Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Trying to get some no-nonsense info on DACs. (Read 7141 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Trying to get some no-nonsense info on DACs.

I've been a on/off lurker here for a while but I've only really taken any interest in things like EAC settings and compression.

Now I'm interested in a "DAC". I've been trying to read between the lines of the audiophile jargon, but it's not been easy. I've got an Auzentech prelude which has served me well and aside from it having no line out, nothing to complain about. I'm just looking for something better, for the sake of upgrading.

I wont bore you all with the models of entry level audiophile DACs I've short listed, but I've been reading that some of the far cheaper pro audio interfaces do a great job, they just don't have the showroom goodlooks. I would just like some advice on what to keep in mind from people with a more scientific, dare I say sceptical standpoint.

How do things like the Transit and ua-1ex compete with devices in the audiophile market? Is Asynchronous USB really that important? Will they give any sound quality upgrade when compared to my Prelude? I don't want to muddle this post with too many huge and sweeping questions and I'm not necessarily trying to save money, just not waste it on the dreaded oil of snake.

Something I would very much like that is quantifiable, a good line out to go into my headphone amp and stereo receiver from the pc.

Trying to get some no-nonsense info on DACs.

Reply #1 –
Quote from: cutshaw on
I've got an Auzentech prelude which has served me well and aside from it having no line out, nothing to complain about. I'm just looking for something better, for the sake of upgrading.


Please define "better". What exactly are you lacking in your current setup? Any specific features?

As for the technical side, Auzentech Prelude has specs that are not so easy to beat (not that it is necessary to beat them, anyway).

Trying to get some no-nonsense info on DACs.

Reply #2 –
Quote from: alexeysp on
Please define "better". What exactly are you lacking in your current setup? Any specific features?


A line out would be nice, I could let my amps handle the amping solely. I just want the clearest most uncoloured signal to my amps. I know there is a mod for removing the opamp from the prelude, but it involves bridging/shorting pins in the socket when the opamp is out and I've not managed to find much info on how to do it.

Quote from: alexeysp on
As for the technical side, Auzentech Prelude has specs that are not so easy to beat (not that it is necessary to beat them, anyway).


This is the most interesting thing. If this card is so good and does so many things, surely a dedicated external DAC that has nothing else to do and costs twice as much is doing something "better", or people just like flexing their wallets and stroking their shiny boxes, which I don't think is a stretch of the imagination.


Trying to get some no-nonsense info on DACs.

Reply #3 –
Perhaps it is just as much that people like to make money. They need some means to that end — thus products where there seems to be a market. The market is perhaps primarily created by some of those sellers who have a talent for it, which means they are good at making “customers” believe they have a need for whatever product they come up with.

And it is generally nice to get your hands on new toys, if you know they probably won’t mean anything special to you after a short time.

Trying to get some no-nonsense info on DACs.

Reply #4 –
Quote from: AndyH-ha on
And it is generally nice to get your hands on new toys, if you know they probably won’t mean anything special to you after a short time.



Very true. I'm tempted to spend £300 on the new Arcam rDac just because it looks amazing and has Arcam in the name. Will it change my world? At this point, I have no idea.

Trying to get some no-nonsense info on DACs.

Reply #5 –
Quote
which has served me well and aside from it having no line out,
It doesn't have a line out????  What kind of output do you have?

Quote
This is the most interesting thing. If this card is so good and does so many things, surely a dedicated external DAC that has nothing else to do and costs twice as much is doing something "better", or people just like flexing their wallets and stroking their shiny boxes, which I don't think is a stretch of the imagination.
Where I work we build some (non-audio) boxes that are perhaps slighly more  complex than a  DAC or soundcard.  But since we build & market these things in small quantities in the USA, we have to sell them  for $300 - $700 USD.  If we were manufacturing these things by the thousands in China, we could probably build them for 1/5th or 1/10th of the cost.

And when I took a marketing class, my professor said, "When in doubt, raise the price".  The idea is (besides making more profit) that the customer will associate quality with price.  If the price is too low, the customer will think your product is "cheap".  In the audiophile market, it's important for the product to have a price that sends a message of high-quality.

Quote
Very true. I'm tempted to spend £300 on the new Arcam rDac just because it looks amazing and has Arcam in the name. Will it change my world? At this point, I have no idea
If you want to buy something that looks nice and has good build quality, go for it!  But, be very skeptical about claims of audiophile sound quality.    It's not that hard to build a good DAC, and once you get to a point where you can't hear any defects, any "audio improvement" beyond that is snake oil.  If you want to hear a difference, buy better speakers...

Quote
Is Asynchronous USB really that important?
I'm not sure if "asynchronous" is even a valid USB mode!  As far as I know, all USB audio devices have their own internal clock (as do PCI soundcards).  The data flowing over the data bus or USB bus is typically flowing at a higher rate than the audio sample rate.  Buffers keep the audio flowing smoothly as the data bus & CPU are interrupted for multitasking.

Trying to get some no-nonsense info on DACs.

Reply #6 –
Quote from: DVDdoug on
It doesn't have a line out????  What kind of output do you have?

What I should have said was I would have liked to have a card with a passive line out, bypassing the opamps on the soundcard, if this makes more sense? This is maybe more nonsense from the audiophile community, but I did buy into the idea that less in the signal path was better. I'm double amping if I use my stereo receiver or headamp and that's not ideal. At least is made sense to me! 
Quote from: DVDdoug on
If you want to hear a difference, buy better speakers

I've managed that part already. I'm a headphone fan and got the best headphones my budget would allow and it's what started me looking into this mess.

Thanks very much for your input, your post was exactly what I was looking for. I like shiny toys, but I've no interest in buying something purely for it's aesthetic value.

Trying to get some no-nonsense info on DACs.

Reply #7 –
Quote from: cutshaw on
What I should have said was I would have liked to have a card with a passive line out, bypassing the opamps on the soundcard, if this makes more sense?


Unbuffered DAC output is generally a bad idea. The DAC chip itself has very limited driving capability. This means that without a buffer it is likely to distort more than any decent operational amplifier will. Also, some DAC chips output current instead of voltage - in this case some sort of active conversion circuitry is mandatory, you simply can not leave it out.

Operational amplifiers used in Auzen Prelude are very good and add infinitisemally small amounts of noise and distortion. Your headphones, however good they are, probably distort tens if not hundreds times more.

The widespread belief that any operational amplifier "colors" the sound is a myth. Whenever this claim is made it is never supported by any reproducible experimental results.

Trying to get some no-nonsense info on DACs.

Reply #8 –
Quote from: alexeysp on
Quote from: cutshaw on
What I should have said was I would have liked to have a card with a passive line out, bypassing the opamps on the soundcard, if this makes more sense?


Unbuffered DAC output is generally a bad idea. The DAC chip itself has very limited driving capability. This means that without a buffer it is likely to distort more than any decent operational amplifier will. Also, some DAC chips output current instead of voltage - in this case some sort of active conversion circuitry is mandatory, you simply can not leave it out.

Operational amplifiers used in Auzen Prelude are very good and add infinitisemally small amounts of noise and distortion. Your headphones, however good they are, probably distort tens if not hundreds times more.

The widespread belief that any operational amplifier "colors" the sound is a myth. Whenever this claim is made it is never supported by any reproducible experimental results.


I'm learning as I go, so bear with me. What is the difference between the headphone out and a line out on a portable player? From what I've gathered the line out doesn't use the portables built in amplification and that's why it's better to use it for a headphone amp and why people with ipods or other players with proprietary connections use the line out dock mods.

Something practical I would like about having this option with my soundcard/dac would be more use of the pot on my amp(s). With the prelude's volume at 67(the installed default) in windows, I can only use the volume on my amps to about 9 o'clock and I'm a huge fan of a having something I can physically turn, more control would be great. Do I just lower the volume in windows? I thought the sound suffered doing that, but that could all be in my head.

Trying to get some no-nonsense info on DACs.

Reply #9 –
Quote from: cutshaw on
I'm learning as I go, so bear with me. What is the difference between the headphone out and a line out on a portable player?


It depends on the particular player schematics. Chances are that there is no practical difference. The headphone output may have increased voltage swing and/or supply more current than the line out. But the mere fact that the player has both headphone out and line out does not mean that the line out is not amplified.

Quote
From what I've gathered the line out doesn't use the portables built in amplification and that's why it's better to use it for a headphone amp and why people with ipods or other players with proprietary connections use the line out dock mods.


I am not familiar with the ipod internals but I honestly doubt that those mods result in any audible benefits. Maybe someone more knowledgeable can provide additional information on this subject.

Quote
Something practical I would like about having this option with my soundcard/dac would be more use of the pot on my amp(s). With the prelude's volume at 67(the installed default) in windows, I can only use the volume on my amps to about 9 o'clock


This is typical. A standalone headphone amplifier is usually able to output more power than any set of headphones could ever need.

Quote
Do I just lower the volume in windows? I thought the sound suffered doing that, but that could all be in my head.


Set the output format to 24 bits in the player you are using. With this resolution even theoretical quality degradation caused by digital attenuation is negligible. Of course there still is the SNR degradation in the analog path, but with the excellent analog specs of your sound card I'd estimate you have over 30 dB of headroom before the DAC/amplifier circuit noise could possibly become audible under typical listening conditions.

Trying to get some no-nonsense info on DACs.

Reply #10 –
A line level output is intended to feed an accurate voltage to a high impedance input, usually 10 thousand  to 50 thousand ohm. Current requirements are minuscule, well under 1 milliamp, meaning essentially no power delivery. Headphones are small speakers, low but varying impedance loads, which require some power to drive them.

Being very small, headphone power requirements are very small relative to speakers. Power requirements to reproduce accurate bass are high relative to midrange and high frequency audio. While a line output may have a significant power reserve, it is not necessarily a given that it does. For at least some line level outputs, a low impedance load may well pull the voltage down significantly, especially at lower frequencies, distorting the signal, or at least lowering its level relative to the audio as a whole. This distortion may be more benign than the clipping and over compression commonly discussed in this forum, but it is real enough.

Measurement of waveforms with the different loads, a genuine line level input vs a headphone, should tell the tale for any given line out.

Trying to get some no-nonsense info on DACs.

Reply #11 –
Quote from: alexeysp on
Set the output format to 24 bits in the player you are using. With this resolution even theoretical quality degradation caused by digital attenuation is negligible. Of course there still is the SNR degradation in the analog path, but with the excellent analog specs of your sound card I'd estimate you have over 30 dB of headroom before the DAC/amplifier circuit noise could possibly become audible under typical listening conditions.


This might sound awfully dense, but do you mean that I should go into Foobar somewhere, set the output to 24bits and then I should be able to lower the volume in windows a lot more without losing any audible quality?



Quote from: AndyH-ha on
A line level output is intended to feed an accurate voltage to a high impedance input, usually 10 thousand  to 50 thousand ohm. Current requirements are minuscule, well under 1 milliamp, meaning essentially no power delivery. Headphones are small speakers, low but varying impedance loads, which require some power to drive them.

Being very small, headphone power requirements are very small relative to speakers. Power requirements to reproduce accurate bass are high relative to midrange and high frequency audio. While a line output may have a significant power reserve, it is not necessarily a given that it does. For at least some line level outputs, a low impedance load may well pull the voltage down significantly, especially at lower frequencies, distorting the signal, or at least lowering its level relative to the audio as a whole. This distortion may be more benign than the clipping and over compression commonly discussed in this forum, but it is real enough.

Measurement of waveforms with the different loads, a genuine line level input vs a headphone, should tell the tale for any given line out.


I don't suppose you could break that down for a complete layman? If you try and picture me looking up at you licking an ice cream, that should help.

Trying to get some no-nonsense info on DACs.

Reply #12 –
Quote from: cutshaw on
do you mean that I should go into Foobar somewhere, set the output to 24bits and then I should be able to lower the volume in windows a lot more without losing any audible quality?


In simple words, yes. This will diminish the quantization noise that is intrinsic to the digitally represented audio signal. This will not help with the analog part of the noise, but as I already said the analog noise of your card should be very low as it is, unless something is broken.

If you are using foobar2000, this setting is in the Playback/Output section of the preferences dialog.

Trying to get some no-nonsense info on DACs.

Reply #13 –
Quote from: cutshaw on
Now I'm interested in a "DAC". I've been trying to read between the lines of the audiophile jargon, but it's not been easy.


No nonsense advcie? Look elsewhere in the audio chain if you want improvements in sound quality.

Read between the lines of audiophile jargon? Useless. All there is between the lines of that noise is more noise.

Trying to get some no-nonsense info on DACs.

Reply #14 –
Quote
I don't suppose you could break that down for a complete layman?


Do we need to start with the most basic relationships in electricity? There are three fundamental aspects of any circuit: voltage, current, and resistance. Voltage is the potential for something to happen, current flow is what happens, and resistance is what must be overcome for it to happen. Or sort-of something like that.

These three elements are related in specific and exact ways. For a direct current (one way flow, no signal)  Ohm’s Law is E=IR, where
E is the voltage difference over the circuit (or some part thereof), measured in Volts,
I is the current, which can be looked at as a flow of power from one part of the circuit to another, measured in Amperes, and
R is the resistence measured in Ohms.

All circuits working above absolute zero (temperature) have some resistence to current flow. This is sort-of like friction. Power is expended in moving current through the resistence. That power can be merely dissipated as heat, as in a simple resistor, or can first do some work, as in a speaker, before being dissipated as heat (except a speaker can’t really do anything on a pure direct current circuit, so a light bulb is a better example.

Anyway, the amount of power is EI, or volts times amperes, termed watts. The amount of current (amperage, I) that flows is determined by the potential (voltage, E) and the resistence (ohms, R) of the load (e.g. the light bulb):  I=E/R gives the current; E**2/R gives the power.

Things become more complex when an alternating voltage is involved, which means that the positive and negative poles of the source, and thus the direction of flow, switch back and forth. An audio signal is a type of alternating voltage.

Ohm’s law still essentially expresses the relationships, but we need to substitute impedance for resistence. Impedance is the combination of resistence and reactance. Reactance is the opposition to alternating current flow by capacitors and inductors, where the friction analogy isn’t quite as good. Reactance depends on frequency. At any given frequence it is some number of ohms for any given capacitance or some other number of ohms for any given inductance. Capaticance is as in capacitor and inductance is as in coil. The physics involved is quite different for the three factors of impedance..

Speakers and headphones have both resistence and reactance. The reactance factors are significant relative to the total impedance.

A given voltage source, such as a battery or a power transformer, or an electronic circuit, is finite. Only a given amount of power can be drawn from it. As the current draw increases, the voltage measured across the circuit decreases, thus the higher current cannot be maintained.

Circuits can be built such that they will maintain a constant voltage over a range of loads, but the requirements are different for different kinds of circuits. An audio power amplifier needs to have a large current reserve because its job is to deliver power in exact proportion to the audio signal presented to its input. Speakers are low resistence (low impedance) devices (e.g. 4, 8, or 16 ohm ratings) and thus from I=E/R we see that a relatively large current is required.

A preamp does not (essentially) deliver power, it only amplifies (increases the voltage of) the signal that is presented to its input. Its output is usually to a rather high resistence load, (e.g. 10,000 to 50,000 ohms), which means the current it has to be able to deliver is very small. Also, the load resistence normally does not vary much, so the current demands are pretty constant, within a relatively narrow range.

Any instrument intended to give an accurate measure of signal, such as a meter or scope, has a very high input resistence so it will not draw enough  current to significantly alter the signal it is trying to measure. This is also the proper characteristics for anything attached to a line level output; it should not alter (distort) the signal being supplied by the line level output.

If instead of the normal high resistence load, a small resistence (or impedance) load is attached, we see from E=IR that the voltage output will decrease (because such circuits are not designed with a reserve current capacity, current cannot successfully increase as R decreases, thus E decreases). Microphone preamps, phono preamps, and line level preamps are examples of this circuit type in the audio world. DACs are another. These are all designed to output to a large and stable resistive load.

Since headphones require little power relative to loudspeakers, they may work on some line level circuits but this is not dependable. A major part of headphone impedance is inductive reactance (the coil around the magnet). Inductive reactance decreases with frequency, so the headphone resistance goes down with bass, which means the power requirement goes up. Bass has to be relatively loud to be heard, meaning a higher signal level, due to the characteristics of human hearing, which also leads to more power required from the circuit.

Without the power reserve of a good power amplifier, or headphone amplifier, the signal level cannot be dependably maintained. Output will not linearly related to the input signal, thus distorted. I don’t believe any professional soundcard is intended to have headphones attached to its output. Many external interfaces have separate headphone outputs, otherwise they need to feed a mixer, headphone amplifier, or some similar load. (Remember, computer speakers are “active,” they have power amplifiers inside the box.) Possibly some of the gaming/multimedia soundcards have higher capacity buffered line outputs so that headphones can attached in place of amplifier inputs.

This non-linearity may be relatively benign. There will not be clipping, excess noise generation, possibly no harmonic distortion. If the result suits you, fine. If it works at all, and you want to find out what is actually happening, you need to record the output directly to a soundcard (to analyze with software), or measure on a scope, under both conditions.

Is this a correct response?

Trying to get some no-nonsense info on DACs.

Reply #15 –
Quote from: AndyH-ha on
A major part of headphone impedance is inductive reactance (the coil around the magnet). Inductive reactance decreases with frequency, so the headphone resistance goes down with bass, which means the power requirement goes up.


Inductive reactance is actually directly proportional to frequency.

Otherwise, what you've said would be true if the coil was not moving. In a real dynamic transducer quite a few electromechanical effects come into play. As a result, real headphone presents a complex load with more or less pronounced resonance peak, as you can see on the graphs here: http://www.headphone.com/learning-center/build-a-graph.php.

Here's an article describing the corresponding model: An Improved Electrical Equivalent Circuit Model for Dynamic Moving Coil Transducers

Trying to get some no-nonsense info on DACs.

Reply #16 –
Quote from: AndyH-ha on
All circuits working above absolute zero (temperature) have some resistence to current flow.

I suppose this is nit picking, but many substances become superconductive above absolute zero, some of them well above absolute zero (though well below room temperature).
 

Trying to get some no-nonsense info on DACs.

Reply #17 –
A great deal can be said concerning every statement in my post. All of it is more complex. The real question is whether or not the questioner can get enough from it to form a useful picture.

Yes, the statement I wrote is backwards for inductive reactance. I guess the generally greater power requirement for bass is only the requirement of a larger amount of radiated acoustical energy.