[DACLadder]

 
So DACLadder, just to confirm - the 1 kHz square waves we're seeing, they're good/acceptable even with the ripple we're seeing?

In the editor I see the same ripple in the 1kHz downloaded file as I observe on the output of the M7 when the file is played back(below).   So the M7 seems to faithfully reproduce what you input.  The mild ripple is due to the limited bandwidth at 48kHz sampling rate when the file is created.   Tomorrow morning I will generate some test signals and put them on SkyDrive for you to use.
 
 

 

[aive]

In the editor I see the same ripple in the 1kHz downloaded file as I observe on the output of the M7 when the file is played back(below).   So the M7 seems to faithfully reproduce what you input.  The mild ripple is due to the limited bandwidth at 48kHz sampling rate when the file is created.   Tomorrow morning I will generate some test signals and put them on SkyDrive for you to use.

You're a champion DACLadder. Note I also created files with 192khz sample rate with little difference.

 

[DACLadder]

You're a champion DACLadder. Note I also created files with 192khz sample rate with little difference.

Here're some test files...  Yes, you learn something new everyday.  I see some mild bandpass ripple with 1Khz sq wave at 44kHz.   Looks almost perfect at 192khz. 
 
Edit:  I almost forgot a warning.  The SQ wave files will blow your tweeters, headphones, and ear drums.  Turn off your amplifiers before playback.  
 
Download test files here....  Find the Public -> HeadFi directory.   Right click on the files for download options
https://onedrive.live.com/redir?resid=636DC4E8060D66A1%21172
 
Master 7 1kHz 0dB 44kHz-24

 
 
Master 7 1kHz 0dB 192kHz-24

 

[aive]

  Download test files here....  Find the Public -> HeadFi directory.   Right click on the files for download options
https://onedrive.live.com/redir?resid=636DC4E8060D66A1%21172
 
Master 7 1kHz 0dB 192kHz-24

 
Thanks mate, I'll give the 192 kHz one a try and see if I get similar results. Last time I just checked casually, and it appeared the same although some sections didn't have ripple, didn't do a one-shot trigger though, will try this time.

 

[sandab]

 
My apologies to everyone.   The web site where I downloaded the squarewave files is performing prefiltering based upon the selected sampling rate.  Therefore the waveforms do not look like squarewaves. 

Ouch.  Easily happens.  The same easily happens when generating white or pink noise, although the opposite: the noise can't contain frequencies > nyquist/2 or it won't have an even spectrum.
 
Here's a python script I created on OSX to generate square-wave WAV files.  (I'd then XLD convert them to ALAC and import to my "Test Data" album in iTunes.)  I'm pretty sure it's straightforward enough to tweak to your ends.
http://www.rockgarden.net/download/square.py
 
Here's one for (properly low-passed) white gaussian noise.
http://www.rockgarden.net/download/noise.py
 
These should run on Linux as well, but for Windows you'd need to install Cygwin or some other Python port.
 
Edit: I see the square wave is pass filtered... but that would be easily removed.  This is nothing fancy, just one-off simple scripts to generate test data.

 

[Josh83]

I have an Audio GD NOS1704, but I'm interested in upgrading to the Master 7.
 
Can anyone tell me what the default jumper settings for the Master 7 are (PLL, Digital Filter Stopband Attenuation, Oversampling, and Dithering)? Along with that, can someone explain PLL, Attenuation, and Dithering to me in layman's terms?
 
Finally, do many of you run the Master 7 in NOS mode? If I wanted to do so, what would be the best settings for the other jumpers?
 
(And am I correct in thinking that, on the PLL, with the jumper off, that the DAC is in NOS mode and all the other jumpers are bypassed, too? Whereas setting the oversampling jumper to 1X gives you NOS mode, but also allows you to play with the other jumers' settings, too?)
 
Thanks, in advance, for the guidance.

 

[DACLadder]

 
Edit: I see the square wave is pass filtered... but that would be easily removed.  This is nothing fancy, just one-off simple scripts to generate test data.

The last 1 kHz SQW scope capture shows bandpass filtering by the M7.  The test signals I generated have no filtering, dithering, etc. - just straight data.  So the scope capture (above) is measuring the M7 output when fed the signal below.
 
Not complaining about the M7.  Looks very good to me at 44.1kHz sampling.
 

 
The files I downloaded online and used originally clearly filtered to fit the passband (below).  Therefore, not good to use when attempting to measure what the M7 is applying to the signals.   Very misleading in my earlier posts...
 

 

[Articnoise]

I really want to hear the CLONES Audio Sheva DAC and see how it will do against M7. It’s not an R2R dac….but nevertheless has anybody heard it?
 
It’s truly packed with features for its price, have received some pleasant review and on pictures it looks very good:
 
·         Es9018 chip

·         2 stereo analog outputs, one on XLR one on RCA

·         AES/EBU, S/PDIF, BNC, Toslink, I2S and USB

·         100 steps fine setting via rotary encoder (digital domain) control main output

·         Dot Matrix display for display of sampling rate, input source, audio level, filter type

·         Rotary encoder with push switch

·         DSD 64/128 and DXD (384 KHz 24 bit PCM) compatible

·         Ultra-Low Phase Noise Femto Master Clock

·         Power supply has separate torroidal transformers for Analog and Digital section

·         AES/EBU, COAX, OPT up to 192kHz

 

Price €1799

 

[cradon]

Are the driver and firmware updates available for Mac?

 

[aive]

Are the driver and firmware updates available for Mac?

No drivers required for Mac. And the firmware update tool is windows only.

 

[aljordan]

  I have an Audio GD NOS1704, but I'm interested in upgrading to the Master 7.
 
Can anyone tell me what the default jumper settings for the Master 7 are (PLL, Digital Filter Stopband Attenuation, Oversampling, and Dithering)? Along with that, can someone explain PLL, Attenuation, and Dithering to me in layman's terms?
 
Finally, do many of you run the Master 7 in NOS mode? If I wanted to do so, what would be the best settings for the other jumpers?
 
(And am I correct in thinking that, on the PLL, with the jumper off, that the DAC is in NOS mode and all the other jumpers are bypassed, too? Whereas setting the oversampling jumper to 1X gives you NOS mode, but also allows you to play with the other jumers' settings, too?)
 
Thanks, in advance, for the guidance.

 
Hi Josh83,
 
You can see this page for information about the DSP-1: http://www.audio-gd.com/Pro/dac/DSP1/DSP1ENspecs.htm
 
The defaults as shipped are all jumpers off except jumper 7, which leaves you with:
PLL - Active
Stopband Attenuation - 130 DB
8 times oversampling
Data Dithering - Active
Oversampling mode - on
 
You are incorrect in thinking that the PLL switch will lead to NOS mode.  The PLL is used to derive a system clock from a reference frequency.  When you turn the PLL off, you are simply using the incoming clock present with the data.  The purpose of the PLL is to even out short term speed fluctuations such as jitter and fast drift.  But nothing is perfect, so whether to keep the PLL on or off really depends on the quality of the incoming clocked data.  I don't have the PLL engaged on my Reference 7 and I've never had a problem, but I feed it with a very good (Berkeley Alpha) USB to SPDIF convertor. 
 
If you want to run in NOS mode, just set jumper 7 to off, then all other jumpers have no effect because you are basically bypassing the DSP-1.  I've tried it once, but did not care for the results, however I was running at 44.1 sampling rate at the time.  It might be interesting to upsample on the PC to something like 88.2 and then try a lower over sampling mode in the DAC.
 
Data dithering is intentionally applied form of noise used to randomize quantization error.  It can improve signal to noise ratio, resolution, and linearity.  I've never tried turning it off in my reference 7, except when it was off by default when setting the DAC to NOS mode.  It makes sense to always have it engaged.
 
Stopband attenuation is the steepness of the filter used to keep alias images out of the audible frequency band.  A steeper filter (higher number of DB) generally has different phase response and more ringing than a gentle filter.  Here you should just try the different settings and pick the one that sounds best to you.
 
Alan

 

[Josh83]

   
Hi Josh83,
 
You can see this page for information about the DSP-1: http://www.audio-gd.com/Pro/dac/DSP1/DSP1ENspecs.htm
 
The defaults as shipped are all jumpers off except jumper 7, which leaves you with:
PLL - Active
Stopband Attenuation - 130 DB
8 times oversampling
Data Dithering - Active
Oversampling mode - on
 
You are incorrect in thinking that the PLL switch will lead to NOS mode.  The PLL is used to derive a system clock from a reference frequency.  When you turn the PLL off, you are simply using the incoming clock present with the data.  The purpose of the PLL is to even out short term speed fluctuations such as jitter and fast drift.  But nothing is perfect, so whether to keep the PLL on or off really depends on the quality of the incoming clocked data.  I don't have the PLL engaged on my Reference 7 and I've never had a problem, but I feed it with a very good (Berkeley Alpha) USB to SPDIF convertor. 
 
If you want to run in NOS mode, just set jumper 7 to off, then all other jumpers have no effect because you are basically bypassing the DSP-1.  I've tried it once, but did not care for the results, however I was running at 44.1 sampling rate at the time.  It might be interesting to upsample on the PC to something like 88.2 and then try a lower over sampling mode in the DAC.
 
Data dithering is intentionally applied form of noise used to randomize quantization error.  It can improve signal to noise ratio, resolution, and linearity.  I've never tried turning it off in my reference 7, except when it was off by default when setting the DAC to NOS mode.  It makes sense to always have it engaged.
 
Stopband attenuation is the steepness of the filter used to keep alias images out of the audible frequency band.  A steeper filter (higher number of DB) generally has different phase response and more ringing than a gentle filter.  Here you should just try the different settings and pick the one that sounds best to you.
 
Alan

 
Alan,
 
Thank you! That was a thorough and easily digestible response. It cleared up almost everything.
 
My only other question is that it seems like there are two ways to do NOS, if I understand correctly. I can leave the 7th jumper off, putting it in NOS and bypassing everything else. Or, I can put jumpers 4 and 5 on, which also puts it into NOS, but would allow me to experiment with the attenuation and other settings, too. Is that correct?
 
Josh

 

[aljordan]

   
Alan,
 
Thank you! That was a thorough and easily digestible response. It cleared up almost everything.
 
My only other question is that it seems like there are two ways to do NOS, if I understand correctly. I can leave the 7th jumper off, putting it in NOS and bypassing everything else. Or, I can put jumpers 4 and 5 on, which also puts it into NOS, but would allow me to experiment with the attenuation and other settings, too. Is that correct?
 
Josh

 
Hi Josh,
 
Yes, you are correct.  Setting jumpers 4 and 5 is a better way of putting the DAC into NOS mode because you can still have the other filters engaged as desired.  Sorry, I missed that question in your earlier post.
 
Alan

 

[Anda]

  My only other question is that it seems like there are two ways to do NOS, if I understand correctly. I can leave the 7th jumper off, putting it in NOS and bypassing everything else. Or, I can put jumpers 4 and 5 on, which also puts it into NOS, but would allow me to experiment with the attenuation and other settings, too. Is that correct?

 
The jumper settings on the Master 7 seem slightly different from the DSP-1. According to the M7 manual you can only choose 8x, 4x and 1x oversampling, where the DSP-1 also has 2x.
Is this correct or is it an error in the M7 manual?

 

[ciphercomplete]

Since I am using I2S I think I will try turning off PLL and give it a go.  Unless I2S needs PLL.