• 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 U-DIO8 USB to AES Converter Review

devteam

Member
Audio Company
Forum Donor
Joined
Sep 26, 2019
Messages
18
Likes
68
Hi there. Thanks for joining and responding. The test is very different. It is capturing jitter and analyzing, not the signal itself as you showed. I can share the project file with you.
Sounds good. Please send us the file to [email protected] and we'll take it from there.

We indeed used a different method, I guess similar method as per APx jitter documention and to what you're typically doing for all DAC I guess e.g. Jitter Noise & Spectrum (E.g. below). Not sure how to get that specific new measurement as we haven't seen it before.. :)

Mola Mola Tambaqui USB DAC Streamer Headphone Amplifier Jitter Audio Measurements.png
 

Trdat

Addicted to Fun and Learning
Forum Donor
Joined
Sep 6, 2019
Messages
967
Likes
396
Location
Yerevan "Sydney Born"
I am not having much luck with minidsp, it seem the products I have bought haven't measured well and I just received the UDIO8. Very disappointing, soon the revised version will come out and need to purchase it again.

In any cause I just want to confirm the conclusion. The jitter shouldn't be audible according to Amir, but in a high end system will it be the weak link? Or still not audible enough to be a problem?
 

g29

Addicted to Fun and Learning
Joined
May 1, 2019
Messages
520
Likes
318
I am not having much luck with minidsp, it seem the products I have bought haven't measured well and I just received the UDIO8. Very disappointing, soon the revised version will come out and need to purchase it again.

In any cause I just want to confirm the conclusion. The jitter shouldn't be audible according to Amir, but in a high end system will it be the weak link? Or still not audible enough to be a problem?

"... We see peak to peak jitter of about 800 picoseconds. From what I recall, 500 picoseconds is the worst you can have (for a sinusoidal) jitter to preserve 16 bits. ...". Since Redbook CD is 16-bits from the 1980's, one would think 16-bits would be the barebones threshold to meet.
 
Last edited:

pkane

Master Contributor
Forum Donor
Joined
Aug 18, 2017
Messages
5,632
Likes
10,206
Location
North-East
I am not having much luck with minidsp, it seem the products I have bought haven't measured well and I just received the UDIO8. Very disappointing, soon the revised version will come out and need to purchase it again.

In any cause I just want to confirm the conclusion. The jitter shouldn't be audible according to Amir, but in a high end system will it be the weak link? Or still not audible enough to be a problem?

Here's what 800ps sinusoidal jitter looks like when applied to a 12kHz tone. 500ps is just about reaching -96dBFS:

1579287997057.png
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,376
Likes
234,529
Location
Seattle Area
"... We see peak to peak jitter of about 800 picoseconds. From what I recall, 500 picoseconds is the worst you can have (for a sinusoidal) jitter to preserve 16 bits. ...". Since Redbook CD is 16-bits from the 1980's, one would think 16-bits would be the barebones threshold to meet.
To be clear, this criteria should be used *after* the DAC has filtered what jitter it can suppress. I should have been more clear. :) The measurements I post show what goes into the DAC, not what comes out of it (this would be DAC dependent naturally).
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,376
Likes
234,529
Location
Seattle Area
Sounds good. Please send us the file to [email protected] and we'll take it from there.
OK, I just sent it you. You are right that I usually use the other measurement but that is for analysis of DAC analog output, not digital streams.
 

Trdat

Addicted to Fun and Learning
Forum Donor
Joined
Sep 6, 2019
Messages
967
Likes
396
Location
Yerevan "Sydney Born"
To be clear, this criteria should be used *after* the DAC has filtered what jitter it can suppress. I should have been more clear. :) The measurements I post show what goes into the DAC, not what comes out of it (this would be DAC dependent naturally).


Do you think we can try the measurements with perhaps two different DAC's after it, in the chain to see how the UDIO8 effects a previously well measured DAC. And if the DAC suppresses any of the jitter at all?

Personally I think it's no excuse for a mediocre design, my trust for minidsp is out the door. But perhaps with the test I mentioned we can be more sure before jumping shit creek.
 
Last edited:
  • Like
Reactions: g29

g29

Addicted to Fun and Learning
Joined
May 1, 2019
Messages
520
Likes
318
Do you think we can try the measurements with perhaps two different DAC's after it in the chain to see how the UDIO8 effects a previously well measures DAC. And if the DAC surpresses any of the jitter at all?

Personally I think it's no excuse for a mediocre design, my trust for minidsp is out the door. But perhaps with the test I mentioned we can be more sure before jumping shit creek.

I would assume poorly designed DACs would have a higher tendency to pass the jitter through or benefit more from a better clock signal (e.g. similar to the Schitt DAC and the decrapifier experiment).
 

pos

Addicted to Fun and Learning
Forum Donor
Joined
Feb 13, 2018
Messages
572
Likes
718
The prototype was using the miniDSP drivers. Their production version is supposedly using their own USB/XMOS board along with a bunch of other changes.
Yes, what I meant by that is that there is probably no issue with the USBstreamer.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,376
Likes
234,529
Location
Seattle Area
Do you think we can try the measurements with perhaps two different DAC's after it in the chain to see how the UDIO8 effects a previously well measures DAC. And if the DAC surpresses any of the jitter at all?
I did that. I thought i mentioned that the Matrix went down a few notches to 116 dB SINAD. It seems like the implementations of AES/S/PDIF is not as good as USB so DACs are not doing as much as they could to eliminate the jitter. 116 dB of course is completely transparent still.
 

Trdat

Addicted to Fun and Learning
Forum Donor
Joined
Sep 6, 2019
Messages
967
Likes
396
Location
Yerevan "Sydney Born"
I did that. I thought i mentioned that the Matrix went down a few notches to 116 dB SINAD. It seems like the implementations of AES/S/PDIF is not as good as USB so DACs are not doing as much as they could to eliminate the jitter. 116 dB of course is completely transparent still.

Can we use the 6db SINAD the was the reduction in the Matrix DAC to 115/116 db with the Minidsp hooked up as a baseline for other DAC's?
 

KMN

Member
Joined
Sep 5, 2019
Messages
43
Likes
20
That second graph seems pretty fascinating. Just trying to makes sense of this. There is a larger higher frequency ripple and then a smaller lower frequency ripple. It kind of looks like the higher frequency ripple may be related to a 48KHz rate and the lower frequency ripple seems coincidentally close to 12KHz. So the conclusion is that both the analog and the clock handling circuitry are somehow electrically polluting the output reference clock circuitry?
 

pos

Addicted to Fun and Learning
Forum Donor
Joined
Feb 13, 2018
Messages
572
Likes
718
I did that. I thought i mentioned that the Matrix went down a few notches to 116 dB SINAD. It seems like the implementations of AES/S/PDIF is not as good as USB so DACs are not doing as much as they could to eliminate the jitter. 116 dB of course is completely transparent still.
Did you try the D10 in this configuration? Maybe with a simple spdif/AES inline passive converter, or directly in the spdif input of the matrix?
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,376
Likes
234,529
Location
Seattle Area
Did you try the D10 in this configuration? Maybe with a simple spdif/AES inline passive converter, or directly in the spdif input of the matrix?
I did. Oddly it produced 1 dB less SINAD than minidsp. The issue is that the D10 outputs S/PDIF and the minidsp produces AES. I think the handling of each interface is different in Matrix so it is not a comparison of source devices. There are enough unknowns here that I did not want to post the results and bring attention to it. But here I am doing that.... :)
 

pos

Addicted to Fun and Learning
Forum Donor
Joined
Feb 13, 2018
Messages
572
Likes
718
There are enough unknowns here that I did not want to post the results and bring attention to it.
The full story is still much better than the partial one :)
Do you have access to canare 110/75 ohms adapters? They even have some with a pad to reduce AES levels to spdif ones.
 

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
7,835
Likes
16,497
Location
Monument, CO
Last edited:

pos

Addicted to Fun and Learning
Forum Donor
Joined
Feb 13, 2018
Messages
572
Likes
718
I do not.
I can drop ship some to you, they are pretty easy to source on ebay.
For example the canare bcj-xp-trb will convert a 75 ohms coaxial input to a 110 ohms XLR one.
It does not alter voltage but any properly implemented AES3 input should be ok with it as the minimum input level is the same as spdif at 0.2V

Another option would be to try and compare the D10 and udio8 on your ADI2 pro, as it has both inputs.
I think it would really be worthwhile to follow through here, because if we expect any properly implemented DAC to reject jitter up to a certain amount, then jitter under that amount should not matter.
 
Last edited:
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,376
Likes
234,529
Location
Seattle Area
@devteam , Any feedback from the new test files ???
Their analyzer doesn't have the advanced option for jitter extraction and analysis as I showed in the review. They were going to see if they could find that can.
 
Top Bottom