I am nowhere near the level yet where I am thinking about implementing this, but I am still curious. In almost all demos I see from game developers, water along a shoreline is clean and unbroken, simply ending at the point that the terrain rises above water level, which is a very clear contrast to what you see in the real world.

Are there any good techniques people are starting to use to create nicer transitions between sea and land? A procedural approach would be of particular interest because it would be universally applicable without having to be hand-crafted. Scenarios would include serene in-and-out lapping of small waves against the beach, right through to violent crashing of waves against rocky shore lines. Slight rising and lowering of the water level near the shore would also be important in order to simulate the waves coming in and out effectively.

This is something I'm very interested in implementing in my own development when I'm further along; hopefully next year at this rate. In the mean time, I'm eager to collect examples, tutorials and papers concerning other people's efforts in this area.

Check out the Unigine Tropics demo, and go to free camera mode. http://unigine.com/products/tropics/

Myself I have ideas that I intend to prototype at some point in the future. In rough terms, I would calculate local wave solutions near the shoreline, characterized by functions that approximate locally the curvature and distance from the shoreline. For LOD, waves would have a variable granularity based on the distance from a reference point (e.g. the main camera position). For rendering, the local wave solutions would be splatted on offscreen accumulation buffers encoding the final waves displacement, gradients, foam level, variance etc. Physics would be handled entirely on the CPU side.

I'd be happy to go to into more details if there's enough interest, and also share ideas about other approaches.

Although I'm far from qualified to say anything definitive on the subject, I sort of have an idea that the topic could be approached using interpolation techniques, much like real-time PBR lighting can be calculated at low resolution and then interpolated for a high quality result. My idea is a simple one; basically you'd use a small set of control points, dynamically simulate the movement of those particles around the shape of the shore, whether sandy or rocky, and heavily affected by whether the sea is calm or choppy. By using control points and control splines, mesh deformation adhering to the control data and particles whose movement is dictated by the same, combined with existing techniques for making realistic oceans and water surfaces, I can't help but think there is a decent solution there that would have acceptable performance. Something I'll look into when I'm a bit further along, I guess.

Axefrog, your idea is brilliant. It shares similarities with the approach I hinted at in my previous post. Both approaches involve estimating the wave path near the shoreline at control points and then generating the final water surface by splatting a wave shape along the curve defined by the points, warped in amplitude and possibly other features based on the local bathymetry. One design question concerns the organization of control points. Some possibilites are a regular grid, a hierarchical structure or free particles. They all have pros and cons depending on the exact tech requirements. And do not forget physics, that also has big implications on the design choices.
For a prototype of the rendering code, I'd start by simulating waves approaching a circular island, for which the solution is known a priori. I'd then move to the computation of local control points, which in my opinion is by far the most challenging aspect of this technology.

Ah, well now I feel validated, thanks :)

I will however point out that I'm far from ready to implement this kind of thing; as I said it's more like something I'll be ready to have a crack at maybe next year, as I have a lot of self-education and groundwork to do before then. This is good though, it's all part of my ongoing process to collect as much information as possible in order to refine my general development road map as I move forward. If you end up implementing something along these lines in your Typhoon engine though, I'd love to see what you come up with.