Rollout times and economy fluctuations

I've mentioned this a few times in various places, but I would like to get a bit of discussion going on this topic in particular, as I've not come up with a satisfactory solution in my own head yet.

I'll start by saying that I very much support the long rollout times for factories, and the playstyle that encourages. I don't want rollout time reduced or removed at all.

With that being said, I really don't enjoy the massive economy fluctuations that result from factories drawing resources for about 60% of the production cycle for a unit and not drawing them for the remaining 40%. For a single factory, this would be fine but the above-mentioned playstyle, namely building lots and lots of factories, means that this fluctuation begins to amplify very quickly.

It makes economy management and balance a lot harder and my poor little brain can't work out whether my economy is healthy or buckling, even with the assistance of YourLocalMadSci's lovely Economy UI Mod.

The obvious solution is to make factories continue to draw resources through the rollout time as well, but this doesn't sit very well with the fact that they would then be consuming metal without spewing out any green particles.

A proposition might be to change the animation for the rollout phase to 'justify' their continuing to consume resources throughout rollout, but I'm not sure exactly how much effort would be entailed in this.

The floor is yours - am I making a big deal out of nothing? Are there any obvious solutions staring me in the face?

Yeah I have thought about this a lot too. The best solution I could come up with is what I do ( or at least try to do). Basically it takes 20 secs to produce an ant and 10 seconds to roll off. I usually take a group of 9 factories making ants, divide them in 3s and start each group of 3 ten seconds after eachother.

The above method works and can be implemented with all factories but is quite tedious. Obviously building a metal storage will help this but at this stage in development I feel mas storage is a waste of mass. Only time I can see me wanting a mass storage is when wreck reclaiming is introduced.

I think the overall problem is just cost. The cost of units is high in some cases and low in others but they cost -10 mass to build no matter what. This could just be the TA side in me comin out. Im not sure how supcom, faf or sc did this. But on TA the mass usage from the Fabber was proportional to overall cost of the building it built. This was controlled by energy needed though.

I feel like you're going to want MORE metal storage once wrecks are introduced. That way you can run a pretty big deficit and just rely on the sporadic bursts of income as your engineers reclaim attack waves. More storage means you won't waste the spike of income and you can let it slowly dwindle while waiting for the next delicious herd of soon-to-be-dead enemy robots.

As far as the OPs idea, I think that something like this would make it easier to balance your economy, but NOT changing it promotes building more metal and energy storage, which I think is a positive thing. So I'd have to say, don't change it.

The temporary and most obvious solution to this is Metal and Energy storage, to buffer the massive fluctuations in resources; which I agree on, it is very hard to tell when your economy is hurting or growing unless you frequently check your amounts in storage, and that's assuming you HAVE anything in storage (which you shouldn't for metal if you are maximizing your efficiency).

The first solution I can come up with is hacky and very weird. On each factory, there would be two construction pads; rolloff time is exactly equal to build time, and one unit is always building while one unit is always rolling off. This keeps the rollout time, and as a result, reduces the efficiency of using Fabbers to assist factories, and encourages use of a multitude of factories to create armies as opposed to one assisted factory.

However it also means that factories must be rebalanced to have "twice" their build-power. It also means you have to pay for two construction pads when in some cases you may only want one. In addition, it means you are loosing a lot of time when the factory stops and starts building, as only one unit will be building instead of one rolling off and another building.

My next thought is to have the factory "charge" metal and energy capacitors before building, maybe to create the wireframe of a unit which it then nanolathes. This is also hacky and, as WrongKitty said, would result in players being confused as to why their factory is eating resources and not building stuff. There would have to be an obvious indicator as to why the factory is consuming resources and not building, and it would result in confusing calculations both for the devs and for competitive players for resource consumption, and it would overall be a very clunky solution.

The last solution I can think of is adding a resource graph to the top of the HUD. That way players can recognize and identify fluctuations in their economy, see patterns, and using Storage (it's good that Storage finally has a use in an "Annihilation" game), can allow their factories to run at peak efficiency without squandering resources or stalling. The downside of this is that it involves tacking a graph onto live gameplay.

I would like it if the factories would show (when selected and in the tooltip in the building-selection menu) how much they actually suck in metal and energy.

this does indeed encourage you to build storage. I think that's what you've been missing. other than that. you can either offset all the factories for a more fluid economy or make them all perfectly rythmic and micro engies on and off during the factories rolloff time on something very big and expensive that you're trying to build very fast.

My answer: Keep building eco until you win.If that doesnt work and you're building THAT many factories you're doing strategy wrong.

It's not quite as simple as that. If you only build eco when your opponent is building tanks you're gonna have a bad time.

It's a strategic choice, do you build an incredible amount of tanks and neglect economic expansion in order to rush and overwhelm your enemy, or do you focus on Eco and hold the line until your production is superior?

It's always been a key choice in TA, one of the few aspects that SupCom retained untainted

I find this a bit of an issue as well. It isn't a huge one, but it does make it that bit harder to gauge the healthiness of your eco. Even with energy storage, the massive fluctuations in energy drain can be really weird.

The best solution I can think of is to make the energy drain constant through out build phase and roll-out time. Ideally you'd want to change the animation so something is going on in roll-out time (prepping for the next build)... but I don't necessarily see that as an issue. As long as you saw units in the build queue, you'd know that the factory was 'live' and not dormant. But I guess that does require selecting the factory. Some other form of visual indicator then. Maybe each factory can have a light on it, heh!

whoa! necro

dw I think the game is edging towards more storage use.

Well I find it makes storage a useful, viable choice. Something that was severely lacking in supcom.

Yeah, zihuatanejo and I were discussing this on the IRC. My main concern as we approach beta is that the fluctuations are going to increase in scale as we find the number of factories growing to fight on several planets at once and make the whole thing very difficult to gauge.

Sounds like a good thing - stops storage becoming defunct lategame.

I'm all for the fact that it makes energy storage useful and necessary (continually so as ZaphodX points out, in the end-game you'll need more energy storage) - so how about this. Could some sort of 'max potential energy consumption' variable be created and tracked? I quite like that idea because it keeps things as they are, but it would just give you that info you need.

If you see that max potential drain is under your total production rate, then you'll know you're ok. If you see that max potential drain is higher than your current production, even if your energy bar isn't drained, you know you need to up your production.

Tricky part would be factories vs fabbers. F actories would be more constant. Perhaps this is too complex. Dammit!

Why not just have an average of input-output over the last minute. If its negative then you need to halt production or build more eco. If its positive then you should build more. If its positive but you still stall then you need to build more storage.

Intial reaction: not bad! That could work nicely I think.

Yeah, thats the real solution. An additional value that shows dampened rolling average. Another alternative is to vary the rollout time with random variation to add noise. The more factories built, the more likely the pool is to average over time.