I've never understood well the retrofit costs, until today. Now I know!



You can find this in the file C:\Program Files (x86)\Steam\steamapps\common\Endless Legend\Public\Simulation\ResourceConverterDefinitions.xml

[CODE]



($(Input) * $Property(../ClassEmpire:GameSpeedProductionToMoneyMultiplier) * 0.75) ^ 1.1



[/CODE]



So, retrofit costs, for normal game speeds, is ( <> * 0.75 ) ^ 1.1 - (<> * 0.75 ) ^ 1.1



Example: Marine with standard equipment costs 80 (unit cost) + 18 (unit equipment cost) = 98.

Marine2 , in full tier 1 equipment plus necklace costs 80 + 66 (unit equipment cost ) = 146



To retrofit a marine into a marine 2 , the cost would be ( 66 * 0.75 ) ^ 1.1 - (18 * 0.75) ^ 1.1 = 55.6. As showed in the game!



One interesting question, now: is it cheaper or is it more expensive to have lots of smaller retrofits instead of a single, larger retrofit? The answer is : neither! Since the costs are always based in the difference between full equipment costs, unless there are some rounding errors, it is the same thing to make many smaller retrofits instead of a single, larger one...



Another interesting question: is it cheaper to build all units as cheaper fits and then retrofit them to better units, or is it better to produce upgraded units from the start ?



That is not a very easy one, and will depend much on the production x dust relative productions for your empire... If you have gold to spare, and you are buying out production, to hurry cheap units and then upgrade them is much better than trying to hurry the upgraded ones directly. If the correlation between production and dust income is lower than 7 x 10, it is probably better to produce cheaper units and then upgrade... Unless you are using dust for other things...





[CODE]



($(Input) * $Property(../ClassEmpire:GameSpeedProductionToMoneyMultiplier)) ^ 1.2



[/CODE]

This means that if you have 98 production and would want to buy the remaining production for the 146 production you would expend (146-98)^1.2 = 104.9 gold...

- Excellent post abmpicoli. It should be conserved somewhere and put in the wiki one day !

Another interesting question: is it cheaper to build all units as cheaper fits and then retrofit them to better units, or is it better to produce upgraded units from the start ?

- I always had intuition that it was equivalent.



- But let me ask you a question : is it bad to often sell the same kind of units ? I got impression that when you sell again and again the same kind of units (marines for exemple), the price is going down. Am I crazy to think that ? ^^

This thread doesn't cover "selling" costs at the marketplace... The idea here is to evaluate hurry costs x retrofit costs, and evaluate if retrofitting too much is a good or bad thing...



Now, regarding units sell price... Since the units tab in the marketplace have a "demand" field attached to it, I think it is possible that units price will go down if you sell units too fast... I really don't like selling units to the marketplace as a standard means to convert production into gold, unless there is absolute no way to do something else... Stockpiles are a good option, if you stop to think about it: one thing I did with my vaulter's game was to build food stockpiles in my main cities: this allowed me to make the starter/recently conquered cities to grow very fast, by activating food stockpiles upon them. At era 4 they have a reasonable conversion ratio. 2500 production generates 900 of (science/ food / production) that can be redirected to those cities that are really lacking...



And notice that 900 food for a recently started city will mean about 5 new workers, that can be redirected to generate dust... And 900 production for a small city will mean probably a "free" borough for those smaller cities, getting yet more dust outcome... Or a RightOfWay generating 20/20 dust income per turn.

Yes, I would not wanted to make an off topic.



I thought your analyse and conclusions were clears : it's always better to retrofit in compare to hurry the units. I find it strange, I thought it could be egual ! But if you observed it, it is the truth.



Off course, if as civ you got special boost to hurry cost, it become more interesting to always hurry.





PS : In my opinion, hurrying units should be twice more expansive than hurring building, because if not, a dust civ as the broken lords can rush an army everywere when they want, without fearing a surprise attack. It is anti game I think.



A more elegant solution should be finded than just doubling the price.

It's cheaper to buy out a naked unit and retrofit it than to buy out a fully kitted unit. But usually, if you're doing unit buyouts, it's due to emergencies that might not leave you the opportunity to retrofit (you'll be under siege and won't have access to strategics for retrofit, for example).



There's another advantage to late retrofits, which is that you can place cheap naked armies in multiple locations and retrofit them into elites when you feel the need. You might not have the strategics to maintain a strong army on both your eastern and western borders, but you can upgrade instantly whereas it might take several turns to get a strong army where you need it.

This will change style for all Broken Lords. Buying naked units and retrofitting is cheaper than buying well dressed units ?



You can even decide which one you want to retrofit after you already send units to explore. First to encounter the enemy can be upgraded. And you can even save yourself dust. This is really awesome.



Do I get it right ?

You did get it right, but in my opinion it's not as big of a deal as that. By the time BL are buying out units rather than creating them with industry, the game should already be close to won.

I want to attract the attention of the devs about this.



If it is cheaper to retrofit than to buy a full equiped unit, it is absolutely anormal. Retrofiting mean flexibility, flexibility should be very expansive, if not it is too easy to have unimproved garrisoned unit which you upgrape when the ennemy is coming.

I must be missing something here.



So let's say I'm playing the Vaulters, and have a fully upgraded Marine unit. I create a Recruit unit using the Marine model, completely barebones. But I can't upgrade the Recruit to Marine, as they're not the same unit.



Do you guys mean editing the Marine unit, stripping it of all gear, producing the units in question, and then editing the Marine again to re-upgrade it in order to finally retrofit the stripped down ones? Every time I want to produce a batch of units? Sounds very cumbersome.

LordShadow wrote:
Do you guys mean editing the Marine unit, stripping it of all gear, producing the units in question, and then editing the Marine again to re-upgrade it in order to finally retrofit the stripped down ones? Every time I want to produce a batch of units? Sounds very cumbersome.

Yes, that is what they mean. Yes, it is cumbersome. The fact that you can gain an economic advantage by doing something cumbersome (i.e. that the UI is clearly not designed to facilitate) seems kind of bad, eh?



(It's not quite as cumbersome as you make it out, though. You can use multiple unit templates (that will sometimes have identical specs) in order to upgrade your existing units while working on a batch of new ones; you can also add a bunch of copies of a unit to your production queue before you edits its template to continue producing the old design even after the edit.)



The basic problem is (or appears to be) that they want buyout to be inefficient (so that you'll try to build things normally when you can), but they also want retrofitting to be fairly efficient (so you won't feel bad about upgrading your old units to newly-unlocked gear), and if both of those things are true, it basically guarantees that building naked units and then retrofitting them will be more efficient than rushing fully-equipped units.



If one wanted to place a high priority on avoiding that, the answer is probably to make it so that retrofit cost is always (buyout cost of new design) - (buyout cost of old design). And then maybe make the buyout formula different for units than it is for buildings so that that's less harsh on retrofitting.



But then, how does one account for buyout cost reduction? Or for units whose equipment costs a different amount now than it originally did when you purchased it?