I think one thing needs to be clear: "working" is too a vague term to be used at this stage concerning AP's MEMS chip.
We know that it plays sound, both individual tones and a frequency sweep.
How far short that is from playing "complex music" we do not know. My guess is that it is not far away from being able to do so. But playing tones means a hell of a lot. One has to assume that for it to be able to play tones at all means that every pixel can be involved as required to play that tone. For all audible tones (i.e. not ultrasonic output at the default frequency) and for doing a full frequency sweep has to mean the device is using the full addressing system to create all of the achievable frequencies. That means all of the pixels are addressable and that the addressing system (in conjunction with the ASIC logic) not only "works" with the MEMS but is able to switch the pixels on and off in the right sequence, at the right time, and for the right duration. That, in my book, is a significant milestone.
What we do not yet know is how the whole thing sounds, but we do know that it is not yet performing up to expectations. But seriously, can anyone really be surprised that there are still niggles with such a revolutionary system?
We don't know, but this being AP, we can certainly assume that the system will not be announced to be ready until they have all of those niggles sorted, and that should also not surprise us.
We have to bear in mind that the initial prototype MEMS system was able to play music: tinny scratchy music to be sure, but recognisably music! That in my mind is a strong argument that the refined, near-release system we are waiting on has to have been able to satisfy its creators that it was capable of exceeding that initial prototypes capabilities. They have seen and heard stuff from it that reinforced their confidence all this time.
However it is probably inevitable that we can expect only a deafening silence from AP until they have something unequivocal to announce. My guess is that they will have to give an update at the end of August, and that it will either be suitably unequivocal or it will downplay the problems as being all under control, until such time that they do have something unequivocal to announce.
What this will tell us, and this is a note of hope, is that they currently have no information that is compellingly price sensitive like "we now know that the thing does not, and cannot work - our design is missing some vital detail we did not know about when we started". Until they reach that conclusion we will not hear any bad news. The longer there is no admission to that effect, the longer and more drawn out the birth is going to be, but sooner or later we are going to hear the fat lady sing.
(20min delay)
