• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). There are daily reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Marantz SACD 30n Measurements (SACD Player, DAC & Streamer)

OP
VintageFlanker

VintageFlanker

Major Contributor
Forum Donor
Joined
Sep 20, 2018
Messages
3,002
Likes
9,380
Location
Paris
And then the resulting measurement
Thanks for your time, @Rja4000! :)

I tried to replicate your exact settings (same for tone generator but 2V, 2.5Vrms peak):
Marantz RCA 2V.png


Honestly, I still have no clue what's the deal with the rise post-3Khz. So I tested my Matrix Mini-i Pro 3 (measured by Amir here), with its fixed balanced output (4.1Vrms):
Matrix XLR 4V.png


Got weird numbers and the same rise. I'm not convinced either by the overall shape of the harmonic signature.

I also replicated your parameters with ARTA (for the Matrix, The Marantz is going back):
Capture d’écran (18).png


Back to RMAA, I re-measured the Matrix with both RMAA and Cosmos in mono-mode, with THD+N test set to 0dBFS (not my usual testing, the latter running in Stereo mode, and THD set to stock -3dB):
thd.png
Right​
THD, %
0.00014​
THD + Noise, %
0.00029​
THD + Noise (A-weighted), %
0.00030​

Oddly, the result seems much closer to what Amir get:
Matrix Audio Mini-i Pro 3 Measurements USB DAC.png


Bottom line is that the Marantz SACD seems to have a similar (but slight) "Hump" between -40dB and -25dB to what we've seen with ESS or other home-made DAC.
More investigation on this may be usefull, in my opinion...
ADI-2 Pro FS into Cosmos ADC seems to produce this hump for me, as well:
Here is the Matrix :
ff978bb0-15fe-4213-a7c0-b3cd8ce2aea1.png


(Cosmos in Stereo Mode) Same hump issue, as you said, so probably on the Cosmos. Besides, Amir measured this:

index.php
 
Last edited:

Rja4000

Major Contributor
Forum Donor
Joined
May 31, 2019
Messages
1,474
Likes
1,666
Location
Liège, Belgium
Honestly, I still have no clue what's the deal with the rise post-3Khz. So I tested my Matrix Mini-i Pro 3 (measured by Amir here), with its fixed balanced output (4.1Vrms):
Matrix XLR 4V.png


Got weird numbers and the same rise. I'm not convinced either by the overall shape of the harmonic signature.
The noise is for sure not where it should be.
You're 85dB below signal.
You should be around -110dB.
Are you in 24 bits ?
Something is definitely wrong here.

Why is your ADC set to 384/32 ?
Try and set it to 44.1/24.
 
Last edited:

Rja4000

Major Contributor
Forum Donor
Joined
May 31, 2019
Messages
1,474
Likes
1,666
Location
Liège, Belgium
Back to RMAA, I re-measured the Matrix with both RMAA and Cosmos in mono-mode, with THD+N test set to 0dBFS (not my usual testing, the latter running in Stereo mode, and THD set to stock -3dB):
thd.png

THD + Noise, %
0.00029​
THD + Noise (A-weighted), %
0.00030​

Oddly, the result seems much closer to what Amir get

Indeed, those figures are much more in line with what we should expect.
Noise is 25dB lower than in your measurements with REW (or with Arta).

I don't know what's wrong with your REW settings.

Maybe you should uncheck this AES17-2015 standard notch filter (I don't have that setting on my version) ?

Are you using ASIO4ALL ?
 
Last edited:
OP
VintageFlanker

VintageFlanker

Major Contributor
Forum Donor
Joined
Sep 20, 2018
Messages
3,002
Likes
9,380
Location
Paris
Maybe you should uncheck this AES17-2015 standard notch filter (I don't have that setting on my version) ?
As for raw numbers, it performs on par with or without (However, it gets rid of sidebands for the Marantz). Which version do you use? Mine is 5.20.17.
Something is definitely wrong here.
On that we agree! :p
Why is your ADC set to 384/32 ?
It's not. This the way Windows sees the device: "E1DA Cosmos ADC PCM 32/384". It is indeed set to 24b/44.1Khz (Both Cosmos and DUT).;)
Are you using ASIO4ALL?
I think that where the issue is coming from. The results I posted earlier where running on Java. I also tried FlexASIO (1.8) and I finally managed to get rid of the HF rise. Yet, results are several dB better, but still capped at 95dB THD+N or so, Noise being, again, the bottleneck. Trying ASIO4ALL (2.15) made REW very unstable and at the end, crashing. Could buffer size be the issue? Needless to say, I can't use the standard DAC ASIO driver, since REW sees it for outputs only... Will have to spend more time to investigate on these...

Very frustrating. I'm getting mad.o_O
 
OP
VintageFlanker

VintageFlanker

Major Contributor
Forum Donor
Joined
Sep 20, 2018
Messages
3,002
Likes
9,380
Location
Paris
Reboot is often the solution :)
As usual with computer topics, once you'll figure out, it should work and remain stable.
Finally, I figured it out!o_O It was a conflict in REW between FlexASIO, ASIO4ALL and drivers from manufacturers. I uninstalled all and tried one after the other.

At last, I got a proper FFT for my Matrix (Distortion profile is a waaay closer match to both AP and RMAA):
REW Matrix Asio Working.png


Sadly, the SACD 30n has been sent back already. As requested tho, I will include REW THD+N graphs in my next reviews, but will also keep the usual RMAA bunch of tests.
 
Top Bottom