Pixel USB-C Audio out measurements (Stepped sine tests added 1/24))

[zed011]

With respect, color me skeptical.
That would not pass THX certification, and the Razer adapter is THX certified.
Testing results can be sensitive to testing configuration and equipment,
and something in your setup might be affecting your results,
especially given your distortion results on other devices.
I've not run rigorous tests,
but my subjective impression of the Razer adapter is quite positive.

Oh dear. Thx certified means nothing , it's a marketing term. My Logitech speakers are thx certified does that mean it's good? No, my bookshelf speakers blow it away. No one serious in audio believes in THX certification, again it's a marketing label thrown out to nonchalant consumers

 

[Galactus]

Thanks, great idea.
On order for next week and some testing.

Your a legend

 

[syunz]

I just got it last night and looked at just a couple of things before getting on a flight.

The output impedance is about 6 ohms, much higher than the old one. Definitely a problem for multi transducer iem's.

Interestingly, the alsa parameters are much different and it supports both recording and playback at 44 and 48 k sample rates. The original only supported 48k playback.


Will give it a through evaluation in about a week.

Will you be looking into the new apple usb-c to 3.5mm that launched with the new ipad pro? It's only $9 and it's looking like a good value comparatively to the others if it performs well.

 

[bjrmd]

Absolutely, should be able to do next week.

 

[JNavas2]

Oh dear. Thx certified means nothing , it's a marketing term. My Logitech speakers are thx certified does that mean it's good? No, my bookshelf speakers blow it away. No one serious in audio believes in THX certification, again it's a marketing label thrown out to nonchalant consumers

So you're accusing THX of fraud? That's pretty serious. Got anything to back that up?
“Extraordinary claims require extraordinary evidence.” —Carl Sagan
FYI, THX isn't a guarantee of excellence -- it's a quality assurance system.
As such, it's directly relevant to the issue of distortion in USB-C to 3.5 mm adapters.
(Also FYI, speaker certification is separated into classes.)

Sure, you are certainly free to believe what you want.
I triple checked everything and compared to the Dragonfly red and Google DAC at the same time, same equipment.

Yes, but as I said, results can be sensitive to the testing environment and methodology, and the fact that you're finding distortion is all non-Dragonfly products suggests that there might be an issue in your results. Do you have any confirmation?
Or explanation of how the Razer adapter could pass THX certification?
Otherwise we just have your unconfirmed data. And I'm not hearing distortion.
"I believe in observation, measurement, and reasoning, confirmed by independent observers." —Isaac Asimov

 

[zed011]

Ever wonder why the top audio brands don't give a hoot about THX? They don't list their test methodologies on their site and use catch phrases on their site. THX means absolutely nothing to me and any serious about audio. It's more like a logo you out on things and hope it sells better because uninformed consumers think it's better. You got any proof to back up that THX makes any difference? You don't have to throw quotes around to sound wise...

Razor owns thx that's why they can slap anything thx on it until some technical people review the device and find out it's probably not better than any decent dac

 

[bjrmd]

and the fact that you're finding distortion is all non-Dragonfly products suggests that there might be an issue in your results.

No.
If you look at the test results for the Dragonfly, you will see they are not much different than the Google DAC, HTC, and some of the other dongles in regards to distortion. The only real major plus for the DFR is the super low output impedance, and better than average crosstalk. So despite owning the DFR, I actually use the HTC dongle on my Senn 800s, the old Google DAC on my se846, as well as the new RHA CL2. I also have the AKG N5005 using the apt X bluetooth adapter, Oriolus Mk2 on the Google Dac.

As far as listening tests, that's another can of worms - if you want that approach check out Head-fi.

I also do not claim that any of the distortion banding is audible.
Many folks just like to know.
More important to me in the testing is the output impedance and crosstalk (under load) which is generally ignored elsewhere.

 

[JNavas2]

Ever wonder why the top audio brands don't give a hoot about THX? They don't list their test methodologies on their site and use catch phrases on their site. THX means absolutely nothing to me and any serious about audio. It's more like a logo you out on things and hope it sells better because uninformed consumers think it's better. You got any proof to back up that THX makes any difference? You don't have to throw quotes around to sound wise...
Razor owns thx that's why they can slap anything thx on it until some technical people review the device and find out it's probably not better than any decent dac

So no confirmation, just more accusations.

No.
If you look at the test results for the Dragonfly, you will see they are not much different than the Google DAC, HTC, and some of the other dongles in regards to distortion. The only real major plus for the DFR is the super low output impedance, and better than average crosstalk. So despite owning the DFR, I actually use the HTC dongle on my Senn 800s, the old Google DAC on my se846, as well as the new RHA CL2. I also have the AKG N5005 using the apt X bluetooth adapter, Oriolus Mk2 on the Google Dac.
As far as listening tests, that's another can of worms - if you want that approach check out Head-fi.
I also do not claim that any of the distortion banding is audible.
Many folks just like to know.
More important to me in the testing is the output impedance and crosstalk (under load) which is generally ignored elsewhere.

So no confirmation, but we agree that Razer distortion is not audible.

 

[zed011]

You never provided proof either and also you're the one who started defending THX like you work there

Anyway THD is hardly audible on any gear that is not complete crap and it's also not considered as one of the most important aspects of sound reproduction that makes a notable audible difference. It also matters at what frequency it's measured at. Flat frequency response, dynamic range and output impedance matter more. None of us unless we work in the field and can afford thousands of dollars of equipment will be able to analyze audio seriously. I suggest reading this very detailed and informative blog for anyone who is serious about audio as it's one of the best sites I have ever seen about audio particularly about dacs and headphone amps with measurements
to back up his work. httpguy.blogspot.com/2011/02/rightmark-audio-analyzer-rmaa.html?m=0

 

[bjrmd]

One of the best sources of testing and quality assessment info is the Nwavguy. His blog is a few years old, but still spot on.

In addition, some people with good hearing, experience and sensitive Iem's, may be able to notice the background noise floor and harmonic distortion. The problem I pointed out originally is distortion banding even at very low gain, so the music is competing with the distortion.

I personally don't see the logic of buying a dongle that has inferior properties for a higher price.

 

[JNavas2]

...
In addition, some people with good hearing, experience and sensitive Iem's, may be able to notice the background noise floor and harmonic distortion. The problem I pointed out originally is distortion banding even at very low gain, so the music is competing with the distortion.
I personally don't see the logic of buying a dongle that has inferior properties for a higher price.

As lawyers say, "Assumes facts not in evidence", as in lack of confirmation of "inferior".
I personally don't see the logic of buying a dongle at a much higher price that doesn't deliver an audible difference (e.g., Dragonfly).
When I purchased my Razer dongle, the Google dongle was priced at $20, with poor latency, which matters much more to me than claimed but inaudible sound differences, and the Razer was essentially the same price. (Google now has an improved version at a lower price.)
"Different strokes for different folks."

 

[BaronInkjet]

OK guys, here is a serious competitor for you: VE Odyssey. Only can get it from China.

https://www.veclan.com/engappliance_sel_one?eng_ApplianceVo.eac_id=32

$10 + $4 shipping. I got mine in about two weeks. See the HeadFi discussion of it here:

https://www.head-fi.org/threads/ve-odyssey-discussions-and-impressions-thread.886989/

I have been using it like I used the Google one, with Neutron USB driver. But where the Google one is often flat and lifeless into my HiFiMan RE600s, the Odyssey has plenty of punch. It's not quite as expansive an experience as Astell & Kern BT, but is very, very good to my ears. Especially for only $14! I can't remember how many of these things I have bought, but this is the only one (besides A&K BT) that I am keeping.

Happy listening!

 

[holtmann]

Apple USB-C to 3.5mm adapter

Out of curiosity I got the USB-C to 3.5mm headphone adapter from Apple (meant for the new iPad Pro). Plugged into an Android device (I tested on a Mi5s) it's recognised as an USB audio device.
Here are some first results:
1) Playing music using Spotify (Android Audio) the sounds is clean, but volume levels are extremely low. Much lower than the Google USB C to 3.5mm adapter or any other adapter I've tested. I tested with low impedance headphones.
2) Playing music using a dedicated USB Audio App (I used USB Audio Player and Neutron) there is only digital noise.

I assume it's working with normal volume levels on an iPad Pro (I don't own one) and in Android maximum audio levels are low due to mixer/digital gain settings.

Here is the USB device info for the adapter in case anyone is interested:
Device Info
Device Path: /dev/bus/usb/001/004
Device Class: Miscellaneous (0xef)
Vendor ID: 05ac
Vendor Name (reported): Apple, Inc.
Vendor Name (from DB): Apple, Inc.
Product ID: 110a
Product Name (reported): USB-C to 3.5mm Headphone Jack Adapter
Product Name (from DB): not found

Additional Info
Interface #0
Class: Audio Device (0x1)
Endpoint: #0
Address : 0x81 (10000001)
Number : 1
Direction : Inbound (0x80)
Type : Intrrupt (0x3)
Poll Interval : 1
Max Packet Size: 6
Attributes : 000000011
Interface #1
Class: Audio Device (0x1)
Endpoints: none
Interface #2
Class: Audio Device (0x1)
Endpoint: #0
Address : 0x02 (000000010)
Number : 2
Direction : Outbound (0x0)
Type : Isochronous (0x1)
Poll Interval : 1
Max Packet Size: 288
Attributes : 00000001101
Interface #3
Class: Audio Device (0x1)
Endpoint: #0
Address : 0x02 (000000010)
Number : 2
Direction : Outbound (0x0)
Type : Isochronous (0x1)
Poll Interval : 1
Max Packet Size: 192
Attributes : 00000001101
Interface #4
Class: Human Interaction Device (0x3)
Endpoint: #0
Address : 0x85 (10000101)
Number : 5
Direction : Inbound (0x80)
Type : Intrrupt (0x3)
Poll Interval : 1
Max Packet Size: 64
Attributes : 000000011
Interface #5
Class: Human Interaction Device (0x3)
Endpoint: #0
Address : 0x86 (10000110)
Number : 6
Direction : Inbound (0x80)
Type : Intrrupt (0x3)
Poll Interval : 1
Max Packet Size: 64
Attributes : 000000011
Interface #6
Class: Audio Device (0x1)
Endpoint: #0
Address : 0x81 (10000001)
Number : 1
Direction : Inbound (0x80)
Type : Intrrupt (0x3)
Poll Interval : 1
Max Packet Size: 6
Attributes : 000000011
Interface #7
Class: Audio Device (0x1)
Endpoints: none
Interface #8
Class: Audio Device (0x1)
Endpoint: #0
Address : 0x02 (000000010)
Number : 2
Direction : Outbound (0x0)
Type : Isochronous (0x1)
Poll Interval : 1
Max Packet Size: 288
Attributes : 00000001101
Interface #9
Class: Audio Device (0x1)
Endpoint: #0
Address : 0x02 (000000010)
Number : 2
Direction : Outbound (0x0)
Type : Isochronous (0x1)
Poll Interval : 1
Max Packet Size: 192
Attributes : 00000001101
Interface #10
Class: Human Interaction Device (0x3)
Endpoint: #0
Address : 0x85 (10000101)
Number : 5
Direction : Inbound (0x80)
Type : Intrrupt (0x3)
Poll Interval : 1
Max Packet Size: 64
Attributes : 000000011
Interface #11
Class: Human Interaction Device (0x3)
Endpoint: #0
Address : 0x86 (10000110)
Number : 6
Direction : Inbound (0x80)
Type : Intrrupt (0x3)
Poll Interval : 1
Max Packet Size: 64
Attributes : 000000011
Interface #12
Class: Audio Device (0x1)
Endpoint: #0
Address : 0x81 (10000001)
Number : 1
Direction : Inbound (0x80)
Type : Intrrupt (0x3)
Poll Interval : 1
Max Packet Size: 6
Attributes : 000000011
Interface #13
Class: Audio Device (0x1)
Endpoints: none
Interface #14
Class: Audio Device (0x1)
Endpoint: #0
Address : 0x02 (000000010)
Number : 2
Direction : Outbound (0x0)
Type : Isochronous (0x1)
Poll Interval : 1
Max Packet Size: 288
Attributes : 00000001101
Interface #15
Class: Audio Device (0x1)
Endpoint: #0
Address : 0x02 (000000010)
Number : 2
Direction : Outbound (0x0)
Type : Isochronous (0x1)
Poll Interval : 1
Max Packet Size: 192
Attributes : 00000001101
Interface #16
Class: Human Interaction Device (0x3)
Endpoint: #0
Address : 0x85 (10000101)
Number : 5
Direction : Inbound (0x80)
Type : Intrrupt (0x3)
Poll Interval : 1
Max Packet Size: 64
Attributes : 000000011
Interface #17
Class: Human Interaction Device (0x3)
Endpoint: #0
Address : 0x86 (10000110)
Number : 6
Direction : Inbound (0x80)
Type : Intrrupt (0x3)
Poll Interval : 1
Max Packet Size: 64
Attributes : 000000011

 

[bjrmd]

Apple USB-C DAC measurements

Note- the 3Khz banding is an artifact present even with silence/Dac off

Max output stock: 80 mV (!!!)
Max output rooted with appropriate gain adjustments is 1 Volt

Crosstalk unloaded 75 dB
Crosstalk loaded 40 dB

Output impedance .2 Ohms (excellent)

Distortion at 50 mV loaded with Shure se846:

Compared to Google Dac (loaded same as above, same sitting):

Compare to Dragonfly red (same load, same sitting):

Will start working on getting some more voltage out of this nice little guy if possible (with tinymix and root).
More comments to follow......

Max voltage on a Win10 laptop = .5 V

Difference in distortion banding using the Apple Dac vs the Google Dac on the Pixel 2 XL (Nov patch) vs a Sony Xperia (Oreo, Sept patch). All tests are with an se846 as the load, all done at the same sitting, voltage out.

Pixel with Google Dac:

Sony (Oreo) with Google Dac:

Pixel with Apple Dac:

Sony with Apple Dac:

Additional comments:
The Apple Dac is capable of .5 V on a PC, but not on stock Android devices (Pixel, Sony).
The Nov patch for Pie has "fixed" the stock gain issues in both line out and headphone mode
As seen in the Sony (android 8.0) tests with the Google Dac, the noise/distortion was there - an Oreo problem that used to plague the Pixel as well.
But, the Sony did not show any distortion increase (compared to the Pixel) using the Apple Dac. So although the driver does not take advantage of full voltage, the gain control appears to be optimized in the Dac itself (?).

Edit as of 11/18 the Apple Dac can put out 1 Volt, it just needs the correct gain settings (exactly like the Dragonfly).

The Apple unit is about half the size of the Google (original) DAC.

 

[holtmann]

So with max 80mV output the extremely low maximum volume is no surprise. I assume output levels when connected to an iPad will be around 1V (typical for most apple devices and also the lighting->3.5mm adapter)
Any idea why the USB AudioPlayers are not working with this dongle - just a driver issue?

 

[BaronInkjet]

A little more on this: I tried it today in the car line in to amp, and don't like it nearly as much as my Pixel 2 device. Am guessing that it's same output regardless of what is connected. Still, for 16 ohm IEM sounds way, way better than Pixel device though.

OK guys, here is a serious competitor for you: VE Odyssey. Only can get it from China.

https://www.veclan.com/engappliance_sel_one?eng_ApplianceVo.eac_id=32

$10 + $4 shipping. I got mine in about two weeks. See the HeadFi discussion of it here:

https://www.head-fi.org/threads/ve-odyssey-discussions-and-impressions-thread.886989/

I have been using it like I used the Google one, with Neutron USB driver. But where the Google one is often flat and lifeless into my HiFiMan RE600s, the Odyssey has plenty of punch. It's not quite as expansive an experience as Astell & Kern BT, but is very, very good to my ears. Especially for only $14! I can't remember how many of these things I have bought, but this is the only one (besides A&K BT) that I am keeping.

Happy listening!

 

[danzel]

Razer phone 2 comes out with new audio adapter and initial impression from gsmarena says that it was "loud and has a perfectly accurate output".

 

[jasenko]

Hi guys, just to chip in with my observations... I got mine (Apple dongle) today and did a few tests. As everyone pointed out, the audio coming straight out of android is low volume but very clean. Tested using USB Audio Player Pro and the sound was amazing. Plenty of volume and very, very balanced sound. Now that UAPP supports GPM, I am sorted...

To my ears sounds exactly the same as the Apple original dongle on the iPhone.

I also tested it on the Pixelbook in Spotify. It is able to produce so much volume, the ears can start bleeding. Again, sounds is much nicer that Pixel dongle. No noticeable distortion. I think my Dragonfly Black will now end up retired in the drawer for good.

 

[brojarbor]

@jasenko
Speaking of "I got mine", which one specifically are you referring to?

I just wish to make sure as your comparison with Audioquest Black makes it even more attractive.

Wysłane z mojego HTC U Ultra przy użyciu Tapatalka

 

[bjrmd]

Apt X HD comparison (new Shure BT2) with wired Google DAC:
Testing setup Pixel 2XL and EARS with the Shure se846 iem
Freq response identical:

Noise and distortion very similar (this would be different than a pure scope tracing of audio jack properties):
Shure BT2:

Google DAC:

Bottom line:
The new Shure aptx HD unit has transparent audio output compared to wired (at least according to the test platform used here). Whether some individuals can hear the difference is a different issue. I could not.