Share
1,416 Posts.
lightbulb Created with Sketch. 254
clock Created with Sketch.
31/01/23
11:58
Share
Originally posted by BobF:
↑
If we read the comments on the 'noise problem' this is described as a 'high frequency component' that depends on the frequency of sound that is being reproduced. High frequency here suggests a background hiss accompanying certain played frequencies. "... During our extensive work with GEN-I chips we discovered that during music play an audible high frequency component (for lack of better word “noise”) was produced when reconstructing certain frequencies. The root cause of this phenomena was traced to inconsistencies / lack of uniformity in fabrication. Specifically, the theories of digital sound reconstruction dictate that when superimposing the time-delayed sound pulses generated by our “pixels“, variances in the anticipated pulse magnitude caused by inconsistencies in manufacturing, can manifest into the improper reconstruction of sound and in certain cases produce an unwanted sound that is audible. ..." I had assumed that the hiss present on the clean room videos was more from the fans in the background than from the chip itself, and I do not know if this 'noise problem' is the same thing as the audible hiss on those videos, or that most of it is still the fans etc. I suspect that most of the noise is from the fans, because we know that they were running and they are certainly noisy, just as we heard. In the case of the unpublished video I decided that what I was hearing was the hiss of the assorted lab equipment used to do the demo when the microphone gain was turned up high. When the sound clip stopped playing the hiss was still present on the video - until the microphone was turned off - at which point the hiss disappeared. So it could still be a mistake that posters on HC have been making by presuming that the background hiss on the videos is the actual 'noise problem' and that this has now been acknowledged. We will probably never know. But it may never matter. "... While industry methodologies involved in mass producing semiconductors [could be] anticipated to reduce if not eliminate such variances over time, nonetheless the company was compelled to explore other possible means of negating the problem. The innovation ultimately conceived involves expanding the electromechanical characteristics of the devices, [such] that when combined with a proprietary software-only approach [this] provide the capability of rendering any such noise inaudible . This solution which has been tried and tested on our prototyping platforms has proven to offer meaningful improvement to the clarity of sound even though it cannot be fully utilized without the electromechanical changes incorporated in the mass-production version of the chip . ..." In other words the problem and the causes of the problem have been acknowledged, explored, understood, and responded to with modifications. These solutions have been implemented, and on the prototyping systems these have been shown to work as required. These solutions cannot be fully evaluated until the production processes and systems and the physical electromechanical structure changes are implemented in those chips that are currently in manufacture, but there is a sound basis for confidence that it will. In the process of the problem solving required for this particular problem, new design knowledge has been accrued that will lead to additional future innovations that will further improve the performance of the chip. With each run of production and the testing that will be done on the chips coming from it, further tuning will be done. All of this adds to the 'secret sauce' that makes the idea into a reality that can be made in millions and sold to customers. The writer mentions the series of problems and issues that have also given the company its share of nightmares to date. This includes those problems that led to the serendipitous and quite unexpected effectiveness of the partial structure that has now given us a chip that is 8 times louder and cheaper to make. Problems are not something to be shunned when you are developing a critical technology. They lead to progress. But from a business case they are a delay and an un-budgeted cost burden. Like many others, it sounds like they are also now behind us.
Expand
If that was the case bob they already have some of the chips now and they have the fix as well so the product can be demonstrated right now and as you can see from the 4c they are simply unwilling to do that which to me means they are not telling the whole truth. Jury still well and truly out.