It's not the depth of the water that matters most, it's the SSP type. But, if the water is shallow, 90% of the time the mission designer will use the bottom limited SSP type. Once you're in BL SSP, there is a drastic difference between sonar performance based on bottom type; if rock it's almost as good as a surface duct, if mud or sand it's absolutely horrible. There is no middle ground.

Sea state does matter, but only by a mile or so. You could consider the SSP and bottom type the 'coarse adjustment' knob and the sea state the 'fine adjustment' knob.

Verified. Bottom type has major effects on sound travel in shallow water. I only did some simple tests. FFG, two surface targets at 5 nm and 10 nm. About 100ft depth.
Rock bottom gives nice signal, mud is worse, lets say double loss, half detection range.
Sand is disaster. Even FFG's own contact on TA deployed at 160ft gave just 2 lines ! 5nm target gave really weak 1 line, easy to miss. 10nm target was perfectly undetectable.
This calls for more testing, especially at what depth this 'shallow water mode' kicks in and if the transition is smooth or sharp.

If by shallow water mode you mean an area near the surface where sonar performance is degraded.... This is something I've been wondering about for some time but never got around to really looking into. Some variables I suspect make a difference are the depth of the contact (not just location along the SSP as you've already documented, but also if they are located in the surface chop zone) and the sea state.

If you meant something that happens in areas of shallow water, then this is not something that I've observed but is still interesting if it plays out.

So i did some more test. And it's simple. Bottom type affects detection ranges in shallow water and deep water by same amount. There is no difference.
Sand will give about 30% detection range compared to rock even with few km of water under the keel.

Also my experience shows target depth does not matter, unless it crosses the layer. Depth of listener may affect detection ranges, if the listener is near surface, depending on sea state. With sea state 1, the noisy surface layer is thick few feet.

Yet one thing which decreases detection range. Extended masts. They create turbulence or so .. the increase in noise is visible in BB waterfall displays.
So with sand bottom, sea state 3, PD and masts out, such detection range seems to be normal (but then you did not tell your conditions with this detail).

When sonars are sufficiently powerful that they can make bottom bounce detections or convergence zone detections then you need some kind of model of how sound travels in the ocean. In the second world war, from what I've read, they mostly only made direct path detections. If I recall correctly, the big advance of the SQS-26 over World War II sonars was that it could make bottom bounce and convergence zone detections.

If all one is worried about is direct path energy, though, then "cookie cutter" range laws are fine with detection.

Yes, we're probably saying the same thing here. My point is that even a "range only" model should be more realistic and not allow contacts to be heard from 15 nm away no matter the conditions. Also, the inclusion of surface noise wouldn't be bad either.

I like SH3 but I love DW
Nedless to say, but I always learn something new from you people and the excellent discussions in this forum.:up:

I could take it or leave it, honestly. In WWII sonar was so crude that it's not really clear how you would deal with much variability. Additionally, there was very little (if any) real attention to quieting. Everything was loud loud loud. That was an age where there were barely even industrial regulations on noise levels, let alone stealth considerations. You only really need to worry about the details of ambient noise or whatever else when the signal gets down near that level.

The Germans were the first to really start thinking about how to make their boats quieter. The innovations they sparked didn't really have an impact until after the war, though.

I can see that you have quite a strong opinion when it comes to people complaining about sensor accuracy or demanding more from the sim. Your arguments surely make sense and since I agree that having a tactical mind is the key to sims, I will leave it at that. :ping:

"In tactics, the most important thing is not whether you go left or right, but why you go left or right" A.M. Gray

Well... the thing is... I'm not sure that including more and more details for the sake of adding details is giving you a better simulation. I suppose for a certain type of techno weenie it is, but I'm not that person.

The thing about all simulations is that their instructiveness tends to hinge on whether or not they capture the essentials of a given phenomena. My experience in both professional wargaming and in the hobby is that there is too much emphasis on including every little detail. None the less, as computers have become more powerful, it seems like the tendency has been to include greater and greater detail simply because the computational capacity is there. The thing is, I've spent many many days tweaking the knobs on these sorts of simulations with every gorey detail in them, only to discover that they didn't matter anyhow, usually for a good reason. A simpler model usually would have sufficed, and provided more insight more quickly. Often the values associated with the knobs in question are subject to such great uncertainty that it's questionable whether one can quantify their impact anyhow.

What I do wish they would do in computer simulations is make them more transparent. I'd really like them to say in the manual, "here's how we decided to model our sonar," for example. From what I can tell in DW, for example, it really just generates a very simple family of transmission loss curves, but it took someone else's experimenting to figure it out. If they made it more transparent, it'd be easier to figure out exactly what the scope of the model was, so one could make scenarios that were more challenging, given what it represents. I don't really care what assumptions they make so long as I know what they are and I can work with them.

:yep:

The only thing about that SeaQueen is, and I'm sure you're aware, is that if they even got a small fraction of anything classified in their sonar model the US Navy and any other nation they provided software for will have them for lunch.

This has been my stance eversince Sub Command and DW's early start. We don't know how much classified material they used for any of their modelling and can't make judgements as such.

DW sound simulation is deep in public area of knowledge. At the moment I could write the same simulation myself based of my measurements (although there may be details I missed, like recently discovered bottom type influence). I mean I could do that in a weekend, it really is simple. My sound propagation utility, based on public information (Ulric's book mostly) has all DW has (minus bottom) and much more.
Simply said .. sound propagation is public, including methods of simulation. Well maybe they have some super-simulation somewhere used for sonar design and so .. but DW is not in this class, by a big margin.
Classified things may be actual noise levels of ships, and sensor sensitivity, lets say 'parameters of actual military warfare' .. but that can be seen from the game itself.

Btw. don't get me wrong. Since 1.04 I think DW sonar modeling is more or less appropriate, considering how many receivers it can simulate in real-time (sonobuoy fields I mean).

Hm .. I thought about it .. why would I not make my sources public .. here are some reasons:

- I can't. Some contract does not allow that. For example I buy some module, I extend it, so it is crucial part of my new product. But I usually can't make public what I bought somewhere, even if I modify it.
Edit: also often when you do your project for someone (the navy), then use the same sources for other project (game), you are usually not allowed to make it public.

- I don't want to. For example I still use that code for my other projects, which I sell.
Also exposing your source codes is like exposing your soul. Also most source codes looks terrible and are far from 'how source code should look like', even in projects which are open source from beginning.

- I don't care. Making sources public .. why ? Will it make my name better ? Better than 'army contractor since 1973' ? Will it bring us more money, more projects ?
Isn't making naval simulation source codes public even in conflict with security policy of the state ? Should we risk it ?


Now all that could apply with SCS. Adding new playables would be little different. It could be trouble, if they plan to make DW 2. But if they don't plan (as I think), there is no reason for that anymore. They could allow that. Only difference would be that those mods, which exists I think, would be posted here, and would meet more audience.