Jutland AAR

[julianbarker]

Have played the Jutland scenario loads of times, but this is the clearest result I have seen. Playing Germany I managed to encounter the BCF from the north. Initially exchanging fire with the 5BS my BCs suffered little, but managed to destroy Warspite with a lucky hit. By the time the 1 and 2 BCS got engaged they were trapped between my battle line and my BCs. My battleline circled round them to the north, and was between them and the bulk of the Grand Fleet. When the British battleline engaged, I was crossing it's T and the British BBs never properly engaged me but took lots of damage. By now darkness was only an hour away, so I turned SE and deployed DDs to help keep the British BBs at a distance. Night came, and  slipped away to the south. Final result, I had two BBs torpedoed  but neither in danger of foundering. The British lost Warspite and all six of the BCF BCs. The 2BS also took a pounding when the T was crossed. 

Definitely the clearest outcome I have had from this scenario, and a point score almost 700,000 in my favour!

[dlacombe65]

Great results! After how many games of Jutland?

[julianbarker]

Probably about ten playings of that scenario. Usually there is a more even distribution of damage. This is the first time as Germany I have managed to isolate the BCF and pound it.

[randomizer]

Good show! Scheer's dream come true.

[vonfriedman]

In the thread subject recently started about a Jutland AAR, played in Admiral mode for the 100th anniversary of that battle, a discussion arose about the strange behavior of AI controlled Beatty's BCF, which failed to turn north after meeting the bulk of enemy's forces.
It seems to me that such strange behaviors could avoided by changing the setting (from attack to withdraw - if it is possible) of minor AI controlled forces while playing the scenario in Admiral mode and not only before the scenario starts.
The same should automatically happens to the AI controlled opposing forces, following the overall battle situation and not only on such issues as damage and so on. This could avoid having some forces wandering around or just avoiding to engage the enemy, as reported in some AAR and as experienced also by myself.
I admit not being familiar with AI issues nor with previous discussions - if any - about this subject.

[andyhall]

There's not much going on in this thread regarding Jutland, so I'll drop in a non-sequitur and trust the moderators to move or delete it if they deem it necessary.



Last night I was playing a Baltic action that put my three Russian dreadnoughts up against two German dreadnoughts and three battle cruisers. At one point the second ship in my battle line took a direct hit from a 12-inch shell that destroyed the bridge and (presumably) killed or incapacitated all the command staff and control gear there.

At it happens, the leading ship in that column had just begun a turn to port. Sure enough, the second ship that had been hit missed the turn, and continued on the same course for a couple of more minutes, before getting back under control and swinging back into line behind the leading ship. The third ship in line missed the turn as well, keeping station on the second ship's stern.

This behavior in the game struck me as being exactly right; it's just what I would expect to happen in a real-world engagement. It's a credit to the developers that it handled this situation in this way, and marks yet another difference between a simulation and a game.

[randomizer]

Probably not the place but it's an excellent observation worth a response. In my (possibly biased) opinion, Fredrik's recreation of machine-age naval combat in SAI and RTW provides more representative results more often than any other model. That is to say that far more often than not, an action will produce results that appear consistent with those that we might expect in the real world. Notice that I am avoiding the term "realistic" or "realism" since these have come to carry so much subjective, extraneous baggage in gaming communities as to have become essentially meaningless. It can be frustrating reading through the Forums, reading post after post where a writer obsesses on the occasions where the games might miss the mark while ignoring all of the many hits. Thank you for the anecdote and the support.

[vonfriedman]

I too very much appreciate this game. In any case: why not making it possible to "change one's mind" during the battle and to become aggressive or cautious following the circumstances or higher command advice?

[randomizer]

Jun 19, 2016 11:40:13 GMT -6 vonfriedman said:

I too very much appreciate this game. In any case: why not making it possible to "change one's mind" during the battle and to become aggressive or cautious following the circumstances or higher command advice?

The following is my opinion only and does in no way reflect the opinions of SAI's designer or any other member of the NWS Team. It is offered up solely as personal opinion and a response to an issue that you seem unwilling to let go of.

Do you have a single substantive action during machine-age naval combat where this actually happened? It will really great to see an example because if there are any, I must have missed them entirely.

Armed forces tend to fight like they are trained and within the doctrinal parameters of that training and professional competence of the leadership. "Changing one's mind", presumably switching from an defensive posture to an offensive posture in action (or vice-versa; this seems to be your big gripe, that doing this on demand in SAI is impossible) would most likely result in chaos and disaster. Exactly how would a commander transmit this "mind change" to subordinates? In the days when wireless was useless for giving tactical instructions, communications were restricted to signal flags, semaphore and searchlights, all problematic in poor visibility in addition to being slow and imprecise except for the pre-arranged signal groups that conformed to the existing fleet doctrine. There's no telepathy or magic mind-meld that would instantly allow subordinates into the now-changed mind of the commander and putting this into SAI where doctrine and postures are at the core of the opposition actions would be gamey in the extreme.

There's a very real military axiom that states Order + Counter-order = Disorder.

The AI in SAI suffers from the same information restrictions as does the Player. You already have tremendous advantages simply because you're human but that's not enough; you seem to want a magical button to instantly place the minds of all of your computer-controlled subordinates completely into your new plans.

Come up with precise examples of when this successfully happened any time between 1900 and 1925, I look forward to it.

Thanks.

[vonfriedman]

Thank you for replying.
My question was based on the thread "Steam and Iron AAR" by cv60. A strange behaviour of AI-controlled Beatty was documented there.
May be that a "change of mind" or, more properly, a change in tactical behavior would have led to a more realistic outcome of the simulation.
At Jutland, the aggressiveness of Beatty ceased when he met the HSF. Idem during the second battle of Helgoland, when the german BBs appeared.

[Fredrik W]

The overall aggressiveness is just one factor influencing how the AI behaves. The AI constantly evaluates its situation based on enemy strength, own strength, damage, ammo levels etc. That means the AI might "change its mind" a number of times. A force with attack stance will just be a little more likely to be aggressive than a normal force, similarly, a cautious force will just be a little more careful, but other factors are more important.