A recent feature on SACD (Super Audio CD (SACD) - is it just a load of noise?) is causing a bit of a stir...
Ref: your article entitled, "Super Audio CD (SACD) - is it just a load of
The math used in this article is severely flawed. Redbook CD has 16 bits of resolution. These bits are absolute value - each of which can be either 1 or zero. Sample rate is 44.1 Khz. This means that each second of a CD has 44,100 x 16, which means that each second of a CD has 705,600 bits. This is a far cry from your claim of 2,900,000,000 bits per second. FYI, there is currently no optical medium that could hold such a huge amount of data; nor is there any system that could decode it.
By contrast, each second of an SACD has 2,882,400 bits. Note that this is substantially higher than the real CD numbers (vs. the false ones you posted). So, purely from a bit-capacity standpoint, SACD has much higher resolution than CD.
Hello: You may want to have a look at an another analysis of the DSD vs. PCM resolution issue you raised in the recent article, "Super Audio CD (SACD) - is it just a load of noise?"
While I will say that this alternative analysis appears more sensible to me at the moment, I don't have a strong like or dislike for SACD or any other format. I have vinyl records that I still play, plenty of CDs, a few SACDs, and a few DVDs.
The alternative analysis offered elsewhere is this...
Posted by Christine Tham (A) on June 03, 2005 at 19:04:24
In Reply to: SACD a red herring? posted by Moorghan on June 3, 2005 at 08:33:49:
take 1 PCM sample at 44.1kHz. this can take on 65536 possible values (2^16).
in this same time period, there are 64 DSD 1-bit values.
the number of possible permutations is 2^64 - a far higher number of possibilities than the 1 PCM sample.
the 2^64 possible "states" within the 1 time period can be decimated back into one PCM sample (at 44.1kHz) or two PCM samples (at 88.2kHz), but as you see, the process will be inherently lossy.
this is precisely what a sigma delta ADC converter does.
PS - the basic flaw in the article's calculation is that the author uses multiplication rather than exponentiation to calculate the number of possible states.
This is getting interesting now. I have to say that 'data combinations' was the wrong term to use. It isn't relevant - ask me if you don't see why. I should have said 'states' (and I didn't say 'bits'). The figures refer to the number of possible states the data can take within a one second window.
Clear now? Or perhaps you still disagree...
Come on the FREE COURSE TOUR