Hi BobF I'm just trying to clarify my thoughts on how near-perfect sound might be achieved from chips that may be less than perfect.
Let's say that a run of chips in production has a slight fault that produces unwanted noise at a particular frequency. Let's also say that the effect of this fault can be negated by applying an algorithm in the ASIC and copying the resultant firmware to the ASIC of every chip in that run. So far so good.
It has been said that such a fault is caused by very small (nanometre) departures from the required measurements. That's all very fine when this error in measurement is absolutely consistent, at least within a run of a mass produced quantity. But I do wonder about consistency and repeatability in a situation where tolerances are very precise and unforgiving.
In the worst case, where the error may be variable within a production run it seems it might require a unique algorithm per chip to apply a correction. This might only be determined by testing each chip to establish and apply the algorithm adjustment to that chip's ASIC. For a mass-produced volume of chips I can see the need for an industrial test process in order to handle the volumes.
What do you think Bob?
- Forums
- ASX - By Stock
- AKP
- Earth Mountain almost there
Earth Mountain almost there, page-41
-
- There are more pages in this discussion • 652 more messages in this thread...
You’re viewing a single post only. To view the entire thread just sign in or Join Now (FREE)
Featured News
Add AKP (ASX) to my watchlist
(20min delay)
|
|||||
Last
$6.20 |
Change
0.000(0.00%) |
Mkt cap ! $181.1M |
Open | High | Low | Value | Volume |
0.0¢ | 0.0¢ | 0.0¢ | $0 | 0 |
Featured News
AKP (ASX) Chart |
Day chart unavailable