• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

miniDSP 2x4 HD Measurements

peniku8

Senior Member
Forum Donor
Joined
Dec 9, 2020
Messages
384
Likes
771
Here is a full set of measurements of the miniDsp 2x4 HD that I purchased a few years ago. The version I have is the old revision.

NQAgbJM.jpeg


I tested the device as ADC, DAC (both in unbalanced and in balanced operation) and as standalone unit with dsp on and dsp off. Digital connection was over USB.


Unbalanced DAC:

THD+N: -96dB


YQ7NOZB.jpeg


Output 1 looks a bit worse than output 2. When only one output was fed, THD+N improved to about -99dB. When all 4 were active, THD+N worsened to -92dB.


THD vs level:

g1omJpx.jpeg



Dynamic Range: ~100dB

6YrSatk.jpeg



Frequency response:

VG5m8c4.jpeg



THD vs frequency:


dZTYJOp.jpeg



IMD:

enmqwht.jpeg




Unbalanced ADC:

THD+N: -84dB

rZ2bo56.jpeg



THD vs level:


6CYomOz.jpeg



Dynamic Range: ~87dB

YFJWhUI.jpeg



Frequency Response:

CewmWYq.jpeg



THD vs Frequency:

9FpWoE5.jpeg



IMD:


hr2R6Jh.jpeg




You can combine two unbalanced outputs into a single differentially balanced one by inverting the phase on the 2nd output. With the correct adapter cables, the miniDsp 2x4 HD turns into a fully balanced 1 IN, 2 OUT device.
This should improve CMRR, but also reduce common mode noise coming from the device itself. Let's see if this is in fact the case...

Balanced DAC:

THD+N: -100dB

Dw5rUyz.jpeg


Wow, did this ever clean up the signal. THD+N improved by 4-5dB but the actual measurement looks a lot cleaner than before. Most of the random noise spikes are now gone!


THD vs level:


n4aoMj1.jpeg



Dynamic Range: 105dB

srg5q8e.jpeg



Frequency Response:
(same as unbalanced)

mcKC28W.jpeg



THD vs Frequency:

Mr3BBcq.jpeg



IMD:


e8UjXIx.jpeg



Balanced ADC:

THD+N: -96dB

wzTvxOL.jpeg


Wow! THD+N improved by 12dB compared to the unbalanced input.


THD vs level:


l2RWJ5K.jpeg



Dynamic Range: 105dB

gp5sTCy.jpeg



Frequency Response:
(same as unbalanced)

0Puhw4B.jpeg



THD vs Frequency:

iSpt0y8.jpeg



IMD:


DAblGqM.jpeg



Unbalanced Round trip ADA: 76dB THD+N

CqHrrq5.jpeg


Some users have commented that the performance of the unit degrades when you activate filers, so I went ahead and dialed in an input EQ as well as an output EQ.
I made sure that the overall gain at 1KHz stayed at unity and measured the channel with EQ active against the 2nd channel without EQ:

H25lJuh.jpeg


We can see that performance itself doesn't change. Most of the noise comes from the ADC, so you will increase that if you dial in positive EQ and decrease that by dialing in negative EQ.
This tells us that correct gain staging is particularly important for the input side of this unit, but the use of DSP itself does not affect the performance of the unit.


Conclusion

The dsp capabilities of the unit are nice and the new software looks a lot sleeker than the old one. When used in balanced mode, performance is quite good, but you would need a second unit if you're feeding it via RCA and you want to run it in stereo. EQ reduces headroom, which technically brings down the performance of the DAC in a theoretical sense, but since the ADC is the limiting factor here, that effect should be negligible.
If you want to use the device on your mains and you feed it via the RCA inputs, it's important to have gain staging on point, otherwise it is possible that the 2x4 "HD" audibly degrades performance. In balanced mode you get some extra leeway and I wouldn't worry much about it then. Still, I would've wished to see better noise performance overall. The unbalanced graphs look really dirty with noise spikes all over the place.
 

Sokel

Master Contributor
Joined
Sep 8, 2021
Messages
6,389
Likes
6,537
Great review!
Is it possible to measure the low's channel with the same filter on against off at say 80Hz (same way as the very last measurement but at low's)
Flex (before the update that fixed some things) showed significant elevated noise this way while the high's channels seem mostly unaffected (same as yours) .
 
OP
peniku8

peniku8

Senior Member
Forum Donor
Joined
Dec 9, 2020
Messages
384
Likes
771
Great review!
Is it possible to measure the low's channel with the same filter on against off at say 80Hz (same way as the very last measurement but at low's)
Flex (before the update that fixed some things) showed significant elevated noise this way while the high's channels seem mostly unaffected (same as yours) .
I can repeat the last test with a lower frequency stimulus, sure.
 

Sokel

Master Contributor
Joined
Sep 8, 2021
Messages
6,389
Likes
6,537
I can repeat the last test with a lower frequency stimulus, sure.
That will be great and we will be able to compare it against the digital measurements of this thread:


 
OP
peniku8

peniku8

Senior Member
Forum Donor
Joined
Dec 9, 2020
Messages
384
Likes
771
Software is Room EQ Wizard.

Hardware is most likely the E1DA Cosmos ADC/APU/Scaler.
That is correct! ADC+Scaler. EX5 as DAC with a low-level adapter that allows me to measure dynamic range of up to 140dB.
 

Tell

Active Member
Joined
Mar 30, 2024
Messages
162
Likes
201
Nice review, confirms why I'm still very happy with mine! ;)
But I had no idea that you could do that hack for running in balanced mode, cool stuff! Too bad you can only run two channels out while doing so.
 
Top Bottom