Review and Measurements of Schiit Modi 3 DAC

[Veri]

My question is, what performance should I expect from a 2018 Mac Mini?

It will work absolutely fine. Just configure the device in the MIDI control panel.

 

[Panelhead]

The software may default to internal speaker sometimes on power up.
Path
Settings
Sound
Schiit Modius (Or whatever it is recognized as)
Seems to remember after a few cycles. Changed DACs a few months back and it would select Soundflower. Now it selects the dac every time.

 

[Aperiodic]

What playback software do you plan on using? There are several options.
Many who have compared feel a PC works bettering playback. Might be true if using iTunes for playback.
A Mac gets around the driver issues a lot of times. And driver breaking with software updates.
I use a Mini that is several generations old. Lots of RAM and a large SSD. I think the last model that allowed swapping both. Works great.

The Tidal app, Idagio app, and Swinsian app. I don't use iTunes much, and not for any serious listening.Tidal and Idagio allow the playback app to directly control the MIDI steeings, as does Fidelia.
I only ask because of the noticeable difference in performance between ASIO and WASAPI. From what I've been able to gather, Core Audio handles the functionality I'm asking about on the Mac. Assuming the flaws seen in the original test are ASIO4ALL artifacts (which is what Amir seems to be suggesting),I assume I have nothing to worry about; but you know what they say about assuming...

 

[AllanDavidson]

Looks like there's a new kid in town

Co-founder Jason Stoddard writes: “Not only did we add our exclusive Unison USB interface, the product of two person-years of development, to our least expensive DAC, we also tweaked it a bit to improve performance even further! Modi 3+ is the highest-performance Modi we’ve ever made…and it also comes with Unison USB…and it’s also still $99.”

https://darko.audio/2020/12/schiit-announces-99-modi-3-dac-w-unison-usb/

 

[Angsty]

Looks like there's a new kid in town

https://darko.audio/2020/12/schiit-announces-99-modi-3-dac-w-unison-usb/

Oh, yeah. We need measurements.

 

[Veri]

Oh, yeah. We need measurements.

Well, the Schiit AP ones have generally been quite like Amir's so that's a good starting point.

 

[Matias]

Oh, yeah. We need measurements.

https://www.audiosciencereview.com/forum/index.php?threads/schiit-modi-3-with-unison-usb™.18054/

 

[DeNyss]

Oh, yeah. We need measurements.

5 x 3.5 x 1.25” (c)

 

[Angsty]

5 x 3.5 x 1.25” (c)

Not exactly 36x24x36, but it’s not made like a brick house, either.

 

[0%_Audiophile]

Modi 3 vs Modi 3+?

 

[0%_Audiophile]

Just saw B-stock on Schiit website and decided to buy Modi 3

 

[DSJR]

Oh FFS - here we go again! Another subjective comment with no real reference other than what he likes the sound of, so absolutely useless -

https://audio-head.com/review-schii...il&utm_term=0_5454ba75df-60eeef3291-592412741


I'm left feeling 'we' subjectively like or even prefer some 'noise' in the background a la FM radio (13 bit in the UK) or good vinyl. Take that away and I wonder of the starkness aluded to digital by analogue-orientated people is the removal of such artefacts. Apologies if this has been covered before.

 

[0%_Audiophile]

I bought 2 months ago Modi 3 and I had Magni 3 little bit earlier, so I sold all creative cards, such as G6, because Schiit has so cool sound, no drivers needed, very good bass etc. even do not wanna explain all details, I do not like only Magni power supply)

 

[vibess]

I bought 2 months ago Modi 3 and I had Magni 3 little bit earlier, so I sold all creative cards, such as G6, because Schiit has so cool sound, no drivers needed, very good bass etc. even do not wanna explain all details, I do not like only Magni power supply)

Currently runnning my DT1770Pro of my G6 and think it does a decent job, but has been looking to add an amp like the Asgard 3 or Magni3+ to my setup. Thought the G6 DAC was good. What improvement would I actually get also replacing the G6 DAC with for example a Modi3+ ?
Most importantly, would I be able to tell a difference?

 

[VintageFlanker]

Just a quick post in order to bring some repeatability. I just got a few years old Schiit Modi 3 (AK4490) in intend to measure it through my Cosmos ADC.

Disclaimer: Please note that SINAD (THD+N) is naturally few dB worse than AP results, due to my test setup limitations (hardware and software). For example, I get 107dB SINAD with my Matrix Mini-i Pro 3 against 112dB in Amir's review.​

RMAA Summary - USB/DirectSound - 24bits/44.1Khz​

Frequency response (from 40 Hz to 15 kHz), dB	+0.01, -0.09​	Excellent​
Noise level, dB (A)	-117.3​	Excellent​
Dynamic range, dB (A)	117.2​	Excellent​
THD, %	0.00047​	Excellent​
THD + Noise, dB (A)	-99.1​	Excellent​
IMD + Noise, %	0.00496​	Excellent​
Stereo crosstalk, dB	-115.3​	Excellent​
IMD at 10 kHz, %	0.00138​	Excellent​

​

Frequency response​

	Left​	Right​
From 20 Hz to 20 kHz, dB	-7.76, +0.01​	-7.77, +0.01​
From 40 Hz to 15 kHz, dB	-0.08, +0.01​	-0.09, +0.01​

​

Noise level​

	Left​	Right​
RMS power, dB	-117.0​	-116.8​
RMS power (A-weighted), dB	-117.6​	-116.9​
Peak level, dB FS	-93.0​	-93.8​
DC offset, %	-0.0​	+0.0​

​

Dynamic range​

	Left​	Right​
Dynamic range, dB	+117.0​	+116.7​
Dynamic range (A-weighted), dB	+117.6​	+116.9​
DC offset, %	-0.00​	+0.00​

​

THD + Noise (at -3 dB FS)​

	Left​	Right​
THD, %	0.00048​	0.00047​
THD + Noise, %	0.00098​	0.00098​
THD + Noise (A-weighted), %	0.00111​	0.00110​

​

Intermodulation distortion​

	Left​	Right​
IMD + Noise, %	0.00497​	0.00496​
IMD + Noise (A-weighted), %	0.00195​	0.00196​

​

IMD (swept tones)​

	Left​	Right​
IMD + Noise at 5000 Hz,	0.00182​	0.00180​
IMD + Noise at 10000 Hz,	0.00099​	0.00101​
IMD + Noise at 15000 Hz,	0.00134​	0.00133​

​

Stereo crosstalk​

	Left​	Right​
Crosstalk at 100 Hz, dB	-115​	-116​
Crosstalk at 1000 Hz, dB	-113​	-116​
Crosstalk at 10000 Hz, dB	-110​	-113​

Let see what happens when comparing USB against S/PDIF inputs:

RMAA Summary (USB VS Coax VS Optical)​

Test	[DirectSound] Schiit Modi 3 USB	[DirectSound] Schiit Modi 3 Optical	[DirectSound] Schiit Modi 3 Coaxial
Frequency response (from 40 Hz to 15 kHz), dB:​	+0.01, -0.09	+0.03, -0.34	+0.03, -0.34
Noise level, dB (A):​	-117.3	-116.6	-116.6
Dynamic range, dB (A):​	117.2	116.6	116.4
THD, %:​	0.00047	0.00050	0.00051
IMD + Noise, %:​	0.00496	0.00415	0.00421
Stereo crosstalk, dB:​	-115.3	-115.7	-114.9

Results are basically the same except for frequency response. I've got a repeatable 0.5dB drop when using S/PDIF :

Multitone Analyzer (100 Tones)

​

Both USB and S/PDIF performed exactly the same. Good results, but a step shy of state of the art. For relative comparison, Topping E30 in the same test got -104dB. In fairness : E30 showed higher harmonics peaks, while Modi did have some higher noise floor...

Overall : Very respectable performance for a 4 years old, 99$ (and USA made) DAC! You still get decent results that translate into audible transparency.​

 

[Angsty]

Thanks for the update! Still quite content with my Modi 3 for 16/44.1 use even though I have a Topping D70s that I can swap out for it.

 

[Beng]

Dear @amirm ,

I just saw that on youtube.

Would you please, if you still have the Modi 3 on hand, MEASURE this statement?: "Modi 3 sounds a bit better"

Thanks.

Crystals which are the source the clock in any system are mechanical resonators based on the piezo-electric. This means that crystals have a sensitivity to mechanical vibration. The sensitivity is a function of the type cut of the crystal. 25 years ago this was an issue because most systems used lower cost crystals which were more sensitive to vibration. The act of potting on of these crystals in resin reduced the sensitivity of the crystals. Many adopted plls which reduced or masked the sensitivity by spreading the spectrum of the induced jitter. That plus the adoption of better crystals made the problem less audible. It is far less of a problem today than it was then. But it could still be audible. The problem is no one is measuring it today.
Ben

 

[VintageFlanker]

But it could still be audible. The problem is no one is measuring it today.

Fine. And how do you measure such a "problem" ?

Jitter? Would be surprised if I measure anything out of margins under common domestic "vibrations" (idea of extreme scenario: blasting my speakers at 90% volume near some DAC).

 

[Veri]

That plus the adoption of better crystals made the problem less audible. It is far less of a problem today than it was then. But it could still be audible. The problem is no one is measuring it today.
Ben

-https://archimago.blogspot.com/2018/08/demo-musings-lets-listen-to-some-jitter.html
-What do you think amir is measuring in the jitter sections of his measurements..?

 

[Beng]

Fine. And how do you measure such a "problem" ?

Jitter? Would be surprised if I measure anything out of margins under common domestic "vibrations" (idea of extreme scenario: blasting my speakers at 90% volume near some DAC).

In the old days I connected a pll to the clock and monitored it. I realize that could wander into the destructive testing. One way might be to use a vibration plate. In my experience crystals seemed to be sensitive to midrange frequencies, middle of the piano, which is beyond most cheap vibration plates.