• 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!

Measurement and Review of Burson Play DAC & Headphone Amplifier

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,388
Location
Seattle Area
This is a review and measurements of the Bursson Play DAC model "V6 Vivid." It is on loan from a member and retails for $549. It is a unique product in that it comes in form factor of a 5.25" drive bay so that you can either slide it into your desktop computer that way, or use it independently on your desk.

Burson Play DAC picture.jpg


Above you see it stacked in the middle of two Topping DACs (D50 and DX7).

The back has both ATA standard power connection for inside computer use and external 12 volt 6 amp (!) power brick.

The unit comes with a bunch of alternate op-amps that you can use to swap out what is in it. I did not play with any and tested it as I received it.

A rotary control changes the volume but requires many turns to go from min to max. Combined with rather slippery knob, made big changes difficult. But then again you get high accuracy for your level.

Speaking of level, this thing puts out some serious voltages/power. In both RCA and headphone out you get over 10 volts unloaded! It however clips at that level with a 33 ohm dummy load. Turning it down to 75% remedied that and resulted in a clean 5.7 volt output with very low output impedance of just 0.4 ohms:

Burson Play DAC Headphone Output Impedance.png


The rear power jack unfortunately did not make a good connection. Touching would cause the unit to power cycle. Not sure if it came broken this way or got that way as it travelled through multiple hands.

The unit worked plug-and-play in Windows 10 Creator's edition, allowing me to use and test it with Roon in bit-perfect WASAPI interface.

Let's get into measurements and see how it does. Please read my tutorial on my setup and how these tests are conducted: https://www.audiosciencereview.com/forum/index.php?threads/understanding-audio-measurements.2351/. Naturally all tests are done using the only input port which is USB so the test results are not as comprehensive as some of the other DACs that have S/PDIF input.

Measurements
I usually don't expect frequency response measurements to be eventful but they were here:

Burson Play DAC Frequency Response.png


As we see there is a 3 dB roll off by the time we get to 20 kHz. And the drop starts at around 6 kHz. This is far more steep than any DAC I have measured. Won't affect older folks like me but if you are still young, you will be cheated out of some high frequency response, resulting in the sound to be more mellow than with other DACs.

EDIT: further tests show that this droop only occurs at full volume. Dialing it down to 60% essentially makes the response flat.

Let's see how it does with noise and jitter:

Burson Play DAC Jitter and Noise Measurement.png


As we see the results are not competitive with similar/lower cost DACs from Topping. The noise floor is high and is modulated by some other signal. There is also an idle tone at around 14 kHz.

Note that the above was at lower volume to more or less match the Topping. If I turned up the volume higher, the noise floor increases substantially with it.

Next let's look at THD+N versus level:

Burson Play DAC THD+N distortion measurement.png


Here the noise and distortion are not only higher than the Topping DX7 but the two channels have varying amount of distortion+noise. I have not seen that much separation in measurements of other DACs.

I cut off the right side but if I let the graph go, distortion+noise sharply increases. So low-distortion listening comes at levels below max (by at least 20 dB).

Let's look at low level linearity by playing a very weak -90 dB (24 bit) sine wave:

Burson Play DAC Linearity Measurement.png


The response here is pretty good as long as volume control is set high. If I match levels to Topping, the sine wave would fall apart (not shown).

Subjective Listening
The much more powerful output stage allowed the Burson Play DAC to easily drive my most inefficient headphone, the Sennheiser HD650 to ear bleeding level. As you see in the first picture, I am listening to music as I am typing this with the volume level at just 26 out of 99. I can push it up to 60 or so but dare not after that.

I don't detect any noise when pausing the audio even when switching to most sensitive headphone I have, the Grado SR60e.

I am on a short time fuse so no time for extensive AB testing. But I say the measurement deficiencies don't seem to translate into audible issues.

Conclusions
Clearly the Burson Play is designed for muscle than beauty (in measurements). There are better DACs with much more refinement. What the Burson does well is to pump out tons of power through its headphone and RCA jacks. And it uniquely works as a drive bay in your desktop computer.

At this price, I can't recommend it but if you have inefficient headphones or want to listen at far higher levels than even I do, I won't stop you from purchasing it. :)

I will be returning this unit soon so if you have any questions, ask them quick! :D

----
If you like this review, please consider donating funds for these types of hardware purchase using Patreon (https://www.patreon.com/audiosciencereview), or upgrading your membership here though Paypal (https://audiosciencereview.com/foru...eview-and-measurements.2164/page-3#post-59054).
 
Last edited:

Jimster480

Major Contributor
Joined
Jan 26, 2018
Messages
2,880
Likes
2,032
Location
Tampa Bay
Awesome review Amir!
The unit is actually on "loan" from me which is on "loan" from Burson who sent it to me for me to review it.

Glad to have measurements from it, I shared the link with Burson now!
 

garbulky

Major Contributor
Joined
Feb 14, 2018
Messages
1,510
Likes
827
You noticed that??? :D
I enjoyed the review! Do you ahve the equipment for power output in to different loads types of measurements? Like how many mwatts it can put out in to a 300 ohm load and the quality etc
We see very little data from headphone manufacturers. They'll either put a power output at only 30 ohms which imo is not helpful. I usually use a HD600. And then they do it in Volts which also isn't terribly informational. I like a 30, 60, 100, 150, 300, 600 ohm measurement.
Also it would be fantastic as to what kind of THD/fr responses we get in to those loads at significant power output.

Also I would love to see more detail on the power supply. Does it have a torroid, switching?
 

Jimster480

Major Contributor
Joined
Jan 26, 2018
Messages
2,880
Likes
2,032
Location
Tampa Bay
I enjoyed the review! Do you ahve the equipment for power output in to different loads types of measurements? Like how many mwatts it can put out in to a 300 ohm load and the quality etc
We see very little data from headphone manufacturers. They'll either put a power output at only 30 ohms which imo is not helpful. I usually use a HD600. And then they do it in Volts which also isn't terribly informational. I like a 30, 60, 100, 150, 300, 600 ohm measurement.
Also it would be fantastic as to what kind of THD/fr responses we get in to those loads at significant power output.

Also I would love to see more detail on the power supply. Does it have a torroid, switching?
The power brick for the Burson is a switching external box, its not linear.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,388
Location
Seattle Area
I enjoyed the review! Do you ahve the equipment for power output in to different loads types of measurements? Like how many mwatts it can put out in to a 300 ohm load and the quality etc
We see very little data from headphone manufacturers. They'll either put a power output at only 30 ohms which imo is not helpful. I usually use a HD600. And then they do it in Volts which also isn't terribly informational. I like a 30, 60, 100, 150, 300, 600 ohm measurement.
Also it would be fantastic as to what kind of THD/fr responses we get in to those loads at significant power output.
Yes I have the gear. It is time which I am in shortage of right now with so much to review. :)
 

trl

Major Contributor
King of Mods
Joined
Feb 28, 2018
Messages
1,967
Likes
2,523
Location
Iasi, RO
Really great site Amir, thank you!

I had tested recently the PLAY myself and found absolutely flat frequency response. However, I suggest you do another measurement or change the device with another one, in case you got a lemon product.

Also, noise is absolutely the same, no matter the volume knob position (min or max); the volume is controlled digitally inside the DAC itself and is not affecting the SNR. Basically, if music is stopped or paused, the noise with volume to the min. is the same as the noise measured when volume is to max. You can also take out the USB XMOS PCB and you'll have the same noise as well. :)

With volume to the max. on 30 ohms resistive load I got about 8V RMS perfect sinewave, no visible distortions, instead of 5.7V RMS as you got in your tests, so again...either you got a lemon product, either something else got wrong.

Changing the original SMPS with a LPSU will not change much the noise-ratio, but some interesting mods could be done inside the box, for those that really like electronics, with dedicated LPSU for each rail from the analog parts and also for the digital parts. :)

At his moment, strictly noise-related, I got in order of SNR the following DACs at home:
- ODAC (the most noisyer from all I have)
- ROLAND UA-M10
- ASUS Essence One Muses MKii (modded, dedicated PSU for the DAC)
- BURSON Play (modded, dedicated PSUs for DAC and analog part)
Basically, PLAY is the less noisy DAC I got and I can hear the -105dB dithered WAV file with ease. In compare, with ODAC+O2 I can't differentiate the -105dB sound from the background noise, instead Play is doing it with ease now and without any background noise. Using Solo2 and K550 for this test, as these are the most sensitive cans I got.
 
Last edited:
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,388
Location
Seattle Area
You need an assistant sir...not raising my hand, just saying...
The thought is front and center every time I get a message asking when their device is being tested! :)

But it is a "good" problem to have, i.e. a lot of gear to test. With winters here not being that much fun, I am able to dedicate a lot of time to it.

Alas, I have been ill the last two weeks with fair amount of pain which gets worse when I sit at my lab workstation. So I am not as productive as I like to be.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,388
Location
Seattle Area
Really great site Amir, thank you!

I had tested recently the PLAY myself and found absolutely flat frequency response. However, I suggest you do another measurement or change the device with another one, in case you got a lemon product.

Also, noise is absolutely the same, no matter the volume knob position (min or max); the volume is controlled digitally inside the DAC itself and is not affecting the SNR. Basically, if music is stopped or paused, the noise with volume to the min. is the same as the noise measured when volume is to max. You can also take out the USB XMOS PCB and you'll have the same noise as well. :)

With volume to the max. on 30 ohms resistive load I got about 8V RMS perfect sinewave, no visible distortions, instead of 5.7V RMS as you got in your tests, so again...either you got a lemon product, either something else got wrong.
Welcome to the forum trl. The unit came with the top part of the enclosure loose. So it is possible other folks have been into it and changed op-amps and such. I have to send the thing back but I will re-confirm the few of the measurements to see if I made a mistake or not.

For now, my noise test was with the 12 kHz J-Test signal which I assume you have not run. I did not try it with the rest of my tests.

Which op-amp do you have in yours by the way? That would certainly impact noise and clipping points.
 

trl

Major Contributor
King of Mods
Joined
Feb 28, 2018
Messages
1,967
Likes
2,523
Location
Iasi, RO
I was doing the J-test from ARTA, but I am mostly relying on my ears when I say that there's no noise difference between volume to min or to max. It also makes sense here, because this is not a volume-in-the-middle between voltage amplification stage and the final stage (or similar). Not sure I can trust my ARTA Jitter test, but I'm 110% sure the frequency response is completely flat; also did A/B test between PLAY and other DACs I own, including signal test files up to 20KHz.

BURSON-PLAY_30ohm.png

That's with a -1dB input signal: 7.4V RMS. With 0dB input signal it almost gets to 8V RMS.


Frequency_response.png

Frequency response is quite flat to my PLAY DAC

PLAY_jitter_48KHz.png

J-test with ARTA

 

Attachments

  • BURSON-PLAY_30ohm.png
    BURSON-PLAY_30ohm.png
    65.8 KB · Views: 257
  • BURSON-PLAY_30ohm.png
    BURSON-PLAY_30ohm.png
    65.8 KB · Views: 248
  • BURSON-PLAY_30ohm.png
    BURSON-PLAY_30ohm.png
    65.8 KB · Views: 261
Last edited:
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,388
Location
Seattle Area
Ok, mystery on frequency response solved. On a hunch, I lowered the volume from max and response became nearly flat:

Burson Play DAC Frequency Response vs level.png


This is as a result of feedback loop at higher maximum volume relative to 20 kHz bandwidth we need.

You can see this nicely in this type of graph about feedback:

NV_0701_Marston_FIGURE6.jpg
 

trl

Major Contributor
King of Mods
Joined
Feb 28, 2018
Messages
1,967
Likes
2,523
Location
Iasi, RO
Probably your PLAY was distorting; please, bear in mind that volume should never get to the max. when doing the tests, just measure the output volume till 2V RMS (RCA plugs are directly connected to 6.3mm plug, so you'll get the full 7-8V RMS on the RCA plugs as well if volume is to the max.!). Something between 50-60% of volume should suffice for the tests.

When I did the above test I had either 3xSS V6 Classic, either 2xMUSE8820+1xMUSES8920 (sorry I can't remember exactly).
Right now I'm using 3xAD8599, as the noise decreased even more.

In case you want to conduct an A/B test between 2 DACs and test the noise, then this is the -105dB dithered wave file I was talking about: https://www.webel.com.au/content/au...-and-online-generators-and-online-audio-tests.
A 4X-gain low-noise headphone amplifier should be good enough to do this test. I was using the Objective2 with a gain of 3.7X that has virtually no noise and a passive A/B test DIY device with switches and I was able to compare all the DACs I have (DACs were not feed with music!).

P.S.: After adding LPSUs I wasn't doing anymore tests (I get a nasty 12KHz on ARTA even with RCA cable not connected to DAC and even if it's in short-circuit).
WP_20180201_08_11_42_Pro.jpg
AdobePhotoshopExpress_1d3d8fd0b960492e89d782bec5d70fbb.jpg
AdobePhotoshopExpress_82c34cf6022143a3aa1c6d3fc08d5686.jpg
 
Last edited:
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,388
Location
Seattle Area
Probably your PLAY was distorting; please, bear in mind that volume should never get to the max. when doing the tests, just measure the output volume till 2V RMS (RCA plugs are directly connected to 6.3mm plug, so you'll get the full 7-8V RMS on the RCA plugs as well if volume is to the max.!). Something between 50-60% of volume should suffice for the tests.
I test all DACs using their RCA jacks without attenuators in the path. It is a mistake for them to connect the headphone output to RCA jacks. All other DACs have independent outputs.

If I start messing with volume level I would have to do that for every DAC and it becomes a huge, messy task.

That said, if this DAC is used with a pre-amp, you will likely need to turn down its volume which I did in my jitter tests.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,388
Location
Seattle Area
I just confirmed the output at full volume and it definitely clips with 33 ohm load:

upload_2018-2-28_10-20-52.png


Green is unloaded, cyan is loaded with 33 ohm. You can see the voltage swings there which is +- 14 volts peak, and 10.3 volt RMS.
 

Jimster480

Major Contributor
Joined
Jan 26, 2018
Messages
2,880
Likes
2,032
Location
Tampa Bay
Try to lower the volume down to 2vRMS and see, its unfortunate you cannot test for ENOB with USB yet.

But yes having the headphone output going to the RCA really makes no sense... Try plugging in a headphone at the same time as plugging something into the RCA.... I got some feedback noise in my 1More Triple over ear headphones with that, and with my Denon Ah-mm400....
 

Frank Dernie

Master Contributor
Forum Donor
Joined
Mar 24, 2016
Messages
6,445
Likes
15,780
Location
Oxfordshire
I am no expert on the digital side of things but analogue volume controls rarely need "full volume" for rated output and in many traditional amps the rated power is achieved at the nominal sensitivity with the volume control well below maximum, allowing a bit of headroom for inputs with a lower output. I don't know whether this is a limitation with this product but "full volume" on a traditional amp may well be well into clipping.
It is 30 years + since I checked myself but the amp I had then went into clipping using the rated input voltage with the volume control around 1 o'clock, where the end stop was around 4 o'clock.
 
Top Bottom