Fosi Audio DS2 Portable DAC & Amp Review

[AndiGee]

4.4mm to 4Pin XLR adapters are entirely safe to use.

That is what I am thinking too. And why should a 4-connector 4.4 -> 3.5 mm adapter behave different?

 

[staticV3]

That is what I am thinking too. And why should a 4-connector 4.4 -> 3.5 mm adapter behave different?

Because you cannot easily convert differential audio output to single-ended.

Many of these types of adapters will short out some of the Amp's outputs, which in the worst case, can permanently break the Amp.

For headphone output:

Single-ended to Single-ended -> Safe
Single-ended to Balanced ->Safe

Balanced to Balanced -> Safe
Balanced to Single-ended ->Unsafe.

Edit:
To be fair, 4.4mm Pentaconn with its fifth GND pin can be made to provide safe unbalanced and balanced headphone output, but the designer would have to specifically design their Amp for that purpose, and sell adapter cables that are guaranteed safe to use.

 

[AndiGee]

Because you cannot easily convert differential audio output to single-ended.

Yes, I was thinking that Fosi wrote the advice not to use adapters just to prevent people from accidentally 'converting' balanced to single ended. But a balanced 4pin 3.5 adapter should do the same thing as my XLR adapter.

 

[staticV3]

a balanced 4pin 3.5 adapter should do the same thing as my XLR adapter.

In theory, yes.

In practice, there's no accepted standard for differential 3.5mm TRRS headphone output, so unless you solder your own adapter and headphone cable, a store-bought 4.4->3.5 adapter is liable to shorting the Amp.

 

[Bongó]

Android está remuestreando el contenido de Tidal, generalmente a 48 kHz, por lo que siempre verás el color rojo en tu Fosi.

Puedes instalar USB Audio Player Pro, por ejemplo (cuesta alrededor de 10$) y verás los precios originales y la combinación de colores.

Muchas gracias por la respuesta Turambar (soy seguidor de tu excelente canal de YouTube). Así lo haré, solo por jugar con un juguete nuevo porque la verdad es que más allá de los 48 khz mis oídos no logran distinguir la diferencia, supongo que es porque la cadena de audio es muy simple. Así lo haré y estaré actualizando mi experiencia junto con el 400se. Saludos.

(Translated by moderator. Please post in English going forward.)

Thank you very much for the reply Turambar (I follow your excellent YouTube channel). I will do so, just to play with a new toy because the truth is that beyond 48 khz my ears cannot distinguish the difference, I suppose it is because the audio chain is very simple. I will do so and I will be updating my experience along with the 400se. Greetings.

 

[AndiGee]

there's no accepted standard for differential 3.5mm TRRS headphone output

OK, thank for the clarification, the thing that I call 4pin 3.5 does not exist, so no balanced 3.5 mm plugs.

This means a 4.4 -> 3.5 mm adapter must terminate the L/R negative channels rather than short-circuiting those. I doubt that you could buy an thing that doesn't do that, but you never know.

... and to get not to off-topic I'd also like to report that the Fosi DS2 is driving my balanced headphones with a authority I just did never expected from such small device. I also tried my HD600 single ended - perfectly acceptable headroom.

 

[Bongó]

¡Bienvenido!

No te preocupes por nada de eso. ¡Todos estamos aprendiendo!

Gracias por la bienvenida. No me preocuparé entonces por hacer preguntas para aprender, gracias. Disfruto mucho de este pasatiempo.

 

[Raex]

Does anyone have the power-measurements for single ended operation, they seem to be missing on the graphs....?

 

[MC_RME]

I will measure it next week with my AP to see how much of that purported THD+N loss quoted in a former post is actually present. I suspect nearly none, based on Amir's measurements. Fosi might use an additional driver to keep the DAC chips more clean at lower impedances...

I finally found the time to start testing and measuring. I use iPhone 11 Pro and 16 Pro for testing, plus a notebook via USB C with the BRAVO ASIO driver. Lots of unexpected issues that keep me busy to find out what's going on. What is already interesting is that I am not able to perform a THD+N sweep of the Fosi without running into complete nonsense values. Reason seems to be that the High Performance Analyzer mode of the APx gets disturbed by the high amount of out-of-band noise, a known property of the whole CS DAC chip series. Turning the High Performance mode off fixes that but limits THD+N to around -105 or so dB. @amirm might be able to fix his sweeps by adding his AP switching amplifier measurement filter. And maybe that was the reason he had an 'instrumentation error'.

Back to the bench...

 

[OldTimer]

Innovations in a deflated people pockets era.

 

[renaudrenaud]

Innovations in a deflated people pockets era.

So I went in holidays riding the bike for few days. But my high quality audio system was with me: DS2 and Zero Red.

A pleasure.

 

[MC_RME]

So here are some more results from a lazy Sunday afternoon.

I measured and wrote down the max values that the DS2 can provide at the edge of clipping - harmonics are about to rise but just do not. Excuse me to have done this first only for balanced (that is what I will mainly use):

600 Ohm: 4.0 V, +14.4 dBu, 118.2 dB THD+N, 27.5 mW

300 Ohm: 3.9 V, +14.1 dBu, 118.0 dB THD+N, 51.3 mW

50 Ohm: 2.9 V, +11.5 dBu, 112 dB THD+N, 170 mW

32 Ohm: 2.29 V, +9.4 dBu, 109 dB THD+N, 164 mW

16 Ohm: 1.2 V, +3.9 dBu, 105 dB THD+N, 92 mW

Note the discussion here about 510 mW output power - that is not there. With 1% THD I measured 2.6 V, +10.5 dBu aka 420 mW @ 32 Ohm still no match..

SNR: Base noise is a fixed -115.5 dBu (rms unweighted), plus the respective output level (for example 115.5+14.4= 129.9 dB, 115.5+9.4= 124.9 dB (all rms unweighted. dBA = +2.3 dB).

In my book pretty impressive so far. So let me first check @amirm 's dashboard to see if my unit can keep up:

Wow, it's even slightly better! Or my measurement setup was (when Amir measured the ASIO drivers were not known). Anyway, what bothered me most is that in this thread talk was CS would be cheating, THD/THD+N getting significantly worse when using lower impedance phones, with values of -80 and even -64 dB! Fortunately I can debunk this for the Fosi DS2, see here, a full sweep at 32 Ohm 2-channel load (in case you missed the above table where this was clearly visible already, with 16 Ohm still at -105 dB):

I mentioned that I had problems to do the sweep. Here is what is there and also what happens: enormous out-of-band junk, which at certain levels (around -25 dBFS) makes the special dual high performance input of the APx fail. The noise floor then is shown wrong, the notch filter fails as well, measurement values are all way off:

I was able to work around this by starting the sweep without the High Performance mode, then switching it on at around -10 dBFS. The result is valid and correct for the whole curve in the sweep picture.

Here is the same at full level (600 Ohm):

If I find the time I'll add some more.

Edit: I had a typo in my table and removed the statement about 510 mW @ 50 Ohm load.

 

[Basilderat]

Unfortunately I have a very shaky connection with the 4.4 mm input of the Fosi. If I insert the plug all the way in, I get no sound. If I pull the plug out again a little, I get the R-channel, the then L-channel - then at some point both. This could indicate that the stop of the socket is not correct.

Can anyone confirm this? It doesn't have to be a fault with the 4.4 socket of the Fosi, it could of course also be due to my XLR adapter. This may be what you get when the plug/adapter is larger than the device to which it is connected to.

Unfortunately, I don't have another device on which I could test my adapter.

View attachment 396517

The balanced cable below works perfectly for me. There were no issues with the 4.4mm socket, the plug just clicks in / seats really nicely, both channels are fine:

4.4mm BALANCED Audio Cable For Sennheiser HD 2.20S 2.30i 2.30g HD 560S HD 620S | eBay

Compatible with subsequent 4.4mm BALANCED connector player. Length 130cm / 4.2ft.

www.ebay.com.au

 

[MC_RME]

[Edit: tired eyes saw the wrong thing...see my post 398]

That isn't a balanced cable at all. If it works then the minus rings of the Pentaconn are not connected at all. It is unbalanced from the Pentaconn to the TRS.

@AndiGee That said I tried several Pentaconn cables (all balanced) and none showed the issue that you describe.

 

[Basilderat]

That isn't a balanced cable at all. If it works then the minus rings of the Pentaconn are not connected at all. It is unbalanced from the Pentaconn to the TRS.

@AndiGee That said I tried several Pentaconn cables (all balanced) and none showed the issue that you describe.

Looks like the genuine balanced cable from Sennheiser??:

Balanced Audio Cable for HD 620S (1.8m, 4.4mm)

Discover True Sound with Sennheiser headphones, earbuds, wireless Bluetooth headsets and soundbars. Sign up & enjoy 10% off your first order.

sg.sennheiser-hearing.com

 

[staticV3]

That isn't a balanced cable at all. If it works then the minus rings of the Pentaconn are not connected at all. It is unbalanced from the Pentaconn to the TRS.

Looks to be a genuine Pentaconn to TRRS cable.

Headphones like the HD560S support this:

Source

 

[MC_RME]

Oops - I need new glasses. Saw a 3 pin TRS...

 

[AndiGee]

Thank you all very much for your feedback, I can relieve fosi as well as the manufacturer of the XLR adapter 'ddaudio' - the problem with the non-functioning connection was ... me. I just didn't had the courage to push the plug into the socket until the final 'click'. It takes some kind of force, and I was a bit afraid of damaging the little device.

In the meantime I can report rare sound dropouts with one of my sources, laptop with Windows 11 and EqualizerAPO 1.3.2 installed. Happens randomly while TIDAL playback. I tried to change the output bitrate and deactivated a mysterious 'sound effect' settings, but the dropouts remain.

 

[mc.god]

It might not even be a firmware thing, Savitech doesn't even have a flag for this feature in their technical documents, so it might be hard-coded in their chips.

(Just checked my old Tempotec Sonata HD Pro, with Savitech's chip inside, it behaves exactly the same as DS2.)

I usually use JRiver, and there's no problem whatsoever, it came with better bit-perfect stream support as well, so I can't say much about Foobar2000's behavior.

Jcally JM20 with the same Savitech doesn't behave in this way (Sonata BHD does), when you start playing after it's gone into muting status the very first samples are not truncated\faded but instead you can hear a very brief background pink noise. What is preferable it's to you.

 

[Eddie]

Jcally JM20 with the same Savitech doesn't behave in this way, when you start playing after it's gone into muting status the very first samples are not truncated\faded but instead you can hear a very brief bakground pink noise. What is preferable it's to you.

Then it seems to be a firmware issue rather than a hardware issue.
In the meantime, I contacted Fosi about this behavior. So far they don't seem to understand the problem.