FreeOrion

Forums for the FreeOrion project
It is currently Mon Dec 11, 2017 6:40 pm

All times are UTC




Forum locked This topic is locked, you cannot edit posts or make further replies.  [ 18 posts ]  Go to page 1, 2  Next
Author Message
PostPosted: Mon Jan 21, 2008 10:45 pm 
Offline
Programming, Design, Admin
User avatar

Joined: Wed Oct 08, 2003 1:33 am
Posts: 12040
Location: Munich
This thread is to discuss how upgrades should work and interact with shipdesigns and individual ships.

The creation of the thread was motived by this post in the Cost and Build Time design thread:
Tortanick wrote:
How flexible will "upgrade existing parts" be, will you for example be aloud to replace an engine with any other engine, including speed optimised engines to stealth engines. Or will every component have direct successors you can upgrade to, or something in-between?

P.S. if you have direct successors, can we give them more imaginative names than mark1, mark2 :)

When trying to answer this, I realized we haven't had an official discussion about it and there's nothing in the design pad.

I've had two ideas for how upgrades should work:

1) A ship design lists the general "class" of a part, and each specific ship using that design could have a different version of that part. A design might specify that a ship has a "laser" in slot 1. The design doesn't specify which refinement of laser to use, but newly built ships generally have the most up-to-date / best version available put into them. (We could allow the player to override this and put in older parts if they want, perhaps for export purposes or to hide their tech level?)

When upgrading a ship, the versions of all its parts are generally updated to the newest versions available (though again, we could allow upgrades to newer but not newest versions, or upgrades of only some of the parts). The newest versions always cost the same or less than earier versions and are equally good or better at whatever they do, so there's no reason to want to build a ship with an earlier version of a part. (Upgrading still costs and takes time, though)

Individual part versions could be called "Mark I", "Mark II", etc. or "basic laser", "repeating laser", "super laser", etc.

Parts wouldn't be upgradable forever; there would be a limited number of versions of each part, after which you couldn't upgrade any more and would have to make a new ship design to use newer parts.

The advantage to this method is that to upgrade a ship, you don't need to create a new ship design. You can have ten ships, all with the same design, but with different versions of the parts in the common design. The disadvantage is that you have many ships of apparently the same "design" with different capabilities due to different versions of their parts with the same class.

2) A ship design lists specific parts. Players can specify an earlier part that is obsolete if they so choose.

There may be part classes like "beam", "shield", etc. but these are mostly for sorting / ease of use. Parts have an "upgrades to" field, which specifies another part that can replace the part when upgrading (vaguely similar to how techs have prerequisites, but in reverse). Upgrades could skip parts and upgrade directly to the upgrade of the current part's listed upgraded, etc.

Newer parts may cost less or cost more, but should always be better than the earlier parts.

More than one part can upgrade to the same other part, so there could be a "gatling laser" and a "heavy laser" that both upgrade to "gatling heavy laser" that combines the advantages of both. Each part has only one upgrade, however.

There would be no built-in restriction about the "class" of part that a part can upgrade to, although in practice we'd only include upgrades to better versions of parts that do roughly the same thing.

Like system 1, parts won't upgrade forever, as eventually you'll reach a part that has no upgrade, requiring building a new ship to use newer parts.

Part names would be less likely to be "Mark I", "Mark II", etc. than system 1, and names could be anything more easily without the need to make the straight-line upgrades clear as with system 1.

This system has the advantage that there's more flexibility in upgrades. The main disadvantage is that every time you upgrade a ship to a combination of parts you don't already have, you'll probably have to make a new shipdesign. You'll also have to update your old shipdesigns to use newer parts, making a new ship design every time you research an upgrade to a part for which you've made a design and/or built a ship. This could lead to many similar ship designs to keep track of, which could be a big pain and get very tedious and confusing.

Thoughts? Is there another option I'm not thinking of?


Top
 Profile  
 
PostPosted: Tue Jan 22, 2008 1:14 am 
Offline
Space Krill
User avatar

Joined: Mon Jan 21, 2008 6:17 am
Posts: 5
Would the actual upgrading of ships be handled something like Civilization, or would you actually have to spend time/PP on upgrading the ship?

Obviously a Civilization-type system is simpler to use, you just send the ship to the nearest capable shipyard, click upgrade and (if you can afford it) you get the newest version of the ship the next turn.

However, if upgrading a ship takes a certain number of turns in the shipyard (perhaps based on the parts used) it adds a bit more strategy to upgrading. Do you upgrade your older ships and slow down production of new ones? When an enemy fleet attacks is it worth taking ships off your front lines for several turns to gain a technological advantage, or do you depend on weight of numbers?

Other than that the only thing I have to say is that whatever method of upgrading gets used, keeping the number of ship designs low is probably a good idea. One of my biggest problems with MOO3 was trying to keep track of what ship designs were up to date.

_________________
Half a pot of coffee, two beers, a Mars bar and lasagna are not a good combination


Top
 Profile  
 
PostPosted: Tue Jan 22, 2008 9:28 am 
Offline
Creative Contributor
User avatar

Joined: Sat May 26, 2007 8:05 pm
Posts: 576
Well I can think of a third method:
Refits, basically this is instead of upgrading, with refits you can turn any ship into any other ship with the same hull type, preserving the experience of the crew and spending less than building a new ship. The formula might be something like:
Moving a part: 1/10th of the parts cost
removing a part: reduce the total cost by 1/3 of the parts cost
adding a part: the entire cost of a part.
Obviously you don't pay for the hull a second time

There was a previously mentioned exploit of building a huge empty hull then refitting it, having to pay for all the components during a refit should be the same as building them all the first time. It might by slightly more effective to build the hulls, mothball them, then refit them but unless its hugely unbalanced that's a strategy of prudent investing not an exploit.


If we are to use an idea where the final upgraded form is based on the current selection of components, like Geoff's then in addition to the points that Geoff made about what gets upgraded, we need to decide how the interface works. Obviously if there is some any flexibility at all you don't want players to have to make the same identical choice for a dozen ships when its time to upgrade you need some way of makeing the choice once and applying it many times.

The way I'd do this is by choosing an existing design in the design screen and click upgrade, you are taken to a specialised design screen where you start with the old ship and can manually add and remove parts within the constraints of whatever rules for upgrades are chosen. Alternatively you can click an auto button to move to the latest and greatest then fiddle from there. Afterwords all ships from the old design have an upgrade button that turns them into the new design. As a slight alternative you could design multiple upgrade choices for a single design.


One thing I'd like to see the upgrade system do is keep ships of a certain name identical, for the entire game*. A player should be able to see 6 "Zeus" ships and not have to check all six to see witch have mark 1 Tesla coils and witch have mark 2, she should be able to know they're called Zeus so they have mark 1. Similarly if a player is in combat and sees those same 6 Zeus ships approaching on his flank he should be able to remember their abilities from the last battle, and know all 6 are like that, no need to manually check all 6.
If a player wants to upgrade his Zeus ships, choose a new name

* IMHO Once a ship name has been used in a game, it should be permanently locked.


Top
 Profile  
 
PostPosted: Tue Jan 22, 2008 5:25 pm 
Online
Creative Contributor
User avatar

Joined: Sat Sep 13, 2003 6:52 pm
Posts: 1396
Well there are two issues here

1) upgrading
and
2) refitting


For upgrading, strategy 1 and strategy 2 are basically the same... the KEY term is "only one upgrade for each part"

If a Laser can get upgraded to EITHER a Gatling laser or a Heavy laser, then you have a semi-refitting

If a Laser can get replaced with a Missile, then you have true refitting.


The Idea of Upgrading, with no refitting, makes sense for simplicity's sake... when you take a ship to be upgraded you can do 'max upgrade' or 'limited upgrade' [if your budget is low or you have a time crunch], and essentially a ship is consumed to allow you to build a 'new' ship.
in that case I would say the Cost is
PP= PP differential between new parts and old parts + X% of old part cost
Time = Time to construct this hull size * Y
where X is between 10-100 (depending on game balance)
and Y is between 0.5 and 0.1 (depending on game balance AND possibly how many parts are being upgraded)

While I like the Idea of Refitting as a way of changing your Ship Strategy, it really depends on the balance between maintenance costs and build costs for ships.
If the Maintenance / Build ratio is high, then it is more reasonable to just scrap the old design and build a new ship
If the Maintenance/ Build ratio is low, then Building a new ship is a major expense and I might really like to refit one.

The other issue is crew experience, and that should NOT be maintained in the case of refitting (although it probably should in the case of upgrading)... in the case of refitting or scrapping a ship, the crew experience should be dumped into the Imperial Experience Pool (which is then used to build up experience in new ships).

As for naming designs, well there should be a default name so the first time an upgrade for the existing 'Zeus' design becomes available it gets a default name of 'Zeus I' or 'Zeus 1' and that name can be changed until the first 'Zeus I' ship is produced (either through upgrade or basic production)


Last edited by Krikkitone on Wed Jan 23, 2008 5:20 pm, edited 1 time in total.

Top
 Profile  
 
PostPosted: Tue Jan 22, 2008 10:14 pm 
Offline
Design & Graphics Lead Emeritus
User avatar

Joined: Sat Sep 23, 2006 7:09 pm
Posts: 3858
Location: USA — midwest
I think the convenience of an "upgrade" is that your ship can be improved without being redesigned.

It should be made clear that our tech-tree has a concept that's not in MoO (or at least not in1 &2): refinements. Most components will have a few slightly better versions which can be researched. But that research is generally a dead-end. For example: if you discover the laser mark1 you have the choice of the relatively quick reward of the laser mark2, or you can research a fundamentally new tech which will take longer but be significantly more powerful that any laser refinement.

So refinements would make pretty sensible candidates for an automatic upgrade.


I think if that's all that was done it would be sufficient, and reasonable.

_________________
—• Read this First before posting Game Design Ideas!
—• Design Philosophy

—•— My Ideas, Organized —•— Get an Avatar —•— Acronyms —•—


Top
 Profile  
 
PostPosted: Wed Jan 23, 2008 6:48 am 
Offline
Cosmic Dragon
User avatar

Joined: Fri Jun 27, 2003 12:37 am
Posts: 2175
Location: Auckland, New Zealand
I think it should work like this:

[Ship Design Screen - Upgrade Section]:
When you upgrade a ship or ships, what you are doing is changing the design. The old design is being replaced with the new design, even if it is just a Mark2 laser over a mark 1. So the player would change the desired design to the upgraded version. If needed, the player can change the design competely. The difference between the old design and new design (hull, parts changes) gives the cost and time to build (upgrade). With this new design, the player can then initiate the upgrade process.

The player would select the desired ships to be upgraded from a list of ships that are of the old design (filters based on ships location, etc would help). The desired ships would travel to the nearest starbase/shipyard (alternativley the nearest selected starbase/shipyard waypoints) if they are not already there and start upgrading as soon as they arive. As ships complete upgrading, they sit idle at their current location, unless some kind of rally point is set. A rally point would be optionally set when the order is given to upgrade the chosen ships.

The player can keep track of the progress of the upgrades by going back to the ship design: upgrade screen and looking at the % complete eg "20 of 50 ships have completed upgrade". Perhaps even mass cancelling/scrapping options would also be available. In such a case where the player had upgraded 50% ships, but the remaining upgrades were taking too long and new ships have become available, the player could choose to scap all remaining ships in the process of being upgraded.


Top
 Profile  
 
PostPosted: Fri Jan 25, 2008 1:21 pm 
Offline
Programming, Design, Admin
User avatar

Joined: Wed Oct 08, 2003 1:33 am
Posts: 12040
Location: Munich
Refitting, or completely changing the design of a ship by replacing parts with unrelated parts, I don't think should be allowed. It would essentially let players build entirely new ships, and so would need to be balanced similarly to how ships are built, with a multi-turn time in which the ship is being refitted, which is somewhat awkward for ships already on the map... as they'd either have to be made immobile, or removed from the map. Also, being able to do so likely means that it would be necessary to do so to win, and essentially it will be adding micromangement to fleet maintenance.

A workable alterantive seems to be not having single-ship upgrades at all. Instead, refinements would cause all ships in a fleet with the refined part to instantly perform better, without requiring the ship to return to a shipyard for an explicit upgrade. There wouldn't be major balance issued with this, as you'd still only be able to upgrade parts a few times, and then would need to build new ships to get newer parts. The ability to "upgrade" a ship, by researching a refinement, without the ship returning to a shipyard can be expliained by the refinement providing new knowledge about how to use the existing part better, rather than an entirely new part being built.

Are there any obvious / major problems with doing things this way?


Top
 Profile  
 
PostPosted: Fri Jan 25, 2008 7:12 pm 
Offline
Creative Contributor
User avatar

Joined: Sat May 26, 2007 8:05 pm
Posts: 576
Geoff the Medio wrote:
Are there any obvious / major problems with doing things this way?


Well if the refinements are balanced against their instant and free nature then that's a fine system. The only "problem" is that it may lead to us not using a better system. Since thus far no one has a better system it seems fine unless an outstanding idea comes along.

I quite like the implication of refinements being a very weak free upgrade, gives a constant choice between researching a new technology then building new ships, or researching a refinement that's less value for science but backwards compatible.


Top
 Profile  
 
PostPosted: Sat Jan 26, 2008 6:55 am 
Online
Creative Contributor
User avatar

Joined: Sat Sep 13, 2003 6:52 pm
Posts: 1396
Well the problem is the nature of the refinement... If they are simple slight bonuses, then it is OK

I still think an update at the shipyards (similar to a repair might be good.... just automatically upgrading everything to the max refinement.


Top
 Profile  
 
PostPosted: Sat Jan 26, 2008 7:23 am 
Offline
Programming, Design, Admin
User avatar

Joined: Wed Oct 08, 2003 1:33 am
Posts: 12040
Location: Munich
Krikkitone wrote:
Well the problem is the nature of the refinement... If they are simple slight bonuses, then it is OK

They'll be balanced... which presumably means not huge.

Quote:
I still think an update at the shipyards (similar to a repair might be good.... just automatically upgrading everything to the max refinement.

Making refinement work automatically for v0.4 doesn't mean we can't later add upgrades to completely new components that requires being at a shipyard, should this be necessary.


Top
 Profile  
 
PostPosted: Sat Jan 26, 2008 11:53 am 
Offline
Space Kraken
User avatar

Joined: Sun Dec 16, 2007 7:00 pm
Posts: 161
Goodmorning all, Colour = edit

I was thinking about the upgrade problem, and i generally agree with Geoff, any 'requirement' to return all your ships to stardock is begging to push the acceptable micromanagement limit for FO.


I think the instantaneous solution is good, but i might suggest a -small- adaptation.

I think upgrades should still take time, not long, but some time, and not necessarily the same amount of time for each ship in your fleet.

proposed solution;

Each ship has a set probability of undergoing spontaneous upgrade each turn (30%chance, for example would give 99% replacement in 11 turns , 40% would give 99% coverage in 9 turns). (for the math inclined this would give an exponential decay on number of ships left un-upgraded) {the reasoning being that not all ships are visited by the maintenance team each turn, and for the most part the maintenance teams would be doing the 'upgrading'}

Each tech would have an incorporation probability associated with it (larger strength differences, smaller probabilities of being incorporated each turn.).

Now, bonuses to this model

One could have race picks, which make them technically handy, and give a bonus to upgrade probability. so all their ships upgrade faster.
One could have ships with a special object on them which gives +n% bonus to probability for upgrading/turn to all 'nearby' ships.
One could have probabilities modified by experience levels. Green crews being much less willing to gut their ship in deep space and self upgrade, preferring to wait for the maintenance crew, vet crews being up for the challenge of surprising the maintenance's crew by having it done before they arrive.
One could set shipyards to 'upgrade mode' which costs a small amount, and gives +n% to the nearest m un-upgraded ships (n and m depending on the cost, and size of the shipyard) [this would auto turn off, when there are no more ships to upgrade]
One could force-refitor upgrade a ship at a shipyard, (one turn-perhaps two for REALLY big, or simultaneous tech advances upgrade).
Physically damaged ships could have a reduced probability while damaged, then an increased one post damaged for several turns. [this is the only one i don't see as being user friendly but it has to do with the fact that once damaged first priority is to get the ship back in functional condition, and being damaged puts a ship higher on maintenance crews ships to visit to replace consumed repair supplies, and while there, upgrade because clearly they need it]
One could have refit probability semi-locationally dependent;
Ships outside your 100% supply lines would suffer in upgrade probability(probability linearly with supply, so 50% supply, on a ship with 40% probability/turn -> 20% probability until it is back in range). I can imagine sending a largish fleet, across a boarder to attack, and before they arrive completing two different upgrades in two different fields. . . . should the attacking fleet already be auto upgraded even though they are engaged in military operations at the time? This would be the last modifier to be applied, so if you have a bonus to probability due to a space doc, or nearby upgrade ship these bonuses would also be diminished.
Ships actively traveling threw star lanes would have 0 or near 0 % chance.

Technicalities:

Ships would only be able to work on one upgrade goal at a time, so as to only need one probability value/ship.
If more then one upgrade is possible, the tech discovered on the earlier turn would be completed first
of more then one tech is completed on the same turn the one with the higher base probability (the weaker tech) would triumph.

User issues, and transparency:

As a user i only really care about two things;
how long will the upgrade take?
how do i make it faster?
although the background cases may be a bit complex, the two questions remain easy to answer.

Question one, 1/probability = likely number of turns, (either the probability and/or the probable turns) [this would not change unless the probability changes].
Question two, to increase rate; one can 'Spend money at doc, visiting or not', build an upgrade ship to move with your biggest armies, have better race picks, or get back into your own space. All of which IMHO make sense. only two of them might need some form of marker, ie, if a ship is being targeted by a spacedoc or not, and if it is suffering from lack of supplies or not. A ship might also be able to get into a fight to increase it's chances . . . if that method is permitted[as i said not the most user friendly of rules and i would understand if it didn't make it into final version . . . well i would understand if none of this made it into the final version . . . it is only an idea].

Wishful thinking:

I just can't help but mention that with this system the 'Geordi La Forge''s of FO are permitted to exist. For non fans, Geordi La Forge, had implemented several changes to the Enterprise's systems, which hadn't been developed or approved yet in the research circles. Star trek: NG 'Galaxy's Child' amongst others.

This would be modeled as a probability for a given upgrade to occur, Before researching it.
This would ONLY depend on racial picks, hero's/fleet commanders(if there are any), experience, and recent battles[which is another reason why I offer the recent battles option necessity being the mother of invention]. [all these probabilities would be very small 1-2 %, max]

In effect, the probability for this to occur may or may not be hidden, and the fact that it happened may or may not be hidden (ship just preforms better). [I am not convinced either way which is easier. if Not hidden, then some reasoning may need to be given why one ship has the new tech, and hasn't shared it with society yet... if hidden it's easier, the only people who know are the ship's engineers, and they're probably so busy they don't even notice's their equipment is more advanced.]

Sorry once again for the long post.

I'm just wordy . . . I'll try to cut down in the future.

Any thoughts are always welcome.

best wishes all
Robbie


Last edited by Robbie.Price on Wed Jan 30, 2008 6:20 pm, edited 1 time in total.

Top
 Profile  
 
PostPosted: Mon Jan 28, 2008 9:57 pm 
Online
Creative Contributor
User avatar

Joined: Sat Sep 13, 2003 6:52 pm
Posts: 1396
Well I think the 'Upgradability' should function very similar to 'Repairability' of a ship.

Essentially, to upgrade a component of a ship should take the exact same steps that would be involved in repairing Major damage to that component (so PP costs, Time, extra supply required, need to visit a Space Station)


I think the 'Supply' concept might work the best, a ship that is 'Operating/Refueling/Reloading' takes one level of supply, a ship that is 'Repairing/Upgrading' takes another level of supply.

(Now 'operating' and 'reloading' are probably available at a distance... I could see 'Repair' and 'Upgrade' require a much shorter range.)

If 'Supply' is some quantifiable resource that gets passed out to ships... then it is simple, each component that is to be Repaired/Upgraded requires X units of Supply. If there is some limit as to how much Supply you can make each turn (or a limit to how fast it can be passed out), then that limits how fast ships upgrade (and repair as well which is a definite bonus) [you should definitely be able to whittle down a force because repairs can't get to them fast enough.]

Race picks (and Technologies and Governments, and cultures) naturally could influence this.


Top
 Profile  
 
PostPosted: Tue Jan 29, 2008 12:56 am 
Offline
Space Floater
User avatar

Joined: Sat Feb 14, 2004 9:15 am
Posts: 22
Geoff the Medio wrote:
Refitting, or completely changing the design of a ship by replacing parts with unrelated parts, I don't think should be allowed.

I don't see any problem with completely changing the design of a ship, except for the size of the ship, of course. I think a good way to handle your concerns is with the cost of upgrading the parts on the ship. The basic idea would be that if ALL the parts on the ship are upgraded at the same time, this would be more expensive than building a new ship. People would generally not want to do it, but could if they really wanted to, and the benefit would be that upgrading would be faster even though it would be significantly more expensive. On the other hand, if only a few parts are upgraded, this would, of course be less expensive than building a new ship.

You could implement this basic idea by first charging a base cost for the upgrade of the ship depending upon the size of the ship, then charge an additional cost for each part that is upgraded. The cost of each part upgraded would be slightly more expensive than a new part of the same type in a new ship, but if only a few parts are upgraded, upgrading the ship would still be cost effective. Only when you begin to upgrade most of the parts all at the same time does it become impractical.
Geoff the Medio wrote:
A workable alterantive seems to be not having single-ship upgrades at all. Instead, refinements would cause all ships in a fleet with the refined part to instantly perform better, without requiring the ship to return to a shipyard for an explicit upgrade.

I don't like the idea of having to upgrade each ship individually as it was done in MOO2, myself. Instead, I like the idea of having a set number of ship designs, and you can then upgrade ships of one design to another design provided that certain conditions are met, such as both ships designs have the same size hull. To upgrade a ship from one design to another, you could give the ship a command to go to the nearest shipyard where it will be added to an upgrade queue where it will become unavailable for use until upgrade is complete. Upgrading would take less time than building a new ship, but could possibly be more expensive if large numbers of parts are upgraded all at once.


Top
 Profile  
 
PostPosted: Wed Jan 30, 2008 6:12 pm 
Offline
Space Kraken
User avatar

Joined: Sun Dec 16, 2007 7:00 pm
Posts: 161
Goodmorning all,

I have been re-reading my post, i should have specified more clearly that the system only works for 'Upgrades' of existing equipment, i used the term 'refit' on several occasions, which is inappropriate. (In my mind) Refits replace parts with parts of different functionalities, or different base technologies, Upgrades simply go from tech level 1 of lasers to tech level 2 of lasers. That process would be automated by my proposed solution, going from laser X to plasma cannon Y, would still require explicate user input at some level. (This could still be included . . . but would require serious thinking for balancing issues and user Interface issues . . .)

I'm going to go re-word the above to correct my error.

Best wishes all
Robbie Price


Top
 Profile  
 
PostPosted: Wed Feb 06, 2008 12:01 am 
Offline
Large Juggernaut
User avatar

Joined: Fri Jan 20, 2006 9:34 pm
Posts: 938
Location: GA
Geoff the Medio wrote:
Refitting, or completely changing the design of a ship by replacing parts with unrelated parts, I don't think should be allowed. It would essentially let players build entirely new ships, and so would need to be balanced similarly to how ships are built, with a multi-turn time in which the ship is being refitted, which is somewhat awkward for ships already on the map... as they'd either have to be made immobile, or removed from the map.
SE5 handles it by giving the ship a crapload of damage markers. Obviously systems that weren't modified don't get any markers. Of course SE5 has a repair system that won't necessarily completely remove the damage markers in one turn.

_________________
Computer programming is fun.


Top
 Profile  
 
Display posts from previous:  Sort by  
Forum locked This topic is locked, you cannot edit posts or make further replies.  [ 18 posts ]  Go to page 1, 2  Next

All times are UTC


Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Group