Battle of Dogger Bank

[tramker]

I made a new scenario, Battle of Dogger Bank.

http://bet.iline.cz/~mk/dl/DoggerBank.zip

Since the AI is dumb, if you want realistic results, set both sides as human controlled and just let the Germans charge ahead at their given course and speed.



http://www.worldwar1.co.uk/dogger-bank.html

[CaptHawkeye]

The AI isn't really "dumb" as much as it's just trying to do its job. IE: Win the war. Dogger Bank basically calls for the AI to think on a widely strategic sense. (IE: Sheer could gain nothing from trying to fight Beatty's cruisers and may just expose himself to the Grand Fleet.) The problem is the battle AI doesn't think that way and has little concept of self-preservation unless it is TOTALLY outnumbered and outgunned.

Besides, none of Jutland's scenarios are designed to play out exactly like they did in history. They start out that way but the beauty of the game is that after the initial dispositions are set, *you* are writing history now. Who'd want to play Jutland if all the battles played exactly like they were supposed to anyway? I'll read a book in that case.

[tramker]

I think a scenario like this shouldn't be hard on the AI, if the game allowed to gave it a little bit of guidance:

- ability to set a goal (preserve the force and escape to Wilhelmshaven).
- ability to keep formations (both division a task force formations). There should be several preset division formations (column, line, echelon) and task force formation (like in the campaign, but tha AI forgets about it as soon as the battle starts).

[Bullethead]

Dogger Bank is a problematic battle to do at present because the Jutland game is set up for 1916. It's not just the obvious thing of having to use a Brit AC for Blucher, it's fire control In very early 1915, things were rather different in that regard, which tended to put Brit and German BC gunnery fairly close to parity. By 1916, however, the German BCs shot way better than the Brit BCs, and that's how things are in Jutland. So that's what you'll see here.

This presents a serious balance problem for a Dogger Bank scenario with the historical OOB. Under the 1916 gunnery conditions modeled in the game, the Germans have a very substantial advantage. They will score significantly more hits in a give period of time than the Brits will, effectively making the Brits outnumbered even though they have a few more BCs. The AI will realize this and fight accordingly.

We hope someday to produce expansion packs for Jutland, extending the timeframe on both ends of 1916. If that ever happens, we'll have early-war versions of the existing ships, plus of course Blucher. At that point, this scenario should work much better.

I agree with you that the scenario editor should allow you to set an objective for a fleet. The game actually supports this already, but it's only available to developers at present, and not via the scenario editor. I've mentioned this myself a few times, but it would probably be good to hear it from customers as well. Why not go to the bug report server and send in a ticket with the SUGGESTION type selected, asking for this?

[tramker]

Acually the first time I played it, I got pretty good results, disabled one german BC and later sunk it with torpedoes and got Lion's turret disabled and severe flooding, but it didn't sink.

But I wonder, what game simulation options are supposed to be realistic ? The presets seems okish, but advanced hits are off and torpedoes seems to me too accurate and maybe gunnery damage too low, but I don't really know.

[Lempereur1]

Originally Posted by tramker

I think a scenario like this shouldn't be hard on the AI, if the game allowed to gave it a little bit of guidance:

- ability to set a goal (preserve the force and escape to Wilhelmshaven).
- ability to keep formations (both division a task force formations). There should be several preset division formations (column, line, echelon) and task force formation (like in the campaign, but tha AI forgets about it as soon as the battle starts).

Why dont you upload your scenario so everyone can take a look at it.

[tramker]

Originally Posted by Lempereur1

Why dont you upload your scenario so everyone can take a look at it.

Of course I did, see the first post.

[Bullethead]

Originally Posted by tramker

But I wonder, what game simulation options are supposed to be realistic ? The presets seems okish, but advanced hits are off and torpedoes seems to me too accurate and maybe gunnery damage too low, but I don't really know.

For historical accuracy, you should leave all the sliders in their default positions and turn on both Advanced Critical Hit (ACH) options.

Turning on "Fagile AP" nerfs the guns that had crappy AP shells in 1916. This affects all Brit guns from 7.5" and up, and a few older German guns (mostly in predreadnoughts and shore batteries). It has no effect on guns of 6"/15cm and below because they only shoot HE anyway (except the 6" BL Mk XVII in Canada). Leaving this setting off basically gives the Brits late-war Greenboy shells, so their 13.5" and 15" guns quickly obliterate the Germans.

Turning on "Poor Cordite Handling" makes all Brit ships of AC size and larger have a 20% chance of blowing up if one of their main turrets is penetrated. This really should be called "Dangerous Cordite", because the Brit ships would explode even if all precautions were followed, due to the nastiness of their propellant. This is an extra and significant chance for Brits to explode over and above the regular, rare explosion caused by the "regular" critical hit. If you leave this setting off, all ships on both sides are still subject to the occasional explosion from the normal mechanism.

So, for the ACH settings....

• Both on: Historical, but can be frustrating for the Brits.
• Fragile AP on, explosions off: Good for MP, because Brits aren't blowing up all the time, and their big guns don't wreck the Germans too fast.
• Explosions on, fragile AP off: Would be historical for 1917-18, so is good for what-if the Brits had Greenboys earlier. Makes things interesting because now both sides are rather vulnerable.
• Both off: Significant Brit advantage in firepower, unbalanced by the German ability to even the odds with lucky hits.

As for torpedo accuracy, that's actually OK when the number of torps fired is historical (as in about 1/2 of the available DD tubes per attack). Most torps come close enough to make the target dodge, but hit rates match those shown in Campbell.

The problem right now is that DDs are firing ALL their torps at once, contrary to doctrine. This makes it rather hard for ships to dodge massive attacks--not enough room between all the torps. We're looking at this at present.

[tramker]

OK, and what about the gunnery damage ? It seems to me that ships (especially BBs) can take too many high caliber hits. I would think about 15-20 high caliber hits should significantly wreck any ship, but thinks like Seydlitz can happily take 30 or 40 hits.
This is coupled with the damage control, I played with it, but couldn't find good solution.

I think gunnery should cause significant 'infastructure' damage (knock out guns, turrets, steering, fires, propulsion, comunication etc), but not that much of flooding. Currently if I increase gunnery damage, it mainly causes more flooding and more fires.

In short I want more damage but less flooding at the same time, what do you think ?

[Bullethead]

Originally Posted by tramker

OK, and what about the gunnery damage ? It seems to me that ships (especially BBs) can take too many high caliber hits. I would think about 15-20 high caliber hits should significantly wreck any ship, but thinks like Seydlitz can happily take 30 or 40 hits.

At Jutland, only 1 capital ship (Lutzow) was sunk by accumulated shell damage, and she took 24 heavy hits. She would likely have survived them had they been less concentrated forward, because she was ultimately scuttled due to her screws being out of the water and no way to tow her home.

Other ships with double-digit heavy hits at Jutland were Seydlitz (22 and a torp), Derfflinger (21), Warspite (at least 15), Lion (13), and Konig (10). Of these, only Seydlitz was in real danger, and that again because much of her damage was forward. The others lost a turret or 2, but little if any speed, and never were in any danger of sinking.

Capital ships, even relatively fragile Brit BCs, were tough things.

In short I want more damage but less flooding at the same time, what do you think ?

I think I've got some explaining to do. This will be fairly long, so bear with me...

In our games, every shell fired is tracked with its realistic ballistic trajectory and where it hits the ship model is where its damage begins to apply. The shells you see in the game are the real things for hitting, hit location, and damage.

When a shell hits, it can burst or break up on impact, or it can penetrate. If it penetrates, its path into the ship is determined, as well as the point where it explodes, if it does. Some shells are duds and AP shells that never hit anything solid can go clear through a ship without exploding. Along the way, shells can hit other armor and be deflected, such as first penetrating the upper belt and then bouncing off the armored deck.

When a shell explodes, it can damage stuff (systems, weapons), start fires, and kill crew within its burst radius. If it doesn't explode, it can still knock stuff over in its direct path. Shells that miss just short can even hit the ship underwater, below the belt. The damage done to stuff in the shell's way depends on the armor of the stuff at risk.

And now a very important point. Ships only sink when water comes in faster than it can be pumped out, just like in real life. Thus, if the shell doesn't make a hole at or below the waterline, it contributes nothing towards sinking the ship, even if it does damage to systems and weapons. However, holes oringinally above the waterline can contribute to sinking later if the ship settles enough from other flooding.

The result of all this is that there are a lot of "meaningless" hits in our games, just as in real life. A 15" shell through the funnel isn't likely to do anything to the ship. Hits that get stopped by armor also don't do much, because the whole purpose of armor was to prevent damage in those areas. IOW, our ships don't have "hit points" that are reduced by every hit. You actually have to do the types of damage at the location you hit. The bottom line, therefore, is that if you're unlucky on hit location, duds, or the effectiveness of fragile AP shells (which varies substantially per hit), a given ship can take a LOT of hits with little or no real decrease in its fighting power. OTOH, if you're lucky on all of this, you can maim ships with just a few hits.

We think our ballistics, penetration, hit location, and damage systems are the most realistic ever put into a naval sim, on a par with anything in the most detailed tank sim. As such, it produces highly variable results in individual trials, but over the long run produces very believable results on average. The only thing is, the graphics for hits always look the same, whether the shell does massive damage or just goes through a funnel. But this is realistic--you might know you hit, but you'd really have no idea of how hard unless the ship blew up, drastically lost speed, or whatever.

[Sailor Steve]

I love the concepts as you've described them. I've been plugging away for years on my own tabletop miniatures WWI game, and came some time ago to conclusions that match yours. In my rules the only British ships to have a real danger of exploding from turret hits are Beatty's battlecruisers, and then only from lessons "learned" at Dogger Bank, and unlearned after Jutland.

I was going to ask if you had read a certain book, but then I got to this part:

Originally Posted by Bullethead

...but hit rates match those shown in Campbell.

You now have my full unqualified support. Once my finances let me make even small casual purchases I'll be getting it. I love that you're using Campbell's descriptions of shell explosion patterns as part of your guidelines.

The problem right now is that DDs are firing ALL their torps at once, contrary to doctrine. This makes it rather hard for ships to dodge massive attacks--not enough room between all the torps. We're looking at this at present.

I remember reading that the Germans criticized the British torpedo tactics as being more haphazard than their own, which involved a group line-abreast approach, turn to a parallel course and fire one tube each, then turn away. Not sure how accurate my description is - just thought I'd toss it out there.

[tramker]

Originally Posted by Bullethead

I think I've got some explaining to do. This will be fairly long, so bear with me...

Thanks for the explanation, this should be put into the manual How does the gunnery damage slider fits in ? It changes blast radius of shells or something like that ?

Anyway, I've liked the gunnery and penetration model from the beggining and that was the main reason I bought the game. But I'm a little underwhelmed by flooding and damage control model, so maybe you should continue you explanation there.

I particullary don't like the idea of ships with more then 100% damage control, which cannot take in any water unless I manage to reduce their crew number by means of starting heavy fire. It seems to me these should be two unrelated things. How did the pumping work in the RL anyway ?

[Bullethead]

Originally Posted by Sailor Steve

II was going to ask if you had read a certain book, but then I got to this part:

Campbell is hard to argue with in terms of hits, hit rates, damage done, and the whys and wherefores of all that. The only quibble on this I have is his count of hits on Warspite.

OTOH, Campbell ain't ideal for where the ships were at specific times because he follows the Beatty school there unquestioningly. Gordon is considerably more more accurate on these points.

You now have my full unqualified support.

Please tell your friends about us

[Bullethead]

Originally Posted by tramker

How does the gunnery damage slider fits in ? It changes blast radius of shells or something like that ?

That, and it increases penetration, and makes the explosions more macho.

Anyway, I've liked the gunnery and penetration model from the beggining and that was the main reason I bought the game. But I'm a little underwhelmed by flooding and damage control model, so maybe you should continue you explanation there.

I dunno, it seems OK to me. Do you have Campbell's book? It's pretty much required reading on this subject.

As best we can tell, Campbell did the best analysis of damage. Thus, we ran thousands of tests (literally) to determine the long-term averages and get them about where he says they were. You have to do a LOT of tests because the whole system has so many variables in it that you can't draw any meaningful conclusions from just a few runs.

I particullary don't like the idea of ships with more then 100% damage control, which cannot take in any water unless I manage to reduce their crew number by means of starting heavy fire. It seems to me these should be two unrelated things.

Not true. Any hit that does flooding damage has 2 components, permanent and temporary. Permanent is basically compartments open to the sea in a big way, which there's no way to pump out. The temporary is in neighboring compartments that can be sealed off and pumped out, or at least corrected (up to a point) by counterflooding.

Damage control rating is dependent on crew size vs. displacement and internal volume. Thus, some ships have more "spare" people (and therefor DC equipment) than others, and can have DC ratings over 100. But this doesn't make them immune to damage.

How did the pumping work in the RL anyway ?

Very differently than many folks have been led to believe from older naval games like GNB.

The bulk of a ship's internal volume was divided into a relative few, relatively large watertight compartments. Most of these ran clear across the ship from side to side. These large watertight compartments were usually (i.e., except for engineering spaces) subdivided into many small rooms, but these weren't "watertight". Sure, they might consist of bulkheads solidly welded to decks, but the bulkheads weren't strong enough to withstand the pressure if 1 side of them was totally flooded. IOW, the could localize an overflowing toilet, but that was about it. This sort of compartmentization, a few large spaces, was about all that anything of CL size and below had, apart from some fuel tanks and scattered void spaces. Thus, they had little or no ability to counterflood.

Larger ships usually had wing compartments outboard of torpedo bulkheads, and double bottoms, for most of their lengths. All these spaces were subdivided into many small watertight compartments. The idea was that most damage would be isoloated in these essentially sacrificial spaces, and similar expendable spaces on the other side of the ship could be counterflooded to maintain trim. But only up to a point, beyond which the ship couldn't take any more water inside.

Where a ship took a hit, it was basically impossible to eliminate the flooding. That area was open to the sea with no way to patch it, so no matter how much you pumped there, the same amount would come back in. Thus, the objective was to wall such areas off by true watertight bulkheads, shored up by big timbers, and run the pumps on the "dry" side of these bulkheads. Bulkheads always leak, either from flying fragments, twisted hatch seals, or pipe/wiring pass-throughs.

[GerritJ9]

Am I correct when I assume that by "Campbell" you mean N.J.M. Campbell's "Jutland: An analysis of the fighting"? Campbell also wrote (among others) "Warship Special 1: Battlecruisers" (ISBN 0 85177 130 0) covering the British and German battlecruisers of the Great War. Apart from the Jutland damage, it also covers the damage suffered at Dogger Bank by the various vessels (info not found in "Jutland"), for instance, though without the detailed sketches found in "Jutland". Unfortunately this most useful work is long out of print, so I'm glad I have a copy.