More freedom with building

[Herpaderp]

I understand trying to keep it historical, but it would be nice to have a bit more freedom in sandbox mode than we have. If I want to put cage masts on a super battleship, I want to be able to do that. If I want to mount an 18" gun onto a light cruiser, I want to be able to do that. Having specific places you can mount things makes sense in the campaign or naval academy, but in sandbox, it limits creativity. I can't make a replica of some historical ships because it won't let me put certain parts in places because there aren't any mounts there. I don't care if the ship is incredibly unbalanced, as long as it still floats. I want to make outrageous ships, but can't because of mountings and allowed parts. It's a great game despite that, and it's fun to make different ships with strange layouts, but they're all still fairly similar because of the mounting issue. I want to put the main tower on the back of the ship, with all guns forward, like the Nelson class, but the main tower can only go so far back on the hull because there aren't any mountings further back, and it requires a secondary tower, which won't go in if there aren't any mountings behind the main tower. Keep it the same for the campaign and naval academy, but let us run wild in sandbox, that's why it's sandbox.

[DougToss]

No light cruiser could withstand the firing of an 18", so I don't understand what the point of the exercise would be. You couldn't even get it out of the yard. 

[Shaftoe]

While I agree that the game should allow far more creative freedom, so we can build historical vessels such as the Nelson-class BB, I cannot agree with the idea of having 18" on a light cruiser (unless it is some sort of 1st April scenario). Realism and design feasibility often walk hand in hand, so there are serious practical reasons why some unorthodox ship designs simply wouldn't work. 

"Strategy should govern the type of ship to be designed. Ship design, as dictated by strategy, should govern tactics. Tactics should govern details of armaments."

-J. A. Fisher,  1904

[disc]

That example is a little extreme, but certainly there were plenty of ideas for building small ships with enormous guns....

...granted, not very good ideas.

[DougToss]

I was just about to say, none of those worked out. Even the Hull there, probably the most successful of the bunch, only had that gun for a while. 
That's with a 8" gun on a 2800 ton ship, an 18" gun is what, 150 tons minimum? That's before whatever kind of ungodly reinforced mount you'd need so that the keel could take the firing and to manage the heel when firing broadside.  The Japanese experiment with Canet guns was not successful, Matsushima didn't exactly cover herself in glory.

[Ruan]

Part of me says let us build those silly designs and just apply the appropriate penalties. However, realistically they are so outlandish that just having restrictions would provide the same result with less development.

Op does bring up a good point though. The designer is still very ridged with placement. The builder should have rules we need to follow. Like CL can only use 6-8 inch guns maximum or 4in armor maximum. That sort of thing but should give as much freedom within those rules as possible.

I honestly feel that the current setup of having a huge number of hulls, super structure, guns, etc to pick from is creating more work for less results. This isn't a play for modular hull building but I like to use it as an example as you can build a few modules and end up with a larger amount of combinations and thus more hulls that what you would get with a pre-built list. Same goes for little things like having barbettes tied directly to guns and modular super structure. At any rate this is just me complaining that wasn't the system worked towards, as far as I can tell, and it's probably to late to change most of it at this point. All we can hope for now is that snap points are removed and perhaps some tools to control the shape of hulls to make the most of them.

[FlyingDolphino]

I think more freedom is only good, however appropriate penalties should be applied so that doing certain outlandish things, will be very difficult to get to work even decently and thus are unviable to build an effective ship. The designer is not very flexible, which is a bit of a shame and i feel like maybe having some more customization of the hull that the player can do, as well as removing snap points would help to allow more creative ship designs. Basically, if I wanna build a ship with some ungodly turret set up which makes no sense physics wise, then i should be able to do it, but obviously the ship would have to deal with a massive instability penalty making the ship pretty useless at aiming.

Theres no harm in allowing creative freedom as long as the penalties mean that you cant make something effective, when IRL it would of just not worked

[DougToss]

I think in a technical sense, there are a lot of ways the ship designs could be freed up. More placement points, altering hull form and so on, I'm all about it. I think players should be able to design anything a naval design board, naval architect or shipyard would allow, plus or minus some eccentric concepts that while not ideal or practical, were certainly possible. That means putting up with reasonable penalties that would have been found acceptable when they were discovered in the design stage, in trials or in combats. 

I think casemated 12" guns for example were just on the edge of technical possibility when the game starts, and while they are terrible and inefficient for a variety of reasons, there were apparently plans for Russian predreadnoughts with large calibre main armament mixed between turrets and casemates. 8" secondary armament would be another example. A terrible idea, considering the British found 6" guns useless as secondary armament, bagged charges are much slower and harder to handle than QF shells and so on, but designers then and players now did wonder if it would be better to have guns that could disable a torpedo boat in a single hit.

Beyond limiting designs that would immediately capsize, sink during sea trials and so on, the reason I would like some limit in the ship designer is that many players are coming in without any knowledge of warships. I don't care about preventing them from designing ships that are bad in gameplay terms, that's part of the fun and learning process, what I mean is that it may be better to make something actually impossible in the game, rather than explain to the player why it was practically impossible. If a ship would evidently capsize as soon as it was launched, nobody would allow the keel to be laid. A 500% penalty to accuracy is essentially "useless in battle" so again, that is never leaving the design stage.   I guess my standard for this would be the designer should reject designs that no shipyard or even the most eccentric designer would sign off on. If a design would evidently be useless in combat and no doctrine could incorporate it, I just don't see how or why it could be built.

Basically, nobody would allow for the construction of the Matsushima in 1900, ship design had already moved on, it had proven ineffective in combat, the design philosophy and doctrine of the Canet gun was refuted by the Spanish-American war and would shortly be even more so in the Russo-Japanese war. However I don't know how you can tell the player that (tooltips, encyclopedia, manual), or teach them through gameplay, so restricting them from doing that is effectively saying "you are playing as the head of your nation's naval design board, and what you want to do would not be possible for someone in your position". 

I know in X-Plane you can design every part of your plane. Do players who don't know about aviation design a plane with one wing, and when it doesn't work get mad quit the game? How does the game teach them how to design a plane, why planes are designed the way they are, and to boil it all down, how airplanes work?

I'm still working out my thought on the interplay between player education/training either through information or gameplay, and the designer. I'd love to hear what others think because it really is a puzzle. How to you allow people with no knowledge of warships to design ships, without them getting frustrated when their ships don't work the way they want or expect, for reasons they don't understand, and quit?

 

Edited May 21, 2020 by DougToss

[lordcmdr]

3 hours ago, FlyingDolphino said:

I think more freedom is only good, however appropriate penalties should be applied so that doing certain outlandish things, will be very difficult to get to work even decently and thus are unviable to build an effective ship. The designer is not very flexible, which is a bit of a shame and i feel like maybe having some more customization of the hull that the player can do, as well as removing snap points would help to allow more creative ship designs. Basically, if I wanna build a ship with some ungodly turret set up which makes no sense physics wise, then i should be able to do it, but obviously the ship would have to deal with a massive instability penalty making the ship pretty useless at aiming.

Theres no harm in allowing creative freedom as long as the penalties mean that you cant make something effective, when IRL it would of just not worked

I agree that you should be able to build whatever you want, and let the design stand for itself. If it isn't feasible- well, you wasted money and now your ship immediately rolled over upon a hard turn.

[AceOfBlades]

For what it's worth, I'd like a bit more freedom to do things ahistorical as long as they are physically possible. I'd like to be able to at least have the option to put obselete technologies on my ship if I so choose. 

[IronKaputt]

6 hours ago, DougToss said:

Beyond limiting designs that would immediately capsize, sink during sea trials and so on, the reason I would like some limit in the ship designer is that many players are coming in without any knowledge of warships. I don't care about preventing them from designing ships that are bad in gameplay terms, that's part of the fun and learning process, what I mean is that it may be better to make something actually impossible in the game, rather than explain to the player why it was practically impossible. If a ship would evidently capsize as soon as it was launched, nobody would allow the keel to be laid. A 500% penalty to accuracy is essentially "useless in battle" so again, that is never leaving the design stage.   I guess my standard for this would be the designer should reject designs that no shipyard or even the most eccentric designer would sign off on. If a design would evidently be useless in combat and no doctrine could incorporate it, I just don't see how or why it could be built.

The only reason to make something actually impossible is to prevent "Camel is a horse designed by AI" situation I presume, othewise limits goes against the game's whole idea. Physical possibility is crusial, no argue with that,

but why we forbidden to have sonars and smokes on large ships for example?

I'm strongly disagree with "it may be better to make something actually impossible in the game, rather than explain to the player why it was practically impossible" - it's simpler, yes, but not better in any sense. "Yeah, the designer kinda there, but since we consider players unable to think and learn, you can't actually do much with it. Please enjoy the game which treats you like ignorant moron."

If player with zero knowledge skips tutorials and ignores designer warnings he'll learn the hard way, but please don't rob people the possibility to experiment (or just have some stupid fun with meme ships).

Edited May 21, 2020 by IronKaputt

[madham82]

Everyone should keep in mind, some of the reasons you can't do X is because the AI has to be able to build with the same rules players do. Short of splitting off the building into 2 sets of rules (player and AI/autobuild), there will be things players can figure out how to place but teaching the AI to do it is another story. 

[disc]

Would it be difficult to implement two separate rulesets?

[Ruan]

47 minutes ago, disc said:

Would it be difficult to implement two separate rulesets?

The problem with this is it could be confusing to the player. Differing rule sets can make it surprisingly difficult to gauge what the AI can and cannot do and work around it. The second problem is differing rule sets could give the player a substantial advantage over the AI in unexpected or unintended ways.

[Wowzery]

While I agree with the OP, I can also see how adding even more options can overwhelm a new player who isn't as knowledgeable in warships.  There is a balance to achieve in giving knowledgeable players more options while limiting newbies.

[roachbeef]

18 hours ago, Ruan said:

Differing rule sets can make it surprisingly difficult to gauge what the AI can and cannot do and work around it.

I do not believe that is a bad thing. AI can do some things that player cannot do, like control their entire fleet at once (yes, I know you can pause but most people will find that too time consuming). You don't need to force symmetry to have a fun game. In fact, limiting the AI even further in terms of what it will build may not be a bad idea. It'll make more believable decisions that way instead of building 49-kn ships with 2 9" main guns and 24 2" pop guns.

 

If someone wants to exploit the AI, they can do it in their own time. Most players won't bother, and the ones who go in-depth into exploiting AI decisions are typically long-term players anyways.

[Ruan]

3 hours ago, roachbeef said:

I do not believe that is a bad thing. AI can do some things that player cannot do, like control their entire fleet at once (yes, I know you can pause but most people will find that too time consuming). You don't need to force symmetry to have a fun game. In fact, limiting the AI even further in terms of what it will build may not be a bad idea. It'll make more believable decisions that way instead of building 49-kn ships with 2 9" main guns and 24 2" pop guns.

 

If someone wants to exploit the AI, they can do it in their own time. Most players won't bother, and the ones who go in-depth into exploiting AI decisions are typically long-term players anyways.

I think you misunderstood what I meant. You and the AI should be able to build the same ships within the same rules. If you can build a 39kn Battleship with 2 9" main guns, so can the AI. If you can build a Nelson class look-a-like on a cruiser, then so can the AI. Rule sets are not the same as algorithms however. The Devs could tell the AI, yes you can build that but maybe you should build something more conventional instead. So the AI might only pop out a wacky design 5% of the time instead of 60% of the time. That sort of thing. I believe this may be what's preventing the changes people have proposed. Like the proposal to have towers placed on a center line instead of single points. You'll need the AI to be able to also use that line to figure out where to place the tower and still build a functional ship with all the other modules.

The best example of differing rule sets I can think of is Rule the Waves 2. You as the player builds ships to a slightly differing set of rules than the AI. The result is the AI can have designs before you would expect them or designs you can't replicate at all. Such as more guns or larger torpedoes, even adding AA or build aircraft. If you want to see it happening you can use the auto design in the ship builder or even better, let the AI build your legacy fleet. Easy example is CL classes. Use the auto to build you one and then attempt to copy it manually. Most of the time it'll spit out an error.

 

[FlyingDolphino]

The AI build rules is certainly something to take into account. Though I would rather have the AI on a different build rules that means that they push out ships similar to how they do now (mind you i have seem some questionable design choices, but hey this can be improved with testing etc) Whereas the player should have way more freedom with what they can do. Freedom in design adds a lot more playability to the game, as it allows you to experiment with more interesting designs, or even just allow for more variations in ship designs. I would like to see stuff like edible hulls, more placement points for guns etc just allow for more variations in ship design. Because as it stands imo the editor is not very flexible at all currently.

More designing freedom may me its harder for new players to learn, however what if the editor cautions the player regarding questionable design choices, eg if the ship has a lat/lon offset it could alert the player *Warning, your ship has a high lon/lat offset, which will lead to penalities to accuracy* and then the player can decide to proceed regardless. Alerts like these could be shown for various things, such as low ship speed for class, low armour for class etc. This will allow freedom to the player, but the game would also advice the player regarding various aspects of the ship which should also help the learning process.

What do you guys think?

[roachbeef]

11 hours ago, Ruan said:

think you misunderstood what I meant.

If you go back to the comment that sparked this discussion:

On 5/21/2020 at 6:01 PM, madham82 said:

there will be things players can figure out how to place but teaching the AI to do it is another story. 

So yes, we're in agreement that there's no need for the AI and player to use different "rule sets." However, practically speaking, the only thing we care about with regards to limiting factors in the game is the algorithm the AI uses to generate its designs. We all know why it must be limiting: turn times and load times can grow exponentially unless we put in restrictions for the AI, or forgo self-generation entirely and make it use some sort of local or cloud-based bank of designs. But we're letting the AI's stupidity to limit player freedom when we can simply place more restrictions via the AI's algorithms.

 

Moving back to my argument on ship design, I'm going to repeat it every now and then until the devs commit to fixing the issue:

Quote

• Complex restrictions  ==   less freedom.
• Simple restrictions     ==   more freedom.

Currently, the devs are implementing a bunch of unintuitive and complex rules on placement, stifling player creativity. That's terrible game design if you're looking for player choice. There should be simple rules that allow player creativity to shine. No more limited mounting points, no more fixed armor schemes, no more arbitrary pre-baked decisions on which mounts can be placed where. Rules should be based purely on real-life factors like structural integrity, funding, and model collision, not an arbitrary decision made by the devs for the player.

The cost of giving player freedom is that it would cost an exponentially higher amount of time for the AI to build something that would be challenging to face in combat. That's why their choices need to be limited to keep turn times and loading times to an acceptable level. Maybe they'll download designs from a cloud of highly upvoted previous user designs, tiered according to the known specifications of the users' ships. Maybe the devs can make some prebuilt—preferably historical—designs that the AI can pick and choose from. The easier option for now is to limit what the AI can build. 

Keeping the current system is out of the question.

• transverse bulkheads
• customizable compartmentalization of engine spaces
• separate magazine protection
• edit placement, length, and height of belt armor
• real-time placement restriction based on model collision and 

You have turrets exploding from hits to the extended belt in AoN schemes because the devs are refusing to give players the ability to edit the height or length of belt armor. You have magazine detonations and destroyed engines way too often because the game completely removed key aspects of ship design by removing critical mechanics such as: You have designs like the County Class impossible to replicate because the idea of forgoing belt armor and creating smaller citadels to cover only the magazines is impossible.

[DougToss]

I'm not arguing against freedom of choice generally, only that without player education of some sort, players will design ships that don't work, won't know why they don't work and become frustrated and quit or come to the forum demanding 10" secondary guns because 3" guns aren't cutting it as far as they are concerned. Again, I think ships should be allowed to have 10" secondary guns, but for innumerable reasons they would be pretty much useless. In RTW 2 I will occasionally put on casemated 10"s just for my own amusement, they hit less than 1% of the time at battle range. If people don't know why those guns are hitting so rarely, they're going to assume that something is wrong with the game, not their design and they won't learn from  the experience. That's because ship design is not intuitive! Most people don't know about naval warfare, and much of what they think they know is wrong! There needs to be some sort of player education influencing their design choices.

Free choice without education or understanding = bad design

Now, I know that some people learn from mistakes so:

repeated trials of bad design = improved design 

Maybe even better understanding of theory, but likely limited to practice, which for most players is good enough. The problem with this is that there is a ceiling. Experiential learning alone will likely lead to players building the best iterations of a conceptually flawed ship, since they are working from their own intuition and observation without questioning the underlying assumptions or being presented with the existing body of knowledge. 

You may be able to design the best possible 3" armoured cruiser, but as long as you still think an armoured cruiser should mount 3" guns, your design can only ever be so good. 

Learning from mistakes requires knowing what mistakes were made, and why they were mistakes,  which means on some level gaining theoretical knowledge. Without theory to inform the analysis of results and determine pathways to improvement, there is no guarantee that players will learn the correct lessons. 

In the above example, determining that armoured cruisers should mount a medium-large calibre armament. That's not intuitive, it took professional naval architects, naval theorists, staff officers and naval officers decades to figure this out. Players are not just going to work it out on their own. 

persistent bad design = bad gameplay experience

Nobody likes being blown out of the water repeatedly without understanding why, and naval design and tactics are not intuitive and not likely to be prior knowledge for many players. When people find themselves repeatedly banging their head against the wall, and there's no guidance they will probably assume something is wrong with the game rather than their own reasoning and design process. People don't know what they don't know, and most people do not know how ships were designed or fought. 

Theoretical knowledge + repeated trails + understanding the practical lessons of those trials through the theoretical knowledge = good design

X-Plane and Kerbal Space Program allow essentially full freedom, but they also work much harder to train the player than this title does. This is done primarily in game terms through well-designed tutorials, excellent UI with tooltips and an encyclopedia. It is also done through huge amounts of player training and documentation so that players understand how and why aircraft and spacecraft are designed the way they are. 

Basically, how do you give players the theoretical knowledge through education, so that they are learning the right practical lessons from experience  in designing ships?

Edited May 25, 2020 by DougToss

[Latur Husky]

I think Dev's should remove that obsolete block function. It makes no sense that game forces you to use modern equipment/propellants/fuel etc... because old one is too old...

Edited May 26, 2020 by Latur Husky

[Xenol]

To everyone saying that allowing unlimited freedom would lead (somehow) to a bad gameplay experience, I invite you to play Kerbal Space Program (Which this game initially advertised itself as being like). You have infinite scope to be stupid, but that hasn't stopped it becoming immensely popular.

You have to think where the money is. Of the two target audiences for this game (amateur naval historian nerds vs people who have a general interest in battleships) the latter is a much larger group. By limiting the players to 'reason' you are alienating the main audience, wheras allowing unlimited scope is beneficial to all parties. 

[DougToss]

Here is the RTW rationale for limits:

"Ships designed will have their ship type determined automatically by the program. This is to prevent unrealistic ship designs and also because the ship types are used by the tactical AI, and a 30000 ton destroyer with 15 in guns will tend to confuse the AI. Very unusual designs will be disallowed by the program as illegal ship types."

"Most development in RTW represents technical progress, but there are also a number of developments that reflect the conservatism of naval establishments or simply the time needed to grasp the utility of new concepts before they are commonly adopted. They are also in the game to prevent the player from using too much hindsight in ship designs. For example, there was no technical reason that battleships with three or more main turrets couldn’t be built before 1904, indeed some such ships were built. But if the game had allowed the building of such ships from start, players would in all likelihood start building dreadnought- like ships immediately"

"After you have designed your ship you will get a report on the design and any problems it might have.  Items labelled ‘Error’ in the report must be fixed, otherwise the design is not legal. Items labelled ’Note’ in the report are just reminders or hints that can be disregarded."

"Ship designs should have their weights kept within the displacement limit. You are allowed to build ships that are slightly overweight, but that will carry a penalty in stability and flotation, and is usually not cost effective compared to increasing the size of the ship. Its main use is if you have a displacement limitation, for example by a treaty or in dock size."

"Appendix 1:  Ship type definitions used by the game 
These are the definitions used by the game for defining ship types. Anything falling outside these parameters is considered an illegal ship type. The reason these type definitions must be enforced by the program is that the AI relies heavily on ship type for decision making. It also prevents players from building unrealistic ship designs. 
These are the main definitions, but there are some exceptions in special cases. 

Destroyer (DD) 
Displacement less than 2000 and speed more than 19 knots. Must have torpedoes. It cannot be armoured. 

Light Cruiser (CL) 
Displacement larger than 2000 and less than 8000. Speed must be more than 16 knots and  main gun calibre cannot be larger than 6 inches, unless it is a protected cruiser in which case it can have guns in single mountings up to 8 inches. 

Predreadnought Battleship (B) 
Displacement must be at least 5000 and it must have belt armour at least 6 inches. Main gun calibre must be larger than 6 inches and speed less than 20 knots. It cannot have more than 2 main turrets.  

Armoured Cruiser (CA) 
Displacement must be more than 4000 and speed greater than 19 knots. It must have more than 2 inches of belt armour but no more than 12 inches. Main guns must be at least 6 inches calibre and cannot be more than 11 inches. 

Battlecruiser (BC) 
Must have main gun calibre larger than 10 inches and speed more than 23 knots, or three main gun turrets and speed more than 21 knots. In some borderline cases armour thickness can be the difference between a BC 
and a BB. Speed requirement rises with time, as fast battleships develop. 

Dreadnought battleship (BB) 
Must have at least 3 main gun turrets, displacement over 8000 and belt armour of more than 6 inches. Main guns must be more than 10 inches. 

Minesweeper (MS)  
(this definition actually includes all kinds of minor combatants without torpedoes). 
Displacement less than 1800 and belt armour 2 inches or less. Gun calibre can be maximum of 6 inches. Cannot have torpedoes. 

Armed Merchant Cruiser (AMC) 
Displacement more than 1700, no armour and gun calibre no more than 6 inches. "

The manual is well worth a read, most, if not all of the concepts carry over to UA:D, and the explanation of doctrine driving design on pages 21-22 was particularly good. I think that those points illistrate the crux of the problem.

As for audicence, I am not a marketeer, but War Thunder and World of Warships already exists for the mainstream, and I don't know how they could stick with this title for very long in competition. It's just not going to give them what they want out of gameplay in the long run, especially in a campaign.  In terms of crossover appeal, following any sim or wargame community will show you quite quickly that people with a general interest often do become enthusiasts. The IL2:BOS community is booming because of War Thunder, but War Thunder is not getting players from IL2:BOS.  It's about drawing them in through intuitive UI and explaining how the game and systems work, lowering your standards to suit what you perceive to be their level is a real lack of faith in your audience.

Where enthusiasts will support titles they like for decades, casual gamers move on fairly quickly, either when the spectacle wears off or when the next thing that catches their eye comes around. 

I just made a point about KSP in another thread. To reitorate it here, KSP is a fun sandbox, but it really got the details right and didn't dumb anything down. You can't achieve orbit by pointing a giant rocket straight up. KSP expands the player's horizons through gameplay and encourages them to build the skills and knowledge to engage with deep systems. More than that, look at community engagement and you will see that the community is full of people who put tremendous amounts of work into the complex and granular systems where the "lol meme rocket go boom" crowd move on pretty quickly. You can argue that the sales of the later subsidized the deep simulation built for the former, but there is no doubt about who is sticking with the title and producing mods, workshop content and youtube content to draw new players in and extend the engagement of the community. Even more specifically, no matter how you play KSP the ruleset is the same, and it is really, really complicated! 

You can spend the whole game not understanding the heat system, but it's there. You don't have to know what conduction, convection and radiation are to mess around in the sandbox, but everything you do in the sandbox interacts with those systems and if you take the time to develop the knowledge, you get more out of it.