It's true what you say, Andreas, about multicore not necessarily improving the performance of single-task execution, and that there's only a gain if an app is tuned (threaded) specifically for multicore. The other thing that strikes me is, hard as multicore programming is in the "regular" computer and DSP space(s), it'll be something fairly new to embedded developers, so they'll have a steep(er) learning curve to come up.
Some other disadvantage not mention quite often is, that developers have sometime only access to software for single-core processor development and that such developed programs finally run in the worst cast much slower on the multi-core platform. Multi-core systems also do not improve automatically the performance of single task execution, so discussing about multi-core platforms, developers should also have a deep understanding of multithreading and multitasking. For example only the question if you want to run multiple threads on different cores can cause a lot of work and problems.
Programming multicore is hard. Traditional IDEs are merely graphical front ends for compilers. This is no longer practical with multicore. The graphical interface must serve to minimize the nuts-and-bolts grind of low-level programming. TI has the right idea with Grace. Cypress also has this partially implemented with the PSoC IDE.
Ideally one should be able to allocate resources, activate peripherals, and set up pinouts by moving around the mouse. Only when, for instance, you want to do a running average, you might actually write some code.
Multicore MCUs have impressive capabilities, which vendors obviously want to highlight. Often unsaid is the fact that multicore parts are more expensive than their single-core cousins. From the users' perspective, it's all about selecting the right part for the job. If multicore capability is required, great. But if not, it's more cost-effective to go with a less powerful part.
Festo's BionicKangaroo combines pneumatic and electrical drive technology, plus very precise controls and condition monitoring. Like a real kangaroo, the BionicKangaroo robot harvests the kinetic energy of each takeoff and immediately uses it to power the next jump.
Design News and Digi-Key presents: Creating & Testing Your First RTOS Application Using MQX, a crash course that will look at defining a project, selecting a target processor, blocking code, defining tasks, completing code, and debugging.
Focus on Fundamentals consists of 45-minute on-line classes that cover a host of technologies. You learn without leaving the comfort of your desk. All classes are taught by subject-matter experts and all are archived. So if you can't attend live, attend at your convenience.