Is airflow modelled? Sailing realism questions.

[tigger]

I just spend first few hours sailing around with the beginner ships and I have some questions.

 

Is the current sailing model the final one? If so, there are some things that seem pretty strange to me,

leading me to think that the existing sail model is perhaps overly simplistic. I might be wrong in my observations, but others likely see the same thing.

 

As far as I can see, the wind speed seems constant, it is not shadowed by large waves, islands, other boats, or other sails. Further,

I believe that wind speed and sea state are not correlated. I have never seen a sail flutter, when pointing into the wind, they are just flat. Further, they generate lift at very small angles of attack. It seems that I can beat almost 30 degrees into the wind on for-and-aft rigged vessels. (Unsure about fully rigged ones yet.) A simple explanation of why this is so seems to be lack of apparent wind: apparent wind direction traverses towards to bow with increased speed.

 

My belief that wind is modelled as perfectly laminar flow of infinite speed but with finite generated lift is further reinforced by the fact that booms and sails appear to have no weight to them - a vessel on starboard tack rolled significantly to starboard will always keep it's mizzen or main fully filled with boom extending to port side.

 

I guess these are the basic ones that bug me immediately. A reasonable airflow model should also fix a number of other ones.

Edited February 17, 2015 by tigger

[admin]

 

 

My belief that wind is modelled as perfectly laminar flow of infinite speed but with finite generated lift is further reinforced by the fact that booms and sails appear to have no weight to them - a vessel on starboard tack rolled significantly to starboard will always keep it's mizzen or main fully filled with boom extending to port side.

 

 

 

Pictures will others to understand what you are talking better (including us).

Sails luffing of flutter will not be implemented in the foreseeable future. Weather, fires and explosions and somewhat improved smoke first - then we will see if we have FPS reserves left.

[maturin]

 

Pictures will others to understand what you are talking better (including us).

I think he's under the impression that the sailing model is based on first-causes physics like the HMS Surprise simulator, rather than a hull just having inherent stats and sailing abilities with certain levels of thrust provided by various sails.

 

 

Tigger, I think you're asking too much of the simulation here. Wind doesn't exist at all as a moving force or object in the game. So far as I can tell it is really just a direction that determines how sails and flags animate, and how much force a given sail is allowed to impart to the hull.

 

For example, the HMS Surprise has a config-defined top speed of around 14 knots, and a main staysail can provide a maximum 11% of that speed if the sail is angled 90 degrees to the wind. This is true no matter how the sail might be blanketed, how the sail might be tilted from heel or a collision, whether there's an island in the way, etc. If you tilt the sail any closer or farther from the wind, it will start losing efficiency, until at 170 degrees from the wind it is at 0% power. If you shoot a hole in it, it loses power. And when you press S to douse sail, the aggregated power of the sails starts slowly decreasing 60%...59%...58%...57% until the speed reaches the percentage that is appropriate for that reduced sailplan. The sail isn't doing anything or interacting with the wind 'in reality.' The game just tracks the angle between wind and canvas, and then tells the ship how fast it can go and quickly fast it can turn. There's no airflow anywhere.

 

In short, this is a videogame. It 'fakes' everything because otherwise 95% of development time would be spent tweaking the physics models, and there would only be time to add one single ship. Basically, it would have to be like DCS: Surprise. But a tall ship simulator is much more complex from a development standpoint than a flight sim. Planes only have two wings and an engine which never changes. A ships has about forty different wings and engines for different conditions. And to bring up the example of HMS Surprise simulator, even that sim's exhaustive physics modelling results in some rather strange outcomes. Even it can't get the intricacies of tacking right, for examples, or balance all the limitations.

 

The game has to mimic real-life behavior with arbitrary stats, shortcuts and smoke and mirrors. This is necessary and correct. Naval Action can reach a very high level of verisimilitude if balanced attentively and imaginatively.

 

I certainly hope that the wind will become more of a dynamic, changeable force with local variations. But this would not be accomplished with any airflow modelling. Rather, you would just have an island, and an arbitrary, randomized area to leeward of that island would have a different coefficient for wind force. There won't be any detailed flow modelling. That would be prohibitively complex for the coder and taxing for the server. 

 

 

 

So could you reiterate your specific feature requests? How should the game's vessels act differently?

 

Certainly the gaff rigs are too effective upwind. In fact, they can sail at 6 knots a mere 15 degrees from the wind. But fixing this doesn't require complex physics. I imagine there's just a spreadsheet with a performance curve defined somewhere in the config. Simply define the efficiency of gaff sails to be 0% at 40 degrees to the wind, and you're done. Same result achieved.