Jump to content
IGNORED

Best (or worst) Digital Interface?


Recommended Posts

Let's suppose I have a recently purchased DAC with a lot of different choices for the digital interfaces.

 

Is there one that is (likely to be) much better than the others?  Or is there one that is (likely to be) worse?

 

I am asking in terms of theoretically the best/worst and also in terms of typical implementations (say with a $5k price cap).

 

I hear a lot about noise, ground loops, jitter, etc. but it isn't clear to me if there are clear differences across brands in these effects or in SQ overall.

 

Choices include:

 

USB

 

WiFi (I use Apple AirPlay from iTunes currently)

 

Toslink

 

Ethernet (I have CAT 6 cable runs)

 

HDMI

 

Thunderbolt 

 

Coax 

 

 

I will leave out bluetooth....

Link to comment

At $5k or less buy a Chord or Benchmark.  Those are effectively immune from jitter and most other effects on any input. 

 

As in terms of how they operate Toslink and SPDIF have to be the worst in regards to the interface itself.  Er well let me back up, HDMI is the worst.   Chances are none of them are audibly different with any decently made gear. 

And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. 

Link to comment
33 minutes ago, esldude said:

At $5k or less buy a Chord or Benchmark.  Those are effectively immune from jitter and most other effects on any input. 

 

As in terms of how they operate Toslink and SPDIF have to be the worst in regards to the interface itself.  Er well let me back up, HDMI is the worst.   Chances are none of them are audibly different with any decently made gear. 

I would consider Exasound or Mytek on your short list, as well.  

 

As far as interfaces/connection methods are concerned, there is USB, and there are all the other also rans, whose future does not look bright.  Properly implemented, all criticisms of USB are simply misinformed and incorrect, and there is no objective, repeatable evidence that any of the alternatives you cite are "better".  

 

USB is widespread, and as a result it has been subject to many excellent, middling or poor implementations in hardware or in software drivers.  But, just because the Yugo or Trabant was a mind bogglingly atrocious example of the genus automobilicus, that does not make Porsches, Mercedes's, etc. garbage.  Good USB, well implemented, is as good as anything available, unless you wish to consider dedicated Ethernet into a Merging NADAC.  But, that is outside the budget, obviously.

Link to comment

Okay.

 

The best possible SQ within that price range (ie, excluding high-end ladder DACs) is as follows:

 

Dedicated audio PC (HQPlayer) -> USB (isolation/reclocking) -> I2S bridge -> DSD512 DAC.

OR

Dedicated audio PC (HQPlayer) -> USB (isolation/reclocking) -> DSD512 DAC (with IS2-over-USB support).

 

If you're stuck on a non-DSD DAC and are willing to aim for less then the optimum SQ:

 

Any PC -> high-grade TOSLink -> DAC

 

Electrical noise vs. self-noise: Self-noise is generated by the receiving device; SPDIF (ie, TOSLink, Coax, etc) doesn't generate much because it's fairly slow. USB controllers generate a lot of self-noise; by feeding a pristine USB signal to it, you minimize how much it has to work and generate noise. Electrical noise is all the garbage being transmitted from your chain into the delicate DAC. TOSLink is immune to electrical noise, and since SPDIF doesn't generate a lot of noise by itself, you don't need to jump through all the hoops you'd have to with USB.

 

Jitter: Not that important. A modern DAC, especially one in this price range, is going to have clocks good enough to clean up a signal that is impacted from such things like SPDIF's normal jitter.

 

Quality of TOSLink cable is key. High-end plastic conductors designed for full bandwidth. Quartz glass conductors. Cheap TOSLink cables don't have enough bandwidth for even Redbook streams, forget about high-definition audio.

Link to comment
3 minutes ago, mansr said:

FWIW, I've tested 6 metres of cheap plastic Toslink for several hours without a single bit error.

 

There is no error correction process in SPDIF so I don't see how you could possibly track any errors. The issue is smearing caused by losses and out-of-order data caused by internal reflections and impurities.

Link to comment
1 hour ago, mansr said:

FWIW, I've tested 6 metres of cheap plastic Toslink for several hours without a single bit error.

I have done the same test.  No errors at 96 khz. I was using of course the tremendously well engineered exotic Monoprice Premium Toslink.  25 feet which is just short of 8 meters.  Currently $8.46 plus shipping.

 

Inexpensive Toslink cable improved while no one was looking.  If the gear has good transmitter and receivers it also is not high jitter like once was the case.  Plus if you have heard music recorded in the last 15 years there is fairly high probability it went over an ADAT interface at some point.  ADAT uses the same Toslink cable, and connectors. 

 

Doing the test with an early DVD player that still works you will get few errors at 48 khz.  But that is a device that is approaching 20 years old and was inexpensive at the time.  The errors are from the DVD unit not the modern cheap cabling.

And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. 

Link to comment
32 minutes ago, esldude said:

I have done the same test.  No errors at 96 khz. I was using of course the tremendously well engineered exotic Monoprice Premium Toslink.  25 feet which is just short of 8 meters.  Currently $8.46 plus shipping.

 

Inexpensive Toslink cable improved while no one was looking.  If the gear has good transmitter and receivers it also is not high jitter like once was the case.  Plus if you have heard music recorded in the last 15 years there is fairly high probability it went over an ADAT interface at some point.  ADAT uses the same Toslink cable, and connectors. 

 

Doing the test with an early DVD player that still works you will get few errors at 48 khz.  But that is a device that is approaching 20 years old and was inexpensive at the time.  The errors are from the DVD unit not the modern cheap cabling.

 

How was this test conducted?

Link to comment

 

 

 

 

Curio

1 hour ago, esldude said:

I have done the same test.  No errors at 96 khz. I was using of course the tremendously well engineered exotic Monoprice Premium Toslink.  25 feet which is just short of 8 meters.  Currently $8.46 plus shipping.

 

Inexpensive Toslink cable improved while no one was looking.  If the gear has good transmitter and receivers it also is not high jitter like once was the case.  Plus if you have heard music recorded in the last 15 years there is fairly high probability it went over an ADAT interface at some point.  ADAT uses the same Toslink cable, and connectors. 

 

Doing the test with an early DVD player that still works you will get few errors at 48 khz.  But that is a device that is approaching 20 years old and was inexpensive at the time.  The errors are from the DVD unit not the modern cheap cabling.

.Yup,newer gear with good electronics can be just as good across optical as coax with an expensive cable at 96khz and below. I'd like to see them update the standard to 192 khz and DSD rates as wire connectors for digital signal is 1980's tech... optical is expected today for reliable data transmission

Regards,

Dave

 

Audio system

Link to comment
1 hour ago, GUTB said:

 

How was this test conducted?

Windows computer using Foobar 2000 WASAPI for bit perfect transmission to a Tact RCS2.0 via USB to SPDIF converter.  The Tact also duplicates digital inputs as digital outputs.  25 ft Toslink cable connected the Tact to a TC IMpact Twin USB recording interface connected to a second computer.  I put together a 30 minute file of music.  Let it repeat play for several hours one evening.  Took the recorded file off the second computer and compared each file of 30 minutes with the original.  All copies were bit perfect copies of the original.  So no errors. 

 

I also did the test another time. I used the Toslink output from a Gigabyte computer motherboard.  That fed the TC IMpact Twin and was recorded to a second computer. Again no errors.  The same cable from Monoprice, but this one was only 6 ft long. 

 

And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. 

Link to comment
Quote

 

Ralf,

You might also consider i2S.  No conversion required.  You would need an i2S card, similar to the bridge card manufactured by PinkFaun.  But there are some contributors on this forum that believe i2S is the most efficient means of audio transfer.  No interface required!

Good weekend

jjk

Link to comment

I compared 75 Ohm coax and USB with my streamer and DAC. I don't think there is much difference. Currently using coax. There is a whole industry on improving USB. That made me a little weary of USB.

 

What I do find ironic is that a certain poster in this topic advocates to convert all PCM to DSD. But he hates class D amps since they convert the audio signal.

[br]

Link to comment
31 minutes ago, mordante said:

I compared 75 Ohm coax and USB with my streamer and DAC. I don't think there is much difference. Currently using coax. There is a whole industry on improving USB. That made me a little weary of USB.

 

What I do find ironic is that a certain poster in this topic advocates to convert all PCM to DSD. But he hates class D amps since they convert the audio signal.

 

I think he has complained about the side-effects, not the conversion per se.

"Science draws the wave, poetry fills it with water" Teixeira de Pascoaes

 

HQPlayer Desktop / Mac mini → Intona 7054 → RME ADI-2 DAC FS (DSD256)

Link to comment
11 minutes ago, semente said:

 

I think he has complained about the side-effects, not the conversion per se.

 

That's because the signal that goes INTO class D is NOT the signal which comes OUT. What comes out of class D is a modulated and filtered facsimile. That's why it has been pointed out that class D is mathematically, physically inferior to linear amplification. Of course, there are benefits to this trade-off, as always -- because class D doesn't amplify the input signal, it's also not amplifying unwanted components of the input signal. 

[br]

Link to comment
59 minutes ago, Ralf11 said:

Well, it sounds like they are all about the same (??)

It is very, very likely they are all audibly the same.  Technically and measurably those which let the clock in the DAC be the master have lower jitter whether you can hear it or not. So Wifi, ethernet and asynch USB are preferred.

And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. 

Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



×
×
  • Create New...