Monthly Archives: January 2020

Rule the Waves 2: To May, 1910

June 1908

We have the chance to establish a protectorate in Iceland, but it doesn’t go well, and a local warlord takes over.

(That’s the fun thing about random events, the randomness.)

The two Mediterranean Gueydons go into the reserve fleet, along with the older Tages and Fauconneaus. Our starting ships are beginning to get the (O) next to their name which indicates that they are obsolete and, more to the point, old—I don’t recall, but that may have reliability-in-battle implications.

July 1908

Tourville and Dunkerque, the other two Duquesne-class battlecruisers, finish their working up and make their way to the Mediterranean.

September 1908

Expanding private shipbuilding and industry yields efficiencies for the Navy, which means another Pascal laid down to replace another Gueydon.

October 1908

Spies get a hold of the blueprints for Italy’s dreadnought.


It’s at least on par with our Devastation class. Similar broadside—the Devastation can bring all eight guns to bear on one target, while the Andrea Doria can’t, and has slightly heavier armor, but the Andrea Doria is a little faster.

November 1908

Tensions are rising steadily with Germany. (It isn’t even my doing—tensions can go up without events.)

On the plus side, our next class of battleship will look almost conventional.


I lay down another two submarines. We’re falling behind somewhat in that realm, but because our interests are mainly close to home, the cheap coastal boats will suffice for as long as we care to invest in submarines.

March 1909

Germany is clearly pushing for a war with us, which doesn’t bode well. It’s apt to be a strongly commerce-raid-y war on our part. Big fleet actions won’t go well.

Per a reader suggestion, I put the finishing touches on a battlecruiser to mount those new 15″ guns.

Trying something new as far as drawing ship designs goes.

Later note: I just realized I forgot conning tower armor on the Lyon. I guess that’ll be a one-off. Costly mistake.

July 1909

The keel of the first Lyon is laid. Devastation is looking like a one-off, especially since we just developed improved 12″ guns. Perhaps a 10-gun ship will follow.

September 1909

Don’t look now, but tensions with Austria-Hungary are rising. Fingers altogether crossed.

January 1910

Catching flak for not deploying enough in Northern Europe to counter German aggression (tensions ahve been rising again), I begrudgingly move the battlecruisers up that way.

April 1910

We’re on the brink of war with Germany, but a new dreadnought design is on the way:

Note the 10-guns-in-4-turrets arrangement. These are quality-0 12″ guns, with a range of about 17,000 yards—better than the 15″ -1 guns on the Lyon/Lille-class, if also less punchy.

I redid the Lyon class with conning tower armor; future ships will be part of the Lille class.

Two-Year Reports

Not much has changed since last time, with the exception of budget (we can just about afford a Redoubtable, a Lille, and the Lyon currently under construction, along with a light ship or two), and high tensions with Germany.


We are still allied with Great Britain, on the one hand; on the other hand, Britain was worth a whopping 100 victory points in the last war. Maybe they’ll be more useful against Germany.

Wednesday What We’re Reading (Jan. 29, 2020)

Workshop Projects of the Month

Wuhan Coronavirus


Guns/SHOT Show

Science and Technology


Grab Bag

Rule the Waves 2: To May, 1908

By shirking other responsibilities, I managed to fit a full two years in.

June 1906

We uncover an Italian spy, pushing the issue for some extra budget.

July 1906

Britain is working hard to win the new dreadnought race. In addition to their one completed dreadnought, they have four more under construction, along with two battlecruisers.

August 1906

Tensiosn with Germany continue to decrease. I think we might have turned the corner on that crisis.

September 1906


A new technology allows us to plausibly design eight- or ten-gun battleships even with our current three-centerline-turret limitation.


Improved engine technology and doctrinal allowances for 700-ton destroyers let us design our best class yet, with two guns, four torpedo tubes, and 31-knot speed.

October 1906


A historical friend of ours comes calling. We say ‘oui’, and I think this calls for pushing for another war with Italy.

November 1906

Speaking of which…


War breaks out. The first battle is a cruiser action where darkness and poor weather guarantee the fleets pass one another in the night.

In addition to the three Duquesne battlecruisers, one Isly light cruiser, and five Pistolet destroyers on the ways, I queue up six corvettes and two armed merchant cruisers equipped with mines, to help take the pressure off of the fleet.

December 1906

Two days before Christmas, three French light cruisers (an Isly and two Tages) supported by a destroyer flotilla embark on a raid on coastal shipping. They’ll arrive south of La Spezia near dusk, doglegging a bit south to hopefully avoid any patrolling Italian ships.

At 2 p.m., the French squadron sights the northern end of Corsica and turns east.


Just after 4 p.m., the squadron sights one ship moving west-northwest at warship speeds, and turns northwest in pursuit. Another appears on the horizon to the northeast.

At 4:30 p.m., Isly‘s crew spots a line of three ships due north of the squadron, probably armored cruisers. The enemy line turns toward the French squadron just as the sun dips below the horizon, and the French ships elect to make a daring run toward the Italian coast, hoping to search out and sink a coastal merchant or two during the night.

The decision pays off, as the French squadron runs across two transports in quick succession, sinking them both. Isly and Lalande combine to take down the first one, while a torpedo from Isly sinks the second, as its crew flees in small boats.

The job done, the French squadron rings up flank speed and dashes for the Riviera. A major victory for France nets us 968 victory points.

January 1907

On the 28th, two Italian cruisers come across a French Atlantic convoy, facing off against Linois (a Tage-class light cruiser) and a pair of Fauconneaus. The weather is clear, but it’s 4:41 p.m., so the Italian squadron has a limited time during which it can press its attack without attracting torpedo fire in reply.

Confused night fighting sees the loss of both destroyers, but Linois survives and drives off the armored cruisers before they can sink very many freighters. In ship losses, the Italians are the clear victors, but because so much of the convoy survived, it goes in the books as a minor French victory.

February 1907

The Italians put out peace feelers, amusingly. I instruct the government to put the screws to them, and negotiations stall.

After a quiet first few months, French raiders come through in a big way, sinking eleven merchants to the Italians’ 2.

On the 25th, two French light cruisers and a squadron of the newer Francisque destroyers chance upon an Italian convoy southeast of Malta, and sink four transports before fleeing in the face of a superior Italian escort. The objective being six transports, it goes down as a French loss. I disagree, but what are you going to do?

March 1907

This month, it’s a convoy defense. Isly and a pair of Tage-class light cruisers drive off an attack from a similar Italian force. They probably could have defeated the Italians outright, but crews aboard our destroyers misidentified one of their light cruisers as an armored cruiser.

April 1907

Troude, a Tage-class cruiser, rather embarrassingly fails to destroy a bombardment target in Eritrea despite emptying her entire magazine into it.

May 1907

On the 11th, the Italians launch a raid on our coastal facilities. The fleet sorties.

The Battle of Nice

Visibility is good, the weather is pleasant, and we’re in that part of the year where the days are the longest. It’s looking good for something decisive.


5:21 a.m.

The French fleet sights the Italian fleet to the west, and turns to engage.

6:00 a.m.

Together, the Italian fleet turns away from the French. We give chase.

6:44 a.m.

Sighting the Italian armored cruisers to the north, we ditch the battleships for now, on the theory that a stern chase is a long chase, and we have the speed to catch the cruisers against the French coast.


4:57 p.m.

It works. Although the Italian battle line escapes more or less unscathed, three Italian armored cruisers now dot the bottom of the Mediterranean Sea. The only French loss is the light cruiser Lalande, torpedoed by an enemy submarine as we steam back to port.



I’m going to recommend zooming in a bit on this one. It’s a busy picture. I’ve put some numbers on it to guide you through the battle.

At 1), the French fleet spots the Italian, and turns to go broadside to broadside. The sharp turn indicates when the Italians began to run.

At 2), the French fleet turns north after the enemy armored cruisers, and at 3), we drive them up against the coast.

West of 3), the Italian cruisers scatter. We pursue a pair of them up to 4), heavily damaging two of them. One, an Amalfi-class, is dead in the water, and destroyers torpedo it until it sinks. Another, also an Amalfi-class, slinks away up the coast, harangued by light forces until I recall them to screen the battle line.

They get recalled because the fleet spots one more Italian cruiser at 5), this one a slower Carlo Alberto type. One or two heavy hits slow it further, so that the battle line can fully catch up. Eventually, the vastly heavier weight of gunfire tells, and it slips beneath the waves at around 13:00.

From that point to 6), the fleet stands to the northwest, in pursuit of the Amalfi which escaped back at 4), coming across it and engaging it further. The brief French turn southwest, near 6), corresponds to the brief Italian sally at 7). After they lose heart, we resume our chase of the Amalfi, ultimately sinking it just before the minefields at Genoa.

It counts as a major victory, which earns us +1 prestige.

June 1907

The Italian navy declines battle over a large French convoy in the Mediterranean, and has no forces in the Indian Ocean to defend against a coastal raid conducted by a French cruiser squadron.

July 1907

An Italian submarine torpedoes the light cruiser Lavosier, a Chateaurenault. Two Italian light cruisers attack a French convoy, opposed by a pair of Tages and five destroyers. Though the French squadron took heavier damage than the Italian one did, the convoy escapes unscathed.

The first Duquesne-class battlecruiser enters service. It’ll need a month or two to finish working up before we can send it to the Mediterranean.

Design studies on the first French dreadnought battleship are finished now. In thirty months, Devastation will come down the ways.


August 1907

Getting vengeance for Lavosier is our submarine Euler, which torpedoes and sinks the Italian light cruiser Nino Bixio.

On the morning of the 27th, three Tage-class cruisers and a screen of six destroyers set out for the Italian coast, in search of coastal shipping. The weather is breezy and overcast, but visibility is good.

They sink a pair of ships merchants, managing to evade a pair of patrolling Italian armored cruisers by a daring run nearer the coast.

September 1907


The Austrians are building a very obsolete armored cruiser.


The French and Italian fleets collide near Nice, very briefly, before twilight intervenes. During the attempted pursuit, Fronde, a Francisque-class destroyer, strikes a mine and sinks. The fleet quickly abandons the chase, and the Italians notch a very marginal victory.

October 1907


The Italians have an interesting semi-dreadnought in the works, approximately similar to our Tridents, if a bit slower and much more heavily armed.

Duquesne and Jean Bart (the latter another Isly light cruiser) finish working up, and will be heading to the Mediterranean for a November arrival. Or, as I think about it, they will instead stay in northern Europe on trade protection duty, where they might run across some of the Italian armored cruisers raiding our shipping in that region.

Three Tage-class cruisers raid the Italian coast, this time destroying two merchantmen while fleeing from an Italian fleet headed by one of their 22-knot Amalfi-class armored cruisers. (Sure am glad our ships are fast!)

November 1907

At around 2 p.m. on the 8th, under cloudy skies and a moderate breeze out of the southwest, scouting units of la Marine Nationale spot an unknown vessel sailing toward Nice. The fleet turns that direction.

3:28 p.m.

The Italian fleet, predictably, is running to the west. This time, we have a few hours of daylight, so the battle line pours on some speed and tries to get downwind of the Italians, so as to avoid having to shoot with smoke in our eyes.

4:36 p.m.


The Italian battle line comes into view, twelve miles to the east-northeast, heading northwest. We’ll see if we can’t bag a few of these cruisers again; it doesn’t look like the battleships are going to stick around to play.

4:57 p.m.


This Carlo Alberto may be a decent armored cruiser. It is not, however, a match for three battleships.


At this range, even our inept gunners can hardly miss.

5:29 p.m.


The Italian battle line returns as twilight settles in. Soon it’ll be dark; perhaps we’ll be able to run down the Amalfi-class cruiser in front of us.

5:57 p.m.

Night falls. The French fleet takes a northward turn, with an eye toward catching some of the enemies in close and setting the destroyers on them.

7:15 p.m.

The opposite happens; Magenta and La République eat a torpedo a piece. Adroit damage control keeps them afloat, limping back to Nice for patching up.


This battle was a bit simpler, and doesn’t need the explanatory numbering. You can se how the French fleet generally kept to the north of the Italian cruisers, and how, to my surprise, the Italian battleships were south of our ships for much of the action, only turning across our bow at about 3:00 p.m. When I first caught sight of them, it was toward the end of that maneuver.

November 1907, cont’d.

The victory point totals stand at 12,096 for the good guys to 4,438 for the Italians.

December 1907

On the 22nd, the Italians attack a French convoy. I had hoped that Duquesne, dispatched to the Mediterranean in pursuit of the Italian armored cruisers formerly of the North Atlantic, would play a role, but alas, she does not. Instead, three La Républiques take to the sea. It turns into a running gun battle that lasts until nightfall—three French battleships (then two, after Friedland took a hit which damaged her engines) against four Italian, plus two Italian armored cruisers.

Surprisingly, the French ships acquitted themselves well—all three battleships took light damage, inflicting light damage in return on two battleships, and medium damage on the two Italian cruisers.

Intelligence indicates that the next-generation German dreadnought has 16″ belt armor—impressive, and four inches thicker than our upcoming Devastation. A quick look at Britain’s dreadnoughts suggests that Germany is an outlier in the direction of heavier protection.

January 1908

The Italians put out peace feelers again. The Navy recommends that we squeeze them as hard as we can, and…

Unfortunately, the ‘considerable war reparations’ appear to be for flavor.

Large territorial gains, though…

Unlike the last one, this war bears fruit. While Sardinia would have been a sweet apple to pluck, I decide instead to build a belt of bases around Italy (and, for that matter, Austria).

Finally, with an eye toward retiring some of the Gueydons and mothballing others until such time as naval aircraft are invented, I start on an overseas-service light cruiser design.


March 1908

Shipyards lay down the first Pascal, and our second battlecruiser enters service, with the third due next month.

April 1908

Germany’s aggressive naval program prompts an event which forces us to lose face or raise tensions. I elect for the former, dropping our prestige from 25 to 24.

May 1908

Another pair of Pistolets enters service, bringing the total to six. The shipyards enter their postwar slumber, building a pair of light cruisers (an Isly class, named Lavosier after one of the ships we lost in the war, and Pascal), and Devastation.

I don’t have time for the full update—it’s currently 9:44 p.m. on Wednesday, so I need to wrap up and get all the images ready.

Two-Year Report: Diplomacy

Mes amis, we bask in the glory of three new colonies, and a Mediterranean increasingly blanketed by the tricolor flag.

Two-Year Report: Finances

We’re at that point now where the fleet we have costs enough to maintain so that the fleet we want is hard to build. Our recently-peacetime monthly budget is 13,071 funds, of which 6,720 funds are eaten up by maintenance, 1,568 by research, and 200 by intelligence. That leaves us around 4,500 to spend on construction, which is not quite two dreadnoughts.

Some ships will probably have to be mothballed or retired going forward, and the Gueydons overseas are prime candidates. As soon as the new Pascals start to roll off the ways, the Gueydons deployed overseas will start coming home, to be stored away for later use or scrapped altogether.

Other candidates include some of our light cruisers—we still have the second-largest light cruiser fleet in the world, behind Britain, although (as I’ve mentioned before) our light cruisers pull some of the same duties that armored cruisers do in other navies.

Two-Year Report: The Fleet

We’re in good shape compared to our Mediterranean peers. We have three dreadnought-style ships in service (all Duquesne battlecruisers) and one dreadnought battleship under construction. Italy and Austria both have dreadnought battleships under construction, one a piece. Another option, one I haven’t used much in previous Rule the Waves games, is the reserve fleet option, which halves maintenance costs for ships without much long-term penalty.

Anyway, I have two questions for the gallery:

  1. Should we make use of the reserve fleet? If so, what should we reserve? One or two battleships? Older light cruisers? Old destroyers? Some combination?
  2. What should our shipyards focus on? Dreadnought battleships (i.e. 22-knot ships which don’t compromise on armor or guns, to the extent possible), or battlecruisers (i.e., 24+-knot ships with respectable armor and fewer or smaller guns, as needed to attain the above)?

Finally, some bonus material, courtesy of a reader at a different site: video of the 13″ guns firing on a predreadnought of the Royal Sovereign class.

Wednesday What We’re Reading (Jan. 22, 2020)

Winter Wargaming may be slightly delayed—I have a lot of writing left to do on that front, and tonight in which to do it.


  • Rainbox Six, a re-read. The last time I read it was also post-9/11, but I’m much better informed about the state of the world today than I was last time, and the starkness of the divergence between that fictional timeline and our real one is worth mentioning.
  • Castles of Steel, a re-read in progress. I’m up to just past the Dardanelles. That particular campaign (the naval side, not the ugly, unproductive land side) strikes me as one we can’t really game out. Could the Royal Navy have forced the straits alone? Maybe so. But, if so, would the Ottoman Empire have collapsed? Nobody knows for sure, and neither can we design a wargame which gives us an impartial answer.
  • The Mauritius Command, which I thought was a re-read, but now that I’m halfway through I’m not so sure. You can’t go wrong with Patrick O’Brian.


  • Return of the Obra-Dinn, a rollicking good nautically-themed murder mystery. The conceit: an empty ship floats into harbor circa 1807. You, an insurance investigator, go aboard, with a watch in your possession which permits you to view the moment at which a person died, given a bit of their body (or a vision of their body in a vision of someone else’s death). Worth the price of entry.
  • A Painted Ocean. It’s not so much a game as a toy, but it’s a brilliant little toy: you’re the skipper of a full-rigged ship with detailed physics, and you sail it around. That’s the whole game. It’s delightful. The only thing that seems a bit off to me is that she has a habit of ‘sticking’ by the wind as you come out of a tack, not making enough way for the rudder to bear her up; I usually have to reef the mizzen-sails and the spanker to get her head to come away from the wind.



Science and Technology



Grab Bag

  1. I’m not much of a gambler myself, beyond the socially acceptable kind where you hold a basket of mutual funds for 40 years, but if I did want to engage in Pennsylvania’s most recently legalized vice, it would be on the back of a neural network trained to predict, say, hockey games. Sports betting is nice in the same way that poker is, in that you’re not playing against a house who is permitted by law to fleece you, but against the rest of the betters. With a little elbow grease, you can be better at it than they are. 

Rule the Waves 2: To May, 1906

We return with another two-year stretch. (Or, at least, hopefully two years. I have company coming over on Saturday, which is my usual play-the-game day. We’ll see how far I get.)

Goals for this entry include designing a battlecruiser (circa early 1905), keeping the naval budget more or less balanced, rebuilding our older battleships to use better fire control, and pushing for moderate tensions with Italy and/or Austria-Hungary to permit us to build more ships.

June 1904

To start with, I place a few ships into reserve fleet status, which cuts their upkeep in half but reduces their maximum crew quality to ‘Fair’. (Ordinarily, the maximum is ‘Good’. I don’t recall offhand if specialized training increases the maximum to ‘Elite’.) Given that I don’t expect any wars in the immediate future, we can afford to.

Another Francisque-class destroyer comes off the ways, completing our initial buy of seven. I could scrap some of the Fauconneaus, but destroyer upkeep is so cheap that it’s hardly worth the effort.

Finally, I start the rebuild process for the La Républiques, updating them from central rangefinding to central firing. The history of battleship fire control is the history of centralizing more parts of the process. Central rangefinding moves the rangefinding away from the individual guns and to a central position, which can be elevated above the guns’ smoke and also made more delicate and (therefore) more precise.

Believe it or not, this photograph is of a ship firing using smokeless powder. Clearly ‘smoke less’, not ‘smokeless’.

Central firing moves the triggering of the guns to a central location, which helps eliminate errors in timing.

Finally, director firing lays the guns automatically—the turret crew no longer controls azimuth and elevation.

August 1904


July passes quietly. The British are still building pre-dreadnoughts—and pre-dreadnoughts which will be a much greater liability in the future naval era than our Tridents.

Italy, too, is refitting its battleships with central firing.

October 1904

The first La République completes her refit.

At the same time, the first design studies on the Duquesne-class battlecruiser begin.


Tallying the votes across all the places where this AAR is running, German-style battlecruisers won the day. This 24-knot ship mounts six 11″ guns, a secondary battery of 6″ guns (+1 quality), and a tertiary battery of 2″ guns (+1 quality). (A gun of +1 quality is approximately equivalent in range and penetration to a 0-quality gun with a caliber one inch larger.) She has a 10″ armored belt, and tips the scales at a hair over 18,000 tons.

In other news…


I was flipping through the almanac to see where we’re going to land in the dreadnought race (second to get one under construction, it looks like!), and found that the Austrians call this a battleship. We have to have a war with them.

January 1905

We elect to refit the Tridents with central firing before they even come down the ways, which saves us a rebuild cycle on them.

February 1905

The first Duquesne‘s keel is laid. She should be ready in early 1908.

The lack of any budget-increasing events has been a bit of a bummer. I’m considering mothballing some of the light cruiser force to free up some more money. As it is, we’re building one Trident, one Chauteaurenault, one of the new Isly light cruisers, and one Duquesne, and still losing money. Ideally, I’d be able to rebuild a La République with better fire control while still keeping up on the dreadnought program.

Advanced gunnery training is a stretch goal, but the budget is too tight to permit it right now.

March 1905

Given that our light cruiser fleet is still enormous compared to everyone except for Great Britain, I decide that putting a few in mothballs (it’ll take about a year to bring them back to combat strength) is acceptable to keep the battleships rolling. Especially now that we’re building replacement fleet light cruisers, keeping all the Tages at 100% operational capacity isn’t as important.

April 1905

A new government wants to cut arms expenditure. I protest loudly and receive a small bump in the naval budget. There are now three La Républiques rebuilding at the same time. (Also, it’s a little cheaper than it appears at first—you don’t pay regular maintenance on ships under rebuild.)

June 1905


Nothing bad can possibly come of this. We stand behind our ally and reap the budgetary rewards.

Upside: we can afford the refit on the rest of the La Républiques. Downside: tensions are up with Germany, who we really can’t fight on even terms.

September 1905

Thinking they’re being helpful, the government votes to increase naval spending given tensions with Germany, which… raises tensions with Germany.

October 1905

The French public raises 50 million francs for a battleship. We lay down one Duquesne because our last pre-dreadnought Trident completes, and one Duquesne with the funds the public so helpfully collected for us.

Six-gun ships are nice, but I’d like to push to eight soon.

November 1905

Thanks to our dreadnought-building program, Britain is forced to raise spending to keep its navy preeminent.

December 1905


The Americans sell us the rights to steam turbine technology, which we’ll take, thank you very much.

Propulsion is one place where Rule the Waves elides a little bit of detail. Steam turbines, in the game, represent a simple decrease in the weight of a ship’s machinery. This is a bit of a simplification.

Shipboard steam propulsion starts with evaporators. Salt, as you’re probably aware, is corrosive, and salt and steam are worse than either in isolation. Marine boilers and condensers demand fresh water, so steamships have to produce fresh water from the materials at hand—heat and seawater. Evaporators distill seawater to fresh water, which is then fed into the boilers.

Boilers do what they say on the tin, turning fuel (in this era, coal or oil) and fresh water into high-pressure steam. The volume of steam a ship’s boilers produce determines how fast it can turn its engines.

In our early-20th-century timeframe of interest, there were two types of engine of note. The first is the multiple-expansion engine, most frequently the triple-expansion type. Steam flows into three cylinders of increasing size, driving a piston in each cylinder. Increasing the size of the cylinder at each step means that each cylinder generates substantially similar force—as the steam flows through the engine, its pressure goes down, so giving it a larger area to act on counters that effect.

The second type is the steam turbine, demonstrated in dramatic fashion by the Turbinia, which showed up at the Navy Review during Queen Victoria’s Diamond Jubilee, and proceeded to outrun the fastest vessels the Royal Navy could send to chase it down. From this beginning, turbines eventually made it into most of the world’s warships by about 1910. (At the end of this tangent, I actually back up my assertion that the elision of detail is important.) Like all turbines, steam turbines are essentially pinwheels writ large—blow through it, or force high-pressure steam through it, and it rotates.

Finally, after steam passes through the engine, it arrives at the condensers, which turn it back into fresh water for recycling through the system again. Reusing water means that the evaporators don’t have to work as hard (although ‘not as hard’ still translates to ‘tons per hour’, in this context). When condensers break, steam-powered ships are unable to generate as much steam (since they have to wait for the evaporators, rather than using water they already have), which slows them down.

Anyway, all that to say that the US Navy, in the early days of steam turbines, waffled between turbines and the older triple-expansion engines. Why? Because turbines are only very efficient near full power, and triple-expansion engines, though larger and bulkier, can run at cruise power much more effectively. As late as USS Oklahoma (laid down 1910, commissioned 1916), the Navy built ships with triple-expansion engines, because for they had better range for a given weight of fuel, and we Americans didn’t build fast battleships until the North Carolina-class in the late 30s. Other American ships (and other shipbuilding nations) experimented with a smaller cruise turbine, which would push the ship at cruise speed when running at full power.

In Rule the Waves 2, you don’t get the choice. You just pick a fuel type and an engine focus (from Speed, Reliability, or neither).

February 1906

An uprising in China presents us with the chance to reduce tensions with Germany, which we gratefully take.

May 1906

Germany takes advantage of our softness and sends a force to occupy Angola, which produces very little of note.

Two-Year Report: Diplomacy


Aside from the aforementioned tensions with Germany, things are quiet enough. Italy is making noise again, and building a few more battleships to boot.

Two-Year Report: The Fleet



Speaking of which, the fleet report! We’re currently operating at a deficit of 1,553 kilofrancs, but the first batch of ships will finish before we run dry (a new light cruiser and the first Duquesne).

Right now, we look pretty good in the Mediterranean Throwdown Power Rankings. We have a small edge over Italy right now in battleships, and given that our battlecruisers are armored well enough to stand in the line of battle, we’ll maintain that edge even given the predreadnoughts they’re still building.

We’re behind in armored cruisers, as ever, but the battlecruisers are, in part, intended to fix that.

Our huge superiority in light cruisers gives us advantages in the commerce raiding game—we can detach a bunch of them to go sink merchants without much fear of losing them or falling behind our chosen opponents in attached-to-the-fleet strength. Ditto destroyers; they’re a great way to fill the trade protection quota while corvettes build. On the downside, we’re a little behind now on submarines. Should we think about building more?

That said, I think there might be room in the schedule and the budget for an updated Chateaurenault class. The Gueydons, which are filling the larger part of our foreign obligations, are expensive to maintain, especially away from home waters. A class of foreign station light cruisers, with medium or long range and equipped for colonial service (the latter makes a ship count for 150% its tonnage when determining how much you have vs. how much you need on a foreign station), would fill the gap nicely. We could mothball or even scrap a Gueydon or two, and put the savings into more shipbuilding.

Another option might be to put some money toward a class of coastal monitors—ships with, say, a pair of large-caliber turreted guns, low speed and short range, and a ton of armor. With some of those, we could limit the ability of foes to blockade our North Atlantic or Mediterranean coasts without diverting units from the main fleet.

Of course, there’s also room for a class of proper battleship-style dreadnoughts—something with 22-knot speed, a bit more armor, and 8 or 10 guns. (The only reason the Duquesnes are six-gun ships is because we don’t have the technology yet to put more than three turrets on a ship such that all of them can fire at a broadside enemy. There are two technologies that allow that: 4+ centerline turrets, and cross-deck firing for wing turrets.) These three Duquesnes will likely be the only three, as well, given that we have steam turbine technology now, and that leads to large weight savings at higher speeds.

Of course, if we wanted to stick with a six-gun ship, we also just developed 14″, quality -1 naval guns, which would go nicely on a dreadnought.


I won’t be able to do my usual weekend play-through, so next week’s update might slip a bit, or perhaps cover less time.

Cadillac Gage Commentaries: Lowers

Fishbreath hasn’t finished up the usual WWRW for today, so I’m running this instead

It occurred to Fishbreath and I that, while we have chatted at length about the What Would Stoner Do (WWSD) project of the good folks at Inrange, I haven’t written about it. And, with Inrange revisiting it, it’s a good time to shamelessly piggyback and comment on it.

The name for my series, of course, is a play on what Stoner actually did after working on the AR-15: go work for Cadillac Gage.1 As a firearms engineer, the real answer to the question “What would Stoner Do?” is that Mr. Stoner would design what you paid him to design. Were he still alive, and still designing things, I would love to see his take on an NGSW proposal.

Snark aside, let’s get to the project. I do love building AR-15s and working out parts lists. The first part they are talking about is the lower. And they chose the GWACS Mk. II polymer lower originally. They now have KE Arms making an improved Mk. III, which is good as GWACS is no longer a going concern.

As a stylistic choice, I would have preferred a sort of Battle Royale where they look at competing options, so we can better gauge why they went with what they did, and what things they didn’t consider. But that would also lead to fewer questions from yours truly so perhaps I should be grateful that they didn’t play it like I did.

In terms of weight reduction in Anno Domini 20202, we can go lower with a standard forged lower and a carbon fiber stock like the one from Smoke Composites. We can also get even more weight reduction with an aluminum-lithium alloy receiver, such as those made by V7 systems or Hodge Defense Systems Inc. Those are expensive, but the WWSD project didn’t set a budget ceiling.

Anyway, I don’t like the whole integrated grip and stock thing3. I grew up in a ban state and purchased my first AR in “Ban compliant” format with a pinned stock. So maybe I have an irrational aversion to a fixed-length assembly. But I like to be able to adjust my stock. I like to be able to choose a stock with adjustable comb height if I so choose. I even like stocks like the Odin Works Zulu or the Tacmod that let me change the angle of the stock’s recoil pad. This is America. We like choices.

And frankly, the stock should be something that works with the optic, not the other way around. Unless you have a very peculiar parts list, your optic cost more than your stock. Which means that if one of the two parts has to give, it ought to be the stock. Not that adjustable is a compromise, frankly.

We also like choices about our grips. And I realize that grip choice is at once very personal and probably doesn’t matter a whole lot on the clock. But a comfortable grip that Joe Civilian likes helps him to practice with his rifle, since he lacks a sergeant to yell at him to do so.

My choice would not be the CAV-15 polymer lower. The AR-15 is as popular as it is in large part because of its massive aftermarket, and I don’t like closing off large parts of that aftermarket. I like picking the grips that I like, and changing them as my tastes change. I like choosing a stock to suit the intended purpose of my builds and to get the balance where I want it. But my goals and priorities are not necessarily those of Ian and Karl, and they may or may not match up with yours. Purchase your lowers accordingly.

Were I picking a lower for another build, I would choose a nice forged lower from a reputable company with a rollmark that I liked. If I wanted to save weight in my lower, and didn’t care about paying through the nose for the privilege, I would purchase a fancy aluminum-lithium lower.

  1. Obvious disclaimer: this post series and this blog are in no way affiliated with Textron Marine and Land Systems, the former Cadillac Gage. Duh. But I’m writing this anyway in case you had any doubt. 
  2. I don’t recall if these were available in 2017. But it doesn’t matter thanks to the revisit. 
  3. Yes I recognize that’s what makes the GWACS/CAV-15 a viable polymer lower. No, I still don’t like it. You might think that this means I don’t like the core concept, and you’d be right. 

OpenTafl 2020: New Tree Search Horizons

That’s right, I’m getting back to a project I’ve left alone for a while: OpenTafl, my hnefatafl engine project. The only one of its kind, it is both a host for other, yet-unwritten tafl bots, and a tafl bot in itself.

Why the return after all this time? Well, two reasons. First, I had to move the whole project to Github recently, because Bitbucket is shutting down its Mercurial repositories. Second, I had a bit of a brain blast on how to do a particular kind of AI project I’ve been wanting to try for a while.

OpenTafl’s current AI uses an alpha-beta pruning search, which is only a hop, skip, and jump away from the original, mathematician’s tree search algorithm, minimax. In simple terms, alpha-beta pruning minimax plays out all possible variations to a certain depth, assuming each player plays optimally, and picks the move which leads to the best outcome. The pruning bit skips searching nodes which are provably less optimal.

Of course, knowing what the optimal move is depends on one of two things: searching the game tree all the way to the end, or searching to a lesser depth and evaluating the leaf nodes in the tree. The former is impossible for reasons of computational power, the latter is logically impossible1. Evaluation functions, as we call them, are always imperfect, and require an awful lot of domain knowledge to do well.

Because tafl is a poorly-studied game, there isn’t a lot of domain knowledge to encode into an evaluation function, which has always limited OpenTafl’s potential somewhat. There are further downsides to the alpha-beta search it uses, too, in particular that it can’t readily be multi-threaded2. So, what’s the answer?

Well, at least potentially, Monte Carlo tree search. Popular among go AIs (and used as the framework for DeepMind’s efforts in computer players for board games), the secret to MCTS is a bit of randomness and a preference for exploring interesting lines of play. Start at the root of the game tree, navigate through the nodes you’ve already seen. When you find a leaf node (that is, one with no children), you generate its children, then play a random game until someone wins or loses. At each tree node, track the win/loss ratio for the tree, and use Mathematics™ to guide your root-to-leaf exploration in future iterations.

Simple! Of course, tafl poses some unique obstacles to MCTS, as Tuireann of discovered. The biggest issue is that random moves in tafl are very, very unlikely to do anything of interest—tafl branching factors are higher than branching factors in, say, chess, and there’s no space pressure like there is in go. (That is to say, if you keep making random moves in go, the game eventually ends.)

Tafl MCTS artificial intelligences need some way to guide the playout process (the process of playing to the end of a game). The modern approach for this is to either train a neural network on a large corpus of existing high-level games (tafl doesn’t have one), or train a neural network by playing it against itself for a long time (which I don’t have the budget for). Given those constraints, I set about inventing a measure which would permit me to make random-ish moves which nevertheless move the game forward.

I’m calling the measure King Distance to Victory, or KDV; it’s inspired by part of the evaluation function from J.A.R.L., one of the entrants to the 2016 OpenTafl Tafl Open (which is on the calendar again for 2020, AI programmers!). For a given game state, the KDV is the shortest straight-line distance from the king’s current position to a victory space, counting spaces occupied by allies as 2 and spaces occupied by enemies as 3 or 4. The defender’s goal is to bring KDV to zero. The attacker’s goal is to maximize the average KDV and minimize the variation in the KDV values3.

It isn’t a perfect measure, but it isn’t meant to be—rather, it’s a measure to push playouts toward states which end the game. On that ground, I think it’ll be successful. I also hope to write a quick classical evaluation function which uses the KDV measure exclusively, to see how it plays on its own, without the MCTS magic behind it.

More news to come as it is made.

  1. This proof is left as a trivial exercise for the reader. 
  2. To prove which nodes can be ignored, alpha-beta search has to evaluate them in order, for a given value of ‘order’. 
  3. This helps encapsulate the idea that the attackers should aim to surround the king and close the noose, rather than simply get in the way of his best move. Note as well the implication that each state has multiple KDVs: an edge-escape game has up to four (the king moves straight from his current position to each edge), and a corner-escape game has up to eight (the king moves straight from his position to an edge space, then from the edge space to one of two corners). 

Rule the Waves 2: To May, 1904

Let’s get right to it.

June 1902

Lalande, a Tage, joins the navy, scientists invent the six-foot rangefinder, and naval engineers work out that double bottoms are a good plan (stolen from the Americans).

Strikes delay the construction of Suffren by one month.


Jeers from the naval engineering community lead the Ministère de la Marine to hastily release a slightly more traditional design, with 12″ main guns and six 10″ secondaries. It loses 200 tons and gains a bit of extra armor in the bargain.

Designers play around with some armored cruiser designs, but the naval community eventually rejects the idea.

July 1902


An opportunity arises to hack off our other Mediterranean neighbors. With the budget increase, I lay down another pair of Chateaurenaults, for a total of four under construction.

Italy commissions one of the armored cruisers our spies stole the details of, and invents the early coastal submarine.

August 1902

In response to our hacking off, Austria-Hungary increases its naval spending.

Italy is building more coastal batteries in the Mediterranean, a 6″ and an 11″. The latter might someday cause us trouble.

Coastal batteries might be worth investing in at some point—not for the guns themselves, but because, I believe, the amount of coastal fortification you have increases the extent and density of your defensive minefields.

September 1902

We lay down Trident, first of her class. Linois (a light cruiser) and Epieu (a destroyer) enter service.

October 1902

Tensions with Italy are at the breaking point. War is likely, if any events go in such a fashion as to push us any further.

Italy’s naval budget goes up, and they lay down another armored cruiser.

Our budget, annoyingly, goes slightly down.

November 1902

The Italians are busy this month.

January 1903

We completed research into improved face-hardening, which will improve our future ships’ armor.

March 1903

The new naval minister wants 15 destroyers under construction, and is willing to bump the budget a bit to achieve that, so I take the deal.

April 1904


April 1904: Battle of Crete


The first sea action of la Marine nationale opens on a calm April morning in overcast weather, as the Italians happen upon a French cruiser squadron steaming west-southwest Crete. Visibility should be excellent.

On our side are the two Gueydons based in the Mediterranean, Bruix and Montcalm, six Tages, and seven Fauconneaus. Italy has more cruisers than that in this region, but it remains to be seen how many will come out to play. The sun is rising behind us, which gives our ships a bit of an edge if a battle happens early.

In the event that it looks bad, our squadron, with a speed of 23 knots, should be able to outrun the Italians, the only difficulty being the relative lack of sea room here in the Middle Sea.


The light cruisers assigned to scouting fan out for a better view.



The light cruiser Lalande spots a ship ahead. I order a turn to the north-northwest to avoid closing too quickly.


The light cruisers spot a half dozen ships. I increase the squadron’s speed to 20 knots and pull the scouting force in to screen the armored cruisers.

Two minutes later, Lalande identifies one of the light cruisers as a Salerno class. It bristles with small-caliber guns.



The light cruisers begin to identify the Italian battle line, which looks to comprise at least four armored cruisers. The squadron stays at 20 knots, pending identification, but we’re probably running.


Another Italian light cruiser is identified as part of the Nino Bixio class, which is a Salerno without dual-purpose guns.

One minute later, wireless signals from a light cruiser identify one of the enemy ships as a Carlo Alberto-class armored cruiser.


Because the Italians accepted a lower speed, their cruisers get more guns. Because we decided on a higher speed, we don’t need to face them, and Bruix leads Montcalm in a turn east, in pursuit of the better part of valor.



It’s the right move. Look at that swarm of slow, poorly-armored cruisers!



This, however, is quite a light cruiser. 6600 tons, 10 6″ guns? Wild. At least it’s slow.


The Italian battle line cruises past our stern and turns away. The heavy cruisers come about to see if we can’t maybe dispose of a shadowing light cruiser before they come back.


The red circles represent the range of our cruisers’ main batteries. The larger gray circles represent the edge of their visual range.

I was wary of some manner of trap, but the only ships we can see are the two light cruisers ahead of us, which will shortly be in range. I’m still prepared to run if the rest of the Italian squadron makes an appearance, but it’s looking like we might draw first blood.


Lalande is the first to open fire.


The two Italian light cruisers are joined by a third.

Also, it takes us nearly an hour to score a hit: Lalande lands a blow from about 5500 yards.


The Italian cruiser turns away from its allies, and our fleet sets off in pursuit.


Regrettably, the Salerno-class target is still alive three hours after the first shot, though burning and badly damaged.


With the Italian cruiser dead in the water, our ships take one more run past it to ensure it goes under, and depart to the west.



The scenario ends, as our ships and the enemy’s are far apart. La Marine nationale acquitted itself relatively well in the face of a superior force, escaping serious damage and sinking an enemy ship.

Bruix won the gunnery medal for the day, with a 2.25% hit rate.

April 1903 (cont’d)

013-new destroyer

The war cancels the naval minister’s ambition for more destroyers, but a new class is in order anyway. These new Francisques don’t sacrifice anything from the preceding Fauconneaus, and have a two-knot speed advantage on them.

Time for wartime dispositions. The Gueydons in the Mediterranean are made commerce raiders. Because of their speed, they’re practically invincible unless caught entirely off guard.

A pair of Chateaurenaults are coming next month. They’ll go on trade protection for a few months, relieving two Tages currently filling that role, until some corvettes currently under construction can take over.

May 1903

A bevy of technologies arrive this month, but none of them are dramatic improvements—lots of slow-and-gradual stuff.

The Italians raid the Northern European coast with three armored cruisers. Three destroyers sally to meet them, but the two forces don’t meet.

Because we don’t have any armored cruisers, the Italians win a bunch of dominance-of-the-sea victories around the Mediterranean. (If the game thinks a battle should happen and one side can’t field appropriate forces, the other side wins by default, as though the under-equipped side had declined battle.) Our commerce raiders and evasion of their coastal raid earn us about the same number of victory points, and in this second month of the war, the tally is 751-623 for the home team.

On the upside, the Chauteaurenault class exceeds its design speed in trials, and hits the same 24-knot mark our existing light cruisers do.

June 1903


Well now. I can live with those odds.

June 1903: The Battle of Bordighera

A leisurely five-hour cruise from Toulon, we encounter the enemy fleet at 8:18 p.m., on a southwesterly course.


The enemy battleships almost immediately turn away.


After a half an hour of ineffective firing, darkness falls over the Mediterranean.

I’m of a mind to push onward toward the Italian fleet. We have a huge speed advantage, are evenly matched in guns, and have more destroyers (since we’re operating close to our bases). Night fights are bloody, but I think we can make something out of this one.



Running side-by-side with a trio of perhaps-cruisers-perhaps-battleships at 4,000 yard, Solferino, La République, and Magenta score hit after hit on one of them.


The Italians bug out toward La Spezia, having dealt some damage to our battleships, and taken some in return. All told, not a bad little battle. La République had a bit of a scare losing electric power and then catching fire, but got both problems under control by 11:30 p.m.

The damage tally calls it a marginal Italian victory. I believe it’s on the strength of Italian gunnery—we scored a lot of hits, but most of them were with the secondary and tertiary batteries on the battleships. The Italians did better with their heavy guns.

July 1903

The government is asking whether to seek peace. I figure we’ll let it go a while longer—they aren’t that far ahead, and we’ll have some submarines joining the fight soon.

July’s battle is a convoy fight. We have a three-ship battle division, three light cruisers, and a bevy of destroyers. Conditions are good, with a moderate breeze, and the sun is high overhead.

Italy brings a similar force: three battleships, a pair of armored cruisers, and some destroyers.

As before, the red circle is the selected squadron’s gun range, and the gray circle is what it can see. Battle Division 5, the lead squadron in the battle line, is selected. Because Light Cruiser Division 10 is out in front of the fleet a little ways, I can still see the transports making their way east by south.

As you probably can’t make out, it’s a tense situation. Let’s break it down a bit.

First, as I mentioned, it’s a daylight battle, so I can’t rely on sunset to mask my convoy from the depredations of the Italian squadron. Second, and of equal importance, the Italians have two squadrons which could seriously threaten the convoy: the battleship squadron, center-left in the picture above, and a cruiser squadron, to their battleships’ southeast.

At the moment of this screenshot, the two battleship squadrons are having an ineffectual gunnery duel—at that range, around 6,000 yards, nobody really expects to hit much. That fight is happening about 30,000 yards west of the transports; I let the Italians pull me off of the convoy, since their cruisers were shadowing their battleships. Eventually, the cruisers peeled off and headed back for the convoy, at which point the speed of the French battle line paid off—we could make it back to the convoy quickly enough to prevent the Italians from doing very much damage to it.

Unfortunately, I didn’t take many pictures of what was by far the most exciting battle to date. The Italians did heavy damage to one of our destroyers, but we did more damage to their ships generally than they did to ours, and they’ll likely be in the dockyards for a month or two.

Rumors of war-weariness and protests reach our spies in Rome.

August 1903

Another fleet battle off the south of France. This time, the Italians bring six battleships to our four (two are in for maintenance). OUt fleet exchanges desultory fire with the Italian van, then falls back on the minefields at Toulon.

September 1903

The French fleet launches a raid on Italian coastal shipping. Unfortunately, before the sixty-mile seach line comes upon any merchants, it comes upon an Italian cruiser squadron, quickly reverses course, and makes it back to the French coast with no losses, briefly stopping to bombard a shore battery.

October 1903

We decline a cruiser action in the northwestern Mediterranean, and accept a battleship brawl in the early morning of the 17th.

This one looks more favorable than most: the ships involved are four French battleships against, by all appearances, two Italian battleships and three armored cruisers.


On closer inspection, it appears the ratio is flipped: two armored cruisers and three battleships. Still not bad odds, and the Italians have turned to run for it.

Not all of the Italians make it. The most reliable weapons on our Les Républiques (the 5″ tertiary guns) slow down one light cruiser enough for our ships to fall on it. A second Italian light cruiser appears to the south of our force as we’re leaving the battlefield. Quick thinking by your admiral, who detaches a light cruiser squadron to run down this second target, wins the day.

After seven months of war, the tally stands at 2,774-2,732, just barely in favor of the home team.

With some of our cash cushion, French shipyards work on kitting out a pair of armed merchant cruisers to raid Italian shipping.

November 1903


The Italian fleet sorties on the 5th, and at 10:46 a.m., the French fleet is there to meet it. It’s the whole French fleet this time, too, all six battleships.

Although the Italians have five battleships and two armored cruisers to our six battleships, they nevertheless turn tail after a brief exchange of fire, during which their gunnery proves more accurate than ours (as has been the case for this entire war).

Sinking a destroyer unlucky enough to take a hit from a 13″ shell, the French fleet turns northward to bombard a shore battery on the coast near Imperia—the battleship guns should make short work of that, at least—and then perhaps eastward to see if any Italian shipping is at sea.

It isn’t, so we go home.

December 1903

A raid on the Italian west coast produces no results and a few hundred victory points for the Italians. Owing to the relatively successful battles over the last few months, and the Italians declining battle a few times, the war score stands at 3,794-3,198.

January 1904

A new year sees an Italian armored cruiser fail to prosecute a night attack on a convoy, driven off by a pair of Tages and a plucky destroyer flotilla. There are reports of widespread civil disturbance in Italy. If we stay the course a bit longer, I think we’ll be in good shape.

February 1904

A large Italian convoy escapes an attack by French light cruisers. Italy regains a small lead in the victory point rankings.

Our submarines make their first major contribution, torpedoing an Italian armored cruiser (but, alas, not sinking it).

March 1904

One year into the war, the Italians put out peace feelers, and the civilian government agrees. The Navy acquitted itself relatively well, sinking three enemy light cruisers and three destroyers in exchange for the loss of one destroyer of its own.


It’s only entering construction as the war ends, but to combat the new Italian 24-knot light cruisers, the Ministry of the Navy solicits designs for a 25-knot light cruiser, the Isly class. They’ll begin to replace the Tages over time.

Lessons from the War

In no particular order…

  1. I probably shouldn’t have ignored armed merchant cruisers and small corvettes in my previous RTW2 games. Since they’re converted civilian ships (liners and trawlers, respectively), they only take four months to build, and they’re a good way to quickly bulk up a navy so you can use your warships for war, rather than trade protection.
  2. Our lack of powerful armored cruisers is a bit of a problem at the moment.
  3. French gunnery was atrocious—the main batteries on our La Républiques were only good for about one hit every 150 rounds in good conditions. The Italian battleships shot better with their heavy guns.
  4. The Italian 12″ guns outrange our 13″ -2-quality guns, which makes the redesign of the Tridents to use 12″ guns look even better.
  5. French commerce raiders served admirably, sinking merchants at about a 2-1 rate over their Italian counterparts.
  6. The Gueydons, with their high speed but small guns, are extremely useful for hunting light cruisers, but not much good at fighting armored cruisers themselves.
  7. The Italians fought this war with extreme cowardice. One-to-one, I think their 1900-era battleships are superior to ours, and they often ran away when they had numerical superiority.

May 1904

Something marvelous happens. French naval thinkers take a drag on their cigarettes, sip their red wine, and ponder: what if we built ships with three centerline turrets?

The way to a dreadnought battleship is open. The money isn’t there just yet, but a Trident finishes in eight months, and I think that’s the time to get one going in the yards.

Two-Year Report: Diplomacy


The war with Italy has drawn to a close, and tensions are low. Intelligence remains focused on the Italians and, to a lesser degree, the Austrians.

Two-Year Report: The Fleet



At present, we have a world-class fleet of light cruisers and destroyers, to the point that we can consider putting some into the reserve fleet, or mothballing or retiring them altogether. We’re a bit over-budget at the moment, but I haven’t canceled any wartime shipbuilds yet, either.

Our battleship fleet is solidly middle of the road. We can’t challenge the three largest powers, but we can meet any of the other three on equal terms.

As for shipbuilding priorities, I have two thoughts. One: join the British in the dreadnought era (or perhaps the jupiter era, as they’ve begun work on the world’s first new-style battleship, HMS Jupiter) with a dreadnought battleship of our own. Two: instead, start by building a dreadnought cruiser (a battle-cruiser, if you will), to help counter everyone else’s massive advantage in armored cruiser count.

If we start with a battlecruiser, the question is, what do we sacrifice? Battleships try to balance speed, protection, and firepower. Battlecruisers sacrifice one of those three to gain an edge in the other two. Historically, the Royal Navy went with firepower and speed over protection, while the Germans went with speed and protection. (You could presumably sacrifice speed for firepower and protection, but I think you’d just end up with a slightly slower battleship in that case.)

Two-Year Report: Meta

I realize that this entry is a bit hard to follow, jumping around from battle to battle in several different styles. I was so pumped about getting to a war that I kind of forgot to walk through a battle from a gameplay perspective. I’ll try to rectify that next time a war comes up, making heavier use of the after-action report map mode, and perhaps taking a bit more time to explain how the game handles battles.

Wednesday What We’re Reading (Jan. 8, 2020)

Winter Wargaming tomorrow. Don’t forget to check back and vote on the direction for our fictional France—I typically play over the weekend.

I have to write the 2019 audience report at some point. The very short version is: slightly reduced traffic over 2019, but counting Discord, massively increased audience engagement. To our regulars, both the commentariat and the lurkers, we’re happy to have you around.


  • Me: Castles of Steel, by Robert Massie. I’m reminded of an era when the government owned the design of its warships.
  • Parvusimperator: Samurai!, by Saburo Sakai.


Science and Technology


Grab Bag

  1. That said, I don’t find credible the claims that GPT-2 is a revolutionary advancement for a variety of reasons that I should probably explain in an article. Here’s one thought, though, from the SSC comments: “The scary version of [artificial general intelligence] is supposed to learn and improve faster and faster, but GPT-2 is the opposite, because the more things it knows in a domain, the larger the chunk of data it needs to learn the next thing. And the more complex a domain is, the worse this problem gets.” Granted, yes, but that’s also not how humans learn. We’re more like an S-curve, I would say—we start out slow as we familiarize ourselves with the vocabulary of a field, go into a zoom climb where the limit on our learning is our pace of reading, rather than our pace of deep comprehension, then level off as we near our peak achievement. 

Rule the Waves 2: To May, 1902

Before diving into the month-by-month update structure to which Rule the Waves 2 lends itself, I realized based on some comments last week that I haven’t explained what Rule the Waves 2 is.

Rule the Waves 2 is a strategy wargame which covers, in roughly equal parts, the design and development of warships between 1900 and 1955, and battles using those ships. The two most important parts of the game are the ship design screen (which you’ve seen a few times already in the first entry in this series, and which you will see twice more today) and the battle interface (which has not yet come up). The strategic layer which glues the two together is primarily there to create reasons to design ships and to start wars in which the ships get used.

That’s pretty much the long and short of it. The game advances in one-month turns; battles play out in one-minute real-time simulation steps.

June 1900

Back to the game. Two fresh destroyers come down the ways, and our spies uncover details on a new class of Italian armored cruiser. Not much to worry about.

001: cruiser Francesco Ferruccio
Much slower and much lighter than ours.

August 1900

Unlike historical France, this is not a democracy, but the weight of public opinion nevertheless leans toward bullying the Mediterranean. Another fellow is doing a Rule the Waves France playthrough right now, and is also bullying the Mediterranean. It seems to be working, but I also hate to be a copycat. I think we’ll keep our eye mainly on the Mediterranean, but we won’t disdain the chance to expand in Southeast Asia if it falls before us. (At the behest of a blog commenter—I’m running this in three places with three relatively active audiences, if I hadn’t mentioned that already.)

With an eye toward Mediterranean domination, the bulk of the navy moves to Marseilles. A squadron of destroyers stays on the northern coast.

October 1900

October sees a handy event. The Prime Minister makes a gaffe, so we exploit it for more money. This has the side effect of bringing war with Italy nearer, but with the increased budget, I order another Tage-class and push our research spending to the maximum permissible 12%. Worth it.

December 1900

A new government decides to raise expenditures on armaments. That’s good news for us. We lay down a new La République, to be named Suffren.

January 1901

The Gueydon-class cruiser Bruix enters service. Replacing her on the build list is one light cruiser and three destroyers.

Light cruisers are particularly important elements of the fleet. They serve as our eyes in fleet battles, typically deployed in an umbrella ahead of the battle line. Given the speed of our battleships and the size of their guns, holding our preferred range is important, and a robust scouting force of fast light cruisers will let us do that.

February 1901

Parliament votes to cut naval spending, upending our carefully-planned budget. We have enough of a cushion to run a deficit until some ships finish building, however, so that’s what we’ll do.

March 1901

003: heavy secondary guns

A breakthrough in ship design! We can now build semi-dreadnoughts, ships with secondary batteries not dramatically smaller than their primary batteries.

That brings us into approximately the semi-dreadnought era. Pre-dreadnought battleships, with a small battery of heavy guns and a larger battery of quick-firing medium ones ones, were predicated on the idea that naval gunnery at long ranges was not possible. You had your big guns to punch through heavy armor, and a good number of smaller guns to wreck your target’s upper works and superstructure as the range closed, keeping up a high rate of fire.

Early on in the 20th century, advancements in rangefinding and fire control proved that this was not the case. Since your big guns could hit targets at long range, it made less and less sense to carry a lot of middleweight guns when you could instead spend that weight on your heavy ones, and deliver knockout blows from further away from your enemies. This update to the received wisdom on shipbuilding ultimately produced the dreadnought battleship.

The technological capability to build ships with, say, three centerline turrets with 12″ guns existed in 1900, but doctrine had not yet caught up. While most of the technologies we’ll develop in Rule the Waves are actually technological advancements, some (in particular, in the Ship Design tree) represent the development of new ways of thinking. We, as players with some historical knowledge, know that the dreadnought battleship is the way of the future, but our 1901-era French have not yet hit upon that idea, so the game sticks some limits on us to force us to stay in character.

April 1901

Spies report that the Italians have also figured out the heavy secondary battery. We’re keeping an eye on their shipbuilding, but they haven’t laid down any new battleships just yet. It appears they’re focusing on their wimpy armored cruisers.

May 1901

More intelligence reports from Italy: they’ll be commissioning a new light cruiser just after the New Year, and their current light cruiser class has a speed of 21 knots and belt armor of two inches. Their armor is heavier than our light cruisers’, but they can barely outrun our battleships, and our armored cruisers could easily run them down and have a massive firepower advantage.

Our Gueydons slot into the hierarchy of naval warfare right around where battlecruisers do a decade later. Battlecruisers made armored cruisers obsolete, by dint of their higher speed and heavy main armament—a battlecruiser could easily catch and outshoot an armored cruiser, and could easily outrun a battleship (whose armor tilts the fight in favor of itself).

The Gueydons to light cruisers are like battlecruisers will be to armored cruisers, in that they’re excellent light cruiser hunters—faster than other armored cruisers with heavier guns, faster than most light cruisers, and much more heavily armed. Too, as someone pointed out, they’ll make great experimental aircraft carriers someday.

004: almanac
One year in, the Italians have nearly caught up to us in naval budget. We’re closing the battleship gap, but won’t have our sixth ship until 1903. They’re building more armored cruisers, but ours are heavier.

June 1901

We may not get the chance to catch up, though. Italy looks to be making waves in Greece.

005: italy ultimatum

The Italians back down, but tensions rise.

July 1901

With tensions between France and Italy running high, we skip hosting an international regatta in favor of keeping the budget focused on shipbuilding.

The Italians raise their naval budget again.

August 1901

We catch a German spy. I briefly debate using the occasion to push for a higher budget, but decide instead to sweep it under the rug. Tangling with Germany would be unpleasant.

September 1901

A disarmament conference in the Hague ends with no concrete results, to my satisfaction.

October 1901

One light cruiser and three destroyers enter service.

We send a force to quell an uprising in China, which increases tension, but also raises the prestige of the French Navy.

One concrete suggestion from a reader was a cheaper (that is, expendable) class of light cruiser for commerce raiding. Enter the Chateaurenault class:

006: chateaurenault design

At 2800 tons, it’s a little more than half the displacement of our Tages. It doesn’t sacrifice much speed—at 23 knots, it’s still one of the faster light cruisers in current production, at least that I’m aware of—but has lighter armament and armor. It’s much cheaper than the Tages, too, so we can bulk out our fleet of light cruisers more quickly.

December 1901

007: the joys of a free press

Thanks to the newspapers, we get a little budget bump. Design studies on the Chateaurenault class are finished, so we order two with some of our current excess.

January 1902

008: san diego class

Spies recently dispatched to America to see what one of the big-spender naval powers is up to bore their first fruit this month, giving us a blueprint for an American armored cruiser of the San Diego class. It isn’t much to write home about. The Gueydons remain clearly superior in weaponry, speed, and armor, as you might expect from ships with an extra 5,000 tons displacement or so.

It’s a good month for the French intelligence services generally. We hear that the Americans are building a Raleigh-class light cruiser with 3″ main guns, outmatched by even our cheap light cruisers. Austria-Hungary has commissioned two armored cruisers this month, and Italy has improved its armor-piercing shells.

February 1902

009: spies get caught

Spies recently dispatched to America now bear negative fruit. One of them gets caught.

As a consolation prize, we invent a number of new technologies this month, reducing the weight of our ship’s hulls and machinery and allowing us to build ships with secondary batteries in double turrets. Ocean, a La République, is coming off the ways in three months. I smell a new battleship design to take her place.

March 1902

The Japanese wish to buy a technology from us. I decide to sell it, given that we’ll shortly be paying the design expenses on a new battleship design.

Well, since we are paying the design expenses on a new battleship design.

010: trident design

The Trident-class battleship features the same four-gun 13″ main battery as the older Les Républiques, backed by a pair of two-gun 12″ wing turrets in the secondary battery. 12 4″ guns in turrets provide defense against small craft. Their armor protects them against their own guns between at least 5,000 and 9,000 yards, and possibly further out or in depending on gun data at those ranges.

Finally, they have a 22-knot speed, practically unheard-of in this era. Even the real-world HMS Dreadnought, still four years off, only managed 21, and that was with turbines rather than the triple-expansion engines we’re rocking.

Anyway, the upshot is that sufficiently fast heavy ships render armored cruisers obsolete, so there’s no real reason to build more Gueydons.

That brings us to a decision point. The Tridents are the largest ships we can build in our current docks. Should we focus on building a shipyard expansion or two over the next two years, building one Trident at a time, or build out the fleet more quickly by building two Tridents at once? A shipyard expansion costs 2,000 per month, I believe, so at this point in our game history, it’s just about equivalent to one battleship.

April 1902

Just as we prepare to render the armored cruiser obsolete, the monthly intelligence report indicates that Austria-Hungary, Great Britain, Japan, and Italy have either laid down or commissioned armored cruisers this month. Our spies in America steal the blueprints for another San Diego-class, which is of limited use to us given that we saw the same blueprint in January.

May 1902

The La République-class battleship Ocean is commissioned. One year from today, the last La République, Suffren, will join the fleet. Next month, the design studies on the Trident type will finish, and we can start producing one of those.

French researchers have developed reliable bursting charges, which will enhance the damage of penetrating hits. A trade mission to Great Britain yields important results: we developed the technology to build coastal submarines. Finally, we learned how to make 600-ton destroyers. A new design there will be in the offing early in the next entry. Speaking of, should we spend our small-ships budget on new destroyers or on some submarines? The coastal submarine design will work well for us, given our focus on the Mediterranean; each submarine costs a little more than half as much per month as the Fauconneau destroyers do (although the Fauconneaus are cheaper in absolute terms, taking fewer months to build).

The intelligence report is very busy this month: Germany commissions a pair of destroyers and a pair of battleships, Great Britain is concerned about its naval superiority and has increased its spending. (That’s their national special ability. Ours? Our government frequently changes its mind on fleet priorities. C’est la vie…) Italy has laid down a pair of light cruisers, but between what we have in service and what we have under construction, we still have a better light cruiser fleet.

Japan has at least two classes of armored cruiser in progress right now, one with 10″ main guns and another with a 20-knot speed and a 5″ armor belt, both of which are entirely outclassed by even our existing battleships, to say nothing of our future designs.

Two-Year Report: Diplomacy

011: diplomacy

Tensions continue to run high with Italy, our most likely foe in the next few years. Despite our efforts, the Austro-Hungarians continue to view us as friends.

Britain and Germany, the two do-not-touch powers, aren’t our biggest fans, but also don’t seem to have much motivation to come after us.

Two-Year Report: The Fleet

012: fleet comparison

Here are the Mediterranean Current Naval Tonnage and Planned Naval Tonnage Rankings. Italy has about a battleship’s lead over us presently in the latter category, with the situation flipped in the former.

  1. FR 190,700
  2. IT 179,600
  3. AH 128,600

  4. IT 238,300

  5. FR 222,100
  6. AH 164,100

Italy still holds a lead in battleship tonnage, and will continue to do so until at least one Trident enters service (which is two years and five months out from when we lay the first one down).

Their lead in the Planned Naval Tonnage category stems from a heavy investment in armored cruisers, to match the Austro-Hungarian plan. Should we could consider a class of inexpensive armored cruisers to match them? Compared to a Trident (which takes 29 months to build and costs about 2100 funds per month), a hypothetical 22-knot cruiser with 9″ main guns, a 5.5″ belt, and a turreted 5″ secondary battery would take 22 months to build and cost about 1,500 funds per month. That’s a little more than our current class of large light cruisers per month, and five months longer. These cruisers would be armed slightly below the standard of their peers, but armored similarly, and would have a speed advantage of a knot or two. I lean against the idea, myself, given that we have a fast class of battleship and a light cruiser class suitable for commerce raiding, but I figured I’d put it on the table.

Under construction right now, we have one La République, which will be finished in 11 months, a pair of Tages, which will finish between one and four months from now, and a pair of Chateaurenaults, which are about a year away, in addition to a single Fauconneau.

Our current budget surplus is 2,120 funds per month, which will be put into a Trident as soon as the design study finishes next month. Out of our total budget of 15,100 per month, maintenance on the fleet costs us 4,652 and construction costs us 6,315.

Don’t forget, when answering, that shipyard expansions are also on the table as options for spending our money.