JCALLY JM20 MAX Headphone Dongle Review

[delta76]

$21 today. $72 tomorrow. $54 the next. It makes me just not wanna buy a great product.

you can either play along, or miss out, it's common practice by Aliexpress. just make sure you never pay more than the MSRP. they have sales twice a month so not a long wait either

 

[ErinPraline]

$21 today. $72 tomorrow. $54 the next. It makes me just not wanna buy a great product.

Crinacle has it on his store for $30 and the price seems to be fixed

 

[MC_RME]

@amirm , are you still gonna be including the 50mv test for this one?

If I may...

This result is not valid, though. The noise floor is at this level when a signal on dither level or no signal is sent. A 1 kHz sine at -144 dBFS gives this result, but from -143 dBFS onward the noise floor sits at 90.6 dB, about 2 dB higher. It stays there and does not change with higher sine levels, unless the sine itself crawls out of the noise floor. One of the already known specialties of the used CS DAC chip...

 

[CedarX]

ngl I wish I had seen this review before I bought a Topping G5

The G5 has quite a few extra features and has even better performances.
Now, if you were looking for a compact USB-C dongle and ended up with a G5… then yes, this review would have been useful!

 

[Robbo99999]

If I may...

View attachment 445542

This result is not valid, though. The noise floor is at this level when a signal on dither level or no signal is sent. A 1 kHz sine at -144 dBFS gives this result, but from -143 dBFS onward the noise floor sits at 90.6 dB, about 2 dB higher. It stays there and does not change with higher sine levels, unless the sine itself crawls out of the noise floor. One of the already known specialties of the used CS DAC chip...

Ok, I don't fully understand that, but are you saying that Amir can't run the test that he mentioned and include it in the review without him being confident of the result? He "always" does 50mv tests for amps & DAC/amps. I mean he said himself he was gonna post up the result. Upon brief reflection, I'm thinking that Amir wasn't aware of the peculiarities of the CS DAC chip used which you've just mentioned.....so I'm thinking we're not gonna be able to get a valid 50mv reading for comparison and therefore it's not gonna be included? At what mv level can you start to be confident of the measurements with this DAC/amp? So SINAD can be confidently determined from which mv upwards? I'm supposing by definition because of this that for instance the DS2 which can be measured with confidence at 50mv (at least Amir put it in the review) would mean that it by definition has to measure better at these lower levels? Sorry, I don't have a quite deep enough understanding to prevent me asking some silly questions or making some wrong extrapolations, they might be!

 

[jkim]

Ok, I don't fully understand that, but are you saying that Amir can't run the test that he mentioned and include it in the review without him being confident of the result? He "always" does 50mv tests for amps & DAC/amps. I mean he said himself he was gonna post up the result. Upon brief reflection, I'm thinking that Amir wasn't aware of the peculiarities of the CS DAC chip used which you've just mentioned.....so I'm thinking we're not gonna be able to get a valid 50mv reading for comparison and therefore it's not gonna be included? At what mv level can you start to be confident of the measurements with this DAC/amp? So SINAD can be confidently determined from which mv upwards? I'm supposing by definition because of this that for instance the DS2 which can be measured with confidence at 50mv (at least Amir put it in the review) would mean that it by definition has to measure better at these lower levels? Sorry, I don't have a quite deep enough understanding to prevent me asking some silly questions or making some wrong extrapolations, they might be!

This is about the effect of Cirrus Logic's dynamic range enhancement technique. Amir once mentioned he has no problem with this technique---i.e., he does not view it as cheating. See this post in my review of the JM20 Max.

 

[jkim]

If I may...

View attachment 445542

This result is not valid, though. The noise floor is at this level when a signal on dither level or no signal is sent. A 1 kHz sine at -144 dBFS gives this result, but from -143 dBFS onward the noise floor sits at 90.6 dB, about 2 dB higher. It stays there and does not change with higher sine levels, unless the sine itself crawls out of the noise floor. One of the already known specialties of the used CS DAC chip...

I am not sure about your measurement setting. Although Amir did not measure this directly, his THD+N sweep:

indicates its SNR @ 50 mV is around -90.5 to -91 dB. As you know, at this low level, SNR is almost always similar to THD+N---SNR is a tiny bit lower than THD+N. In fact, Amir's THD+N sweeps have always been consistent with his 50 mV SNR measurements.

The SNR @ 50 mV measurement in my JM20 MAX review is also about 90.8 dB:

This was with the help of the APU's 34 dB gain alone, without its notch filter. But I don't think it would be very different with the filter in place.

 

[orangejello]

Any problems using this with an iPad? They mention Windows, Android, and MacOS. Nothing about iPad.

Can anyone answer this?

 

[dark132]

Yeah, they're both fine, I'm just pointing out the differences, and if you want to choose the better measuring one then you'd get the DS2.

EDIT: and the fact the DS2 is balanced out.

For USB output, is balanced output important?

 

[Robin L]

So you can pay $1500 for an Astell & Kern Kann Ultra DAP, and have the hassle of an extra device to carry around, or you can pay only $25 for the JM20 (that performs better) and simply connect it to your smart phone, which 95% of the population carry at all times.

Am I missing something?

I miss the FiiO M3K DAP. I have one, but they're no longer being manufactured. They went for about $75 dollars, have a halfway decent GUI and can hold a great big Micro SD card full of files. Mine has .5 TB of ALEC files. I've also got a Samsung smartphone and a subscription to Tidal. Have to admit, the Smartphone running Tidal sounds better. However, the DAP has plenty of music I'm never going to find on Tidal. Having a DAP that reasonably priced and capable of true high fidelity is possible, but thanks to streaming and smartphones will probably not happen. The FiiO M3K first appeared in 2018.

 

[CedarX]

I miss the FiiO M3K DAP. I have one, but they're no longer being manufactured. They went for about $75 dollars, have a halfway decent GUI and can hold a great big Micro SD card full of files. Mine has .5 TB of ALEC files. I've also got a Samsung smartphone and a subscription to Tidal. Have to admit, the Smartphone running Tidal sounds better. However, the DAP has plenty of music I'm never going to find on Tidal. Having a DAP that reasonably priced and capable of true high fidelity is possible, but thanks to streaming and smartphones will probably not happen. The FiiO M3K first appeared in 2018.

You may want to take a look at the HiBy R1 or Shanling M0S…

 

[dark132]

I just bought one today to match my 8gen2 phone. I found it is very suitable for zero2. Compared with my usual 2qute+L30II+zero2 system, it is only slightly inferior. As a result, I listened to it for an afternoon. This is very suitable for playing some jazz music while walking.

 

[Robin L]

You may want to take a look at the HiBy R1 or Shanling M0S…

I still have two FiiO DAPs, both are working as well now as they did when I first bought them.

 

[Miley_B]

Any problems using this with an iPad? They mention Windows, Android, and MacOS. Nothing about iPad.

Can't say I've noticed any issues with iPad for general use. I'm using it now and it seems just dandy.

 

[Robbo99999]

For USB output, is balanced output important?

It can help prevent ground loops can it not? Plus it allows for 4V output normally. I remember in my previous place I lived I had ground loop issue with JBL 308p speakers and how it was setup with my TV and laptop, but I ended up being able to solve that by using somekind of isolator between the TV aerial and the TV, I don't know how likely it is that people have ground loop problems that can be solved by using a balanced DAC though, but it's one of it's features.

 

[Miley_B]

It can help prevent ground loops can it not? Plus it allows for 4V output normally. I remember in my previous place I lived I had ground loop issue with JBL 308p speakers and how it was setup with my TV and laptop, but I ended up being able to solve that by using somekind of isolator between the TV aerial and the TV, I don't know how likely it is that people have ground loop problems that can be solved by using a balanced DAC though, but it's one of it's features.

May I suggest watching this video.

Balanced interconnect topology is different to balanced headphones which is not balanced as I know it. It is a shame the term balanced caught on WRT headphones as it leads to much confusion. Differential would have been a much more helpful name.

 

[staticV3]

Balanced interconnect topology is different to balanced headphones which is not balanced as I know it.

Balanced Line connections, as well as balanced headphone connections, are typically both differential and impedance balanced.

They're as balanced as can be.

 

[MC_RME]

I am not sure about your measurement setting. Although Amir did not measure this directly, his THD+N sweep:

View attachment 445587

indicates its SNR @ 50 mV is around -90.5 to -91 dB.

So that matches my value (with signal above -143 dBFS) quite well.

As you know, at this low level, SNR is almost always similar to THD+N---SNR is a tiny bit lower than THD+N. In fact, Amir's THD+N sweeps have always been consistent with his 50 mV SNR measurements.

The AP SNR measurement is a fixed, quick measurement process, changing between active output level (adjustable values, here 50 mV, 1 kHz sine) and silence, where this silence (digital zero) is typically not zero, but a very low level dither noise (this default can be switched off, but is typically not changed). That is what I tried to explain. With a signal that is at least a very small one, and more than standard dither, the CS gives a 2 dB lower result in SNR. I wouldn't call that cheating either, it is too small to be of use or to give exaggerated numbers for advertisement. It seems more like a byproduct of whatever other process is going on (maybe internal power saving - I really don't know).

 

[MC_RME]

Balanced Line connections, as well as balanced headphone connections, are typically both differential and impedance balanced.

They're as balanced as can be.

Yes. And while the output is not galvanically isolated using an 'electronically balanced' output (99% of devices have that) into any amp that has an balanced input typically solves ground loop issues.

 

[goldenears]

Is there an ASIO driver available for the JM20 Max?