• 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 daily reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Sound Devices MixPre-3 II Multichannel Recorder Review

Rate this portable recorder

  • 1. Poor (headless panther)

    Votes: 84 78.5%
  • 2. Not terrible (postman panther)

    Votes: 14 13.1%
  • 3. Fine (happy panther)

    Votes: 5 4.7%
  • 4. Great (golfing panther)

    Votes: 4 3.7%

  • Total voters
    107

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,224
Likes
232,527
Location
Seattle Area
This is a review and detailed measurements of the Sound Devices MixPre-3 II multichannel recorder and audio interface with headphone output. It is on kind loan from a member and costs US $895.
Sound Devices MixPre-3 Audio Recorder Interface SD Card Review.jpg

It is a cute, compact and rigid recorder. Gain control for Channel 2 has a stiffer feel which I imagine is designed to let you find it in the dark by feel. The LCD on the right is bright and touch sensitive. It is a bit small for comfortable manipulation but can be done. The user interface is a bit odd in that there is no scrolling and you have to learn to click on "..." to go to the next page. There are a myriad of inputs and a few outputs:

Sound Devices MixPre-3 Audio Recorder Interface SD Card XLR inputs and outputs headphone amp R...jpg


I dread testing this class of device as it takes hours to figure out how to do even simple things. The MixPre-3 was no exception. I installed its ASIO driver but it would keep complaining that it was not seeing it. Windows wouldn't recognize it either. Eventually found a reference that said if that cable connection icon is orange, it means it is not getting power from USB cable. This, despite the fact that it worked fine locally with USB power. I switched USB cables and now Windows recognized it, but its own ASIO driver would not. Swapped out for a third and beefier cable and this time, the driver found it. Strangely, the icon remained orange even though everything worked.

Beyond the USB interface issue, there was fair amount to learn as to how to configure the unit and get it to work the way you want. It took an entire evening and I "think" I figured it out but without an app or good indication from the device what it is doing, it is hard to really know.

For testing, I treated the unit as an audio interface. Recording and playback from SD-card should result in the same performance.

Sound Devices MixPre-3 II DAC/Headphone Amplifier Measurements
Using the 3.5mm output, we get this performance:
Sound Devices MixPre-3 Audio Recorder Interface SD Card Line Out Measurement.png

Trying to get anything beyond 1 volt results in clipping. That is disappointing as I expect to at least get 2 volts out if not 4 volts. Using what we have, SINAD which is sum of noise and distortion lands the MixPre-3 II in the last place of all line interfaces tested:
Best audio interface portable recorder review.png


Fortunately there is a headphone output which generates higher voltage and better performance:
Sound Devices MixPre-3 Audio Recorder Interface SD Card Headphone Out Measurement.png


Still short of 16 bit quality though. Excluding distortion where the problem is, noise performance is good at max volume (left):
Sound Devices MixPre-3 Audio Recorder Interface SD Card Headphone Out Dynamic Range Measurement.png


You could hear noise with sensitive IEMs though given the 75 dB on the right.

The real head-scratcher came when I tried to sweep the unit to measure its power into 300 ohm load with headphone gain set to max:
Sound Devices MixPre-3 Audio Recorder Interface SD Card Headphone Out Power 300 Measurement.png

Looking at the output voltage directly, we get this limiting behavior:
Sound Devices MixPre-3 Audio Recorder Interface SD Card Headphone Out Power 300 Output Voltage...png


I don't know what is going on here. Going with what we have, output level is quite low (spec is 300 mwatts!) to get you much volume. In other words, it is the typical checklist item in audio interfaces. Let's move on to ADC.

Sound Devices MixPre-3 ADC/Microphone Pre Measurements
The odd behavior continued when I was not able to get the MixPre-3 to produce higher level than -7 dBFS:

Sound Devices MixPre-3 Audio Recorder Interface SD Card Measurement.png


If you turned up the gain more, nothing would happen to captured audio levels. Fortunately our measured value matches company's spec for SINAD. Unfortunately, this is not very good number:
Best portable audio interface recorder.png


Frequency response is nice and flat as it should be:
Sound Devices MixPre-3 Audio Recorder Interface SD Card frequency response Measurement.png


Dynamic range is very good:

Sound Devices MixPre-3 Audio Recorder Interface SD Card dynamic range Measurement.png


Not so good is the frequency variability as far as distortion:
Sound Devices MixPre-3 Audio Recorder Interface SD Card THD vs frequency Measurement.png


I grabbed an FFT at 7 kHz to see what is going on:
Sound Devices MixPre-3 Audio Recorder Interface SD Card 7 khz fft Measurement.png

There is a lot of unwanted noise but most of it is ultrasonic.

Sweeping IMD level shows distortion rising pretty early:
Sound Devices MixPre-3 Audio Recorder Interface SD Card IMD vs Level Measurement.png

Noise level is good though as indicated earlier.

I have recently started to measure the microphone preamp "effective input noise (EIN)." This is a draft measurement as there is some fogginess in how to measure this number in the context of a line interfaces that digitizes input (as opposed to pure analog preamp). Here is a comparison to a few other interfaces:
Effect Input Noise Microphone Digital Preamplifier Review.png

As you see, the performance of MixPre-3 II is fair bit worse. While the front panel says it goes to some 90+ dB of gain, anything over 49 dB does nothing.

Conclusions
Objectively, there is little to hang your hat on that you are getting a performance optimized recorder/line interface here. None of the tests show what I would want to see in a professional product. Headphone output is quite bad. DAC has a lot of distortion and too little output. The ADC falls in the poor category of all interfaces tested but is not as bad as its DAC.

Someone is going to say for recording a microphone this is all fine. I am not a user of these products but as long as you don't pay a premium for performance, then you may be right.

------------
As always, questions, comments, recommendations, etc. are welcome.

Any donations are much appreciated using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/
 

Joachim Herbert

Senior Member
Forum Donor
Joined
Jan 20, 2019
Messages
447
Likes
659
Location
Munich, Germany
This is really disappointing. I expected at least decent performance given price and reputation of sound devices.
 

Rja4000

Major Contributor
Forum Donor
Joined
May 31, 2019
Messages
2,660
Likes
4,197
Location
Liège, Belgium
Very interesting review
Thanks !

A detail: The figures are missing for the RME in your EIN bargraph
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,224
Likes
232,527
Location
Seattle Area
A detail: The figures are missing for the RME in your EIN bargraph
Which RME? ADI-2 Pro? Babyface is there.

I forgot to reserve a post for the products specifications. Here they are, and it’s a long list of them:

Specifications​


Audio Inputs​

  • Frequency Response​

    • 10 Hz to 80 kHz +/- 0.5 dB re 1 kHz @ 192 kHz sample rate
  • THD + Noise​

    • 0.005% max (@1 kHz, 22-22 kHz BW, gain= 20 dB, -10 dBu in)
  • ADC​

    • 32 bit precision; 142 dB dynamic range min (A weighted, gain = 10 dB, fader = 0 dB)
  • Equivalent Input Noise​

    • -130 dBV (-128 dBu) max (A-weighting, gain = 76 dB, 150 ohm source impedance)
  • Inputs​

    • Mic: XLR active-balanced; 4k input
    • Line: XLR active-balanced; 4k input
    • Aux/Mic in: 3.5 mm TRS, 2-channel unbalanced; 2.2k
    • USB Audio: 2 channels
    • 5 Total analog inputs: 3 mic-line inputs, 2 on Aux In
    • All inputs fully RF-filtered and overload protected.
  • Gain​

    • Mic input: +6 dB to +76 dB
    • Line input: -20 dB to +30 dB
    • Fader: -inf to +20 dB
    • Headphone: -inf to +20 dB
    • Total, Mic-to-recording (max): +96 dB
    • Aux In (Mic): Gain = +10 dB to +40 dB
    • Aux In (Line): Gain = -10 dB to +20 dB
    • Total Aux Gain including Fader (Mic-to-recording): +60 dB
  • Maximum Input Level​

    • Mic XLR: +14 dBu (limiters on or off)
    • Line XLR/¼-inch: +28 dBu (limiters on or off)
    • Aux In (Mic): -10 dBu
    • Aux In (Line): +10 dBu
  • Low Cut Filters​

    • 40 Hz to 160 Hz (adjustable), 18 dB/oct. First stage analog, subsequent stages digital.
  • Microphone Powering​

    • Mic XLRs: 48 V via 6.8k resistors, 10 mA each
    • Mic 3.5 mm: 3 V @ 3k source
  • Limiters​

    • Limiter at all gain stages, range > 40 dB. First stage analog, subsequent stages digital. Adjustable Threshold, Ratio, and Release.

Audio Outputs​

  • DAC feeding Stereo Out, Headphone Out
    • 32 bit precision; 115 dB dynamic range (A-weighted)
  • L/R Out
    • 3.5 mm TRS stereo unbalanced, 500 ohm output impedance, +7.8 dBu max output level
  • Headphones Out​

    • 3.5 mm TRS stereo unbalanced, 300 mW + 300 mW, for use with any impedance headphones

USB​

  • Audio Interface (USB-C): 6-in/2-out (from computer); 44.1 kHz/48 kHz/96 kHz; 16/24/32-bit integer, 32-bit float; Class-compliant USB 2.0 high speed driver
  • Mass Storage (USB-C): USB 2.0 high speed
  • Keyboard (USB-A): Text entry and control
  • Thumbdrive (USB-A): Manual or auto-copy to drive
  • Midi Control Surface (USB-A)

Recording​

  • Maximum Record Tracks​

    • 5 Tracks Stereo mix + 3 ISO’s
  • Sampling Frequency​

    • 44.1 kHz
    • 47.952 kHz
    • 48 kHz
    • 48.048 kHz
    • 96 kHz
    • 192 kHz
  • Bit Depth​

    • 16
    • 24
    • 32 Float
  • Recording Storage​

    • Type: SD, SDXC, SDHC Card, & USB thumbdrive (copy only)
    • Max Storage Size: 512 GB (SDXC)
    • Card format: exFAT
  • File Type​

    • Polyphonic WAV

Timecode​

  • Timecode Modes​

    • Free Run
    • Time of Day (file stamped w/current ToD)
    • Rec Run
    • Ext LTC (file stamped w/incoming LTC on 3.5mm Aux In)
    • Camera TC (file stamped w/incoming TC from cameras that output TC)
  • Timecode Frame Rates​

    • Auto-detects (fps):
    • 23.98 (same as 23.976)
    • 24
    • 25
    • 29.97DF
    • 29.97ND
    • 30
    • 30DF
  • 3.5 mm Aux In​

    • Timecode Input
    • Timecode: 20k ohm impedance, 0.3 V – 3.0 V p-p (–17 dBu – +3 dBu)
  • 3.5 mm Stereo Out​

    • Timecode Output
    • Timecode: 1k ohm impedance, 3.0 V p-p (+12 dBu)
  • HDMI​

    • TC Reader via HDMI
  • Sample/Timecode Accuracy​

    • 0.2 ppm (0.5 frames per 24 hours)

Remote Control​

  • Bluetooth LE: Wireless control using Wingman app
  • HDMI (micro): Auto-record start/stop trigger from cameras that output record flag over HDMI
  • Timecode: Auto-record start/stop trigger via Aux In timecode

File Delivery to Cloud​

  • Compatible with Viviana Cloud

Touch Screen​

  • 320×256; 1.6-inch, color, sunlight-viewable IPS LCD

Power​

  • AA Batteries: 4x AA sled (included); 8x AA sled (optional accessory). NiMH recommended
  • From computer: Bus-powered via USB-C port; optional MX-USBY cable, connects one USB-C to two USB-A ports
  • L-Mount: Optional sled for 2x (hot-swappable) Li-ion batteries
  • From computer: Bus-powered via USB-C port
  • AC Adapter: MX-PSU (included) Power supply wall adapter w/ USB-C connector; 15 W; has 4 adapter plugs for US, UK, AU & Europe

Environmental​

  • Operating: -20°C to 60°C, 0 to 90% relative humidity; (non-condensing)
  • Storage: -40°C to 85°C

Dimensions (H x W x D)​

  • 3.6 cm x 14.4 cm x 11.0 cm
  • 1.40” x 5.68” x 4.35”

Weight​

  • 0.48 kg (unpackaged, without batteries)
  • 16.8 oz (unpackaged, without batteries)
 
Last edited by a moderator:

Sokel

Master Contributor
Joined
Sep 8, 2021
Messages
5,609
Likes
5,529
Nope,that's bad.
That's a 2001 average performance and at a way lower price range.

Thanks Amir!
 

MaxwellsEq

Major Contributor
Joined
Aug 18, 2020
Messages
1,536
Likes
2,279
I have recently started to measure the microphone preamp "effective input noise (EIN)." This is a draft measurement as there is some fogginess in how to measure this number in the context of a line interfaces that digitizes input (as opposed to pure analog preamp). Here is a comparison to a few other interfaces
Excellent news! Well done for kicking this off.
 

DSJR

Major Contributor
Joined
Jan 27, 2020
Messages
3,215
Likes
4,325
Location
Suffolk Coastal, UK
So - and I'm not baiting - there *are* differences in USB cables? Just not in data transfer, but maybe cable construction in getting the power supplied properly?
 

Sokel

Master Contributor
Joined
Sep 8, 2021
Messages
5,609
Likes
5,529
So - and I'm not baiting - there *are* differences in USB cables? Just not in data transfer, but maybe cable construction in getting the power supplied properly?
There are certified ones and not certified ones as long as I have messed with them.
Certification protocol even states the plug-unplug number of times.

Like anything else there's junk and there's proper stuff.
 

BeerBear

Active Member
Joined
Mar 9, 2020
Messages
264
Likes
252
Thanks for the review!

According to the specs, the A-weighted dynamic range for the inputs is 142dB. Is that something you can confirm?
 

Rja4000

Major Contributor
Forum Donor
Joined
May 31, 2019
Messages
2,660
Likes
4,197
Location
Liège, Belgium
Thanks for the review!

According to the specs, the A-weighted dynamic range for the inputs is 142dB. Is that something you can confirm?
The Specs says

Audio Inputs​

  • ADC​

    • 32 bit precision; 142 dB dynamic range min (A weighted, gain = 10 dB, fader = 0 dB)
So, first, the dynamic range is for the ADC only. It will be hard to measure it, since we need to use an analog input to connect to the ADC. It's just a theoretical value, IMO.

And the specs for the analog inputs is not in the same league.
The best approximate to the actual dynamic range is this

Sound Devices MixPre-3 Audio Recorder Interface SD Card dynamic range Measurement.png

This is for line IN.
Mic IN might be a bit better.
But not night and day better.
 
Last edited:

RandomEar

Senior Member
Joined
Feb 14, 2022
Messages
331
Likes
761
So - and I'm not baiting - there *are* differences in USB cables? Just not in data transfer, but maybe cable construction in getting the power supplied properly?
Yes, of course. Some may not be spec compliant, badly shielded and so on. But there are also significant capability differences in spec compliant cables. That is especially true for USB C, where you can have 100 W of USB PD but only USB 2.0 speeds (480 MBit/s) in a "legal" cable. Additional complications arise with Frankenstein-solutions like A-to-C cables, adapters and "illegal" extension cables, as well as worn, dirty or damaged connectors.

But to get to the possible bait point: If the device gets sufficient power and a digital connection is established without dropouts, there is no difference in the transmitted data packets. The last thing that you can still run into at that point is interference and HF noise via the ground connection.
 

BadAudioAdvice

Active Member
Joined
Sep 3, 2021
Messages
109
Likes
167
Location
CAN / USA
@amirm I wonder if there were a few things going on contributing to the low measured performance.

1. The orange power icon indicates that its operating in low power mode.

Perhaps that limited the line-out level and the power on the headphones.

However, SoundDevices doesn't make mention of reduced output levels or power, but might still have an impact.
From their website:
"To fully power the MixPre-3 II or MixPre-6 II from a USB power source not listed above, you will need BOTH of the following:
  1. USB-C power source that is compliant with the USB powering specification and is able to output 7.5W
  2. USB-C cable that incorporates the correct pull-up resistors to identify itself as able to provide 7.5W
If either of these conditions are not met, the MixPre recorder will operate in low power mode: maximum 2x 48v phantom channels, no HDMI, no USB-A. You can still use all of the MixPre’s channels, but only a maximum of 2 can be phantom powered."

2. The ASIO driver seems to be very problematic, and not really necessary unless needing to use lots of IO channels and 32-bit float. When I tried using the ASIO drivers the device barely worked.
Could the performance be better with the default plug-and-play Windows drivers?

3. There are a few places to adjust gain, control the levels going into the output, and select the input type (line/mic). I was surprised that during the testing performed you mentioned that a few times it didn't increase the output level, or it didn't reach the level it said it could.
 
Last edited:

Bergante

Member
Joined
Sep 9, 2022
Messages
54
Likes
61
Location
Bilbao, Spain
Did you disable the limiters? I ask because these things have a bootload of options, and it can be confusing.

Notably it works as a combination of microphone recorder and mixer. So the gain staging has several elements.

As for USB powering it needs a proper USB-C cable. If it doesn't detect a proper cable it will fall down to low power. The problem is, the world is full of dodgy cables. I remember a mention somewhere of fake ferrite cores as "filters".

In USB audio mode it also has two options: class compliant multitrack audio device or just stereo.
 

617

Major Contributor
Forum Donor
Joined
Mar 18, 2019
Messages
2,384
Likes
5,259
Location
Somerville, MA
I'd contact Sound Devices. Their reputation is very good; they are the standard for field recorders, although I'm sure Zoom and Tascam and Sony are nipping at their heels.

I don't think headphone performance is important for this product, but low noise mic preamps, ADC and battery life are important, as is warranty support and durability.
 

Bergante

Member
Joined
Sep 9, 2022
Messages
54
Likes
61
Location
Bilbao, Spain
I'd contact Sound Devices. Their reputation is very good; they are the standard for field recorders, although I'm sure Zoom and Tascam and Sony are nipping at their heels.

I don't think headphone performance is important for this product, but low noise mic preamps, ADC and battery life are important, as is warranty support and durability.
Actually headphone performance is very important and I have found it to be very good.

Maybe @amirm can save a configuration preset to the SD card (it's a XML file) and share it?

(I don´t work for Sound Devices but I own a first generation MixPre-3)
 

Bergante

Member
Joined
Sep 9, 2022
Messages
54
Likes
61
Location
Bilbao, Spain
Possible trouble areas:

It has three configuration modes: Basic, Advanced and Custom.

In advanced mode you can configure routing and enable/disable the limiters. The limiters not only act on the inputs, but can act also on the headphone output if I remember well.

Routing can be very important for a test because, by default, the recorder works as a recorder+mixer.

It records 5 tracks, 3 individual unprocessed inputs (to which only the preamplifier gain and, if engaged, limiter and HPF applies) and two more tracks for a mix. In that case the unit knobs are mixer faders. Depending on their position, headphone and "line out" outputs may have an additional gain applied: the mixer fader.
 
Last edited:

lolol

New Member
Joined
Nov 29, 2023
Messages
2
Likes
2
As a happy owner of a MixPre 10 II, I suspect the power supply to be involved in these poor measures. Not to mention the 60 Hz hum on the FFT
index.php


The limiter section may also be involved in the poor results you have. Could you please provide your settings for this test campaign ?
 
Top Bottom