This is almost impossible. There's always going to be latency keeping the devices in sync, and audio is very sensitive even to the lowest amount of latency.

I know this is an old answer, but it is essentially a non-answer. The first point with isSuperiorThan rather than isGreaterThan is moot, it will take less than 5 seconds for people to figure it out. The second point is also moot, the compareTo method can stay, there's no issue there.

If you can't articulate what's wrong with the code then there's nothing wrong with the code.

@Daneolog I've edited the answer.

@cher-nov it is quite simple to deduce why hardware devs won't do it, they simply don't have a market for this, so they won't do it. They already covered the bits of the market that exists (CUDA and co) and that's fine. OpenGL works on a much smaller scale and it has to be extensible due to the nature of open source software and the lack of proper manufacturer support. There really is no comparison of the economics involved in openGL and those of DirectX.

You are overthinking this. What you are dealing with here are known as constraints. These constraints should be integrated in your architecture decisions and should also be visible at low level. Don't try to engineer something that's supposed to be straightforward and simple to understand for anyone. Simply reference these constraints wherever necessary.

Why not immediately cache a new object after it has been created by the DB?

Agile is essentially a series of small waterfalls.

Unless you have a specific use case in mind, simply wrap the C structs into C++ classes and then add the C functions which take one such struct as parameter and make them methods inside the class. Functions that alloc memory become ctors, functions that free memory become destructors

Do note that the underlying connection usually has 2 unidirectional streams, the unified interface is just an abstraction.