Request for Comments: Ship Design

[Geoff the Medio]

...in SE you can do research to improve the capacity of your hulls, thus allowing you to stick more components into the hull instead of having to use better components.

This is poorly defined, but if you mean that improving hull capacity should be a many-stage, ongoing incremental process, this is not a good idea. Suppose each time you researched an improved hull capacity, you got something like 10% more space on that hull type. This would have many of the same problems as miniaturization as a refinement for ship parts: in particular, it would mean that every time you refine a hull type, you would have to redesign all your ships to make use of it. This would not be not a good system, as it would require lots of busy work redesignings, produce lots of similar but slightly different ship designs to keep track of, and make it more difficult to define equilvalence for ship upgrade purposes

It's unlikely that we'll have a great variety of hull sizes, or a continuum of hull sizes. More likely, there will be a small number of discrete, fixed, fizes... perhaps 4 or 6, with names like Tiny, Small, Medium, Large, Huge, Colossal. This has the advantage of making the choice of size significant when designing a ship. It becomes necessary to pick between a small number of choices, each with well-defined and easy-to-understand properties, benefits and penalties. There is no way to incrementally enlarge hulls, so there's no need to change the layout of parts to do upgrades; instead, each part can be replaced with the newer version, when avaialble.

As to whether there are hull layouts separate (but within) the discrete sizes, which determine what type or layout of slots are avaiable, this is undertermined and up for debate. Having lots of hull layouts would likely mean having fewer sizes (likely just 4) to compensate. Each layout would be a distinct upgrade path for ship designs; you could upgrade the parts within a particular layout, but not change the layout of the hull. Each layout would have a set number of slots of particular types, with would never change for a particular layout.

Another issue is, assuming we base ship design around filling slots, where the slots have restrictions on the types of parts (engine, gun, armour) that can go in them, and or part size (small, medium, large), what's the rough number of slots in a design? There are two extremes:

1) SMAC-like designs with 1 weapon/functional, 1 defense, 1 engine, and 2 "special" slots. Each slot would have a size, and each part would have a size. The number of the chosen part on a ship would be qual to the size of the slot divided by the size of the part.

2) Effectively unrestricted, with (technically), limited slots, but in practice there are 30 or 50 slots on a ship, with these being all generic, with no per-slot restrictions. In this case, slots would essentially be stand-ins for mass or volume limits in the ship. Each part would require some number of slots to be added (ranging 1 to 5 to maybe 10 for really big parts). Ships would be required to have an engine, and perhaps one or two other "essential" parts, but there would be no restrictions other than this. It might not be necessary to even keep track of individual slots' contents in this case, with just a total number and number of occupied slots.

Somehting in between these extremes is also possible.

There could also be geometric restrictions, where each part consume some number of slots in some geometric configuration, with the available slots laid out in some hull-layout-specific pattern.

Also, note that as in the first post, long lists of how you'd like to put this or that type of turret on or descriptions of how turret size relates to chance to hit and ship size, etc. are not very useful. Please provide reasons for your proposed designs, or at least a few alternatives to consider.

[marhawkman]

...in SE you can do research to improve the capacity of your hulls, thus allowing you to stick more components into the hull instead of having to use better components.

This is poorly defined, but if you mean that improving hull capacity should be a many-stage, ongoing incremental process, this is not a good idea. Suppose each time you researched an improved hull capacity, you got something like 10% more space on that hull type. This would have many of the same problems as miniaturization as a refinement for ship parts: in particular, it would mean that every time you refine a hull type, you would have to redesign all your ships to make use of it. This would not be not a good system, as it would require lots of busy work redesignings, produce lots of similar but slightly different ship designs to keep track of, and make it more difficult to define equilvalence for ship upgrade purposes

I never had that problem. Why? there is no requirement to actually fill all the slots in a hull. I design and build a carrier, and then research a carrier hull improvement to allow me to design bigger cariers in the future. this doesn't have any negative consequences either in the short term or long term. If I decide to upgrade my existing carrier I can make it bigger, but I don't need to.

It's unlikely that we'll have a great variety of hull sizes, or a continuum of hull sizes. More likely, there will be a small number of discrete, fixed, fizes... perhaps 4 or 6, with names like Tiny, Small, Medium, Large, Huge, Colossal. This has the advantage of making the choice of size significant when designing a ship. It becomes necessary to pick between a small number of choices, each with well-defined and easy-to-understand properties, benefits and penalties. There is no way to incrementally enlarge hulls, so there's no need to change the layout of parts to do upgrades; instead, each part can be replaced with the newer version, when avaialble.

that's just it, there is no continuum of hull sizes in SE. SE has hull types and you research them to make them bigger.

As to whether there are hull layouts separate (but within) the discrete sizes, which determine what type or layout of slots are avaiable, this is undertermined and up for debate. Having lots of hull layouts would likely mean having fewer sizes (likely just 4) to compensate. Each layout would be a distinct upgrade path for ship designs; you could upgrade the parts within a particular layout, but not change the layout of the hull. Each layout would have a set number of slots of particular types, with would never change for a particular layout.

Another issue is, assuming we base ship design around filling slots, where the slots have restrictions on the types of parts (engine, gun, armour) that can go in them, and or part size (small, medium, large), what's the rough number of slots in a design? There are two extremes:

1) SMAC-like designs with 1 weapon/functional, 1 defense, 1 engine, and 2 "special" slots. Each slot would have a size, and each part would have a size. The number of the chosen part on a ship would be qual to the size of the slot divided by the size of the part.

2) Effectively unrestricted, with (technically), limited slots, but in practice there are 30 or 50 slots on a ship, with these being all generic, with no per-slot restrictions. In this case, slots would essentially be stand-ins for mass or volume limits in the ship. Each part would require some number of slots to be added (ranging 1 to 5 to maybe 10 for really big parts). Ships would be required to have an engine, and perhaps one or two other "essential" parts, but there would be no restrictions other than this. It might not be necessary to even keep track of individual slots' contents in this case, with just a total number and number of occupied slots.

2 is similar to what the SE games had. It seemed to work quite nicely.

Somehting in between these extremes is also possible.

There could also be geometric restrictions, where each part consume some number of slots in some geometric configuration, with the available slots laid out in some hull-layout-specific pattern.

That sounds cool. mixing that with 2 would provide for some interesting designs....

Also, note that as in the first post, long lists of how you'd like to put this or that type of turret on or descriptions of how turret size relates to chance to hit and ship size, etc. are not very useful. Please provide reasons for your proposed designs, or at least a few alternatives to consider.

You mean my illustration of how you'd unlock the ability to research certain hulls by researching certain other hulls?

[eleazar]

Another issue is, assuming we base ship design around filling slots, where the slots have restrictions on the types of parts (engine, gun, armour) that can go in them, and or part size (small, medium, large), what's the rough number of slots in a design? There are two extremes:

1) SMAC-like designs with 1 weapon/functional, 1 defense, 1 engine, and 2 "special" slots. Each slot would have a size, and each part would have a size. The number of the chosen part on a ship would be qual to the size of the slot divided by the size of the part.

2) Effectively unrestricted, with (technically), limited slots, but in practice there are 30 or 50 slots on a ship, with these being all generic, with no per-slot restrictions. In this case, slots would essentially be stand-ins for mass or volume limits in the ship. Each part would require some number of slots to be added (ranging 1 to 5 to maybe 10 for really big parts). Ships would be required to have an engine, and perhaps one or two other "essential" parts, but there would be no restrictions other than this. It might not be necessary to even keep track of individual slots' contents in this case, with just a total number and number of occupied slots.

Something in between these extremes is also possible.

I'd go with something in-between. More specifically, the smallest hulls would have a minimum number of slots, and increasingly large hulls would have more (and possibly bigger) slots. It's hard to pull numbers out of the air, but i believe that providing too many slots negates much of the usefulness of slots, as a means of simplifying the design process.

For the purposes of the models, external slots will need to be "kept track of". I think it's important that you can mount a "laser cannon" on the front or on the side, or possibly on the rear, etc. External slots with fixed locations seem the obvious way to do that. Those choices in construction have a natural and obvious result in the appearance of your ship.

I like the idea of large components which take up multiple slots, but in practice that may be difficult to implement graphically. (need to consider it more)
It may be that "slot clusters" (possibly 2 and 4 paired slots) could be used for larger component mounts, or for multiple smaller components.
Suppose that Hull X had a 4-slot cluster on the nose. The player could fill it with a single massive weapon, 2 two-slot weapons, or 4 single-slot devices. Obviously my justification is again graphical.


Internal slots could arguable be "indefinitely" located. I'm leaning towards that. There would be no need to divide the available slots into clusters for large components. Nor would you have to rearrange the innards to make everything fit. You can't see it, so it's location doesn't matter. ....The catch is a split UI dealing with internal and external parts differently. It's probably too soon to tell for sure if that complication is worth the trouble saved in designing a ship.

[marhawkman]

Erm.... In SE you have white squares that you put components in. the idea Geoff suggested was that you might have some components that don't only take up a single square.

[Yeeha]

Thats great idea, it would help doing paper, stone scissors effect all weapons could be divided into 3 groups - light weapons with 1slot, standard 2 slots and heavy with 4 slots. That way balancing is easyer & designing would be easyer too for player. Otherwise slots have to be marked where u can place heavy weapon and where u cant.

Edit: although such a system makes question - how missile weapons would work. Are they allways 4 slot missile batteries? or are they 1 slot weapons which require equipment slot for missile storage? or are there different kinds of missiles as well?

I would like such system - there are 3 types of missiles aswell like beam weapons but all missile platvorms require equipment slot for missile storage. Ofcourse this would mean that missile weapons would have to be littlebit superior to beam weapons to balance out. And that superiority would be missile weapons range.
And if u want more missiles just add more missile storages to equipment slots.

[marhawkman]

Thats great idea, it would help doing paper, stone scissors effect all weapons could be divided into 3 groups - light weapons with 1slot, standard 2 slots and heavy with 4 slots. That way balancing is easyer & designing would be easyer too for player. Otherwise slots have to be marked where u can place heavy weapon and where u cant.

Edit: although such a system makes question - how missile weapons would work. Are they allways 4 slot missile batteries? or are they 1 slot weapons which require equipment slot for missile storage? or are there different kinds of missiles as well?

I would like such system - there are 3 types of missiles aswell like beam weapons but all missile platvorms require equipment slot for missile storage. Ofcourse this would mean that missile weapons would have to be littlebit superior to beam weapons to balance out. And that superiority would be missile weapons range.
And if u want more missiles just add more missile storages to equipment slots.

For missiles I'd make it a function of Rack size.

[eleazar]

Thats great idea, it would help doing paper, stone scissors effect all weapons could be divided into 3 groups - light weapons with 1slot, standard 2 slots and heavy with 4 slots.

Perhaps, but i imagine that a single slot weapon near the end of the tech tree would be much more powerful than the earliest four slot weapon.

[Geoff the Medio]

A smaller-scale example of multi-slot geometrically limited ship design in in the game StarScape, as seen here:

http://www.moonpod.com/graphic/StarscapeShots/6.jpg

It can be somewhat annoying in practice though, at least with as few slots with as strict (and complicated) of geometric limitations as are in that game, due to issues where you have enough slots, but just not quite in the right configuration for a particular part shape, so you need to fliddle around and try to make room, etc. It's not easy to see from the screenshot, as the shown ship has no parts in it, but each part has a fixed size and shape... So, if you put a few 1x1 parts in the wrong space, you could still have 4 or 2 empty slots, but not be able to fit a 2x2 or 1x2 due to the way they're laid out between the filled slots.

Tweaking details like this might not be where we want to focus the effort of ship design in FO. It's not *so* bad for StarScape, as you only have four or so ships, of which you use only one at a time and thus care quite a bit about the details of its layout and don't mind tweaking. With FO though, we probably want to keep the part arrangement fiddling to a minimum for individual ship designs amongst dozens per game. Is there a way to do geometrical layout, with variable sized parts, that doesn't have such issues? Or do people really want to do such fiddling? Or if they do, why don't they play StarScape instead of FO, given FO's different focus, and thus why don't we design FO differently?

[eleazar]

It can be somewhat annoying in practice though, at least with as few slots with as strict (and complicated) of geometric limitations as are in that game, due to issues where you have enough slots, but just not quite in the right configuration for a particular part shape, so you need to fliddle around and try to make room, etc. It's not easy to see from the screenshot, as the shown ship has no parts in it, but each part has a fixed size and shape... So, if you put a few 1x1 parts in the wrong space, you could still have 4 or 2 empty slots, but not be able to fit a 2x2 or 1x2 due to the way they're laid out between the filled slots.

Tweaking details like this might not be where we want to focus the effort of ship design in FO. ... With FO though, we probably want to keep the part arrangement fiddling to a minimum for individual ship designs amongst dozens per game. Is there a way to do geometrical layout, with variable sized parts, that doesn't have such issues? Or do people really want to do such fiddling?

I think it could be done in such a way that fiddling is minimized. If not, i would probably turn against the idea. I don't want ship-building to become a tetris mini-game. The "challenge" of ship design for FO shouldn't be fitting arbitrary puzzle pieces together. There should be plenty of interest in balancing power, cost, speed, and defense etc.

That's part of why i suggested the internal pieces should be done differently (items in a list) to minimize the fiddling. If 4 slots were available internally, the player could add a 4-slot component, without any trouble, or shuffling.

External slots probably won't be arranged continuously around the ship. So if 2x2 was the biggest size of external component, 2x2 would be the biggest group of slots available. Little fiddling should be required.

[marhawkman]

One very simple way is to make all parts have a rectangular outline. And the areas you stick them in rectangular as well.

[utilae]

I think this idea would be best if it was like in Diablo / Titan Quest / Buldars Gate, etc

Where the ship internal layout is like a characters inventory in those games. Basically you would have a X by X grid of squares. A small ship might have 3 x 3, but a large ship might have 10 x 10. Components would take up X by X squares, eg a shield in titan quest typically takes up 2x2 squares, and a heavy axe takes up 3x2, but a spear takes up 1x6.

In those games they have auto arange systems, but generally the player would have to try best to fit desired components and number of components into the ship. There probably isn't to much strategy to this ship design method, other than trying to get the best efficiency of lay down components, and certainly miniturisation helps as well to be more efficient.

But I too would rather a list system where you could choose components based on the component. Though a grid system may be a good representation of how much space you are using and have left in any case.

[marhawkman]

that IS the basic idea here.

[Sandlapper]

I agree with the slot/allowable space setup as presented so far. Personally, I want total micromanagement of a ship's design; however, I wouldn't always do that with every ship.

For the sake of the masses, and minimalising micro, Utilae's hint at auto configuration would fit the bill. As implied, simply have an auto config button. The computer offers the "best" design based on current tech.

As an intermediate compromise toward micro-lovers, perhaps offer a checkbox option to auto config that allow "best" beam ship, "best" missle ship, "best" combo lean to beam, "best" combo lean to missle, etc... We also need to make allowances for point defense, scouts, etc...

[marhawkman]

BTW a simple way to do the AI for the auto design:

Just have it look at big components first! :p from years of experience I've learned that the easiest way to solve those mini-riddles is to simply arrange the big items first.

[utilae]

Yep thats how you do it. Try and fit the big ones in first, and try and fill in any holes, etc.