Ship Design Rules of Thumb

[QuantumPete (OP)]

Hi all,

There's lots of good advice on the forum, particularly in the Design Bureau on how to design good ships (especially warships).  I was wondering whether you'd all share your favourite rule of thumb when it comes to ship design? Things like: "warships should be 25% engine", "the ideal engine-to-fuel ratio is 3:1", "4 layers of armour for every 4k tons", "20b km range is enough for most warships", etc.  These have helped a great deal when I first started, to come up with fairly solid designs that I could then adjust to my needs.

Thanks!

[Jorgen_CAB]

In my opinion there are not that many optimal ways you can design a ship but a few loose rules like the ones you mentioned.

I think the most important thing is that you build and design your ships based on your needs not out of some odd template. If you need 30bkm range on a ship because of system terrain then that is an optimum design constrictions you should aim for.

How much maintenance life, deployment times, speed, armour, shields etc do you need. Remember that every ton you dedicate for anything but weapons usually make you weaker as it is weapons that will ultimately destroy defeat the enemy. But at the same time if you don't provide enough scouts or defences you can be ambushed and defeated before you manage to fire of a single shot.

I usually divide up war ships in terms of mission tonnage and none mission tonnage. Mission tonnage is weapons, magazines, hangars and sensors while the rest is armour, engines, fuel, crew quarters and engineering sections. If you can have about 40-50% mission tonnage you are  probably alright for a balanced main capital or escort warship.

The amount of armour you have on a ship have to be weighted on how much space it take versus what weapons you expect the opponent to use and how much shields you intend the ship to utilise. There are no perfect way to say you need this or that level of armour protection for a specific size of ships.

Fuel is also important, there is an optimal design with 3 part engines and 1 part fuel... but in many instances that will make you simply consume too much fuel, especially if your ranges only are between 10-20bkm range. You don't want your fleet to burn hundreds of millions of litres of fuel for each mission or simply moving from one base to another inside your empire.

The more engines you put in your ships the less armour and maintenance you will get or you will have to simply place allot less weapons which mean you need more ships and this quickly will make your fleet very expensive.

[Lord Solar]

I got:15% of mass weapons on a warship. Build your combat fleet at the same speed. Don't just add more layers of armor just because the ship gets bigger, another player showed recently there are diminishing returns to scale with that. I always do at least 40% engine but that's somewhat arbitrary. I think that smaller than 1/3 engine does not make good warships.

[nuclearslurpee]

General rules of thumb that tend to work well:

• Ships should usually have between 30% to 40% of their mass as engines, with missile-based fleets needing less and beam-based fleets needing more. Most NPR designs fall into this range, so you will ensure that your ships are competitive against similar-tech opponents while having adequate room for other components like weapons. For commercial ships you can have a smaller engine fraction than this but I tend to find that this rule of thumb still works pretty well.
• A 3:1 engine to fuel mass ratio is optimal for performance, but a larger engine mass ratio is usually also fine as it conserves fuel. If you use engines with the base 1.0x EP modifier, usually a ratio of 8:1 or 10:1 works fine. The key thing to avoid is to have an excess of fuel (2:1 ratio, for example) as this gives you no performance gains and increases fuel consumption.
• Armor should be set up based on what kind of weapons you expect to face and want to be proof against. For example, if you expect to face warhead-9 missiles, these penetrate to a depth of 3 so you need at least 3 armor layers to avoid internal damage from initial missile salvos. If you expect to face 15cm lasers, these can penetrate to a depth of 4 and the spinal versions can penetrate to a depth of 5, so you need at least 4 armor layers to be proof against 15cm lasers and 5 armor layers to be proof against the spinals. So on, so forth. Of course more armor is good to protect from second and following volleys, but this is a way to estimate your mandatory minimums.
• Shields should be mounted in multiples of the armor width. If your ship has, say, 4-50 armor and 20 shields, your shields are way too weak to be anything more than a speed bump to the enemy. Consider 50 or 100 shield strength instead. Shields also become drastically more effective with higher tech, I think once you get past Epsilon Shields or so armor becomes almost obsolete in a practical comparison.
• Range should be mission-specific. In a conventional or low-tech start, your initial Sol defense fleet only need enough range to cover the Sol system (12.3 billion km is the maximum you would need for this - JPs are at most 6 billion km from the Sun, and Earth orbit is 150m km, add these up for a worst-case and double for a round trip). Once you start exploring new systems then you need ranges that increase over time so you can defend your entire empire or attack anywhere. Fuel efficiency tech will make up most of the difference for you in practice
• You should usually plan on 25% to 33% of your military ship tonnage being dedicated to their primary mission payload - main weapon, frelloff-huge sensor, hangar bays, or whatever else. In addition to that 35% to 50% being dedicated to engines+fuel, the rest will be taken up by armor, shields, engineering, and crew quarters.
• Missile range should usually be about 75% of your fire control range, which allows you to use fire-and-retreat tactics while maintaining a target lock. This probably more than any other rule of thumb is open to argument, but I think it is a helpful way for someone new to missiles to conceptualize what is a suitable range when learning to design their first missiles. I usually use 1-HS MFCs for my anti-ship missiles and 2-HS MFCs for anti-missile missiles, at least on large ships which can afford to use that much space for fire controls. My sensors are then also suitably large to take advantage of this range.
• Anti-ship missiles should be size 4 to 6. Anything larger will be detected at greater than minimum ranges, making it easier to intercept, and anything smaller will not be able to do enough damage until you get to fairly high tech levels.
• It is usually a good practice to mount a PD turret or a couple of 10cm railguns on every ship, so that your fleet is not crippled by the loss of a couple PD escorts. It is better to have too much PD than not enough, and you always need more PD than you anticipate.
• Every sufficiently large warship should have an Auxiliary Bridge, as the crew training benefit and giving your junior officers something to do are both valuable contributions. "Sufficiently large" varies, and depends mostly on how many officers you have compared to ships, but generally any ship above 5,000 tons is a viable candidate if you have officers to spare, and any ship above 10,000 tons should definitely have an AUX.
• Deployment time should be thought of as "time on station", most military ships will not be flying at full-burn for their entire deployment so deployment time should be longer than fuel endurance by a good margin. I usually set 12 months as my standard deployment time for most navy warships. Maintenance life should exceed this, as it is better to have a restless crew than a critical breakdown, and maintenance failure RNG can be harsh if you cut too close of a margin. Typically for my 12-month deployment ships I aim for a maintenance life of 2 years, and on long-deployment ships like survey ships I will settle for ~2 years excess maint life as a margin of safety.
• Maintenance life should be reached by using engineering spaces only. Add MSP bays after adding engineering spaces to provide extra MSP to repair weapons which break while firing. I must emphasize this is a rule of thumb only, and extreme ships (very small or very large) may require a different approach, but for most ships this works well and using engineering spaces instead of MSP bays reduces your overall MSP consumption and saves resources in the long run.

Of course with all rules of thumb these are only generalizations, and with experience a player can and should develop their own doctrines and techniques, but I generally find each of these rules to be a good guideline when you don't have a good idea what to do otherwise. A lot of players will tell you things like "there's no rules, just do whatever you want", but for a newer player starting out or an old veteran jumping back in it is helpful to have a starting point even if it is a little arbitrary.

[xenoscepter]

 --- I always put at least 3 layers of armor on a warship except in extraordinary cases.

 --- I tend to put 1 full Engineering Space per 1,000 Tons. This varies on a ship to ship basis, but it's my yardstick... Anything above or below is measured relative to it.

 ---- My carrier's should have enough fuel, MSP, and ammunition for at least three sorties. The logic of this being, 1 sortie of attack, 1 sortie of pursuit, 1 sortie of opportunity.

 --- Non-carrier vessels with parasite craft carry enough fuel, ammo and MSP for at least one sortie, though there are usually exceptions for fuel-hungry fast scouts or short range crafts where 5,000~10,000 litres can be tantamount to 6 or 7 sorties.

 --- Two engines on anything military, sans stations of course. It sucks to be stranded.

 --- Enough MSP to battle damage repair the most expensive component at least once, preferably thrice. Varies a lot on a ship to ship basis, but this is my yardstick.

[misanthropope]

err on the side of firepower. 

[Polestar]

A. My "rules of thumb" are few. Here are some of them:

For commercial ships:
1. Engines should usually be ~40-50% of the size of the ship. Why so large? Engine cost is proportional to the square of its power rating, up to 100% power, and so - for most commercial ships - it pays to drop the rating and upsize. Other advantages include: faster construction of ships that are cheap compared to their size, more fuel-efficient engines, and reduced engine research costs.

2. Look for opportunities to off-load the engines to dedicated tugs. For starters, anything that stays in one place for years at a time really ought to be a station.


For warships:
1. Most warship designs benefit from a focus on both a single mission and a single combat range. Field ships that know what they're about.
1a. That is not to say that putting point-defence on a beam battleship is wrong - it's not. However, if you put most of your point-defence on dedicated ships, you can tailor fleets to either be better at stopping missile spam, or better at lighting up enemies, depending on local opposition. Indeed, you can send in the point-defence craft (or heavily shielded and/or armoured stuff, whichever) to beat off the missiles, with the beam boys kept safely out of trouble until things settle down.
1b. Same story with combat ranges. Move weapons having sufficiently different ranges to different ships. Move the ships to their optimal combat ranges - when conditions are favorable.
Warning: adopting this proposal, while it makes ship design easier, does mean that you need to be more mindful of both fleet organization and battlefield positioning.

2. Look for ways to off-load non-mission components and costs to other ships or stations. Your ship design might not need...
a. A full sensor suite. Sensors are among the most expensive items around for their size, and you only need to see the enemy once, so be smart about how you achieve redundancy.
b. Armament (especially if expensive) that doesn't carry out the primary mission.
c. Luxurious deployment time, multi-year maintenance life, abundant fuel, or quite that many missile reloads. Why burn fuel carrying more fuel?
d. Engines. If it doesn't have to move often, can a tug tow it around? Even if you do want it to move, can you park it behind a jump gate or Lagrange point and wait in ambush? Or just give it extra long-range missiles and hit first?
e. Heavy armour. At high tech levels, shields are king - as pointed out previously by nuclearslurpee.
f. Jump drives. Alternatives include dedicated jump squadron ships, unarmed jump ship support, or building a gateway.

3. For mobile warships, start with engines totaling 30% of the mass of the ship ... but then focus on the mission, not the rule. Reduce size (down to 20-25%) for slower ships, and also for lower-range ships with little need for fuel and thereby reduce costs. Increase size (up to ~40%) for faster ships, and also for ships needing more range.


B. Considering the OP's examples of advice others might proffer:

"warships should be 25% engine": My warships are either fast or slow - and they know which one they are. If fast, they have something like 30% to 45% of space devoted to engines. Superior speed means a better chance of imposing your preferred combat range on the enemy, and that might be half the victory right there. However, I also build a lot of "monitors" - warships with puny engines (often ~20% of total size) that tank and dish out a lot of damage for their cost.

"the ideal engine-to-fuel ratio is 3:1": I seldom put anything like that amount of fuel on my ships, because they either will (almost) always have tanker support (mobile ships), or hop from station to station (defensive ships). Exceptions might include survey ships, scouts, and long-range raiders - but even for these I tend to find ways to get a tanker involved.

"4 layers of armour for every 4k tons": For me, armor is all about the mission (and the combat range). Early on, I put a lot of armour on. Later, it's mostly about the shields.

"20b km range is enough for most warships": Range on my warships varies dramatically. 20b km would be at the upper end for my "typical" design, but I have had occasion to go much higher for armed raiders and convoy protection ships.


C. Considering a few Rules of Thumb from other posters in this thread:

Jorgen_CAB:
"I usually divide up war ships in terms of mission tonnage and none mission tonnage. Mission tonnage is weapons, magazines, hangars and sensors while the rest is armour, engines, fuel, crew quarters and engineering sections. If you can have about 40-50% mission tonnage you are  probably alright for a balanced main capital or escort warship.":
I'd second this thinking.

Lord Solar:
"15% of mass weapons on a warship."
I would almost never have so few weapons on a warship. My warships are born so that others might die and might easily devote 30-40% of total space to either "open fire" or "time to reload" (with apologies to Schlock Mercenary).

nuclearslurpee:
Agreed with most, especially on shields, maint. life, and armor. One point of difference is with missile design; there are major advantages to large, multi-warhead missiles, the most important of which are range and closing speed.

xenoscepter:
Sage advice on always having at least two engines.

[somebody1212]

For commercial ships:
Having a small thermal and EM sensor on each ship gives you a lot more information on anyone attacking your commercial shipping. How important this actually is varies game-to-game, but I've generally stuck by it on most ships.
Much like military ships, if you're not expecting to survive an engagement for long enough for the military to get there, don't bother defending the ship. Just because you can put CIWS and multiple layers of armour on commercial ships doesn't mean you should.

For all military ships:
Ships should have a long enough range to get from one FOB to another, but remember that your range includes any additional fuel being carried by fleet support tankers, not just the ship itself. Conversely, once you have a standard fleet range defined your FOBs should never be spaced further apart than that.
Even without support tankers, non-parasite ships should be able to cross a system. This is so that you don't need to bring your tankers into a hostile system.
Defences are all-or-nothing - if you don't expect a ship to get into a fight (support tankers, support colliers) or you don't expect it to survive damage (fighters) don't bother.
Anything other than fighters should have at least two engines and at least as much MSP as your largest repair cost.
Deployment time and maintenance life should approximately match. If you have to bring your ships back for shore leave anyway you can give them an overhaul at the same time.
Your damage control rating only becomes important if you expect to take internal damage and survive. If you're using heavily boosted engines and reactors, don't bother.
Exception to the above: Heavily armed beam warships should be bringing a lot more MSP than other ships since they'll be draining it rapidly from weapon breakdowns in combat. This is especially true at the moment for any ship using single-shot railguns, but this won't apply after the next patch since SSRs are being changed.
All non-fighter warships should bring at least a small active sensor. There's nothing worse than losing a fight because your sensor ship got destroyed/crippled.
Where DPS for a given weapon size is important, the most effective beam weapon is the largest railgun you can achieve a 5s reload with. Large particle lances are king against moderate levels of armour but struggle against armourballs, and large lasers are optimal against thicker armour. A good beamship shouldn't rely on a single weapon.
Only AMMs should be using full-size launchers. Salvo size is far more important than reload time for anti-ship missiles.

For military fleets:
Keeping the entire fleet the same speed avoids inefficiency.
Beam fleets need to be faster than their opponents. Missile and carrier fleets don't.
A few AMMs are worth nothing. If you're not bringing at least a few thousand, you're better off not bringing any and bringing more PD ships instead.
Jump drives - a defensive fleet doesn't need one. An offensive fleet only needs one.
A good beam fleet shouldn't rely on a single weapon. In particular, a couple of microwaves can ruin the day of anyone who forgot about sensor redundancy, and a few mesons can give people who believe that armour becomes completely redundant at higher tech levels pause for thought.
Adding tractor beams to a few ships in the fleet can make the difference between being able to drag everyone back home and having to leave crippled ships behind for later. On larger ships, they're easily worth the cost.
Box launchers make for a battle-ending alpha strike. Once. Make sure you have enough fire control to let you choose your targets appropriately, since you won't get a second chance.
CIWS has no place in a fleet of more than three ships.
Against an enemy with point defence, missile alpha strikes are far more effective than trickles. If your fleet doesn't have enough missiles to get through enemy PD, it's better off with no missiles.

For military solo ships:
Operating individually, CIWS is king. You only need to protect one ship, after all.
All of the redundancy arguments apply just as much to a large solo ship as to a fleet. A small solo ship doesn't have the tonnage to handle that level of redundancy.
A solo ship is, by definition, operating unsupported. Beam weapons are far superior to missiles for solo ships because of this.
There comes a point where you're better off bringing a stabilisation module than a jump drive. This generally comes when your ship is too large for a jump drive to handle.
Likewise, there comes a point where you're better off bringing a fuel harvester module rather than additional fuel tanks.

For fighters/parasite FACs:
Parasites act more like weapons than ships. Set your range the same way you'd set a missile's range, not the same way you'd set a fleet's range.
Parasites are disposable. This is particularly true for beam fighters.
Sensor fighters can do a lot to extend a squadron's combat-effective range.
Never ignore the potential of an engineless box missile fighter. They're a great way of giving warships a sudden surprise.

Note that a lot of these are based on PvP experiences (both multi-faction runs and the duels run on the Discord) so in a standard game against the AI some of these will be irrelevant and some of these will be unnecessary overkill.

[misanthropope]

gonna hafta disagree with you on one point, somebody.  a dash of AMM support can make a big improvement in your ability to weather a serious alpha strike.  reduced size AMM launchers fit the niche better, if you're heavily committed to point defense.

[nuclearslurpee]

gonna hafta disagree with you on one point, somebody.  a dash of AMM support can make a big improvement in your ability to weather a serious alpha strike.  reduced size AMM launchers fit the niche better, if you're heavily committed to point defense.

While this does depend on your strategy, doctrine, etc., full-size AMM launchers firing as rapidly as possible are generally going to be the most effective way to use AMMs, especially if you extend your AM sensor net with either a big honkin' AM sensor or more practically with a few sensor fighters that can string out in the direction of the missiles. As long as you can extend your sensor net far enough to get off ~15 or 16 volleys of AMMs, you will outperform box launchers against the incoming wave and likely retain capability against the next wave if there is one.

Box-launcher AMMs are most effective at lower tech levels when your missile reload rate is too slow to make this work effectively, but once you have Reload Rate 3 it is feasible to use a full-size launcher strategy and once you have Reload Rate 6 (30k tech level, so solidly into the midgame) there is really no comparison. Reduced-size reloading launchers, even 33% size, are just not good enough except maybe at very low tech, i.e., prior to Reload Rate 3, and this only if you refuse to use any strategy to extend your sensor net.

So basically: box launcher AMMs are useful at low tech or if you don't want to invest into Reload Rate techs, e.g., you use missiles as a secondary/tertiary weapon. Otherwise, full-size AMM launchers will be the most efficient. Of course it goes without saying that if your priority is roleplaying a fleet doctrine, anything you like is fine. 

[misanthropope]

i think i'd like to see your work there, dr slurp. 

[nuclearslurpee]

i think i'd like to see your work there, dr slurp.

Yeah sure:

Let's assume that we are trying to defeat a massive enemy volley of ASMs, presumably from box launchers (defeating a fleet which uses full-sized ASM launchers is trivial, we've all fought the NPRs enough times to know this). A size-1 box launcher is 0.15 HS (7.5 tons), and a full-size launcher for the same size of missile is 1 HS (50 tons). Additionally, a 1-HS magazine at base tech level can contain 15 additional size-1 missiles, so the full-size launcher plus 1-HS magazine brings 16 missiles by itself for a total of 2 HS plus some amount for crew space. In that same 2 HS allocation we can mount roughly 14 box launchers, maybe 15 depending on how much space the crew quarters will cost us. So, if (and this is an important if) we can launch all 16 of our AMMs from full-size launchers before the enemy missiles hit us, the full-size launchers will break even (and slightly outperform) the box-launcher solution.

For sake of example, consider enemy missiles at roughly ion tech moving at, say, 30,000 km/s (for sake of example; building even faster missiles is not difficult at ion tech but missile doctrine is a whole other can of worms). If we have reload rate 3 through 5, we can launch a AMM volley every 10 seconds during which the enemy missiles cover 300,000 km of our sensor envelope. To launch 16 volleys we need 155 seconds (first volley at t=0, second at t=10, and so on until 16th at t=150, plus one 5-sec increment to actually intercept the enemy missiles), which comes out to a sensor envelope of 4,650,000 km. For fleet-standard RES-1 sensors this is a tall order, requiring some fairly advanced tech, but it is a fairly easy sensor envelope to create with a few AWACS fighters for instance (especially since you get to use their sensor diameter rather than only the radius as for ship-based sensors), and you can create some extra distance by piloting your fleet in the opposite direction as well.

Once we hit Reload Rate 6 and 5-second AMM launchers, this envelope basically gets cut in half and becomes quite easy to achieve, and in fact becomes feasible even without fighters if we use a very large AM sensor. Additionally, since magazine efficiency tech improves as well it is possible to get a few more AMMs into each 2-HS launcher/magazine combo. This is not even getting into the strategic flexibility of being able to reload from a collier between battles (deep AMM magazines are a bit questionable since this limits your volley size compared to shallower magazines but more launchers).

It is trivial to also show that reloadable reduced-size launchers are simply inferior, as they cannot match the salvo size of box launchers nor the fire rate of full-size launchers. The main use of these would be using the 33% size launchers as inefficient box launchers which can be reloaded from magazines for purely strategic/logistical reasons, but frankly if you can get away with doing this it is because you are facing NPR fleets which pose no challenge for a veteran when it comes to missile defense anyways.

[Bluebreaker]

Really wish NPR would use reduced size ASM launchers. I have completely dropped AMM because they don't, no point on wasting galicite on them when a simple railgun or gauss turret on every ship will negate any NPR ASM.

[nakorkren]

One other shipbuilding tip:

Put the largest, most reduced power engine you can on your cargo and colony ships that your tech allows, and then build the largest ships your shipyards support. It's counter-intuitive, but large, reduced power engines actually are cheaper and less mineral intensive, and also build faster, so you can build more of them and negate the fact that they're (somewhat) slower by moving more cargo at a time.

Obviously sometimes you just need (or want) your stuff there asap, so also make a few fast "couriers"; small ones for minerals that are urgently needed to resume production of MSP (asking for a friend) and moderately sized ones for emergency transport of that mass driver you forgot to bring along with your asteroid miners (again, asking for a friend).

[nuclearslurpee]

Put the largest, most reduced power engine you can on your cargo and colony ships that your tech allows, and then build the largest ships your shipyards support. It's counter-intuitive, but large, reduced power engines actually are cheaper and less mineral intensive, and also build faster, so you can build more of them and negate the fact that they're (somewhat) slower by moving more cargo at a time.

This isn't really a general truth. For colony ships in particular, but also other commercial ships such as transports which have expensive mission modules, the engines are often a minor part of the total cost, so reducing their cost may not reduce the cost of the class very much. For example, a size-60 commercial ion drive with 0.5x modifier (375 EP) costs 93.75 BP. A 40,000-ton (approx.) colony ship at this tech level can cost roughly 1500 to 2000 BP, with four engines making up only 375 BP of that cost. A significant fraction but firmly the minority.

If we have the same engine with a 0.25x modifier (187.5 EP; 4000 RP for the tech, reasonable for early Ion tech) each engine costs only about 23.44 BP, and four engines cost 93.75 BP. This is indeed a 75% reduction in the cost of engines, but probably only a 15%, maybe 20% reduction in the overall cost of the ships. It's questionable I think whether it is worth building 15%-20% more ships which travel at only half the speed.

A better argument for this strategy would be the savings in gallicite, which can still be significant even though commercial engines are much cheaper than military engines already, and the savings of fuel if your refining or tanker capability is not too impressive. Especially later in the game, large fleets of very large ships can place significant demands on these two resources, along with duranium as always, and in my opinion this is a much better reason to prefer the lower-efficiency engines in the mid to late game.

It is worth noting, though, that the advice to build commercial ships as large as possible is generally very good as large shipyards are more efficient - a single large ship will take longer to build than a single small ship, of course, but it will be built more quickly than the same tonnage in multiple smaller ships.