• 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!

Budget DAC Review: Schiit Modi 2 ($99)

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,386
Location
Seattle Area
OK, with my wife out on a trip, I have some time to catch up on my review chores. In this wonderful episode, you are treated to the measurements of the Schiit Modi 2. This guy retails for $99 which fit the budget nicely.

It uses the Analog Devices AD 5547 DAC silicon. This is an older DAC (circa 2012) and quite pricey at some $20. The say it is the same chip used as in their Bifrost DAC.

EDIT: that was a mistake. The DAC chip uses is from AKM: AK4490EQ. See later posts.

Some of you may recall me having a poor experience with one of their $400 DACs a few years ago showing one of the worst correlated jitter measurements I have ever made. And my son could readily hear his computer activity through that DAC. Performance that lives to their name nicely.

So it was with some ambivalent that I entered this test. Would this be a repeat of that incident or decent performance? Well, you decide.

Before we get into the measurements, let me say that I went through hell getting this unit working with its drivers. They would throw an obscure error and nothing I tried worked. So I gave it to my son who quickly discovered that there is a toggle switch on the back. If you don't set it to "E" for expert, it will refuse to install its drivers! Cotton picking son of $#&^^%. Why not say that in the driver install package??? Wasted so much of my time. Anyway, these tests are with those drivers installed.

As always, the weapon of choice is the 24-bit/48 Khz J-Test signal. This is a 12 Khz tone with a bit toggling. Ideally you should see a single sharp spike at that frequency and nothing else. Here is how the Schiit Modi 2 did as compared to my reference, iFi iDAC2 ($350):

iFi Schiit Comparison.png


Oh man. We get a whole bunch of deterministic jitter in the form of those spikes which run at 1 Khz and other frequencies. In addition to that, we have random jitter almost occupying the entire audio spectrum widening the "skirt" in our 12 Khz tone.

But the story does not end there. I ran it a second time and got different results! My son and I played with different CPU profiles on the computer connected to it (different from my measurement computer) and they all impacted the output of this DAC!!!

Modi 2 multiple runs.png


Notice how the skirt shape changes and so does the amplitude of all the deterministic spikes. This shows clear lack of isolation from computer/USB bus.

Either they don't measure their DACs, don't care or don't know how to design a clean DAC. Yes, -85 db distortion products is not likely audible and hence the reason they get away with such poor performance. But the engineer in me wants to throw up on it. Let the no-name Chinese vendors produce this stuff.

Actually that might be an insult to Chinese vendors because this is what I got for the Origen+ DAC:

index.php


See how whistle clean it is compared to the Schiit at the same price point?

I sure hope their higher priced stuff is better because my experience so far indicates anything but that. Please don't buy this DAC.

Edit 2: more measurements and data starting here: https://audiosciencereview.com/foru...eview-schiit-modi-2-99.1649/page-6#post-54623 (same conclusions though).

Edit 3: late testing shows the Schiit Modi 2 to be highly sensitive to USB power: https://audiosciencereview.com/foru...eview-and-measurements.1829/page-2#post-46217

index.php


This means that depending on what PC you plug it into, you will get different output! It may be worse than mine or better. Really bad form to not filter USB power better prior to use inside the DAC.

As always, I welcome comments and corrections.

EDIT: See the hardware teardown here: https://www.audiosciencereview.com/forum/index.php?threads/hardware-teardown-of-schiit-modi-2.2082/

EDIT: See additional measurements showing PC sensitivity here: https://www.audiosciencereview.com/...eview-schiit-modi-2-99.1649/page-8#post-57395
 
Last edited:

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,198
Likes
16,981
Location
Riverview FL
It uses the Analog Devices AD 5547 DAC silicon

The AD5547 is a 16bit DAC.

http://www.analog.com/media/en/technical-documentation/data-sheets/AD5547_5557.pdf

As always, the weapon of choice is the 24-bit/48 Khz J-Test signal.

Does that matter?

---

"Modi 2: Improve Any Computer’s Sound
Modi 2 plugs into virtually any computer—Windows, Mac, popular Linux distros, Intel Chromebooks, as well as iPhones and iPads. Just connect via USB and go with no drivers in Standard Mode, up to 24/96 output. Switch to Expert Mode (and install Windows drivers) for extended high-res capabilities to 24/192. "

Maybe not.
 
Last edited:

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
7,835
Likes
16,497
Location
Monument, CO
The AD5547 is not a delta-sigma DAC; it is a conventional R-2R (plus unary MSB) architecture. Some claim this architecture provides better sound than delta-sigma designs due to conventional noise transfer function and no need for a high-speed clock. Trimming is required to achieve high linearity and that probably contributes to the high cost. Precision resistors and trim flags take a lot of area so it is probably a relatively large die, again increasing cost. The on-chip multipliers could be used to extend the dynamic range, or several could be combined to increase the resolution, or it could simply reduce a 24-bit data stream to 16 bits.
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,198
Likes
16,981
Location
Riverview FL
The on-chip multipliers could be used to extend the dynamic range, or several could be combined to increase the resolution, or it could simply reduce a 24-bit data stream to 16 bits.

It only has 16 data lines in the pinout.

The word "dither" is not in the spec sheet.

Nor is "audio".
 
Last edited:

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,522
Likes
37,050
Might be worth doing the 16 bit Jtest. Perhaps 24 bits are simply truncated. Incompetent design yes.

Currently it appears a sub-$40 HDMI switcher fed over HDMI has better jitter performance. It is always something with these Schiit designs.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,386
Location
Seattle Area
I shutdown the system. If I have spare cycles, I will do a 16-bit test. For now, system dependent noise and random jitter won't be caused by truncation so best case is still not good.
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,522
Likes
37,050
This from Stoddard when they were picking the DAC for the Bifrost multibit.

The Bizarre DAC
“I’ve got it,” Mike said, one day when we were both at Schiit. “The AD5547.”
I frowned. “What the hell is that?”
“It’s a DAC. Look it up.”
I did. “16 bit?” I asked, doubtfully.
“Look harder,” Mike said.
I did. And I realized that the specs on this DAC were really amazing. So amazing, I should call out the relevant points with numbers.
1.It features an integral nonlinearity (INL) plot that's better than +/-0.5LSB. This is a spec they never provide for audio DACs, because (a) it would be terrifying in the case of audio multibit DACs and (b) it is not possible to measure delta-sigma DACs in this way, since the output depends on the preceding and following samples.
2.The THD performance is actually scary good for a 16-bit DAC—far better than 16 bits. 16 bit level THD, from a theoretically perfect 16-bit DAC, is -96dB. Most 16 bit DACs from the Jurassic Age of Digital didn't hit this number. The AD5547 is -104dB, much better than 16 bit.
3.Its noise level is down at the 22 bit level. Think about that for a bit. Why would you need 22-bit equivalent noise for 16-bit audio? The answer is, you don’t. And in the old days, you couldn’t. Many, many DACs from the early years of the PCM era couldn’t do 16-bit noise levels.
And, it was a dual DAC. As in, we’d only need one chip per stereo Bifrost. This meant less board space, and an easier implementation…except for one thing: the AD5547 is a parallel input DAC.
“Huh?” you may be asking. “What’s that, and who cares?”
Well, most DACs take their data in serial form—data goes in on one single pin. You send it that stream of data, plus a few clocks, and it’s happy. Most audio DACs are built so they easily interface with the output of typical USB and SPDIF receiver chips, so it’s kinda like building blocks. Hook up a few pins and go. Even I can do it.
But the AD5547 takes the data in parallel, on 16 separate pins, plus other clocks to ensure it converts samples at the right time, and both at once. Bottom line, the clock logic and data input to this DAC are fundamentally different than audio DACs—so there’s no easy paint-by-numbers implementation. Mike and Dave would be figuring out how to interface a sample rate converter (or, I still hoped, the full burrito filter) to another bizarre DAC—one completely different than the AD5791 and AD5781.
There are many days I’m happy to be an analog designer.
 

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
7,835
Likes
16,497
Location
Monument, CO
There are some technical errors in that...

To convert from 24 bits to 16 you can truncate or apply a filter function (effectively choose the most-likely nearest 16-bit number). That would be done before the DAC.
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,522
Likes
37,050
Though not conclusive there are some comments that lead you to believe Schiit is using one of these dual DAC chips for balanced operation, but multiplexing the one DAC for right and left channels. I wonder if that is how the jitter is being introduced? So many declare their multi-bit DACs sound like no other. Hmmmmm?

Certainly looking at the board on the Schiit website you can see there is only one AD5547 on the board though whether they multiplex it or use one half for each channel isn't clear.
 
Last edited:

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,522
Likes
37,050
There are some technical errors in that...

To convert from 24 bits to 16 you can truncate or apply a filter function (effectively choose the most-likely nearest 16-bit number). That would be done before the DAC.

Well there is the issue where he complained Schiit didn't truncate, they rounded the bits.
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,522
Likes
37,050
It is not clear which filter they used, though it appears they used some version of their custom burrito filter. One of the Schiit principles has said that filter must be reset if it encounters a signal that is not at first all zeroes should sample rates change. I wonder if this caused the odd behaviour when Amir ran the test a second time?
 

Cosmik

Major Contributor
Joined
Apr 24, 2016
Messages
3,075
Likes
2,180
Location
UK
It is not clear which filter they used, though it appears they used some version of their custom burrito filter. One of the Schiit principles has said that filter must be reset if it encounters a signal that is not at first all zeroes should sample rates change. I wonder if this caused the odd behaviour when Amir ran the test a second time?
Very reminiscent of the audiophile amplifier that progressively increases distortion on steady tones, and then eventually shuts down completely. Or the audiophile DAC with zero crossing glitches.

I have been there... when I was about 18 years old. I used to think that I knew enough to chain together two 8-bit DACs to make 16 bits etc. It wasn't engineering, but perhaps an early version of 'Maker culture'. Controlling noise, grounding etc. came a distant second place to simply creating something that 'worked' and had a nice box and logo. All good fun, but I wouldn't want to connect anything expensive to what I made.

The maker culture is a contemporary culture or subculture representing a technology-based extension of DIY culture that intersects with hacker culture... and revels in the creation of new devices as well as tinkering with existing ones.

Just about sums it up, I think.
 
Last edited:

Sal1950

Grand Contributor
The Chicago Crusher
Forum Donor
Joined
Mar 1, 2016
Messages
14,070
Likes
16,599
Location
Central Fl
Very interested in that second round of testing.
As a layman I am quite shocked at this measured performance from a couple of gentleman with such good background and credentials?
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,522
Likes
37,050
Very reminiscent of the audiophile amplifier that progressively increases distortion on steady tones, and then eventually shuts down completely. Or the audiophile DAC with zero crossing glitches.

I have been there... when I was about 18 years old. I used to think that I knew enough to chain together two 8-bit DACs to make 16 bits etc. It wasn't engineering, but perhaps an early version of 'Maker culture'. Controlling noise, grounding etc. came a distant second place to simply creating something that 'worked' and had a nice box and logo. All good fun, but I wouldn't want to connect anything expensive to what I made.



Just about sums it up, I think.

Well Mike at Schiit is Mike Moffat who designed Theta digital gear. So I don't know it is just a make something and be proud problem. He does seem to like design's off the beaten path. They just all seem to have some issue. I do think the entire company from its name, name of models and the whole approach is literally rubbing people's face in it. There is always something with every product they make. Amps that slowly grow distortion and shut down, spiky, notchy noise floors, high jitter, substandard performance, 16 bit DACs, or 20 bit DACs, filters that go crazy if you don't feed it with care, and people just talk about the unique sound then buy add ons and upgrades to fix what should not be broken. They made a DSD only DAC and surmised since it didn't sell well DSD was dead. They make gear without a remote despite customers and every review complaining about it. I guess it works.

Now in their defense, they did cover the little E(xpert) switch for driver install in the 4 page manual on page 2.

https://schiit.com/public/upload/PDF/modi_2_manual.pdf

Of course it is unclear if you should leave it in E for 192 khz after installing drivers or should it go back to S? Or do you need to switch back and forth for various sample rates? And if you do switch does that confuse the FPGA based burrito filter requiring a reset? And how do you reset it? Turn power switch on and off or is an unplug replug necessary. Or does unplugging and replugging USB do it? I tell you this company really is Schiit.

I hope Amir tests the Behringer next. I think it will perform better than this.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,386
Location
Seattle Area
I hope Amir tests the Behringer next. I think it will perform better than this.
Behringer is on the bench. Do you want me to measure that next or mess more with Schiit?
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,522
Likes
37,050
Behringer is on the bench. Do you want me to measure that next or mess more with Schiit?

Hey man you don't work for me. It would be nice to investigate the Schiit were I doing it I would be curious to find out. On the other hand I expect the Behringer will simply outlcass it and not be a broken design.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,386
Location
Seattle Area
As a reference, these are the measurements I made of the Schiit Bifrost:

Schiit BIFROST.png


See the massive jitter/modulation spikes on one of its inputs. We are talking distortion component rising to some -50 db or so! This is horrific.

Next was investigating computer activity influence on the DAC. Here is Media Player Classic:

Schiit BIFROST usb Media Player Classic.png


As you see the above problem is not there and output looks clean. But now let's play the same file using Windows Media Player:

Schiit BIFROST usb WMP.png


Notice the new skirt around our main tone. This happened for the duration that WMP was buffering the audio file. Once it finished reading it (a few seconds later), the activity went down and the output became the same as Media Player Classic! Clearly there was little system isolation in this $400 DAC.

My son also bought their headphone amp and short RCA cables. One channel went out and I traced it to one of the RCA cables failing. Opened it and it had horrible solder job and no strain relief so after a few plugs and unplugs, the center pin disconnected from the RCA jack. Had to send it back to them before 30 day return period expired so no pictures of that.

Not saying this new unit is just as broken but the dots connect and don't paint a good picture.
 
Top Bottom