Need for Turning Acceleration

[Poyraz]

The main idea here is to modify the behavior of ships during the turning maneuver, so that the turning would be simulated more realistic and intense like in real life, which would also bring much more variety and another dimension into the sailing, specifically in turning maneuvers.

 

Turning Acceleration vs Linear Acceleration

Currently ingame, the maximum turning speed of different ships vary according to the ships features and different sizes. The turn speed variance makes different ships feel like turning faster or slower. A Santisima is turning slower than a Cutter, which makes sense.

 

However, there is a problem with the current turning dynamics. Currently in game, when you initiate a turning with A or D keys, the ship waits like 1 second and then suddenly reaches its maximum turn rate. This is the same for all ships from Cutter to Santisima.

 

On the other hand in real life, there is another dimension than the speed itself. It is the acceleration. In analogy with linear acceleration, when an object starts turning, it builts its turning speed in a definite period of time according to the torque/moment applied on it. The similarity between linear and angular motion is also listed in below formulas.

 

Now, If we compare the current turning speed with linear speed of the ships, it would mean that ships would reach their maximum speed in 1 second after they set full sails. A Santisima reaching 10 knots and a Cutter reaching 12 knots in one second after pressing the W key, which results in unrealistic and arcadish behavior. In other words, when you set sails, it takes some time like 40-50 seconds to achieve the maximum speed depending on the acceleration of the ship. As in linear motion, it is similar in angular motion. So, when you begin your turn, it would take some time up until you reach your maximum turn rate.

 

 

Turning acceleration is what we are lacking right now. This is the exact reason why turning ships in Naval Action feels kind of arcadish.
 

Real Life
In real life, it would take some time as shown in the graph below to reach a constant turning rate( r ), which is phase 3. During the phase 1 and 2, the turn rate increases for a definite amount of time.
 
The d_h graph here is the rudder angle. The turning behavior currently ingame, represents actually the change of rudder angle. As you can see from the graph, you won't reach to your maximum turning rate, as soon as the rudder is turned to its maximum angle. Some period of time had to pass by, until the ship responses on the deflected rudder angle.

Comparison
Finally, I would like to add a comparison between the current turning behavior and the suggested one. As seen in animation below, the ship on right builds slowly its turn rate and after some time it reaches its maximum turn rate and it begin to turn with a constant turn rate. On the other hand, the ship on the left begin to turn with constant turn rate slightly after the start of the turn, which is the current state in NA.
 
Turning acceleration feature if added, would add:

• much more realistic ship behavior while sailing and turning,
• much more variety to the ships, and each ship will have its own characteristics for turning. (Even similar ships with same maximum speed and maximum turn rate, one has higher turning acceleration than other. Even this would make the ship with higher turning acceleration more agile and nimble ==> much more variety)
• a gain/loss factor for all ships, for instance a ship of line would bring more gun power in return of slower turning, whereas, frigates and smaller ships would act more agile and nimble in return of less armor and guns. (Not everyone would rush for biggest and strongest ship, which is Santisima right now. Some could use the maneuvurability of the frigates, some could chose sheer firepower of rated ships => much more variety and less monotony like Santisima spamed PB's)
• Captains would need to plan their turns and moves in advance bringing more strategy and depth to sailing.

 

This suggestion is actually the last part of the feedback for sailing dynamics in this thread.
 
There were many other parameters mentioned in that thread, however, in my opinion turning acceleration is one of the most important concept which needs to be implemented to NA.

 

Edited June 4, 2016 by Poyraz
Removed unnecessary poll.

[Galileus]

• much more realistic ship behavior while sailing and turning,
• much more variety to the ships, and each ship will have its own characteristics for turning. (Even similar ships with same maximum speed and maximum turn rate, one has higher turning acceleration than other. Even this would make the ship with higher turning acceleration more agile and nimble ==> much more variety)
• a gain/loss factor for all ships, for instance a ship of line would bring more gun power in return of slower turning, whereas, frigates and smaller ships would act more agile and nimble in return of less armor and guns. (Not everyone would rush for biggest and strongest ship, which is Santisima right now. Some could use the maneuvurability of the frigates, some could chose sheer firepower of rated ships => much more variety and less monotony like Santisima spamed PB's)
• Captains would need to plan their turns and moves in advance bringing more strategy and depth to sailing.

 

• True. No arguing that.
• Disagree. Adding variety on top of variety (turn acceleration on top of max turn speed) suffers from diminishing returns and thus ends up barely noticeable. Ships keep their "nible" or "bricky" status, as is now.
• Disagree again. As above - the change wouldn't be big enough to change the current system. There are already more and less nimble vessels, claiming varying their turn acceleration will somehow completely change meta is ignorant.
• Disagree again. With the speed of gameplay now, the need to plan ahead is already obvious. Again, diminishing returns - adding another layer on top of existing layer will not hugely impact the amount of strategizing required now.

 

And due to all that I raise a question - are there really any improvements brought by this system that would excuse further slowing down of the gameplay?

 

I raise another one - how does that "time to reach steady phase" translate into time-warped nature of NA? Would it even be noticeable within sped up time?

[Hethwill]

While this is something that needs revieweing I would first wait to test the Wind system changes.

 

And yes some ships have some abnormal attitudes on the turns.

[Banana_Joe]

Great post, would love this in the game. This would add a lot more skill. 

[Ned Loe]

I agree, turning feels flat on all ships. I hope devs will review this post. Only one problem here, stern camping when ship slowed down will be very easy. This will make rage boarding popular. 

[Pagan Pete]

Pagan Pete can't argue with Physics. +3 votes!

[CaptVonGunn]

Manual Sails are your friend

[Poyraz]

• True. No arguing that.
• Disagree. Adding variety on top of variety (turn acceleration on top of max turn speed) suffers from diminishing returns and thus ends up barely noticeable. Ships keep their "nible" or "bricky" status, as is now. 

 

It has been done before. Have you ever played PotBS?

 

All the time, it has been linear and angular acceleration, which were giving the characteristic and unique sailing profile to every single ship, not the linear or angular speed limits.

 

If you make games with speed as control parameter and input, you will end up in arcade games.

 

If you make games with acceleration (aka force) as control parameter input, you will end up in simulation games which are closer to real life pyhsics.

 

• Disagree again. As above - the change wouldn't be big enough to change the current system. There are already more and less nimble vessels, claiming varying their turn acceleration will somehow completely change meta is ignorant.

 

I am not just making some random number crunching. As I said, it was done before, compare trying to maneuver a SOL in PotBS and you will understand what I mean. In NA, rudder combined with manual sails, the SOL's are maneuvering like a charm at the moment.

 

However, I know it is a matter of taste. Some people might like the concept of a truck accelerating as fast as a Formula1 car.

 

On the contrary, I find it quite unreasonable and unrealistic.

 

And for the meta concept of the game, soon everyone will be sailing Victories and Santisimas (somehow already) and currently there is no incentive to sail other ships than those first rates, but the fear of loosing the only one durability. Turning acceleration is one of the concepts which can create the dilemma of firepower-maneuverability for more variation.

 

And due to all that I raise a question - are there really any improvements brought by this system that would excuse further slowing down of the gameplay?

 

I raise another one - how does that "time to reach steady phase" translate into time-warped nature of NA? Would it even be noticeable within sped up time?

 

Variety and diversity!

 

If we read the turning graph in OP, we can see that the time up until the rudder fully deflected is much more shorter than the time until steady phase. We can assume it takes 10-15x more time than turning the rudder completely.

 

Here we have already made a basic concept. Currently, turning the rudder takes approximately 1-2 seconds in game. This leads to the time until steady phase might be between 10 and 30 seconds. However, this is unique for the ship described in this graph. Every single ship would have its own turning graph in other words turning characteristics.

[PL_Harpoon]

1st. I don't think this would slow gameplay at all. You'd just learn to start the turn sooner, that's all.

2nd. It would add a lot of plausibility to how the ships move.

3rd. Since the turn rates aren't that big anyway, I don't think it would change meta in any way, unless the differences between ships would be huge, which I doubt.

 

In other words, I don't see any reason why this shouldn't be applied other than the fact that it would put a lot more work on dev's hands.

 

But even using one universal turn acceleration model for all ships would be better that the current system.

[admin]

on accel the changes are obvious and they will make ships better

 

the main question that we are dont have an answer is this

slow turning vs fast turning on frigates and heavy vessels - how did it work in reality and what was for example the difference in turning radius and turning speed at 5 knots vs 10 knots vs 15 knots

 

we tried everything to find how it worked then (or how it works now on large sailing vessels) but to no avail

[Banana_Joe]

Even if there isn't any data a simplified version tied to the overall mass of the ship would be an improvement.

[Poyraz]

on accel the changes are obvious and they will make ships better

 

the main question that we are dont have an answer is this

slow turning vs fast turning on frigates and heavy vessels - how did it work in reality and what was for example the difference in turning radius and turning speed at 5 knots vs 10 knots vs 15 knots

 

we tried everything to find how it worked then (or how it works now on large sailing vessels) but to no avail

 

You are right, there are so many parameters that even nowadays turning behaviors are defined after real sea trials. (Unfortunately no chance to sail Santisima or Victory today, but their replicas might give some insight) For instance, the rudder is only a trigger and has the duty to initiate the turning in a turning maneuver. After that the hull at bow below waterline acts as a huge shovel itself and rotates the ship's huge mass. The rudder ratio and hull shape are just two of many parameters in turning which are also mentioned here.

 

Although the simulation is complex, one might say that the correlation between ship speed and turning could be defined as a second degree quadratic function. That is because the forces acting on rudder and ship's hull are changing depending on square of speed, V².

 

For example:

5 knots => 1x turn rate

10 knots => 4x turn rate

15 knots => 9x turn rate

[Galileus]

Here we have already made a basic concept. Currently, turning the rudder takes approximately 1-2 seconds in game. This leads to the time until steady phase might be between 10 and 30 seconds. However, this is unique for the ship described in this graph. Every single ship would have its own turning graph in other words turning characteristics.

 

... and there goes my ignorance. That long? That would actually leave an impact and diversify smaller ships from bigger ones a lot.

 

I withdraw my concerns.

 

Out of curiosity, because that part of physics is not my strong suit - how about turning deceleration? How does that work? Would you be forced to leave the rudder in neutral or even turn it to opposite side to realign faster? What would be - in your knowledge or guess - the time to realign from steady phase in a corvette and in a sol with and without countering with rudder?

[Fuzzyhead12]

... and there goes my ignorance. That long? That would actually leave an impact and diversify smaller ships from bigger ones a lot.

 

I withdraw my concerns.

 

Out of curiosity, because that part of physics is not my strong suit - how about turning deceleration? How does that work? Would you be forced to leave the rudder in neutral or even turn it to opposite side to realign faster? What would be - in your knowledge or guess - the time to realign from steady phase in a corvette and in a sol with and without countering with rudder?

 

here comes the mass, and shape of the hull (flat, kite round), the friction resistance  in the turn, so yes you can "overturn" to a bit, however with mauel sails you would be able to counter that since you have keel on oyur ship that reduces the overtunr effect, or maybe neutralises it, as op said the hull under the waterline acts like a shovel while the rudder is just a trigger to make the ship roll in its axis, so most of the turning is made by the shape of the hull and not the rudder. As soon the rudder is on 0 position the ship rolls back and the keel will neutrelise or work against the overturning

 

What you need to understand is, everytime i try to get a object into a diffrent state of movement i need to acclerate it, if i try to turn a moving object i need to acclerate it, if i out it back in a linal movement i need to acclerate it, these forces are actual simple math and can be easily calculated, by the current engine, For the game you dont even need certain parameters to create realistic turns, cause then you still put the used parameters on all ships and they would still have diffrent relistic turnrates

[maturin]

on accel the changes are obvious and they will make ships better

the main question that we are dont have an answer is this

slow turning vs fast turning on frigates and heavy vessels - how did it work in reality and what was for example the difference in turning radius and turning speed at 5 knots vs 10 knots vs 15 knots

we tried everything to find how it worked then (or how it works now on large sailing vessels) but to no avail

High speed = fast turn, large radius

Low speed = slow turn, small radius

With manual sails, anyways. Slow speed rudder-only turns should have large turning radius as well.

Ryan confirmed this via PM once upon a time.

And of course some ships weren't trimmed right and like to ignore the helm. Many French frigates loved to tack but hated to wear.

[Hethwill]

If i'm not mistaken some ships had way more lift than others resulting in less water working against the pressure put by the hull on a turn. But I think we have that.

 

Would the high resistance point from the water move along the ship as it turns instead of being always constant at a position ? Considering wing relative to sails of course.

[Poyraz]

High speed = fast turn, large radius

Low speed = slow turn, small radius

With manual sails, anyways. Slow speed rudder-only turns should have large turning radius as well.

Ryan confirmed this via PM once upon a time.

And of course some ships weren't trimmed right and like to ignore the helm. Many French frigates loved to tack but hated to wear.

 

Condisering the hydrodynamics (hull underwater) and ignoring the aerodynamisc (sails);

 

If the interaction between water and ship hull would be constant, the turning radius would be same both for slow and high speed levels. At slower speed the ship would cover the same route but in longer times, which means slower turn rates. (degree/second). At higher speeds the ship would cover the same circle in shorter time, which would lead to higher turn rates. That is because turn rates describes the covered angle in a definite amount of time.

 

However, in reality it is much more complex. While the speed gets higher, the interaction between hull and waterflow changes between laminar and turbulent flow range. This effecting the drag at hull and rudder, the turning circles might differ at maximum speed.

 

To define the changing of the turning circles with the current model, the concept drift has to be introduced as well followed by turning around pivot point.

 

At this point, I would assume going for constant turning circles together with varying turning speeds is the easy and lean way to go.

 

PS: Some sea trial results could be found here. The turning radius is approximately same for all speed but maximum speed.

[Poyraz]

on accel the changes are obvious and they will make ships better

 

the main question that we are dont have an answer is this

slow turning vs fast turning on frigates and heavy vessels - how did it work in reality and what was for example the difference in turning radius and turning speed at 5 knots vs 10 knots vs 15 knots

 

we tried everything to find how it worked then (or how it works now on large sailing vessels) but to no avail

 

Turn Rate:

Describing the turning maneuver is quite complex, including also drift in real life and turning around pivot point instead of weight or geometrical cenre. Considering, ships in NA would perform a regular and circular turn maneuver, the turn rate would be defined by the linear speed and characteristic turn radius as in the following formula.

 

ω = V / R

 

The linear speed part V, varries according the ships maximum speed, set sail amount, hull resisstance, etc.

 

The turning radius R is the distance gained during the turning maneuver of the ship. However, it is mostly unique for every ship's hull design. Some of the factors affecting the turn radius are:

 

1. Structural design and length of the vessel

2. Draught and trim of vessel

3. Propulsion power of sails

4. Distribution and stowage of cargo

5. Even keel or carrying a list

6. Position of turning in relation to the available depth of water

7. Amount of rudder angle required to complete the turn

8. External forces affecting the drift angle

 

The only case turning radius R differs from original value is close to the maximum speeds, due to the interaction between hull and water.

 

According to those basic rules, there are two approach for implementing the variation of turn rate of a ship for different situations.

 

A. Constant turn radius:

With a constant turn radius, turn rate only depends on the speed. Slower speeds slower turn rate, higher speeds higher turn rate. Turn radius stays the same for the given ship. To implement the below formula can be used, where speed V is variable and turn radius R is constant.

 

ω (V)  = V / R

 

 

B. Turn radius depending on speed:

Like in real life turn rate depends both on speed and turn radius, which also changes according to speed. High speeds means higher turn radius, slow speeds means shorter turn radius. This approach is little bit complex, but on the other hand it might add the realism and depth in turning maneuvers. For this approach both speed and turn radius are variables in the formula to define the turn rate.

 

ω (V, R)  = V / R

 

The speed variable depends on the given ship, its maximum speed, its set sail percentage, hull resistance, ships polars (wind direction), etc, most of which are already implemented ingame.

 

The turn radius variable depends on if the ship is sailing close to maximum speed or not. To define this change from the data here with a regression analysis, I would propose the following formula to define the turn radius over speed to get a curve as following. However, custom values can be added to modify a curve like this one, too.

 

R (V)  =  ( -V² / 125) + (V / 4)

 

 

 

 

This simulates the turning radius at max speed larger and at slower speeds smaller. The turn radius and also turn rate can be shifted up at 0 speed to allow turning at land crashes, fully demasted situations, stucked at shallow positions and for other gameplay issues.

 

With a varrying turn radius depending on speed, captains can adjust their turn circles according to the speed of their ships. It is more realistic, and also indirectly simulating the drift at speeds closer to the maximum speed. It would allow much more depth compared to the constant turn radius.

 

If the turn rates would be according to the constant turn radius, the ship would sail on the same circle undepended from its speed. The speed would just change the time it takes to complete the maneuver.

 

 

 

Conclusion:

 

Finally, the change of turn rate over time should be similar to the graph below including acceleration and deceleration like in linear movement. I would presume a time between 2 to 8 seconds to gain maximum turn rate from Cutter to Santisima would be a good point to starts. This time is depending on the turn acceleration of the specific ship.

 

Turning acceleration combined with linear acceleration would describe how agile a ship would be. Maximum speed at specific polars and turn rate are another criteria for defining the ship's limit speeds. Those two combined together with armor and gunpower values would allow to differentiate every type of ship with its advantages and disadvantages.

[Horror]

on accel the changes are obvious and they will make ships better

 

the main question that we are dont have an answer is this

slow turning vs fast turning on frigates and heavy vessels - how did it work in reality and what was for example the difference in turning radius and turning speed at 5 knots vs 10 knots vs 15 knots

 

we tried everything to find how it worked then (or how it works now on large sailing vessels) but to no avail

 

These differences would be small compared to the model offered by the OP.

Fantastic suggestion and should not be difficult to implement.

[Horror]

  .

Fantastic suggestion and should not be difficult to implement.

Taking this back.

The AI would need to be retrained and, worse yet, EVERYONE will gripe about not being able to instaturn to get their cannon on target.

[Hethwill]

So, and from that Poyraz exposion ( thanks! ), more speed provokes more pressure, water being water provokes immediate resistance so the turn radius is prolonged to win against that pressure. Lower speeds, less pressure less distance needed. Not sure about the timings if they do the 180º at the same time in reality.

I do not think is correct is the circle motion and giving wind direction and reverse pull. Should be more egg'ish ?

Drift (water or air )might be hard as hell to code in. The new IL-2 has it for airflow and sometimes results in odd behaviours.

[Poyraz]

So, and from that Poyraz exposion ( thanks! ), more speed provokes more pressure, water being water provokes immediate resistance so the turn radius is prolonged to win against that pressure. Lower speeds, less pressure less distance needed. Not sure about the timings if they do the 180º at the same time in reality.

I do not think is correct is the circle motion and giving wind direction and reverse pull. Should be more egg'ish ?

Drift (water or air )might be hard as hell to code in. The new IL-2 has it for airflow and sometimes results in odd behaviours.

 

For the sake of simplicity, I would go with a circle

 

If you would go for more realistic turn route, I would say it would look like more like a spiral shape instead of an ellipse, due to the variying speeds.

[Poyraz]

Good to see some tweaking after so many years...

https://store.steampowered.com/news/app/311310/view/3005581668786510905

Edited June 30, 2022 by Poyraz