Re: Re: Re: No, it won't "bother" it.....
Without actually hearing what you hear, I can't possibly comment but as the first image occurs at 352.8kHz when 8x oversampling is applied, I doubt its an alias thats to blame.
An error in your construction of the D1 output stage would seem to be a more likely candidate.
dieringe said:
Without actually hearing what you hear, I can't possibly comment but as the first image occurs at 352.8kHz when 8x oversampling is applied, I doubt its an alias thats to blame.
An error in your construction of the D1 output stage would seem to be a more likely candidate.
Re: Re: Optimizing a CS8414 + DF1704 + 4x PCM1704 DAC?
I was referring to the case NOT using shifregisters.
The circuit by Hattori Hanzo will degrade the sound for sure and also contains too many parts.
Fabian said:
It will have exactly the same volume as when it is fed with 24bit data. You have to align the MSB of the 16bit data stream with the MSB of the DAC input. A working basic circuit can be found by clicking the link provided by Hattori Hanzo. Modestly said, I think the PCM1704 is a good alternative to TDA 16bit DACs.
I was referring to the case NOT using shifregisters.
The circuit by Hattori Hanzo will degrade the sound for sure and also contains too many parts.
Aliasing and Filtering
Hi Børge,
An analog Bessel low pass filter has linear phase or constant group delay in the pass band and is not hard to design.
The only problem is the Bessel does not have a sharp cut-off at the crossover frequency. This makes the case a difficult one if you go non-oversampling as you want to filter all high frequency products above 20kHz, yet want to preserve a flat response in the 20-20kHz band. In the very early days of the CD we had high order Butterworth brick-wall filters that inevitable had a lot of ringing on the pulse response. Common knowledge in DA-conversion is that the highest frequency of interest should be about 10 times lower than the sampling frequency in order to avoid aliasing distortion and filtering problems. In the case of NON-OS this simply cannot be accomplished as you would end up with a 4,4 kHz filtered audio signal. With 8x oversampling no problem. Unfortunately we cannot change the format of the redbook CD.
Yes and Jocko is right about the brick-wall filter at the recording end before the AD converter. Here obviously we can do nothing about it.
borges said:
Hi Børge,
An analog Bessel low pass filter has linear phase or constant group delay in the pass band and is not hard to design.
The only problem is the Bessel does not have a sharp cut-off at the crossover frequency. This makes the case a difficult one if you go non-oversampling as you want to filter all high frequency products above 20kHz, yet want to preserve a flat response in the 20-20kHz band. In the very early days of the CD we had high order Butterworth brick-wall filters that inevitable had a lot of ringing on the pulse response. Common knowledge in DA-conversion is that the highest frequency of interest should be about 10 times lower than the sampling frequency in order to avoid aliasing distortion and filtering problems. In the case of NON-OS this simply cannot be accomplished as you would end up with a 4,4 kHz filtered audio signal. With 8x oversampling no problem. Unfortunately we cannot change the format of the redbook CD.
Yes and Jocko is right about the brick-wall filter at the recording end before the AD converter. Here obviously we can do nothing about it.
Re: Re: Re: Re: Optimizing a CS8414 + DF1704 + 4x PCM1704 DAC?
they still don't get it
The PCM1704 has a clear digital side (pins 1 to 10) and an analog side (11 to 20) hence should be laid out 90 degrees rotated compared to what Levinson does
some things are so simple......
Peter Daniel said:
they still don't get it
The PCM1704 has a clear digital side (pins 1 to 10) and an analog side (11 to 20) hence should be laid out 90 degrees rotated compared to what Levinson does
some things are so simple......
Peter Daniel said:I wouldn't mind if you translate this, as I'm curious about the microscope technique myself
"Another highlight is the internal upsampling of the 360S. Madrigal
prefers to generate integer multiples of the received data rate
instead of "inventing" intermediate steps. The 44.1kHz of a cd for
instance are calculated to 352.8kHz, 48 or 96 to 386kHz. And as we
deal with a Mark Levison, maximum smoothnees in calculation and sonic
result are practically self-evident."
....
no you don't want to read the rest of it. the microscope is not really explained. only something like trimming the resistors to 0.0002% for over-all minimum distortion of the whole converter circuit
"the ernormous list of technical delicacies could be continued for pages..."
For me this is not an option!
which paper is this from?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.