• 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!

Review and Measurements of iBasso DX120 DAP

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,381
Location
Seattle Area
This is a review and detailed measurements of the iBasso DX120 portable Digital Audio Player (DAP). It is on kind loan from a member and costs USD $319 from Amazon including Prime shipping.

This is a chunky device if you are used to portable players from likes of Apple or your mobile phone:

iBasso DX120 Digital Audio Player DAP Review.jpg

The user interface is attractive and quite fast in browsing menu options. This was better than I expected as was the nice color scheme and graphics.

I usually go to company website to learn more but had a heck of a time doing that with this product. The moment you select DX120 you get this graphics laden page that is a pain in the neck to navigate. You can't scroll through it and have to keep clicking on radio button on the right. Some web designer went crazy designing this page.

I was very pleased to find that using the USB cable, I could treat this as a Windows DAC. This enabled me to run my full suite of measurements which I can't normally run when I have to play content manually in the player. Yes, I tested both modes of operation to make sure DAC mode is representative of DAP playback. See measurements later.

There are three sets of "headphone jacks" at the bottom. First one is 3.5mm and is for line out. The second is also 3.5 mm but is for headphone. The last is 2.5 mm which is for "balanced" headphone output. I tested all three.

As far as cabling, there is an oddest inclusion: a "burn-in cable!" Yes, it does what it says. You are supposed to plug it into the headphone jack and let the unit play for many hours to "form capacitors and such." It seems to have a small dummy load it and you are warned to not run the player at full volume into it.

Let's get into measurements and see how she does.

Measurements
I started with USB DAC mode and line out. Here is our dashboard view:

iBasso DX120 Digital Audio Player DAP Measurements.png


Output is a bit low at 1.8 volts (like to see 2 volts) but otherwise, performance seems pretty good for a portable device. By dialing the digital level by 1 dB I could get a SINAD of over 100 dB. Here is how it ranks as is:
iBasso DX120 Digital Audio Player DAP SINAD Measurements.png


It barely misses our tier 2 performance of all DACs tested.

Dashboard view with headphone out generated similar results (high gain/max volume):
iBasso DX120 Digital Audio Player DAP Local Player Dashboard Measurements.png


Here, we do get to 101 dB which would put it in tier 2. So far so good.

Dynamic range is very good too:
iBasso DX120 Digital Audio Player DAP Dynamic Range Measurements.png


Intermodulation test is pretty decent too:
iBasso DX120 Digital Audio Player DAP IMD Measurements.png


Jitter is unusually clean:

iBasso DX120 Digital Audio Player DAP Jitter Measurements.png


Surely linearity would be bad, no?
iBasso DX120 Digital Audio Player DAP Linearity Measurements.png


No, that is excellent too!

How about multi-tone?
iBasso DX120 Digital Audio Player DAP Multitone Measurements.png


Nothing bad jumps out.

I usually skip frequency response tests because digital products do so well here. In this case, I decided to run it because of a handful of "sound modes" in DX120:
iBasso DX120 Digital Audio Player DAP Frequency Response Measurements.png


What the heck happened here? Not only do the different modes do nothing, but there is a roll off that starts at just 1 kHz! I measured and remeasured with the same results. To make sure my instrumentation is OK, I ran the same test on Topping D10 (in dashed red) and the output is as one expects: essentially ruler flat response. I am at a loss as to what is happening here. And if correct, how people have not noticed such a sharp roll off.

FYI to make sure firmware bug was not in play, I updated the unit to the latest available (2.4.59) release just this month. Yes, I checked the EQ and shut it off but it made no difference. I will do some more testing at different sample rates and see if that makes a difference.

Next unusual result was THD+N versus frequency:

iBasso DX120 Digital Audio Player DAP THD vs Frequency Measurements.png


This is when I started to go crazy. I just saw the identical sharp peak starting at 5 kHz when I measured the Marantz AC8805 processor. Let's dig in just the same to see if the issue is lack of proper filtering in the output of the DAC:

iBasso DX120 Digital Audio Player DAP Digital Filter Measurements.png


It is. The default which is the slow roll off is horrendous (magenta). But even fast roll off takes its time. So is my instrumentation broken? Let's run the identical test on Topping D10 again (red):

iBasso DX120 Digital Audio Player DAP Digital Filter Compared to Topping D10 Measurements.png


The D10 is performing as we expect so instrumentation is fine. But how could two radically different products have the same problem? Could it be that they both use the same DAC? They do! Both the Marantz AV8805 and iBasso DX120 use the AKM AK4459 dac chip! Fortunately unlike ESS, AKM publishes their spec so we can look up the performance metric for the different filters. Here it is for Sharp roll off:
iBasso DX120 Digital Audio Player DAP AK4495 Filter Specs Measurements.png


We are supposed to have 100 dB of attenuation once we get to 24.1 kHz. I put the blue cursor there and we are only seeing 8 dB of reduction!

Can both of these companies be programming the AKM DAC incorrectly? I am at a loss as to what is going on here. Happy to take suggestions for additional tests to figure out what is going on. Hopefully we can get iBasso to investigate this (no hope of getting through to anyone at Marantz).

Let's keep going and see how well the headphone amp works starting with performance at 300 ohm:
iBasso DX120 Digital Audio Player DAP Power at 300 Ohm Measurements.png


We have decent level of noise although that may be due to roll off of the high frequencies. Power though is very low at just 11 milliwatts. For perfection I like to see 100 milliwatts so we are quite far from that.

At 33 ohm, the situation does not improve much:
iBasso DX120 Digital Audio Player DAP Power at 33 Ohm Measurements.png


Testing balanced output though gives us hope:

iBasso DX120 Digital Audio Player DAP Power at 50 Ohm Measurements.png


As expected, we get four times the power with lower noise and distortion to boot.

Volume control is digital so channel balance is perfect:


iBasso DX120 Digital Audio Player DAP Channel Imbalance Measurements.png




Output impedance is a very good 1 ohm:

1551391489568.png



Listening Tests
I still have not had time to fashion a balanced headphone cable for my Sennheiser HD-650. So had to use the normal 3.5 mm jack. As expected, there is just no "oomph" there. I would be listening at near max volume for a lot of music in my library.

The situation improved some with the Hifiman HE-400i. Depending on music, there was decent volume and bass. In some cases though, I was back to full volume and wishing there was more.

If you have the option of using balanced connection, you should be in much better shape. Or alternatively use more sensitive IEMs or headphones.

Conclusions
The iBasso DX120 starts the show well with good distortion and jitter metrics. Then we hit a major snag with frequency response that has very large high frequency roll off. That then got compounded with digital filter settings which seem quite wrong. If they can fix these two bugs, I would be able to put the DX120 on my recommended list. As it is, it has missed the basics.

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

Just collected all of my pennies to afford the nice Carne Asada we had for dinner tonight at the local restaurant. Before my pocket develops a hole, would appreciate some replenishment of those funds 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).
 

Guermantes

Senior Member
Joined
Feb 19, 2018
Messages
484
Likes
561
Location
Brisbane, Australia
That super slow filter response looks like a classic zero-order hold sinc function at a sampling rate of 48 kHz (i.e. no filter at all). Is the DAP resampling 44.1 kHz to 48 kHz?
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,381
Location
Seattle Area
Can we also expect to see measurements of smartphones in the future?
I have done a couple before. I just don't have good access to more.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,381
Location
Seattle Area
Is the DAP resampling 44.1 kHz to 48 kHz?
That would be one explanation of what is going on. What is strange is that we see the same thing in Marantz so I am thinking it has something to do with programming the AKM DAC.
 

Jaimo

Active Member
Joined
Feb 16, 2018
Messages
198
Likes
179
Location
Toronto, Canada
AKM DAC's need a first order Low Pass Filter at their output. Could it be that this is missing from these designs?
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,381
Location
Seattle Area
AKM DAC's need a first order Low Pass Filter at their output. Could it be that this is missing from these designs?
The recommended filter is for very high frequencies though:

1551398542461.png

It wouldn't solve the issues I am seeing.
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,198
Likes
16,981
Location
Riverview FL

Biblob

Addicted to Fun and Learning
Joined
Sep 13, 2018
Messages
633
Likes
600
In the subjective listening test there is no mention of the frequency roll-of at all. Is it because you can't hear it anymore, or is there something else at play? ;)
 

tecnogadget

Addicted to Fun and Learning
Joined
May 21, 2018
Messages
558
Likes
1,011
Location
Madrid, Spain
Great to see an iBasso product here Amirm :D !!! Just a shame it has such a slouch performance concerning FR and roll-off filter.
Hopefully, my iBasso DX90 that uses dual mono ES9018K2M does not measure with such artifacts.
 

JohnBooty

Addicted to Fun and Learning
Forum Donor
Joined
Jul 24, 2018
Messages
632
Likes
1,555
Location
Philadelphia area
In the subjective listening test there is no mention of the frequency roll-of at all. Is it because you can't hear it anymore, or is there something else at play? ;)
Yeah, I'm curious about this too. The frequency response graph looks borderline unlistenable. Like putting some thick blankets between your ears and the headphones.

Yet the subjective listening impression seemed fine, aside from a lack of volume on the Sennheisers?
 

Frank Dernie

Master Contributor
Forum Donor
Joined
Mar 24, 2016
Messages
6,445
Likes
15,779
Location
Oxfordshire
Yeah, I'm curious about this too. The frequency response graph looks borderline unlistenable. Like putting some thick blankets between your ears and the headphones.

Yet the subjective listening impression seemed fine, aside from a lack of volume on the Sennheisers?
I wouldn't expect it to be unlistenable and my guess is that anybody who likes record players may well love it and say it has a lovely analogue sound. That FR is very similar to an expensive Koetsu pickup cartridge.
 

headwhacker

Active Member
Forum Donor
Joined
Dec 25, 2018
Messages
165
Likes
224
Over at HF mentioned it’s an xmos firmware bug and supposed to be fixed on the next update.

Ibasso advise to set the filter to “super slow” when using it as usb dac. Could you verify that Amir?
 

chashaodan

Member
Joined
Oct 28, 2018
Messages
7
Likes
3
What about the balanced out performance with 33ohms loads and no loads?
Here is the official specs:
006qjSbmgy1fuicjya29xj31ba0kujtj.jpg
006qjSbmgy1fuimgxo3z4j31hb0njafj.jpg
 
Last edited:

chongy

New Member
Joined
Jan 10, 2019
Messages
1
Likes
0
Over at HF mentioned it’s an xmos firmware bug and supposed to be fixed on the next update.

Ibasso advise to set the filter to “super slow” when using it as usb dac. Could you verify that Amir?

Apparently it's changing the filter to super slow roll off before using it as a DAC
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,381
Location
Seattle Area
In the subjective listening test there is no mention of the frequency roll-of at all. Is it because you can't hear it anymore, or is there something else at play? ;)
It appears the problem is only in dac mode. Company says bug in xmos firmware.
 
Top Bottom