Why do people recommend using a streamer between the PC and the DAC?

[guigui]

I've seen many people saying that using a streamer, even a cheap one, between the PC and the DAC improves the sound quality coming out of the DAC. Pardon my ignorance, but I don't understand it.

I understand when people recommend a DDC before the DAC, because of the USB isolation and the reclocking. But what a streamer does if you don't use its internal DAC?

Cheers!

 

[voodooless]

A streamer does not improvement is sound quality. It can only be a nice convenience in some cases.

 

[guigui]

A streamer does not improvement is sound quality. It can only be a nice convenience in some cases.

Would you say a DDC could improve sound quality?

 

[Berwhale]

You might want to use a streamer, instead of connecting the PC directly to the DAC for several reasons...

• Electrical noise generated by the PC is audible through your speakers and you don't know how to/can't or don't want to fix it.
• You want to control the streamer from another device like a phone or tablet without having the PC on.
• You want to place the DAC, amps, speakers, etc. away from your PC.

 

[Berwhale]

Would you say a DDC could improve sound quality?

Only if something is broken.

 

[Aerith Gainsborough]

Would you say a DDC could improve sound quality?

The only instance where external filters to help with electrical "PC noise" might make a difference is when you have a really, really crappy DAC that can't do the filtering on it's own.

You want to control the streamer from another device like a phone or tablet without having the PC on.

This is pretty much the only reason for a streamer.
If you listen to music for many hours, a general purpose PC will gobble up far more energy (even when idle) than a small, dedicated device.

Ofc, when you spend a kilobuck+ on a streamer, you might as well have stuck to the PC as a source device. :'D

 

[MaxwellsEq]

I have a streamer in the family room controlled by phone or tablet. It means I don't need to have a laptop or PC in line of sight (which would not be acceptable to other householders). It connects to the DAC via S/PDIF.

 

[antcollinet]

The only instance where external filters to help with electrical "PC noise" might make a difference is when you have a really, really crappy DAC that can't do the filtering on it's own.

Would you say a DDC could improve sound quality?

Ground loop currents will pass through the ground connection of any DAC (or any other audio device) no matter how good. If they are high enough, they will then generate an audible noise signal, added to the audio signal across an unbalanced interconnect downstream of the DAC. A USB to Toslink DDC (Or USB isolator - or other isolating DDC) provides galvanic isolation (Perfect in the case of toslink) that will block this ground current if it is present in the system, and thereby prevent the noise being added to the signal.

PCs are frequently a source of ground currents due to the high power high frequency components within - especially graphics cards.

 

[Aerith Gainsborough]

Ground loop currents will pass through the ground connection of any DAC (or any other audio device) no matter how good. If they are high enough, they will then generate an audible noise signal, added to the audio signal across an unbalanced interconnect downstream of the DAC. A USB to Toslink DDC (Or USB isolator - or other isolating DDC) provides galvanic isolation (Perfect in the case of toslink) that will block this ground current if it is present in the system, and thereby prevent the noise being added to the signal.

Wouldn't that only work if the devices are on completely different circuits in the house, otherwise they would share a ground connection at the power outlet, no?

I certainly never measured anything problematic from my PC.

 

[phofman]

Wouldn't that only work if the devices are on completely different circuits in the house, otherwise they would share a ground connection at the power outlet, no?

Actually these PC-induced ground loops issues are typically not caused by noise on PE lines , but by large currents running inside the PC through ground traces adjacent to the 0V terminals of the interfaces (USB ports, analog audio outputs) and the ATX PSU 0V to which the PE power supply line is connected. A portion of these currents then takes a detour through the connected equipment, running through their return-signal path, creating noise voltages which get superimposed onto the signal itself.

I certainly never measured anything problematic from my PC.

I would tend to say that your equipment does not create ground loops (isolation class II devices, balanced connections, galvanic isolation) or your specific PC happens to not pass large currents internally between ground points of the interfaces you measured and the internal ATX PSU.

 

[Aerith Gainsborough]

I would tend to say that your equipment does not create ground loops (isolation class II devices, balanced connections, galvanic isolation) or your specific PC happens to not pass large currents internally between ground points of the interfaces you measured and the internal ATX PSU.

Possible. I had a PC once back in 2003 (Jesus, that was more than 20 years ago :X) with an onboard sound card and you could clearly hear when the CD ROM spun up, the HDD started seeking or the GPU got taxed. In the headphones, as a kinda grizzly stactic mind you.

Using a bog-standard internal soundcard (Audigy2 ZS back then) eliminated these distortions completely. Haven't heard any distortions from any PC ever since then, be it HDMI, SPDIF, internal soundcard or USB. So I do wonder how widespread this issue really is.

 

[antcollinet]

Wouldn't that only work if the devices are on completely different circuits in the house, otherwise they would share a ground connection at the power outlet, no?

I certainly never measured anything problematic from my PC.

In this case, the loop is power outlet > PC > DAC > Amp > Back to power outlet.

When you have ground noise at the PC, current will flow both directly to the outlet from the PC - but also from the PC round the loop, and back to the outlet.

I've never measured anything problematic either (because I haven't attempted to measure it). But I have had ground loop noise in audio systems with PC Source. The noise was also obviously sourced from the PC, because it was related to screen/graphics activity and mouse movements. It was completely eliminated by using Toslink to the DAC instead of USB.

So I do wonder how widespread this issue really is.

Widespread enough that I regularly advise people here to use Toslink if they can when they ask about ground loop noise. Looking at my posting history, about once a month or so. Probably equally often talking about balanced or pseudo balanced interconnect for ground noise rejection.

 

[JeremyFife]

A streamer *instead* of your PC makes sense, adding one into an existing chain less so.
If you want features (perhaps a particular streaming service) only available on the streamer then fair enough.

Some people just like to overcomplicate what should be simple.
Some people really do have complex needs (I'd suggest they are in the minority)

Don't overthink - enjoy the music

 

[Aerith Gainsborough]

Widespread enough that I regularly advise people here to use Toslink if they can when they ask about ground loop noise. Looking at my posting history, about once a month or so. Probably equally often talking about balanced or pseudo balanced interconnect for ground noise rejection.

I really wonder what's the actual cause here. Can't be the PC designs themselves, otherwise the issue would be far more widespread to the point that PC design would have to be re-thought.

Also, the moment you want Multichannel out of a PC, Toslink is out. Unless you fancy using DDLive or some such.

 

[antcollinet]

I really wonder what's the actual cause here. Can't be the PC designs themselves, otherwise the issue would be far more widespread to the point that PC design would have to be re-thought.

Let's say all PC's are noisy. (They're not, but a lot are)

The issue then becomes audio system susceptibility to the noise. First there needs to be a loop through which the ground currents can conduct. Then the characteristics of that loop need to be such that it is suspeptible to picking up the noise - that will be very dependent on where and how the system grounds are connected. For example if there is no device with a ground downstream of the DAC, then no current will flow in that direction. If the ground impedances are such that most currentt flows from the PC to ground - not a problem. Then even if there is a ground current, the analogue cables have to have sufficient impudence in their ground conductor to allow the noise to reach a level of audiblity - or annoyance. Typical short cables as you would find in a typical stack are unlikely to do this. Other systems - especially (for examples) those with active speakers or subs with long RCA cables feeding them - or other reasons for long cables - are going to be more susceptible.

It's not quite that the stars need to align just right to get noise - but there are a lot of variables which make it not so common to occur. But once you have the problem it can be very difficult to eliminate, except by introducing isolation or balanced interconnect.

Also, the moment you want Multichannel out of a PC, Toslink is out. Unless you fancy using DDLive or some such.

This is true - but the majority of multichannel applications are with HDMI directly into an AVR. AVRs (assuming you are using the internal amplifcation) don't have downstream analogue connections to develop the noise

 

[Mnyb]

Eh I have a music server and four streamers computer and server are in other rooms . I can use the convenience of controlling everything with my phone or tablet and mix local files with Spotify tracks in the same playlist.

Streamers connect to the server via network ethernet or WiFi . Whole house audio

My pc is kind of a gaming pc fugly with tons of fans having that in the living room no ?

I do listen to my pc directly to a DAC when using headphones, then a streamer would not make sense . So no instead, both

 

[Joe Smith]

I don't use a streamer with my main workroom system, PC to DAC to Preamp to Amp.

I do use a streamer, and various BT antenna hooks, for listening in other rooms.

The WiiM mini works really well in the living room main downstairs system, where I use my iPhone as control and the streamer connects directly to my Yamaha amplifier.

In the dining room, I use a small SMSL AO100 amp that has BT as one source input along with RCA. BT connection is fine for low level/background music.

 

[Aerith Gainsborough]

This is true - but the majority of multichannel applications are with HDMI directly into an AVR. AVRs (assuming you are using the internal amplifcation) don't have downstream analogue connections to develop the noise

Okay, I shamelessly admit that you lost me there. :'D
What do you mean by "downstream analogue connections"?

Doesn't an AVR work the same as e.g.: my ADI-2 DAC?
Digital Input -> DSP -> DAC -> AMP -> Analog out

Interesting stuff about XLR, I always assumed it was deliberately designed as a symmetrical connections (3 poles for one channel) to make long runs past stage lights etc less problematic.

 

[antcollinet]

What do you mean by "downstream analogue connections"?

Ground noise is not typically added to the signal inside a device. Instead when you get ground currents in the loop they get added to the signal along the length of an analogue interconnect - typically between a DAC and Amp/preamp, or between Preamp and amp - but can be on any analogue interconnect in the loop.

So take the loop mentioned above Socket>PC>DAC>amp>Socket.

Noise is generated onto the PC ground by PC processes. It flows around the loop, through the USB Cable, Dac Ground, Analogue interconnect ground (normally screen), into the amp, and back to the socket. The signal out of the DAC if you measure it at the DAC RCA Out is perfect. However there is a noise current flowing through the RCA cable screen. Because the screen is not perfect (it has resistance and inductance) that current generates a voltage accrross the screen impedance. So the ground voltage of the two devices (DAC and AMP) is different. One has a noise voltage on it with respect to the other. At the input to the amp, the audio signal is perfect when referenced to the DAC ground - but from the point of view of the amp - the DAC ground has noise on it, so the amp sees that noise on also on the signal.


In the case of the AVR, there is no analogue interconnect between DAC and Amp (downstream of the DAC), quite likely they will share a common ground plane, or there will be a very short wire - so there is no way for the ground noise to generate a differential ground voltage between DAC and amp.

 

[antcollinet]

Interesting stuff about XLR, I always assumed it was deliberately designed as a symmetrical connections (3 poles for one channel) to make long runs past stage lights etc less problematic.

They will help with noise directly coupled to the cable too - but that IME is far less common (at least in domestic audio) than ground loop noise. Plus the important part of the balanced connection isn't voltage balancing - it is impedance balancing. You need each of the two signal connections to have identical output impedance, and identical input impedance. And you need one of the signals to be subtracted from the other at the input end. (Differential amplifier)

With matched impedances, then the noise adds to each single equally (regardless of what voltage is on them - one can be the inverse of the other, or one can be carrying a 0V reference - or anything in between). As long as the noise added to each of the two signals is the same, then when one signal is subtracted from the other the noise cancels out.

This is described far better than I am doing it in Bruno Putzeys' "The G Word". Read from the paragraph titled "Balance"