Behavior in Waves

Welcome, Guest. Please login or register.
Did you miss your activation email?

[mporter]

This is a response to the video Pjotr just posted of a supply vessel in waves. (I can't see any way to respond in the FAQ section, which maybe makes sense).

This vessel is pitching (hobby-horsing) too much for the relatively light sea that is running. Marcstrat is dead right about that. 

It would be interesting to know how the relationship between sea and movement is derived.  If it is like the ship dynamics in SS2006 (ie depending on more-or-less arbitrary parameters) I fear we will be in for many of the same problems.

Cheers,
Michael Porter

[LucAtC]

Good evening Michael,
marcstraat is indeed quite right.
I tried to put numbers on the concern... once more 
The Fairmount Sherpa is some 75m long, and the wave length is more or less this same length. The wave period being around 7sec, the relative speed of tug vs wave is 20.8 kts.
I have the impression that the height of the waves is equivalent to the draught (between 6 and 6.6m?), what would amount to a 30 kts wind (having lasted enough over sufficient fetch?), consistent with such a steep sea. The speed of the tug over ground is then very low, what would also be consistent with the ventilating screws and the fact that the tug pitches at the same frequency than the passing waves.
Only problem, there is apparently almost no wind, the wave system is monochromatic, thus coming from very far?
Does such a swell not look much too steep for the locally prevailing wind conditions? Gone is the wind. (What does that remind me?) 
Dont you also think that the vertical movement of the tug lacks  "ploughing (?)". Is it the right word for the vertical accelerations of the CG?
If there will be a possibility to adapt the parameters of the ships, one can derive that it will also concern the movements in relation to the waves?

Regards,
Luc

[mporter]

Greetings Luc,

I almost thought I might provoke aresponse from you.
To me the waves appear very short in relation to their depth (steep, as you say), and this has to suggest very shallow water, but not quite shallow enough for breaking.

The English word you are looking for is "heave" which is the movement up and down of the vessel, and you are correct that itis missing here, which creates the effect of extreme pivoting about the CG (?) or what is called by the derogatory term "hobby-horsing" (although in the real world it is never seen to this extent).

The pitching acceleration is also not very realistic -- it would appear that the vessel has almost no mass, and this is consistent with other parameters of other vessels, with accelerations impossible for their stated displacement, etc.

Sigh.

Best regards,
Michael

[AriesDW]

What is missing is the lack of intertial forces, which would also mean there is a lack of mass considerations. This problem is evident in SS2006. Take the cruise ship for example. The real life equivalent of that vessel weights about 90k tons. When you view the vessel in ss2006, it heaves slightly in water as calm as those in SS06. I always found this to be very disturbing. A vessel of this size might rock or heave, but extremely lightly - so much it would be practically un-noticable.

In the video for SS08, yes, the ship does pitch and roll in a manner that indicates the waves are moving a hollow structure - which thereby causes the appearance of a lack of mass and interial forces on the vessel. The calculations are missing that consider vessel mass and realstic intertial forces. Should these calculations be entered, or improved, I believe the interaction between the "water" and the vessel would behave differently.

A perfect example of how a vessel should behave in rough waves, is expressed in the video in the link below. You will find that the wave, vessels, and vessel type are all very similar to the video of SS08.

http://www.ulsteingroup.com/kunder/ulstein/cms.nsf/pages/shipdesign.htm?open&qnfl=flash#orcaninaction/film.itm

[LucAtC]

There is also an interesting post about the question from CapFlyer in http://www.shipsim.com/ShipSimForum/index.php/topic,156.0.html
The most intriguing fact, more than the movements of the ship, is this short and steep swell.
I viewed and reviewed many times the 25 first seconds.
There is a total lack of ( :-[aaarrrggghhh "path" or "furrow" or "trough" or "wake"?) water movement around the tug. A ship moving through water is supposed to move water, and in such a swell, it ought to be impressive.
And by the way,  in cauda venenum, the weight of the Ocean Star cannot be 90k tons...
 
Regards
Luc

[AriesDW]

in cauda venenum, the weight of the Ocean Star cannot be 90k tons...
 
Regards
Luc

Why not? Her real life equivalents are . . . 85,900 tons. Not fully laiden. Therefore, I am close.

Nonetheless, the vessel should still not bob, or heave in such waters.


Also, yes, that is something that bothered me in the video as well, there was no "breakwater" or wake at the bow of or surrounding the vessel as she went through the wave - and yes, they should be impressive.
I went on a cruise aboard Carnival Pride, the real life equivalent of the Ocean Star, and we were in 12 waves and she made breakwater at the bow ranging from 12-30+ feet, pending on time of impact in the wave cycle.

[elangeland]

Tonnage of ships is confusing, as there are many measures. Gross tonnage, the most commonly quoted number, is actually a measure of the ship's volume; 1 gross ton = 100 cubic feet.

The displacement (actual weight) of passenger vessels is much smaller than their gross tonnage, since they are mostly empty space inside. Unfortunately, I can't find any displacement figures for Carnival Pride. Anyone?

[Bottman]

Maybe it helps:

"Carnival Pride"´s measurements
GT 85920
NT 53613
tDW 7200
Cheers

[Captain|Austin]

Why not? Her real life equivalents are . . . 85,900 tons. Not fully laiden. Therefore, I am close.

Nonetheless, the vessel should still not bob, or heave in such waters.


Also, yes, that is something that bothered me in the video as well, there was no "breakwater" or wake at the bow of or surrounding the vessel as she went through the wave - and yes, they should be impressive.
I went on a cruise aboard Carnival Pride, the real life equivalent of the Ocean Star, and we were in 12 waves and she made breakwater at the bow ranging from 12-30+ feet, pending on time of impact in the wave cycle.

I went on a cruise on the Norwegian Star. And I noticed the same effects at the bow.

[AriesDW]

Maybe it helps:

"Carnival Pride"´s measurements
GT 85920
NT 53613
tDW 7200
Cheers

These technicalities are getting annoying. Is there something specific you are trying to fight about? I corrected my numbers, which is still in the ballpark vacinity.

[AriesDW]

I went on a cruise on the Norwegian Star. And I noticed the same effects at the bow.

You mean the breakwater intensities?

[mporter]

The displacement of any given ship doesn't matter -- what does matter is that they are not weightless and do not simply bob over the top of the water like a cork.

In addition, Luc is quite right -- these seas are very short for their height.  It seems to me that a sea this short would not show the trachoidal form of deep-water seas but would be cresting and breaking, so showing an asymetrical shape. 

When a ship, moving forward, runs into such a wave there is a significant impact, and that results in lots of flying spray and sometimes flying green water.

Cheers,
Michael

[LucAtC]

Hello AriesDW,
I owe you an explanation about the technicalities of weight, that are not irrelevant -although not quite fun.
Some lack of realism of the sim has to do with a "feeling" of incorrect masses and distribution of them, just like you wrote in your posts. For a player, it is of course meaningless, but for the developers to create realistic models, they need correct figures, you surely agree.
It is so that from the beginning, the data published about the ships are the main dimensions of the ships and what is called "weight" instead of "tonnage (volume)". Unluckily enough, these data are internally inconsistent when you read "weight" (or mass), and for professionals, such inaccuracies suggest a form of negligence or ignorance of basic facts. I had liked to see it being corrected, to no avail.
Take the case of the Ocean Star. The product of the main dimensions of the underwater box Loa x Width x Draft is 298.4 x 36.1 x 6.5 = 70019.56 expressed in m³. Surely the box is smaller, because I took the Loa instead of Lpp. The underwater part of the ship is a fraction of this box, called the block coefficient, probably some 0.63 for today's cruiseliners, and is designed by the modeler. At any rate, less than unity!
That is why the displacement (weight) of the Ocean Star lies probably around a maximum of 44000 tons (not taking the seawater into account, etc...).
Make the same calculations for the other ships, and you will be astonished of the results. This is quite basic and underlines the importance of masses, like you rightly told in your post.
Michael's guess (and mine's) is that such inconsistencies of the physics could explain some poor dynamics of the ships.
Voilà, I should have said from the beginning "the weight of the Ocean Star is far less than 90k tons..."?

Regards,
Luc

[AriesDW]

Well, displacement surely is different than the weight of the vessel itself. And I do see your calculations and I appreciate them, surely. However, if we are talking mass versus deadweight versus displacement, then we can really start getting technical. I should of been speaking in regards to displacement instead of tonnage - yes, I agree to that. However, the vessels deadweight also takes account into how it behaves in heavy sea conditions, as do it's displacement. This is because the weight of the vessel is that is interacting with the other forces of the ship, while the displacement is acting mostly in part with the water itself. So if the ship is falling from a peak into a trough, it is falling with intertial forces acting on the weight of the vessel while water is acting against the vessels displacement . . .

Am I wrong? We are getting very technical here - but who knows, we could be onto something.

Has anyone seen the newest video released by Pjotr on YouTube? The vessel appears to behave somewhat better, but sadly, no breakwater is calculated into the motions of the vessel . . .  Only the image map of a wake pattern.

[LucAtC]

Ok, my gobbeldygook about the weight of the ships aimed only to tell that tonnage was a measurement of volume, and that the displacement of a ship is its weight, usually expressed in tons (and wrongly so, it ought to be in N (Newton) or its multiples, but that is once more jargon of mechanical engineers).

But I think that I understand what you mean, and I agree with your idea, although not quite classical in its expression. You are indeed referring to moments of inertia and distribution of masses of a moving ship in relation to waves. The subject is vast, and as far as I know, is difficult to compute, because waves move the ship which in its turn modifies the wave, and so on.

The video is very encouraging, but indeed the mass seems (no, gives the impression) to be underestimated.
IMO, such a swell does not exist (or is so rare...), and that fact increases a feeling that something is wrong, more especially as the swell is undisturbed by the ship. It looks slightly like a light plastic model hobby-horsing (thx Michael) on artificial waves not far from the coast.

But I think to have understood that parameters can be adjusted, and of the ships, and of the weather, waves included, like in Virtual Sailor? It would make the simulation uneasier to "parametrize(?)" to get nice results, but on that question we have no clue today.

Sorry to be so confusing 
Regards
Luc

[mporter]

Well, to be precise, "weight" (or "deadweight") and displacement are the same thing.  They are often measured in tons (or tonnes, which are different  ) or pounds, but in reality they are forces, being the
effect of gravity acting on the mass of the ship.

"Tonnage"-- at least registry tonnage, which is how large ships are taxed, billed for harbor services, etc, is different altogether and is a rather arbitrary measurement of enclosed volume as modified by many factors.

In a significant seaway, the mass of the ship (its displacement divided by the gravitational constant 32 or 9.81 depending on whether one is in English or metric units) is acted on by gravity (as in still water), the varying buoyancy afforded by the locally rising (or falling) water as the ship moves into a sea, and by the impact of the bow of the ship running into (or falling down onto) a wave.  A precise description of the motions of a ship in a seaway would  be extremely complex and would probably only yield to an iterative numerical solution, given the extent of the interactions between the ship and the waves mentioned by Luc.  But a linear approach can give a simplified but still pretty good analysis. 

The bottom line is that for long waves (such as would normally be encountered at sea) the heave amplitude approaches the wave amplitude and the pitch amplitude approaches the maximum wave slope.

This is all rather boring, but I do wish we had some idea of how these interactions are modeled/executed/parameterized in Shipsim.

Cheers,
Michael

[LucAtC]

A realistic rendering of this kind of movement, ie a ship moving in a seaway, has not yet been done, as far as I know, in any game. This video on Youtube, of a "REAL shipsimulator" looks convincing http://www.youtube.com/watch?v=oooP-YKxwbo
but it is difficult to know if it is a fully computed 3D rendering on basis of a PC.
Silent Hunter 3 (or 4?) is also very good at it, but VStep surely aims still higher than that?
I dont think it has yet been realistically implemented at all anywhere, for a simple reason: Studies of parametric rolling are still underway, and they would not be needed if a true simulation existed.
Having looked closely to the movements of the yacht and of the tug in the seaway, my fear is that the periods of rolling/pitching/heaving, periodic in essence, use the periods of the waves and not the natural frequencies of the ship as a basis. That ruined the rendering in other simulations. Hopefully, VStep will succeed in this rendering for our greatest pleasure!
@AriesDW, to give credit to your Nonetheless, the vessel should still not bob, or heave in such waters., it is probably due to this same reason, the synchronisation of the movements of center of gravity and of the center of buoyancy.
It is not so easy to see, as the pictures are so tiny, as is sometimes the patience of the viewer?
And the sound of the video being what it is, I tried to listen simultaneously to the music of a harpist playing Greensleeves. I forgot her name? I very much liked the play (out of topic perhaps, but still periodic).

Regards
Luc

[AriesDW]

Very good point . .


I think one thing that would make this all seem more realistic is as follows. It is apparent to you and I that if the vessel reponds immediately to the movement of the water, as the vessels seem to do in SS, that the simulation is lacking realism. In the end, I think what we are all saying the game is missing is the feel of "intertial forces" acting on the ship. I am wondering if it would be possible in a heavy swell situation such as we are discussing, and to place the vessels pitching and rolling patterns on the identical pattern as the water, however, place this "plane" or "pattern" in a slight delay to the water map. In essense, find a way to program the vessel to move to the waves, as it does now, but of course put a delay on the motions. I think for the time being it would be a crude way to simulate the vessel acting with some intertial forces, but I think it might look better than it does now - it would also be a good temporary fix until more advanced programming where to be put into place . . .

No?

RE The Video - That is where I received the idea. You can see the vessel has the feel as if it is being acted upoin by SOME intertial forces - however, they are not entirely realistic in feel. I was wondering if they used a technique similar to what I mentioned above, hence the idea . . .

[LucAtC]

That is indeed what has been done in other simulations, lagging the movements of the ship vs the swell, which is of course partly correct. The external wave forces acting on the ship are the sum of the wave trains(?), with some varying pseudo-period. The reaction of the ship is lagging, of which each induced movement has its own period, independe(a?)nt from the wave frequency.
So, lag is normal, but dissimilar for heave ( ), pitching and rolling. Well, also yaw, transverse and longitudinal accelerations in a far lesser measure.
A great deal of linearity could also be camouflaged if the heavy sea would be made much more irregular?
Also, the question is liaised with the ship editor.
I dont mind if there is no possibility to incorporate locally "personal" ships, as long as there is a ship of each category. But the editor would have to enable the modifying of manoeuvrability (if the original model is not satisfactory), and of the movements. Preferably in a consistent way, because all parameters are interdependa(e?)nt.
Perhaps this question will be elucidated by the moderators/developers?

Regards
Luc

[AriesDW]

So to clarify, the lag I am speaking off only is most effective in terms of longitudial ( SP ) forces of the ship? yaw and roll would respond much more quickly due to it being a side acting motion, which has less area and mass to deal with? I am just trying to make sure I am following you correctly. I think we are on to something and I think you and I could continue to develope this further and then present it to VStep - possibly - as outside input.


So going with my lag suggestion, mostly regarding the pitching or heaving motion of the vessel, what do you think would be an easy solution for yaw or roll?

[AriesDW]

As for the irreegularity suggestion - yes . . . I think in the programming code it could be possible to plug in a string of code that would work with a variety of frequency calculations or of course it could also be programmed to generate a slightly varying frequency for wave timing and swell size if max and min parameters were provided, no?

[mporter]

It seems to me there are several problems here that there is no easy fix for.  The first is that I know of no simulator, even the high-end professional ones, that show ships form the outside (so to speak).  This is really a difficult problem, and it's not as simple as a bunch of parameters and a string of code.  Completely realistic action in waves would be something to be worked out on a case-by-case basis because there simply isn't any formula for it.

That said, it would be really positive if Vstep could factor in a mass parameter in their movement equations (Dave's"inertial" effect).

Roll is less of an issue -- generally speaking, the rolling of medium-sized vessels is much more dependent on external (wave) effects than on their own roll period.  The mass of the ship is naturally always there (constant, of course  ), but the tilt of the "playing field" becomes the primary effector.

Again, the issue is very complicated and not at all susceptible to easy solutions. With some simplifications, an approximation of realistic movement might be achieved -- and I wish, not for the first time, that we knew how the Vstep physics engine works.

Cheers,
Michael

[AriesDW]

Michael;

   We have three constants for each vessel in this situation - bouancy (in which we must know the vessels weight and displacement), gravity, and the third of these is actually related to bouancy. It is apparent, however, that VStep has utilized a string of code that has generated these waves.

Knowing that a basic code is presently in use for the current swell intensity levels of SS2008 I wonder:

1. Is there a way to make the attributes of that line of code variate based within two extreme parameters (base and ceiling)?

2. Would that we be able to include in the calculations the vessels displacement as well as bouancy? Gravity I imagine is already assumed or entered into Ship Simulator. I know that VStep does program certain levels of intertia - we see this in the manner they program the roll levels of each ship as it turns at varying speeds and rudder application. This makes me ask if these algorhythms can be applied in this situation, calculating the other motions.

I think what everyone needs to know as that we are not asking for perfection, however, we are asking for a higher sense of realism both in an aesthetic point of view but also in terms of gameplay.

I know something that would definately help in a generic and slight manner would be in delaying the motions of the vessel as it travels through rough water.

[LucAtC]

Hello Dave,
Like you say, it is the difference between the apparent weight (ie mass x actual vertical acceleration)  and the buoyant forces that results in a vertical acceleration and thus movement. Some people suffer from this acceleration...

Varying the height and frequency of the swell, according to arbitrary laws, is usually quite simple. I am not at all sure that the rendering would be as easy: Most other renderings in "seagames" show this too great uniformity of waves at a distance, there must be somewhere a difficulty?
Also, the apparent turmoil of the sea comes two main sources.
The first is the mixing of many wavetrains, differing in direction, frequency and height. Long waves, of longer periods, run quicker than short ones, beating  them and beating with them. Hm.
The second is the wind acting locally on these waves, changing their forms, pushing the tops, ignoring the troughs.
Strong winds of course also make spray, what reduces the visibility (from the point of view of a sailor, not from a pilot).
That also means that more than one swell is needed, and I think it is so in Vstep's code.

Your second question supports the doubts and hopes at the start of the thread. In fact, we dont know if inertia as such is integrated at all in the simulation. I think so, because I am an optimist, but there are many reasons to think otherwise. Surprising movements of the ships, flying ships, memory effects  , bizarre rudders, unrealistic water resistance, etc.. are not consistent with a code based on parameters of a ship.
Certainly, Quest3D incorporates all the dynamics needed, with the exception of hydrodynamic ones. These are probably tailor made (my tailor is rich), but a smooth integration of hydrodynamic laws in a rigid world is not necessarily easy (my tailor is not rich) or quick. And patently, it is improving!

Strange are some mistakes that could not have been made by people acquainted with shipbuilding or shipping, inland or sea (think of rudder angle, steering à la Z-drive,..). But there is enough knowledge and competent people - be it testers- around VStep and R'dam that it could be corrected.
Worrying is that the fundamentals of hydrodynamics are apparently missing, but it could be that SS2008 will change it. Knowing how the relation is made between waves and ship would answer the question, but more importantly would facilitate the suggestions early enough.
It is not commonly known that hydrodynamics of ships is made extremely complex by the water surface...

All that is of course somewhat late, think of  the "kick effect" introduced by a patch, and its integration. Fine tuning is one thing, but if you miss important parts? If the developers had missed all these points, I think it doesnt matter now. More probably, they are now fighting with their not well adapted environment, with the tools and debugging,... and if they needed help, they would have asked for it (they already did it!).

The question still remains open.. What do you think of it?

Regards
Luc

[AriesDW]

What do I think of what, exactly?