the feedback, everyone

[Diazo]

Nice.

If it does get out I'll start a new game to see how it works out for me this evening.

D.

[Eternaly_Lost]

Multi-Homeworld? Do we get collectors = homeworld numbers per planet? Homeworlds/2?

My current 4 homeworld game, I am using about 18 million energy to power all my stuff.

Sorry, forgot to mention: collector output is multiplied by the owner's homeworld count.  Converter output is not impacted, but there's no cap on them so it doesn't matter.

So in a 4-HW game a collector outputs 600,000 energy, which means it would take 30 collectors to power all that stuff.  Knowing you, you may well already have 30 planets   Or do you have less?

At this game, I am at 60 planets and well on the way to taking all 120 of them.

[keith.lamothe]

At this game, I am at 60 planets and well on the way to taking all 120 of them.

Yea, that would give you twice as much energy as you actually need right now, with zero m+c expenditure.  Which means your rebuilds would probably be even more ridiculously fast (60 planets worth of harvesters is probably pretty insane already, though).

[Eternaly_Lost]

At this game, I am at 60 planets and well on the way to taking all 120 of them.

Yea, that would give you twice as much energy as you actually need right now, with zero m+c expenditure.  Which means your rebuilds would probably be even more ridiculously fast (60 planets worth of harvesters is probably pretty insane already, though).

More so when you throw the 20 spire cities I have active right now. The goal with this play though is to have as many of them as I can, so I did 120 planet snake map with my starting homeworlds so that I could place as many of those are I can.

There is something nice about 24 Spire dreadnoughts attacking the next system together, although the EXO waves are starting to get rather large.