Buffers for ADC design with AK5394A and other ADC's ?

[HpW]

Did you mean ESS? ARDA still only lists the ADC and I don't know of anyone who has put it in a product. ESS has some solutions. Still no commercial products I know of with their ADC.

Well, see the measurements on Stereophile... what are not so promising 😀

Ayre Acoustics QA-9 USB A/D converter Measurements | Stereophile.com

Its possible to get 384 through the XMOS interface both ways, on paper. Can you get the Windows sound layer to work at those sample rates?

If the HW and HW driver allows this sample rate and the underlying CPU & HW interface has enough power ... no problemo

Currently as already mentioned I support any SR up to 1 MHz what is currently only an internal SW reduction using custom sample rates.

However 384 is old news. Version 2 of HDMI supports 1536 KHz sample rates and 32 channels. No one in the alliance has any idea why, however, but they hope the big numbers will win sales, somewhere.

May look at RME & MADI interface HW, where more than 100 channels are used using the ASIO sound card interface...

You may have to understand that the HW chain: ADC > PCM (SR/Bits) > HW driver determinate the over all solution.

OK, the SW application itself should calculate in the required bit's (64/80 floating as my SW deals with), SMP (symmetric multiprocessing to separate each channel calculation to each HW CPU) and high performance FFT Window functions, all this do my SW. Validate this using the evaluation version and read the documentation.

HpW

 

[HpW]

The levels of each tone (19KHz and 20KHz) are approx -8 dB. the 1KHz IM product measures -134 dB. This is using the LME49710's. Using the JRC5534's I get -126 dB for the IM product. In either case that is performance close to what can be had with the best of analog equipment.

How this compare using a sound card based DAC & using related SW. While your spectrum shows a bit a large phase noise (FFT Window limitation or HW issue) ... just my opinion.

HpW

 

[gentlevoice]

Hi ... just a quick reply ... the Arda ADC I assume could be interesting sound-wise, however, its -3 dB point (as of the datasheet) is a little above 100 kHz. My guess is that they've implemented a "soft" filter and therefore need more bandwidth to lower out of audio range noise.

Besides Ayre I have heard that the Merging Horus also uses it, also with DSD256 recording capability.

@HpW: Sounds very interesting what you are doing. Will check it out.

Cheers,

Jesper

 

[1audio]

I'm using Praxis and a Juli@ card for the digital interface. I know how to use it so I can concentrate on the analog side. Its way better than I can get from the onboard AK5385A ADC. I'll try it with the RME card tomorrow if I can get the controls sorted. It has the same ADC. The interfaces on the RME Hammerfall drive me nuts. I'll fire up yours in the next few days and see what I get.

The sources are analog ultra low distortion generators. The close in phase noise is not as low as digitally generated sources. They also drift. These are 500K point fft's to get the noise floor down. Since the generators are isolated through 600 Ohm resistors and have a pretty low internal output Z the actual IM in the source should be extremely low, mostly from the resistors and passive connections. -134 IM for the LME49710 is within its spec-I think. They are using the 7K/60 mix so it doesn't have the slew rate of the CCIF test I used but they spec .00005% and show as little as .00003%.(-135dB range).

 

[JensH]

Well, see the measurements on Stereophile... what are not so promising 😀

It certainly looks like the Arda ADC is not the one we need for this purpose! Some of the distortion may of course come from the "zero-feedback analog topology", but the ADC is specified at -105 dB THD+N, so not in the same league as AK5394A or ES9102.

It is a bit of a mystery how the reviewer gets to the conclusion "offers superb measured performance that correlates with its equally superb sound quality." 😕
But there can of course be many good reasons for positive reviews. 😀
And the start is promising, since it preserved absolute polarity!

 

[1audio]

pro/studio recording/play/post prod app.

-RM

But 1) HDMI is not particularly usable in those environments won't go far enough and doesn't support such thinks as external locking and even has latency problems since the audio is packetized into the video frames. And 2) 1536 KHz sampling is beyond anything anyone could conceivably make use of. Most commercial production is done at 48KHz, even today.

 

[HpW]

I'll try it with the RME card tomorrow if I can get the controls sorted. It has the same ADC. The interfaces on the RME Hammerfall drive me nuts. I'll fire up yours in the next few days and see what I get.

Use the ASIO interface and may also set:

- HW driver latency to the maximal possible buffer size
- Also you have to understand the HW mixer & matrix settings
- HpW-Works ASIO Input option: skip at least 3-4 ASIO buffer switches to have a steady source signal

Hp