Okay guys, let's make sure we're on the same page: User has made the Constellation class star ship. They crank out 20 of these ships and they're all over space. 40 turns later they're at Phaser Mark IV technology and decide it's time to upgrade their Constellation ships. They can either A) Upgrade them all at once or Go to individual ships and have them upgrade to a particular ship type.

That's my primary suggestion. An "All" option with a "Individual" option.

I can understand it might be challenging for the AI to have to send its ships back to a starport (or military starbase that has "Stardock" capabilities). If that is so, I can accept allowing an "in the field" refit, so that humans and AI have the same choice. Of course, this would just mean that humans would tend to rotate back their units in a staggered rotation, to get cheap upgrades, until late game, where they don't care about costs, and just do an all in place upgrade. But that's just life in the big galaxy.

Any game mechanic for upgrading ships should take these factors into account: 1) The cost difference between the original ship and the upgraded ship to prevent cheese. 2) The distance from the planet it was constructed on (to prevent cheese with building super fast ships and then upgrading them to mega death ships easily). 3) The MANUFACTURING capacity of either your planet or your civilization to reward players who have built up a manufacturing capacity -- MONEY should NOT come into play unless it's at a massive penalty.

You cannot stop the player from cheesing you in the field, if you allow in-field upgrades! All you will be able to do is slow them down to a certain extent.

Now, are you really married to an in-field upgrade solution? If so, you are providing players and AI alike a production bonus, because their starports can be busy making new, cheap units, and let the in-field upgrade process finish the work! You can raise the cost up (ie, every bc/shield difference in the upgrade versus the build new cost is x5 upward (depending on whatever techs you want to use to adjust costs in the field). Refitting at a starbase or starport should be at just x2 construct new costs (because you are just refitting an existing base, the piecemeal updating should be cheaper to step up then scrap and build new, but it should cost more then just building an empty hull and then refitting so the optimal path is to build the best you can when its new!.)

If, however, you change it so all work has to be done at a starport/starbase, then the balance is back on the side of the biggest empires with the best industry. That would be the MoO2 system, btw. You could automate this approach, sit a button on a ship in the field that when clicked, sends it to the nearest StarPort and has it queue itself into that starport after whatever is being built. Then the ship would rally/auto-launch/notify as any build.

What kind of cheese do you want to stop? The SMAC minimum shell to instant uber unit? This is a way to do that. You could limit in-field upgrades to an empire's Logistics. Have any and all upgrades suck up military spending (stopping all production at home if you are upgrading a huge navy in the field, for instance), and use that production build, up to your Logistics rating (x something reasonable after game balancing). This means that in mid-game, it would take you a long time to upgrade 100 Serpents in the field, because your Logisitics just isn't high enough to rework everything quickly. This means that sending them back to be reworked at a planet's star dock wouldn't matter in a logistical sense, as that still requires the same resource that is being used to upgrade everyone. Only the actual cost is changing (based on range modifier). And I'd make the Range Mod itself based on Logistics.

For example, I have a Logistics of 50. This lets me upgrade 50 x4 (current game balance setting) shields worth of my ships (wherever). I've got 50 old battleships I want to upgrade with new defense and some meaner weapons, and the total shields worth of work, per ship, amounts to 150 each. This means I can upgrade 4 ships in 3 turns (1.33 ships per turn) That means it will take 37 turns to be able to process all those ships in the first place. Now, range could really eat into my costs (sucking yet more military production up from my empire). Under this set up, an empire with a large military budget would be able to better afford upgrading its in field units. And Logistically advanced societies would be able to more quickly and more efficently re-outfit the ships in field. (I'd suggest you modify the cost for range after determining the maximum rate of upgrading, so that the amount of ships done per turn remains the same, if the empire can afford it. This would help both players and AI figure out how long it will take to upgrade their fleets, although the range cost could make humans miscount what they get done at home while doing the upgrades).

Sound reasonable? Of course, as I said, if you just send the ships to a shipyard, you get around all this, as the same mechanism that builds new ships will upgrade old ships. But that can be troublesome to worry about when playing with massive fleets (made worse on large maps as out lying worlds might probably won't have the production capability to put in those big upgrades).

A better alternative is to have them treated as if they are a damaged version of the new ship! Let them be upgraded in the field as if they are being repaired! If Brad and his team allow ship components to become "damaged", then everything will gel together. Then your upgrading in the field capability is actually your REPAIR IN FIELD tech!

Now that idea I really like, though obviously it would require per-component HP tracking and it's likely too late to do something admittedly fairly major like that...

I think it's safe to say we won't come up with a solution that satisfies everyone.

The reason the ships are supposed to be useless during the refit is precisely as a big negative to the cheese of building unfinished ships.  They should not be usable at all while they're refitting.

So here is what I propose, using a derivative of Saber's suggestion:

When you choose to upgrade a given ship class (or individual ship) it'll pop up a dialog telling you how long it would take along with a slider bar of what % of your industrial resources to allocate towards the refit.  As you change the slider, the base time required to do the refit would change.

The further the ship is from a friendly planet, the longer it takes for it to refit.

While I do like the cargo ship concept too, I think it woudl get far far too complicate to implement in any sort of way that is intuitive to players and doesn't add potential micro management pain.

a slider bar of what % of your industrial resources to allocate towards the refit

for the love of god, not another slider

Brad, there isn't any way to make everyone happy. You know that. But hopefully, refits can be made as minor an issue as possible.

I like the consistancy of having refits be merely repairs. Whats the difference in fixing a broken destroyer that has been shot up, had half its weapons destroyed, lost 2 engines, etc and upgrading to a newer design of the same hull with upgraded weapons and engines? If you can repair in field, you certainly can perform many forms of upgrading in the field.

After that, I rather like the mini-freighters chasing down your old ships, bringing out the parts/material to be used. When the mini-freighter finally reaches the ship to be upgraded, it then starts it actual upgrade process (being frozen in the field until done). It might be heck to code, but it would provide for the players and the AI to have more raiding targets then just the trade mini-freighters. Again, a strategic opportunity put into the game, adding new options to utilize. (Indeed, if you combine this with repairing, you have a serious line of supply that can always be attacked in a time of prolonged war.)

Well, I'm done pitching options for your team to think about.

As long as the micro-management is minimized as much as possible while allowing the options of individual unit upgrades, as well as all, you'll make most people satisified with the system. I think that's the best that can be hoped for. It does seem that this thread has clarified that you want to have an independant/parellel refit capacity for the empires, in addition to their ship making, but that you want them both to pull from the same empire resources so that you cannot get 100% refitting while having 100% making new.

I think it's safe to say we won't come up with a solution that satisfies everyone.

That's true. Many of the ideas people have proposed sound good. I can't say that I HATE any of them. We trust your judgment.

BuzzardBlaster, I rather like that idea.  Also, Brad mentioned the other day that in the OS/2 version of GalCiv, there was a "Go Home" button on the ship list entries that made the ship return to its home planet.  Currently, the button on the ship list entry sends the ship to wherever the center of the screen happens to be, but I think that a "Go Home" button might be more useful. 

Last 3 posts, very nice points.

Another good point.

Brad has already stated that upgrading/refitting will be a seperate matter from new ship construction. However, the cost will come out of your normal costs, so that if you want to upgrade quickly, you'll have to take that money from another cost (ie, miltary, social, research). Whether it will be 4 cost sliders instead of three, or a split in your military slider (between new ships and refitting ships), it all means that you cannot max out your building and refitting at the same time.

As for the reason for factories... you only need factories on worlds that build stuff. Primarily, they are needed on new starship makers. On smaller worlds, I'll buld a couple of factories, then build my labs and banks, and then go back and slowly replace the factories with more useful tiling (ie, labs and banks). Factories aren't useful for anything else, so don't waste tilage on them if you aren't going for a big production world chunking out products. That was true in GC1, and its still true in GC2.

Excellent idea, ProudCanadian!