Hydrogenaudio Forums

CD-R and Audio Hardware => Audio Hardware => Topic started by: Yahzi on 20 February, 2013, 12:38:48 PM

Title: DAC IV stages
Post by: Yahzi on 20 February, 2013, 12:38:48 PM
How important are DAC IV stages in the design of a DAC? I've heard they can cause audible differences. Is this true? I'm not a DAC designer, but I assume there are those who understand the technical merits. Is it a big deal?
Title: DAC IV stages
Post by: John_Siau on 20 February, 2013, 02:48:30 PM
The I-V (current to voltage) conversion stage of a digital to analog converter is just one of several critical analog stages that can impact the performance of a DAC.  Two products may share the same D/A conversion IC, but may have very different performance.  Each of the analog stages that follow the DAC IC contribute noise and distortion to the audio.  With careful design, these artifacts can be minimized to levels that are well below audibility.  For example, Benchmark's new DAC2 converter uses a ES9018 DAC IC and achieves a 129 dB SNR A-weighted, and distortion that is better than -109 dB.  This implies that distortion is below the threshold of hearing until peak music levels reach 109 dB SPL.  Furthermore, noise would not exceed the threshold of hearing until peak music levels reach a damaging 129 dB SPL.  For all practical purposes, the noise and distortion produced by the DAC2 are inaudible.  The DAC2 probably defines the current state of the art.

However, I have measured consumer DVD players with ES9018 DAC ICs that are as much as 20 dB nosier than the DAC2.  Distortion measured about -80 dB on the DVD player.  The vast differences in performance are due to differences in the analog output stages.  The DAC2 and the DVD player share the same DAC IC, but the measured performance is very different.

Analog output stages include:

1) I-V conversion
2) Low-pass filters
3) Gain stage
4) Output buffers
Title: DAC IV stages
Post by: Yahzi on 20 February, 2013, 03:47:19 PM
Thanks for the reply John. I'm getting a few pro-DAC users citing IV stages in DAC design as some holy grail cause of significant audio differences. I'm trying to attack the argument in as many ways as I can. The stumbling block right now is the lack of published DBTs. At least I can't seem to find any positive or negative results.

Title: DAC IV stages
Post by: John_Siau on 20 February, 2013, 04:02:24 PM
Designing an I-V converter with a 130 dB SNR is not trivial.  Impedances need to be very low, and this puts significant demands on the opamp.  The opamp must have very low EIN and must be able to deliver high current drive while maintaining low distortion.  There are about 2 opamps that will work in this application.  Resistors must be metal film (to keep distortion low). 

One complication is that high-frequency switching noise is always present on the current outputs of sigma-delta converters.  This high-frequency noise must be bypassed with rather large NPO or COG capacitors before entering the I-V converter.

There are many opportunities to cut costs, and there are many opportunities to make mistakes when designing an I-V conversion stage.  The I-V conversion stage may be more important than the choice of conversion IC.
Title: DAC IV stages
Post by: Yahzi on 20 February, 2013, 04:28:53 PM
But it is largely irrelevant if it is competently designed? Modern DACs... should largely not have any issues in this department? I realise that is pure speculation on my part.
Title: DAC IV stages
Post by: benski on 20 February, 2013, 04:44:39 PM
But it is largely irrelevant if it is competently designed? Modern DACs... should largely not have any issues in this department? I realise that is pure speculation on my part.


I remember a user posting here on hydrogenaudio years ago with scope traces of a DAC that didn't appear to even have a reconstruction filter (or perhaps very poorly designed).  This was on the line-out of a name-brand laptop, if I remember correctly.  I'm not familiar with the ins and outs of the I/V circuit, but I suspect it's easy to get wrong either due to poor design or more likely cost-cutting.  We can debate the audibility of the noise and distortion generated by an I/V that's merely "good enough" and not "excellent".  But practically speaking, if the analog stages of the DAC are significantly noisier than the converter itself, you've overspec'd the converter in your product.
Title: DAC IV stages
Post by: John_Siau on 20 February, 2013, 04:52:41 PM
Take a look at published specifications for D/A converters, DVD players, sound cards, iPod, etc.  Many only have an 80 to 90 dB SNR.  This means that the noise will be audible in an A/B/X test if the peak music levels exceed 80 to 90 dB SPL (easily achieved).

Worse yet, most D/A converters are preceded by a digital volume control.  If 20 dB of digital attenuation is dialed in (very typical), the SNR will degrade by 20 dB.  In many systems, the user will need to apply 20 dB of digital attenuation to achieve a playback level that peaks at 100 dB SPL.  Under these conditions the converter noise may reach 30 to 40 dB SPL.  It is not hard to hear the noise floor drop from 30 of 40 dB SPL to less than 0 dB SPL in an A/B/X test. 

I have an ABX tester, and it is very easy to pick out D/A converters that have high noise floors.

If one wishes to use a digital volume control, the D/A converter will need an extra 20 dB SNR to keep noise inaudible at loud playback levels.  If we want to achieve "CD" quality and use 20 dB of digital attenuation, then 116 dB SNR is required.  Under these conditions, noise may still become audible if playback peaks exceed 96 dB SPL.

D/A converters with >116 dB SNR are few and far between.  Noise alone is a dead giveaway when comparing D/A converters.
Title: DAC IV stages
Post by: Yahzi on 20 February, 2013, 05:09:26 PM
So your view is that it is a potential issue but it can be rectified with careful design. So a competently designed DAC should not pose issues significant enough to be audible? It would seem to be similar to jitter which is also often cited left, right and center. Jitter can be a potential issue if the quantities are large enough but in a good unit there is no reason to assume it should pose an issue. Am I on the right track?
Title: DAC IV stages
Post by: saratoga on 20 February, 2013, 05:51:02 PM
Take a look at published specifications for D/A converters, DVD players, sound cards, iPod, etc.  Many only have an 80 to 90 dB SNR.


You're exaggerating.  Most quality players (iPods included) are very nearly quantization noise limited over line out.  80dB is quite rare and would be exceptionally bad for a name brand device.  The Apple stuff will give you 15.5 effecitve bits over headphone out with full analog volume control for not all that much money. 

D/A converters with >116 dB SNR are few and far between.  Noise alone is a dead giveaway when comparing D/A converters.


Yeah but in reality it doesn't make much difference since you never use that dynamic range.  CDs using > 80dB dynamic range are far rarer then D/A converters with < 80dB dynamic range.
Title: DAC IV stages
Post by: Soap on 20 February, 2013, 09:37:09 PM
For example, Benchmark's new DAC2 converter uses a ES9018 DAC IC and achieves a 129 dB SNR A-weighted, and distortion that is better than -109 dB.  This implies that distortion is below the threshold of hearing until peak music levels reach 109 dB SPL.


Does not mean something completely different?  Does that not mean that distortion, being 109 dB down, isn't crossing the threshold of hearing  when peak levels are 109dB up, but rather crossing the threshold of existence?  I can't hear 20dB sounds, much less 0.

Title: DAC IV stages
Post by: John_Siau on 21 February, 2013, 08:57:51 AM
80dB is quite rare and would be exceptionally bad for a name brand device.  The Apple stuff will give you 15.5 effecitve bits over headphone out with full analog volume control for not all that much money.


15.5 effective bits when the volume control is set at maximum, and to the best of my knowledge, none of the Apple devices have analog volume control (please correct me if I am wrong).  With 20 dB digital attenuation, 15.5 effective bits becomes 12.2 effective bits.  The problem is that significant amounts of digital attenuation are often used at normal listening levels, and the converter noise is often audible.

Yeah but in reality it doesn't make much difference since you never use that dynamic range.  CDs using > 80dB dynamic range are far rarer then D/A converters with < 80dB dynamic range.


CD and high-resolution low-noise recordings do exist, and if these are used in an ABX test, the noise floors of many converters should be audible (on the basis of calculated SPL of the noise).
Title: DAC IV stages
Post by: John_Siau on 21 February, 2013, 09:20:45 AM
For example, Benchmark's new DAC2 converter uses a ES9018 DAC IC and achieves a 129 dB SNR A-weighted, and distortion that is better than -109 dB.  This implies that distortion is below the threshold of hearing until peak music levels reach 109 dB SPL.


Does not mean something completely different?  Does that not mean that distortion, being 109 dB down, isn't crossing the threshold of hearing  when peak levels are 109dB up, but rather crossing the threshold of existence?  I can't hear 20dB sounds, much less 0.


My point was that nobody can rationally argue that the distortion is audible if it is below the threshold of hearing.

Distortion will become audible at some level significantly above the threshold of hearing.  Music masks much of the distortion, and this limits our ability to detect it.  Our ears are also very tolerant of low-order harmonic distortion as many musical sources have significant harmonic distortion.  For these reasons low-order distortion can often reach very high levels before it becomes noticeable.  However, non-harmonic distortion components are often much easier to hear, and may be detectable at relatively low levels if they are not well masked by the music. 

A published THD+N number does not tell us much about the audibility of the distortion (or lack thereof), unless the number is so low that audibility is very unlikely (as with the -109 dB performance of the DAC2).  The -109 dB THD+N number provides a good indication that the distortion should be inaudible.
Title: DAC IV stages
Post by: Arnold B. Krueger on 21 February, 2013, 10:00:19 AM
Take a look at published specifications for D/A converters, DVD players, sound cards, iPod, etc.  Many only have an 80 to 90 dB SNR.  This means that the noise will be audible in an A/B/X test if the peak music levels exceed 80 to 90 dB SPL (easily achieved).


That is far from being a universal truth.  The actual answer depends on the test.

The answer given above probably presumes that we set up the system to play music at a desired high SPL, and then mute the music player. If we hear any noise even if that requires sticking our heads inside the loudspeaker, then the test is failed.  That will demonstrate the need for a DAC whose SNR is equal to the peak SPL observed during the listening phase. This should be less than 120 dB or we are talking pretty severe ear damage.

The first step towards what I think is a more reasonable experiment would be to not allow the listener to stick his head inside the speaker cone, but rather constrain himself to a normal listening position. If then we presume a typical quiet room in a residence, then we can probably still do the same basic kind of test as above and get away with a DAC whose SNR is the peak SPL minus maybe 20-30 dB.  If we cherry pick the room, maybe only 10-15 dB. 

If we back off from a very high peak SPL to a more typical and comfortable SPL that is not the one we use with visiting firemen, then we may pick up another 10 dB or more.

The above rationalizing moves as I would categorize them with some personal bias, probably gets us below 100 dB.

The opposite extreme can be obtained by doing a different experiment that is IMO no less "real world".  In this experiment, we set up the system to play music at a desired high SPL, and then wait for the music to stop playing, but continue to play the recording which would then contain the sound in the performance room that remains including the noises made by the people while they are trying to be quiet, the HVAC, etc.  This test gets us below 90 dB, and may get us down as low as 60 or 70 dB.    75 dB is IME a 90% solution. 85 dB is a 99% solution.

As a practical matter, the vast majority of people find players and systems with 90 dB dynamic range to be highly satisfying, to say the very least. 

Remember that allegedly picky listeners like vinyl, and a vinyl system with 75 dB "Needle Raised" dynamic range is about as good as that technology ever got.
Title: DAC IV stages
Post by: Arnold B. Krueger on 21 February, 2013, 10:04:43 AM
D/A converters with >116 dB SNR are few and far between.  Noise alone is a dead giveaway when comparing D/A converters.


I guess that TI, Crystal Semiconductor, and AKM have been pulling our chains because they all have fairly inexpensive chips that do 115-120 dB dynamic range. ;-)

I've had a LynxTwo since what, 2002 that met that criteria and they are hardly rare. That isn't specsmanship because I've measured it myself. The same converters or competitive converters are now in the < $5 per channel price range.

Title: DAC IV stages
Post by: Arnold B. Krueger on 21 February, 2013, 10:09:01 AM
For example, Benchmark's new DAC2 converter uses a ES9018 DAC IC and achieves a 129 dB SNR A-weighted, and distortion that is better than -109 dB.  This implies that distortion is below the threshold of hearing until peak music levels reach 109 dB SPL.


Does not mean something completely different?  Does that not mean that distortion, being 109 dB down, isn't crossing the threshold of hearing  when peak levels are 109dB up, but rather crossing the threshold of existence?  I can't hear 20dB sounds, much less 0.


My point was that nobody can rationally argue that the distortion is audible if it is below the threshold of hearing.


Equating the threshold of hearing with audible distortion requires a fair amount of naivete if we are leaving the synthetic land of listening to isolated pure tones, and condescending to listening to more complex sounds such as music, speech, and sound effects.  ;-)

Title: DAC IV stages
Post by: Arnold B. Krueger on 21 February, 2013, 10:11:02 AM
Designing an I-V converter with a 130 dB SNR is not trivial.  Impedances need to be very low, and this puts significant demands on the opamp.  The opamp must have very low EIN and must be able to deliver high current drive while maintaining low distortion.  There are about 2 opamps that will work in this application.  Resistors must be metal film (to keep distortion low). 

One complication is that high-frequency switching noise is always present on the current outputs of sigma-delta converters.  This high-frequency noise must be bypassed with rather large NPO or COG capacitors before entering the I-V converter.

There are many opportunities to cut costs, and there are many opportunities to make mistakes when designing an I-V conversion stage.  The I-V conversion stage may be more important than the choice of conversion IC.


Does one have to design anything? Aren't the application notes reliable?  Can't I lift the design of the evaluation board?
Title: DAC IV stages
Post by: John_Siau on 21 February, 2013, 10:28:02 AM
D/A converters with >116 dB SNR are few and far between.  Noise alone is a dead giveaway when comparing D/A converters.


I guess that TI, Crystal Semiconductor, and AKM have been pulling our chains because they all have fairly inexpensive chips that do 115-120 dB dynamic range. ;-)

I've had a LynxTwo since what, 2002 that met that criteria and they are hardly rare. That isn't specsmanship because I've measured it myself. The same converters or competitive converters are now in the < $5 per channel price range.

Most commercial products do not achieve the published specifications provided by the IC manufacturers.  These specifications are an indication of the best performance that can be achieved by these parts on an ideal evaluation board, using lab power supplies.  Lots of things go wrong when manufacturers cram a bunch of parts in a small package while trying to meet cost and schedule targets.  The LynxTwo is a pro product and was not subject to the same design goals as consumer products.

The audio performance of the D/A converters in consumer devices such as TVs, DVD players, portable audio devices, and cable boxes do not come close to 115 dB SNR.  Of equal importance is the fact that nobody listens to these devices with the included digital volume control set to 100%.  Furthermore, the included digital volume controls are usually not dithered (they truncate).

A meaningful ABX test should attempt to replicate the most demanding typical use:

1) Digital volume control at typical setting
2) Loud but reasonable playback levels
3) Reasonably quiet room
4) Accurately matched levels
5) Low-noise source material

If A and B cannot be distinguised under the above conditions, then it is fair to say that neither offers a sonic advantage over the other in typical use.
Title: DAC IV stages
Post by: John_Siau on 21 February, 2013, 10:47:51 AM
Does one have to design anything? Aren't the application notes reliable?  Can't I lift the design of the evaluation board?


Great question!  The answer is no.  It is not quite that simple.

1) The evaluation boards do not have power supplies and magnetic components in close proximity to the converter and analog circuits (a power supply is a necessity in AC-powered commercial products).
2) Battery powered devices must use lower power opamps and higher impedances than those found on an evaluation board.
3) Evaluation boards use very generous amounts of PCB real-estate (often not available in a practical product).
4) Evaluation boards may use components that are outside of the budget of consumer products.
5) Evaluation board often fail to remove the high levels of common-mode distortion and noise that is created by the D/A IC.  These evaluation boards only measure well when feeding the precisely balanced inputs on audio analyzers.  Unbalanced measurements of evaluation boards are often very poor for this reason.  A differential amplifier should be added to remove the common mode components before driving an unbalanced output.  This omission is nearly universal with all of the evaluation boards I have seen.
Title: DAC IV stages
Post by: Arnold B. Krueger on 21 February, 2013, 12:28:56 PM
D/A converters with >116 dB SNR are few and far between.  Noise alone is a dead giveaway when comparing D/A converters.


I guess that TI, Crystal Semiconductor, and AKM have been pulling our chains because they all have fairly inexpensive chips that do 115-120 dB dynamic range. ;-)

I've had a LynxTwo since what, 2002 that met that criteria and they are hardly rare. That isn't specsmanship because I've measured it myself. The same converters or competitive converters are now in the < $5 per channel price range.


Most commercial products do not achieve the published specifications provided by the IC manufacturers.


Given that I have measured about 100 different such devices, let's just say that my evidence says that they often come very close.

Quote
These specifications are an indication of the best performance that can be achieved by these parts on an ideal evaluation board, using lab power supplies


Lab power supplies being easily matched by IC regulators...

Quote
Lots of things go wrong when manufacturers cram a bunch of parts in a small package while trying to meet cost and schedule targets.


Not exactly. Lots of thing can go wrong is true, but that's not saying that they do always go wrong which is the clear meaning of the above sentence.

Quote
The LynxTwo is a pro product and was not subject to the same design goals as consumer products.


Again, I've tested 100s of consumer products and often the line between pro and consumer is blurred, to say the least. Time is part of the story. When the Lynxtwo first came to market the designer told me that the converter chips he used were the most expensive parts on the market and my research put them over $20 each.  Today is a decade later, the parts are still on the market as of sometime in the past year, but in small volumes they cost only a fraction as much.

Quote
The audio performance of the D/A converters in consumer devices such as TVs, DVD players, portable audio devices, and cable boxes do not come close to 115 dB SNR.


As I have pointed out with colaborating evidence, there is no practical need to achieve 115 dB as a rule. It may be fun to thrown out numbers without real-world substantiation, but that leaves one open to challenges from people who have at least one foot in the real world.

Quote
Of equal importance is the fact that nobody listens to these devices with the included digital volume control set to 100%.


Agreed.

Quote
Furthermore, the included digital volume controls are usually not dithered (they truncate).


Again, that appears to be an alleged fact with no real-world evidence to support it.


Quote
A meaningful ABX test should attempt to replicate the most demanding typical use:

1) Digital volume control at typical setting
2) Loud but reasonable playback levels
3) Reasonably quiet room
4) Accurately matched levels
5) Low-noise source material

If A and B cannot be distinguished under the above conditions, then it is fair to say that neither offers a sonic advantage over the other in typical use.


Agreed.
Title: DAC IV stages
Post by: John_Siau on 21 February, 2013, 12:44:30 PM
Take a look at published specifications for D/A converters, DVD players, sound cards, iPod, etc.  Many only have an 80 to 90 dB SNR.  This means that the noise will be audible in an A/B/X test if the peak music levels exceed 80 to 90 dB SPL (easily achieved).


That is far from being a universal truth.  The actual answer depends on the test.


I agree, but the test conditions may be closer to typical applications than you describe below:

The answer given above probably presumes that we set up the system to play music at a desired high SPL, and then mute the music player... 

If we back off from a very high peak SPL to a more typical and comfortable SPL that is not the one we use with visiting firemen ... probably gets us below 100 dB ...

As a practical matter, the vast majority of people find players and systems with 90 dB dynamic range to be highly satisfying, to say the very least.


Using your numbers:
100 dB SPL (slow)
90 dB SNR D/A converter

And adding the following assumptions:
10 to 20 dB music crest factor
30 to 40 dB SPL ambient (quiet room)
10 to 20 dB digital attenuation prior to D/A conversion


system output noise = (average SPL) + (crest factor) - ((D/A SNR) - (digital attenuation))

Worst case given the above assumptions:
(100 + 20) - (90 - 20) = 50 dB SPL

Worst case, the converter noise is 10 to 20 dB louder than the room noise.  This should easily be noticeable between tracks, and may be audible in quiet passages of music, even if the noise is white.

Best case given the above assumptions:
(100 + 10) - (90 - 10) = 30 dB SPL

Best case, the converter noise is equal to the room noise, or 10 dB quiter than the room noise.  It may be very hard to hear this noise between tracks IF the noise is white.

But, please notice the "IF"

The converter SNR may be limited by a single spurious tone and not by white noise.  If so, the noise will be much more noticeable.

Our ears have the ability to hear a 3 kHz tone 30 dB lower in amplitude than the surrounding noise (I encourage the skeptical reader to try this test for themselves).  The 30 dB at 3 kHz number can also be derived from masking theory.

If one of the converters under test has a noise floor that is limited by a single tone (AC hum, power supply switching frequencies, or crosstalk from other parts of the system), then this noise will be much more noticable (and much more objectionable than the above calculations would suggest).

Bottom line, it is quite reasonable to expect noticeable differences in a converter's noise floor in a moderately demanding application.  It is unreasonable to assert that all converters are good enough to be indistinguishable.
Title: DAC IV stages
Post by: saratoga on 21 February, 2013, 02:00:02 PM
80dB is quite rare and would be exceptionally bad for a name brand device.  The Apple stuff will give you 15.5 effecitve bits over headphone out with full analog volume control for not all that much money.


15.5 effective bits when the volume control is set at maximum, and to the best of my knowledge, none of the Apple devices have analog volume control (please correct me if I am wrong).


All portable electronics have analog volume control, Apple's included.  You couldn't buy an mp3 player that had digital volume control if you wanted to.  They aren't made.

With 20 dB digital attenuation, 15.5 effective bits becomes 12.2 effective bits.  The problem is that significant amounts of digital attenuation are often used at normal listening levels, and the converter noise is often audible.


Yeah but no one does this so it doesn't matter.

Edit:  Although I don't mean to imply that analog volume control imposes no SNR hit.  I'm sure at the lowest volume the SNR on an ipod is much worse.  But at reasonable volumes its more then sufficient, which is what I was trying to say.

Yeah but in reality it doesn't make much difference since you never use that dynamic range.  CDs using > 80dB dynamic range are far rarer then D/A converters with < 80dB dynamic range.


CD and high-resolution low-noise recordings do exist, and if these are used in an ABX test, the noise floors of many converters should be audible (on the basis of calculated SPL of the noise).


Yes, but people don't buy CE to do ABX tests of specially selected samples.  They buy them to listen to music.  With enough effort I can find samples that break MP3, AAC, Vorbis, etc.  Doesn't mean I dislike those formats.  It just means I can break them if I try.
Title: DAC IV stages
Post by: [JAZ] on 21 February, 2013, 02:36:31 PM
system output noise = (average SPL) + (crest factor) - ((D/A SNR) - (digital attenuation))


I have several doubts on what you say an definitely, that formula doesn't help.

The part that confuses me especially is why do you imply that lowering the volume digitally, increases the noise level.

What i mean is, decreasing the SNR does not increase the noise SPL.


Quiet room: 30dB SPL.
Let's take a reasonable device (you keep talking about consumer products after all), so output: 90dB SPL
Let's also say that the device has a SNR of -91dB ( 1, just to make it different than 90).

Now, decrease the volume by 20, so max is 70db SPL.
The SNR of the device also now increases from -91dB to -71dB
The difference between max and noise floor (the SNR on this playback situation) is 40db SPL.

Is the room noise floor hearable now, compared to the signal? Maybe.
Is the noise produced by the device audible? I still wonder how.


Let's take a setup with more SNR.
Let's take some in-ear phones. : 100db SPL
Now, since they are in-ear, the room noise floor is reduced, let's say 20db SPL.
Taking the same device that has SNR of -91dB

Let's attenuate by 20dB, so peak signal 80db SPL.
The SNR of the device increases from -91dB to -71dB
Difference between max and noise floor (SNR of playback situation) is now 60dB.

So... what is hearable, after all? The SNR of the device is still below the noise floor.

Title: DAC IV stages
Post by: Ethan Winer on 21 February, 2013, 02:43:41 PM
The first step towards what I think is a more reasonable experiment would be to not allow the listener to stick his head inside the speaker cone

Oh Arny, you are too practical!

I'd apply the same constraint for playback volume. Yes, you can hear quantization distortion on reverb tails at 16 bits if you crank the volume by 40 dB during the fadeout. So don't do that!

--Ethan
Title: DAC IV stages
Post by: John_Siau on 21 February, 2013, 04:49:17 PM

system output noise = (average SPL) + (crest factor) - ((D/A SNR) - (digital attenuation))


I have several doubts on what you say an definitely, that formula doesn't help.

The part that confuses me especially is why do you imply that lowering the volume digitally, increases the noise level.


Decreasing the volume with the digital volume control is necessary to reduce the volume to a normal listening level.  This decreases the signal, but does not decrease the noise level.  Most systems are designed to provide "normal" listening levels at something less than full volume (often 20 dB less than full volume).  If you use 20 dB of digital gain reduction, the SNR degrades by exactly 20 dB. 

If you do the math, the noise floor of consumer D/A converters is often higher than the ambient noise level in a quiet room when audio is playing at "normal" levels.
Title: DAC IV stages
Post by: John_Siau on 21 February, 2013, 05:03:46 PM
All portable electronics have analog volume control, Apple's included.  You couldn't buy an mp3 player that had digital volume control if you wanted to.  They aren't made.


Absolutely incorrect! 

Almost all of these products all have buttons or cap-touch buttons (or sliders) that control volume up and down.  In almost all cases, the volume control is a digital multiplier.  Very few consumer products still have analog volume controls (variable resistors) in the audio path.  DSP is cheap and is already required to do MP3 decoding.
Title: DAC IV stages
Post by: saratoga on 21 February, 2013, 05:23:47 PM
All portable electronics have analog volume control, Apple's included.  You couldn't buy an mp3 player that had digital volume control if you wanted to.  They aren't made.


Absolutely incorrect! 


Name one such product.  Just one.

Almost all of these products all have buttons or cap-touch buttons (or sliders) that control volume up and down.  In almost all cases, the volume control is a digital multiplier.


You've obviously never reverse engineered any of these devices if you think this.

Edit:  I mean think about what you're claiming.  Any modern device can set the line out and headphone out volume independently.  Do you really think theres two stereo DACs included just so that they can use independent digital volume adjustments?  No, that'd be nuts, the cost would be huge.  They have independent volume control because the headphone amp and line out amp volume is software controlled.
Title: DAC IV stages
Post by: [JAZ] on 21 February, 2013, 05:45:23 PM

That the usual listening volume of most electronics is at 1/3rd or 1/4th of the max volume is not a reason to say that digital volume wouldn't have enough SNR.

I guess we all are used to hear the background noise of amplifiers when moving the analog volume up. Yet, that denotes te SNR of that device.

The ideal situation is always an analog volume at the end of the chain, but that only lowers the background noise.
If the amp is able to have a low enough noisefloor, it wouldn't be necessary. Of course, that makes it costlier.
Title: DAC IV stages
Post by: saratoga on 21 February, 2013, 06:05:50 PM
I think SNR is probably not the best way to think about consumer electronics anyway.  Most of these devices have excellent, nearly quantization noise limited DACs that are far more accurate then is actually useful.  Particularly so for anything portable (and thus battery powered).

However, while they'll all give you nearly 16 bit limited performance into a line out, they tend to have fairly limited amplifiers (ignoring Apple, Sandisk which are very good).  They also have essentially fixed noise floors that are independent of volume.  So a more useful approach is to think about them in terms of the impedance and sensitivity that will give good performance.  The noise floor puts a limit on sensitivity, since very high sensitivity headphones will produce more acoustic noise, while the finite output impedance limits how low of an impedance can be driven without distortion. 

That said, quality has been increasing.  Apple and some other companies will now sell you devices with integrated amps that are better then what passed for entry level dedicated headphone amplifiers 5 years ago.
Title: DAC IV stages
Post by: John_Siau on 22 February, 2013, 09:15:03 AM
All portable electronics have analog volume control, Apple's included.  You couldn't buy an mp3 player that had digital volume control if you wanted to.  They aren't made.


Absolutely incorrect! 


Name one such product.  Just one.

Almost all of these products all have buttons or cap-touch buttons (or sliders) that control volume up and down.  In almost all cases, the volume control is a digital multiplier.


You've obviously never reverse engineered any of these devices if you think this.

Edit:  I mean think about what you're claiming.  Any modern device can set the line out and headphone out volume independently.  Do you really think theres two stereo DACs included just so that they can use independent digital volume adjustments?  No, that'd be nuts, the cost would be huge.  They have independent volume control because the headphone amp and line out amp volume is software controlled.

Wrong, but your conclusion that two DACs would be required is correct!

The Apple devices use Cirrus CLI158881 Stereo Codecs which are similar to the CS42L73.  These devices have two stereo DACs, one stereo ADC, and a digital mixing engine.  Such chips form the audio core of almost all portable devices.  Digital gain control is ubiquitous. 

See:

http://www.cirrus.com/en/products/cs42l73.html (http://www.cirrus.com/en/products/cs42l73.html)
Title: DAC IV stages
Post by: scuttle on 22 February, 2013, 09:49:31 AM
All portable electronics have analog volume control, Apple's included.  You couldn't buy an mp3 player that had digital volume control if you wanted to.  They aren't made.


Absolutely incorrect! 


Name one such product.  Just one.

Almost all of these products all have buttons or cap-touch buttons (or sliders) that control volume up and down.  In almost all cases, the volume control is a digital multiplier.


You've obviously never reverse engineered any of these devices if you think this.



From NWAVGuy's dissection and review of the Clip:

http://nwavguy.blogspot.co.uk/2011/02/sans...p-measured.html (http://nwavguy.blogspot.co.uk/2011/02/sansa-clip-measured.html)

DAC linearity is important because most portable devices have digital volume controls that reduce the signal before the DAC


And I'm new here and have never taken a DAP apart, but requiring that someone you disagree with undertake the entire burden of proof while you should be assumed correct by default seems a bit much. If you have dissected DAPs with analog volume controls, you should have said which ones and built a case from there.
Title: DAC IV stages
Post by: Arnold B. Krueger on 22 February, 2013, 10:10:25 AM
All portable electronics have analog volume control, Apple's included.  You couldn't buy an mp3 player that had digital volume control if you wanted to.  They aren't made.


Absolutely incorrect! 

Almost all of these products all have buttons or cap-touch buttons (or sliders) that control volume up and down.  In almost all cases, the volume control is a digital multiplier.  Very few consumer products still have analog volume controls (variable resistors) in the audio path.  DSP is cheap and is already required to do MP3 decoding.


Agreed.

What I find interesting is the fact that 100% of all of the contemporary AVRs that I have examined have both DSPs and digitally controlled analog volume controls.
Title: DAC IV stages
Post by: Arnold B. Krueger on 22 February, 2013, 10:27:16 AM

system output noise = (average SPL) + (crest factor) - ((D/A SNR) - (digital attenuation))


I have several doubts on what you say an definitely, that formula doesn't help.

The part that confuses me especially is why do you imply that lowering the volume digitally, increases the noise level.


Decreasing the volume with the digital volume control is necessary to reduce the volume to a normal listening level.  This decreases the signal, but does not decrease the noise level.  Most systems are designed to provide "normal" listening levels at something less than full volume (often 20 dB less than full volume).  If you use 20 dB of digital gain reduction, the SNR degrades by exactly 20 dB. 

If you do the math, the noise floor of consumer D/A converters is often higher than the ambient noise level in a quiet room when audio is playing at "normal" levels.


That would be related to mathematical models whose relevance can and has been disputed for several decades, starting no later than Fielder's  ca. 1995 AES paper.

Trouble is that in the real world, being disturbed by noise from the DACs is exceedingly rare. We've got ower 10 years experience with consumers playing video DVDs with 16 bit sound at high listening levels and no widespread complaints about background noise. We have  fewer but still a significant number of years of experience with portable digital players with similar results.

When presented with 3 independent well-publicized opportunities to obtain recordings on media with > 16 bits all of the products failed in the mainstream marketplace. It was subsequently found that a very high proportion of those recordings were upsampled recordings that started out with 16 or fewer bits of dynamic range. There were no listener complaints based on just listening until the results of unfavorable technical tests were publicized.

I understand the need for high quality low volume product offerings with far greater capabilities than the minimum that is required for the mainstream market. If nothing else they facilitate the creation of media with the required quality levels since a product can't be better than the means used to create it.  These products need not be justified based on the needs of the mainstream market but can make sense even when they vastly exceed the minimum need.
Title: DAC IV stages
Post by: Arnold B. Krueger on 22 February, 2013, 10:33:40 AM



All portable electronics have analog volume control, Apple's included.  You couldn't buy an mp3 player that had digital volume control if you wanted to.  They aren't made.


Absolutely incorrect! 


Name one such product.  Just one.


The Apple devices use Cirrus CLI158881 Stereo Codecs which are similar to the CS42L73.  These devices have two stereo DACs, one stereo ADC, and a digital mixing engine.  Such chips form the audio core of almost all portable devices.  Digital gain control is ubiquitous. 

See:

http://www.cirrus.com/en/products/cs42l73.html (http://www.cirrus.com/en/products/cs42l73.html)


Agreed. The usual chips we find in mainstream portable digital audio player and smart phones and tablets generally has at last 4 DACs and 2 ADCs.  Even though they are not all used by the specific products, this is also true of the SOC  that is the heart of the economically-priced Sansa Clip+ and Fuze.
Title: DAC IV stages
Post by: saratoga on 22 February, 2013, 11:48:10 AM
Wrong, but your conclusion that two DACs would be required is correct!

The Apple devices use Cirrus CLI158881 Stereo Codecs which are similar to the CS42L73.  These devices have two stereo DACs, one stereo ADC, and a digital mixing engine.  Such chips form the audio core of almost all portable devices.  Digital gain control is ubiquitous. 

See:

http://www.cirrus.com/en/products/cs42l73.html (http://www.cirrus.com/en/products/cs42l73.html)


From the datasheet:

Quote
Analog volume control (+12 to -50 dB in 1 dB
steps; to -76 dB in 2 dB steps) with zero-cross
transitions

Digital volume control (+12 to -102 dB in 0.5 dB
steps) with soft-ramp transitions


Quote
Headphone Analog Volume Control:
Channel A (Address 1Eh) and Channel B (Address 1Fh)
Sets the volume of the signal out of the channel x Headphone (HP) amplifier.


So yes, that DAC have a digital mixer (all modern devices do), but analog volume control. 

Title: DAC IV stages
Post by: saratoga on 22 February, 2013, 11:58:27 AM
From NWAVGuy's dissection and review of the Clip:

http://nwavguy.blogspot.co.uk/2011/02/sans...p-measured.html (http://nwavguy.blogspot.co.uk/2011/02/sansa-clip-measured.html)

DAC linearity is important because most portable devices have digital volume controls that reduce the signal before the DAC


ha, hes wrong.  I somehow missed that reading the review.

And I'm new here and have never taken a DAP apart, but requiring that someone you disagree with undertake the entire burden of proof while you should be assumed correct by default seems a bit much.


How exactly do I prove that something doesn't exist?  I can point out that I've worked on drivers for lots of these devices and not one of them uses digital gain control, but I can't really prove it since I haven't worked on every single device in existence, just a lot of them.

Agreed. The usual chips we find in mainstream portable digital audio player and smart phones and tablets generally has at last 4 DACs and 2 ADCs.  Even though they are not all used by the specific products, this is also true of the SOC  that is the heart of the economically-priced Sansa Clip+ and Fuze.


No, those are 1 stereo DAC devices.  Two is actually rare in low power devices, and off hand I can't think of any dedicated MP3 players that do that.  I think that Cirrus part above is aimed more at tablets and higher power devices, so probably they can afford a little more logic.
Title: DAC IV stages
Post by: John_Siau on 22 February, 2013, 12:17:17 PM
Trouble is that in the real world, being disturbed by noise from the DACs is exceedingly rare...

I understand the need for high quality low volume product offerings with far greater capabilities than the minimum that is required for the mainstream market. If nothing else they facilitate the creation of media with the required quality levels since a product can't be better than the means used to create it.  These products need not be justified based on the needs of the mainstream market but can make sense even when they vastly exceed the minimum need.

Yes, and yes.

But let me bring this thread back to the question of detecting D/A converter differences in an ABX test.  I believe the SNR performance of many common audio products, combined with the ubiquitous use of generous amounts of digital volume control is sufficient to expose detectable differences in an ABX test.  This does not mean that the general public is dissatisfied with these devices, and it does not imply that the converter noise is objectionable, but it does indicate that converter differences should be detectable in an ABX test.  In fact, this has been my experience.

In the past, I posted the results of 2 ABX tests I conducted here at Benchmark:

[a href='index.php?act=findpost&pid=0']ABX test - DAC1 vs MacBookPro via headphones[/a]
[a href='index.php?act=findpost&pid=181']ABX test - DAC1 vs MacBookPro via speakers[/a]

In the two above tests, truncation in the MacBook's digital volume control was a dead give-away and resulted in a perfect score in the ABX tests. 

The calculations I have presented in this current thread suggest that converter noise floor differences can also be sufficient to give away the identity of "X" in an ABX test (when using typical amounts of digital attenuation).  SNR reduction and truncation are two distinct issues that can be caused by digital volume controls.

When digital volume control is used, it takes more than 16 equivalent bits to reproduce a 16-bit noise floor (1 extra equivalent bit is required for every 6.02 dB of digital attenuation).  Again this does not mean that the casual user will be dissatisfied with the audio performance.  However, the casual user MAY be able to detect the difference between two D/A converters.
Title: DAC IV stages
Post by: John_Siau on 22 February, 2013, 12:32:48 PM
The following quote from NwAvGuy's Sansa Clip+ review sums up the situation:

"AUDIBLE HISS (added 2/23/11): Playing back a very low level signal with my most efficient headphones (the UE SuperFi's) the Clip+ has some very slightly audible hiss. Interestingly it seems (subjectively) slightly worse with the Rockbox firmware but I need to investigate that more. With more typical headphones there's zero audible hiss and even with the SuperFi's the hiss in the recording itself and/or background noise leaking past the headphones usually masks the slight hiss. So, in my opinion, it's not a problem unless you have uber-efficient headphones, listen to pristine recordings, and hate even a tiny bit of hiss."

see:

http://nwavguy.blogspot.co.uk/2011/02/sans...p-measured.html (http://nwavguy.blogspot.co.uk/2011/02/sansa-clip-measured.html)
Title: DAC IV stages
Post by: saratoga on 22 February, 2013, 12:40:31 PM
In the past, I posted the results of 2 ABX tests I conducted here at Benchmark:

[a href='index.php?act=findpost&pid=0']ABX test - DAC1 vs MacBookPro via headphones[/a]
[a href='index.php?act=findpost&pid=181']ABX test - DAC1 vs MacBookPro via speakers[/a]

In the two above tests, truncation in the MacBook's digital volume control was a dead give-away and resulted in a perfect score in the ABX tests.


I'm curious how you know that its digital volume control?  Digital control introduces quantization noise, which is white.  Analog volume control introduces thermal noise, which is also white.  Short of plotting the noise power as a function of gain and checking the slope, how would you be sure you heard one or the other? 

FWIW, most PCs use analog volume control (as the Intel onboard specifications use it).  Obviously Apple can do whatever they want, but I'd be surprised if they did otherwise.
Title: DAC IV stages
Post by: greynol on 22 February, 2013, 12:53:44 PM
Since when has NwAvGuy been the ultimate arbiter on such matters?

Since when have we accepted anecdotes as evidence?
Title: DAC IV stages
Post by: scuttle on 22 February, 2013, 01:01:20 PM
From NWAVGuy's dissection and review of the Clip:

http://nwavguy.blogspot.co.uk/2011/02/sans...p-measured.html (http://nwavguy.blogspot.co.uk/2011/02/sansa-clip-measured.html)

DAC linearity is important because most portable devices have digital volume controls that reduce the signal before the DAC


ha, hes wrong.  I somehow missed that reading the review.



No. You have failed to understand your own post. You wrote:

You couldn't buy an mp3 player that had digital volume control if you wanted to. They aren't made.


That the chip used in the Clip (assuming  this is your evidence for your contention - you didn't bother saying why anybody should listen to you) has analog AND digital volume control still makes NWAVguy right and you wrong. Read what you wrote!

And if that chip isn't your evidence, then what is? Are people supposed to say "Yes; this man is angry and can't use apostrophes - clearly he must be in the right!"? No.



Quote
And I'm new here and have never taken a DAP apart, but requiring that someone you disagree with undertake the entire burden of proof while you should be assumed correct by default seems a bit much.


How exactly do I prove that something doesn't exist?  I can point out that I've worked on drivers for lots of these devices and not one of them uses digital gain control, but I can't really prove it since I haven't worked on every single device in existence, just a lot of them.


Yes: your claim was inherently unprovable - which makes it silly - but if you could name a couple of major players then that would at least be indicative.

And you can claim to have written drivers, but so what? The Internet is full of people who claim to be special forces, CIA agents, and aliens - especially when they think it will help them win some silly debate. NWAVGuy is a reasonable expert witness because everyone here knows that he is competent enough to design the  ODAC; all we know about you is that you can type.
Title: DAC IV stages
Post by: scuttle on 22 February, 2013, 01:08:23 PM
Since when has NwAvGuy been the ultimate arbiter on such matters?


You're using a strawman via hyperbole argument.

The question is not whether he is "the ultimate arbiter" but whether an ee engineer who has designed a DAC and who has take several DAP's apart is a reasonable source? Especially compared to none at all.

Which rather answers itself, doesn't it?
Title: DAC IV stages
Post by: John_Siau on 22 February, 2013, 01:10:55 PM
I'm curious how you know that its digital volume control?  Digital control introduces quantization noise, which is white.  Analog volume control introduces thermal noise, which is also white.  Short of plotting the noise power as a function of gain and checking the slope, how would you be sure you heard one or the other? 

FWIW, most PCs use analog volume control (as the Intel onboard specifications use it).  Obviously Apple can do whatever they want, but I'd be surprised if they did otherwise.


Digital volume control introduces noise with a white spectrum if TPDF dither is properly applied.  The spectrum is not white when truncation occurs.  Amplitude linearity is also lost when dither is not applied.  We verified that truncation was occurring.

Most newer media players and newer operating systems have 24-bit volume controls.  These may or may not be dithered (depending on the OS or player selected).  One problem is that some hardware imposes a 16-bit bottleneck between the DSP and the DAC.  The interface between 24-bit DSP (or higher) and 16-bit hardware is still poorly addressed.  Computer systems with newer 24-bit hardware are less-likely to have truncation problems (provided all of the software is up-to-date).
Title: DAC IV stages
Post by: greynol on 22 February, 2013, 01:18:06 PM
You're using a strawman via hyperbole argument.

If you insist.

Quote
The question is not whether he is "the ultimate arbiter" but whether an ee engineer who has designed a DAC and who has take several DAP's apart is a reasonable source?

Appeal to authority much?

BTW, there are a few of us "ee engineers" around these parts.

Quote
Especially compared to none at all.

If you insist.

Title: DAC IV stages
Post by: saratoga on 22 February, 2013, 01:39:47 PM
No. You have failed to understand your own post. You wrote:

You couldn't buy an mp3 player that had digital volume control if you wanted to. They aren't made.


The Clip does not have digital volume control, so even if you take that statement hyper-literally, its still correct (although its also not what I meant).

That the chip used in the Clip (assuming  this is your evidence for your contention - you didn't bother saying why anybody should listen to you) has analog AND digital volume control still makes NWAVguy right and you wrong. Read what you wrote!

And if that chip isn't your evidence, then what is? Are people supposed to say "Yes; this man is angry and can't use apostrophes - clearly he must be in the right!"? No.


Well I wrote the volume control code used in that review, and its open source, so you can read it if you want:

http://git.rockbox.org/?p=rockbox.git;a=co...81ad204720e0eb0 (http://git.rockbox.org/?p=rockbox.git;a=commit;h=103eabd31fb4a5b21e183a7a081ad204720e0eb0)


Yes: your claim was inherently unprovable - which makes it silly - but if you could name a couple of major players then that would at least be indicative.


I don't follow you.  Why is pointing out that Apple, Sandisk and virtually all other companies use analog volume control unprovable or silly?

And you can claim to have written drivers, but so what? The Internet is full of people who claim to be special forces, CIA agents, and aliens - especially when they think it will help them win some silly debate. NWAVGuy is a reasonable expert witness because everyone here knows that he is competent enough to design the  ODAC; all we know about you is that you can type.


You can easily google my username and verify that I've written this stuff.  Or just search on these forums.  I post about this stuff a lot, and plenty of people here know me, NWAVguy included.
Title: DAC IV stages
Post by: John_Siau on 22 February, 2013, 01:41:22 PM
What I find interesting is the fact that 100% of all of the contemporary AVRs that I have examined have both DSPs and digitally controlled analog volume controls.


There is a good reason for this (as I am sure you are aware):

The digitally controlled analog volume control allows control of analog sources without inserting an A/D converter and D/A converter (codec).  The analog volume control chips have certain performance limitations (and costs) but some engineers have chosen this option.  In many cases the A/D and D/A components are available in the box, but input signal level variations can make it difficult to use a digital gain solution.  Low-level input signals place a significant burden on the performance of the A/D as the converter noise may get amplified by digital gain.

We take a similar approach with the DAC2 converter, except that the analog gain control is accomplished with a digitally-controlled motor-driven pot.  We did this to avoid inserting a codec when selecting analog inputs.

In the DAC2 the signal flow is:

Analog input > motor-driven pot > analog output
Digital input > digital gain control > D/A > analog output

We chose to avoid the following:

Analog input > A/D > digital gain control > D/A > analog output
Title: DAC IV stages
Post by: greynol on 22 February, 2013, 01:42:42 PM
That the chip used in the Clip has analog AND digital volume control still makes NWAVguy right and you wrong.

Prove that the digital volume is being used, otherwise you have no room to claim that either person is right or wrong.

Yes: your claim was inherently unprovable - which makes it silly - but if you could name a couple of major players then that would at least be indicative.

Based on the other contention, namely that almost all players implement volume digitally, you would think his claim would be easily falsified.  There's quite a disparity from that and inherently unprovable, no?  I mean, you already think you disproved it if I am to take your contention seriously that NwAvGuy is right and Saratoga is wrong.  Of course at the time of this edit, I think it has been conceded by John that both Apple and Sansa use an analog volume control.  Are these major enough players for you?

The Internet is full of people who claim to be special forces, CIA agents, and aliens - especially when they think it will help them win some silly debate.

Exactly!

all we know

I would be careful in your use of the word we.  You're new here, remember?
Title: DAC IV stages
Post by: saratoga on 22 February, 2013, 01:45:07 PM
I'm curious how you know that its digital volume control?  Digital control introduces quantization noise, which is white.  Analog volume control introduces thermal noise, which is also white.  Short of plotting the noise power as a function of gain and checking the slope, how would you be sure you heard one or the other? 

FWIW, most PCs use analog volume control (as the Intel onboard specifications use it).  Obviously Apple can do whatever they want, but I'd be surprised if they did otherwise.


Digital volume control introduces noise with a white spectrum if TPDF dither is properly applied.  The spectrum is not white when truncation occurs.  Amplitude linearity is also lost when dither is not applied.  We verified that truncation was occurring.


I don't follow you.  How does truncation imply digital volume control?  And how do you know you were hearing it?  You were apparently convinced it was happening on iPods, and that is not the case.  How certain are you that you're not mistaken here?

One problem is that some hardware imposes a 16-bit bottleneck between the DSP and the DAC.  The interface between 24-bit DSP (or higher) and 16-bit hardware is still poorly addressed.  Computer systems with newer 24-bit hardware are less-likely to have truncation problems (provided all of the software is up-to-date).


What hardware is this?  PCI, I2S, etc all handle 24 bit fine and have for many years.  Do you mean USB DACs?  Be specific.
Title: DAC IV stages
Post by: John_Siau on 22 February, 2013, 02:13:17 PM
Well I wrote the volume control code used in that review, and its open source, so you can read it if you want:

http://git.rockbox.org/?p=rockbox.git;a=co...81ad204720e0eb0 (http://git.rockbox.org/?p=rockbox.git;a=commit;h=103eabd31fb4a5b21e183a7a081ad204720e0eb0)


Rockbox uses the MAS3539F MP3 decoder and Codec.

This chip has only one D/A and one A/D as Saratoga has been claiming. 

One block diagram (page 7 of the data sheet) shows a volume control feature in the headphone amplifier (following the D/A converter).  This implies that the unit has some analog volume control capability.

However, a more detailed block diagram on page 8 shows digital mixing and audio processing before the D/A converter (in the digital domain).

The truth is that the MAS3539F uses both analog and digital volume control.  The following is a quote from page 10 of the data sheet:

"To minimize quantization noise, the main volume control
is automatically split into a digital and an analog
part. The volume range is ?114...+12 dB with an additional
mute position. A balance function is provided."

Register 00 11 controls the main volume.  This is "split
between a digital and an analog function" (see page 51 of the data sheet).

Bottom line, some MP3 players may use a combination of analog and digital gain control (see MAS3539F) .  Others only use digital gain control (see CS42L73).
Title: DAC IV stages
Post by: greynol on 22 February, 2013, 02:27:00 PM
So rockbox uses the analog volume control, which is the preferred method, yet NwAvGuy says it sounded worse?

I don't think you get to have your cake and eat it too.
Title: DAC IV stages
Post by: Arnold B. Krueger on 22 February, 2013, 02:36:16 PM
That the chip used in the Clip has analog AND digital volume control still makes NWAVguy right and you wrong.

Prove that the digital volume is being used, otherwise you have no room to claim that either person is right or wrong.


Actually its pretty easy to show that the analog volume control is being used. Reference:

http://www.ams.com/eng/Products/Mobile-Ent...trollers/AS3525 (http://www.ams.com/eng/Products/Mobile-Entertainment/Analog-Integrated-Microcontrollers/AS3525)

download the data sheet PDF and analyze figure 1.

Seems like the only route from the DAC to the headphone amp goes through an analog volume control.
Title: DAC IV stages
Post by: saratoga on 22 February, 2013, 02:36:35 PM
Bottom line, some MP3 players may use a combination of analog and digital gain control (see MAS3539F) .


Like I said above, modern devices do have digital mixers.  So what?  However, the MAS isn't a modern device.  Its a 1990s hardware MP3 decoder with no CPU.  It dates back to a time when mobile CPUs were too slow to process audio. If you're really going to try and argue this, do you want to pick something from at least this millenium

Others only use digital gain control (see CS42L73).


Did you see above where I showed that this is not true? 

Like I said before, you're just wrong about "almost all" of these devices using software volume control.  I've never seen a portable device that uses software volume control, and apparently you have not either.  Or if you have, feel free to tell me about it . . .
Title: DAC IV stages
Post by: John_Siau on 22 February, 2013, 02:38:06 PM
I don't follow you.  How does truncation imply digital volume control?  And how do you know you were hearing it?  You were apparently convinced it was happening on iPods, and that is not the case.  How certain are you that you're not mistaken here?


Truncation is easy to spot on a spectrum analyzer.  Truncation was occurring, but we never determined exactly why.

Truncation can also occur when 16-bit USB interfaces are connected to a computer (as you point out).

I did not say that truncation is happening on iPods, but it may happen on some portable players.  There may be someone on the forum who has some data on this.
Title: DAC IV stages
Post by: saratoga on 22 February, 2013, 02:42:44 PM
I don't follow you.  How does truncation imply digital volume control?  And how do you know you were hearing it?  You were apparently convinced it was happening on iPods, and that is not the case.  How certain are you that you're not mistaken here?


Truncation is easy to spot on a spectrum analyzer.  Truncation was occurring, but we never determined exactly why.


Truncation is ubiquitous on low power devices.  Rockbox is the only portable device I know of that bothers to dither (and even then its off by default since its basically pointless).  But the fact that we (typically) truncate does not imply that we're using digital volume control.  In fact, we use truncation and analog volume control.

So now you've edited the claim to read:

Quote
In the two above tests, truncation in the MacBook's digital volume control was a dead give-away and resulted in a perfect score in the ABX tests.


But this makes even less sense to me.  Why does truncation have to happen in the digital volume control?  How do you know theres even digital volume control at all?  It sounds like you're just assuming . . .


I did not say that truncation is happening on iPods, but it may happen on some portable players.  There may be someone on the forum who has some data on this.


It absolutely happens.  No one dithers.  Its a waste of power.  Bu truncation has nothing to do with volume...
Title: DAC IV stages
Post by: greynol on 22 February, 2013, 02:43:00 PM
Actually its pretty easy to show that the analog volume control is being used.

I think you need to refer to Sansa's implementation before jumping to the conclusion, though common sense is definitely on the side of analog control.
Title: DAC IV stages
Post by: John_Siau on 22 February, 2013, 02:46:37 PM
Like I said before, you're just wrong about "almost all" of these devices using software volume control.  I've never seen a portable device that uses software volume control, and apparently you have not either.  Or if you have, feel free to tell me about it . . .


A digital volume control can be implemented in hardware or software.  It is often implemented in hardware and the software merely sends register commands to the hardware (as in the Rockbox).  Rockbox code exists for MAS3539F hardware, and this is why I used this chip as an example.  I assumed your code was written to support this chip. Did you write code for a different chip?  If so, which one(s).  What chip is used in the clip?
Title: DAC IV stages
Post by: saratoga on 22 February, 2013, 02:53:15 PM
Like I said before, you're just wrong about "almost all" of these devices using software volume control.  I've never seen a portable device that uses software volume control, and apparently you have not either.  Or if you have, feel free to tell me about it . . .


A digital volume control can be implemented in hardware or software.  It is often implemented in hardware and the software merely sends register commands to the hardware (as in the Rockbox).  Rockbox code exists for MAS3539F hardware, and this is why I used this chip as an example.  I assumed your code was written to support this chip. Did you write code for a different chip?  If so, which one(s).  What chip is used in the clip?


Rockbox is an operating system that runs on many dozens of different devices each with different DACs.  The MAS is the first device it ever ran on, being the hardware decoder of one of the very first MP3 players ever made.  But you misunderstand the meaning of those registers.  The MAS does not provide software access to it's PCM data.  There are hardware registers for scaling data because its impossible to do it any other way, and scaling data is often needed (for example, volume normalization).  Volume control is supposed to be implemented using a variable gain amplifier (and in fact we do implement it this way).

There are many different clip players.  The first used the AS3525 which had a DAC similar to the AS3514 integrated.  The newer ones (that are being discussed here) use an unknown chip dubbed (by us) the AS3525v2, which uses a DAC that is related to the AS3543 (although the register maps are not quite identical). 

Title: DAC IV stages
Post by: Arnold B. Krueger on 22 February, 2013, 02:58:12 PM
But let me bring this thread back to the question of detecting D/A converter differences in an ABX test.  I believe the SNR performance of many common audio products, combined with the ubiquitous use of generous amounts of digital volume control is sufficient to expose detectable differences in an ABX test.  This does not mean that the general public is dissatisfied with these devices, and it does not imply that the converter noise is objectionable, but it does indicate that converter differences should be detectable in an ABX test.  In fact, this has been my experience.

In the past, I posted the results of 2 ABX tests I conducted here at Benchmark:

[a href='index.php?act=findpost&pid=0']ABX test - DAC1 vs MacBookPro via headphones[/a]
[a href='index.php?act=findpost&pid=181']ABX test - DAC1 vs MacBookPro via speakers[/a]


I did a little research into this topic and found that there seems to be  a general agreement that the Macbook uses an ALC888 sound chip. Looking around the web I found this interesting result:

http://forums.anandtech.com/showthread.php?p=26725904 (http://forums.anandtech.com/showthread.php?p=26725904)

Basically, they show the results of their technical test suite when run on 3 different system boards.

Two of the test results are similar and can be summarized as  dynamic range = 92.2 dB.

The third is significantly different and shows a dynamic range = 88.8 dB

This makes the point that the performance of the chip is as you suggested in an earlier post, dependent on the actual on-board implementation.

The MacBook is a laptop if memory serves, and conventional wisdom is that laptop sound is often poorer than desktop sound using the same basic components.  Given the variation already in evidence, the Macbook implementation might even be considerably worse that the three examples noted above.

I guess what I'm saying is that I see a presumed diagnosis, but I don't see enough evidence to have an opinion of it, one way or the other.
Title: DAC IV stages
Post by: saratoga on 22 February, 2013, 03:07:20 PM
But let me bring this thread back to the question of detecting D/A converter differences in an ABX test.  I believe the SNR performance of many common audio products, combined with the ubiquitous use of generous amounts of digital volume control is sufficient to expose detectable differences in an ABX test.  This does not mean that the general public is dissatisfied with these devices, and it does not imply that the converter noise is objectionable, but it does indicate that converter differences should be detectable in an ABX test.  In fact, this has been my experience.

In the past, I posted the results of 2 ABX tests I conducted here at Benchmark:

[a href='index.php?act=findpost&pid=0']ABX test - DAC1 vs MacBookPro via headphones[/a]
[a href='index.php?act=findpost&pid=181']ABX test - DAC1 vs MacBookPro via speakers[/a]


I did a little research into this topic and found that there seems to be  a general agreement that the Macbook uses an ALC888 sound chip.


If thats true, then I suppose I was right to be skeptical about the above claims given that the ALC888 has analog volume control.
Title: DAC IV stages
Post by: John_Siau on 22 February, 2013, 03:13:45 PM
Rockbox is an operating system that runs on many dozens of different devices each with different DACs.  The MAS is the first device it ever ran on, being the hardware decoder of one of the very first MP3 players ever made.  But you misunderstand the meaning of those registers.  The MAS does not provide software access to it's PCM data.  There are hardware registers for scaling data because its impossible to do it any other way, and scaling data is often needed (for example, volume normalization).  Volume control is supposed to be implemented using a variable gain amplifier (and in fact we do implement it this way).

There are many different clip players.  The first used the AS3525 which had a DAC similar to the AS3514 integrated.  The newer ones (that are being discussed here) use an unknown chip dubbed (by us) the AS3525v2, which uses a DAC that is related to the AS3543 (although the register maps are not quite identical).


I fully understand the purpose of the registers in the MAS, and I understand that the MAS does not give the CPU access to the PCM data.  But, digital "scaling" is digital volume control.  It may not be the primary volume control system, but it still performs part of the volume control function.  6 dB of gain reduction via "scaling" will move the audio 6 dB closer to the noise floor.  This digital scaling can also cause truncation if it is not dithered.

A variable gain amplifier may also attenuate the audio without having much impact on the output noise.

In either case, if the signal is decreased without also decreasing the output noise, then the effective SNR is reduced.

An analog pot (not found on many modern devices) has the ability to reduce the signal and the noise simultaneously (thus preserving the SNR of the audio).  Digital volume controls and variable gain amplifiers do not replicate the function of a true analog gain control.
Title: DAC IV stages
Post by: greynol on 22 February, 2013, 03:30:11 PM
1) 6 dB is not 20 dB

2) Downward scaling in order to achieve equal loudness will not be done on the type of source material you would use to demonstrate audible problems arising from the use of a digital volume control.  So AFAICT, the only potential issue at play here regarding scaling (in any amount or direction, either downward or upward) is noise caused by truncation.  Furthermore, the end-user generally (always?) has the ability to elect not to use this functionality.
Title: DAC IV stages
Post by: saratoga on 22 February, 2013, 03:38:13 PM
Rockbox is an operating system that runs on many dozens of different devices each with different DACs.  The MAS is the first device it ever ran on, being the hardware decoder of one of the very first MP3 players ever made.  But you misunderstand the meaning of those registers.  The MAS does not provide software access to it's PCM data.  There are hardware registers for scaling data because its impossible to do it any other way, and scaling data is often needed (for example, volume normalization).  Volume control is supposed to be implemented using a variable gain amplifier (and in fact we do implement it this way).

There are many different clip players.  The first used the AS3525 which had a DAC similar to the AS3514 integrated.  The newer ones (that are being discussed here) use an unknown chip dubbed (by us) the AS3525v2, which uses a DAC that is related to the AS3543 (although the register maps are not quite identical).


I fully understand the purpose of the registers in the MAS, and I understand that the MAS does not give the CPU access to the PCM data.  But, digital "scaling" is digital volume control.  It may not be the primary volume control system, but it still performs part of the volume control function.  6 dB of gain reduction via "scaling" will move the audio 6 dB closer to the noise floor.  This digital scaling can also cause truncation if it is not dithered.


You're mixing up truncation and scaling.  Truncation (with or without dithering) happens no matter what.  All of these devices operate at 32 bit internally (except truly ancient devices like the MAS).  You always truncate your 32 bit PCM coming out of your MP3/FLAC/whatever decoder down to 16 bit (or 24 bit in the case of PCs).  Note that this happens no matter how volume control is implemented.  So its no surprise you see the signs of that under a scope.  A quality 16 bit device should basically be expected to do that since dithering is uncommon.

Now you can also scale the waveform.  In rockbox we do this too, although only by small amounts (a few dB for EQ precut, replaygain, etc).  This doesn't change the situation with truncation, since that still has to happen.  Finally, at the end you have volume control, which is done by changing gain to preserve SNR.


A variable gain amplifier may also attenuate the audio without having much impact on the output noise.

In either case, if the signal is decreased without also decreasing the output noise, then the effective SNR is reduced.

An analog pot (not found on many modern devices) has the ability to reduce the signal and the noise simultaneously (thus preserving the SNR of the audio).  Digital volume controls and variable gain amplifiers do not replicate the function of a true analog gain control.


Yes of course.  Like I said before:

Quote
However, while they'll all give you nearly 16 bit limited performance into a line out, they tend to have fairly limited amplifiers (ignoring Apple, Sandisk which are very good). They also have essentially fixed noise floors that are independent of volume. So a more useful approach is to think about them in terms of the impedance and sensitivity that will give good performance. The noise floor puts a limit on sensitivity, since very high sensitivity headphones will produce more acoustic noise, while the finite output impedance limits how low of an impedance can be driven without distortion.


The fixed noise floor kills you.  So while analog gain is better then digital, it doesn't save you running into the noise floor at low volumes.  So SNR does decrease, although slower then 6dB/bit of digital.  The flip side of this though is that going to more effective bits doesn't really help you (unless you do it by reducing the device's analog noise floor).  My point, therefore, was that since we operate under these constraints at all but the very highend, worrying about DAC design is not very productive.  You need better analog amplifiers after the DAC or its all for nothing.
Title: DAC IV stages
Post by: John_Siau on 22 February, 2013, 04:21:24 PM
The fixed noise floor kills you.  So while analog gain is better then digital, it doesn't save you running into the noise floor at low volumes.  So SNR does decrease, although slower then 6dB/bit of digital.  The flip side of this though is that going to more effective bits doesn't really help you (unless you do it by reducing the device's analog noise floor).  My point, therefore, was that since we operate under these constraints at all but the very highend, worrying about DAC design is not very productive.  You need better analog amplifiers after the DAC or its all for nothing.

I agree, and yes amplifiers are sometimes a limiting factor.

If digital gain control is used, the quality of the converter must increase to maintain the same level of performance that could be achieved with an analog gain control following the converter.  Every 6.02 dB of gain reduction that is used in normal listening will require 1 additional bit of effective resolution.  If we start with 129 dB SNR (as in the DAC2), then we can use generous amounts of digital volume control without impacting the overall performance of our playback system (the power amplifier almost always becomes the limiting factor).  But if we start with a laptop that has a 95 dB SNR and use the digital volume control on iTunes or Windows Media Player or the OS then the situation is very different.  iTunes and WMP do not control the gain of an amplifier following the internal DAC, they control the gain in software (digital gain control).  These players and most other computer audio applications place significant demands upon the performance of the built-in DACs.  In a typical media server system, (line out to amplifier, or headphone jack to headphones) the overall performance will suffer if significant use of the software volume control is necessary to achieve a normal listening level.  You can't start with a 95 dB DAC and apply 20 dB attenuation and expect 16-bit performance.  In a practical system, few users want to operate at 100% volume to reach a normal listening level (even though this would give the best SNR).  The sensitivity of the amplifier or headphones demand that adjustments are made. 

Many users are perfectly satisfied with the resulting (95-20)=75 dB effective SNR.  However, it is not unreasonable to expect that people will notice a difference when connecting a 129 dB converter to the laptop media server and amplifier.  The differences should be detectable in ABX tests.  I posted ABX test results showing the effects of truncation (different issue), but similar tests could and should be run on consumer products operating at typical volume settings.
Title: DAC IV stages
Post by: saratoga on 22 February, 2013, 04:24:00 PM
But if we start with a laptop that has a 95 dB SNR and use the digital volume control on iTunes or Windows Media Player or the OS then the situation is very different.  iTunes and WMP do not control the gain of an amplifier following the internal DAC, they control the gain in software (digital gain control).


Is that actually true?  Seems like such a dumb design choice I doubt it, but I don't know much about how the guts of Windows work.

Title: DAC IV stages
Post by: John_Siau on 22 February, 2013, 04:38:46 PM
But if we start with a laptop that has a 95 dB SNR and use the digital volume control on iTunes or Windows Media Player or the OS then the situation is very different.  iTunes and WMP do not control the gain of an amplifier following the internal DAC, they control the gain in software (digital gain control).


Is that actually true?  Seems like such a dumb design choice I doubt it, but I don't know much about how the guts of Windows work.

Yes, the gain control is software-based DSP.

Software-based digital gain control is easy to implement on a fast processor.  A stereo gain-control function places almost no load on the CPU.

From our discussion, it looks like some MP3 players have attempted to do things a little differently, but I am not sure the results are much different (due to the noise limitations of the variable gain amplifiers).  If noise does not change when the volume control is adjusted, then the results are nearly identical (in terms of effective SNR).
Title: DAC IV stages
Post by: greynol on 22 February, 2013, 04:48:54 PM
Software-based digital gain control is easy to implement on a fast processor.  A stereo gain-control function places almost no load on the CPU.

Is there any reason to suspect that software media player volume controls operate at 16 bits or that there will be audible issues with volume controls that operate at 24 or more bits?

From our discussion, it looks like some MP3 players have attempted to do things a little differently

I think "some" is being bit over-cautious, but thankfully (hopefully?) this discussion has moved away from "none" or "a few".
Title: DAC IV stages
Post by: Arnold B. Krueger on 22 February, 2013, 04:56:15 PM
An analog pot (not found on many modern devices) has the ability to reduce the signal and the noise simultaneously (thus preserving the SNR of the audio).  Digital volume controls and variable gain amplifiers do not replicate the function of a true analog gain control.


Most measurements of the performance of an analog pot are made under unrealistic assumptions. They presume that everything following the pot is practically noiseless which is most definitely the the case in the real world.

Slide 13 in http://www.esstech.com/PDF/digital-vs-anal...ume-control.pdf (http://www.esstech.com/PDF/digital-vs-analog-volume-control.pdf) is an example of this unrealistic assumption. In fact the equipment following the pot has far more noise than is shown in the right hand FFT plot.  The actual system noise in the equipment following the pot may be the same or even 20 dB higher than the left hand plot!
Title: DAC IV stages
Post by: John_Siau on 22 February, 2013, 04:57:00 PM
Is there any reason to suspect that software media player volume controls operate at 16 bits or that there will be audible issues with volume controls that operate at 24 or more bits?

Yes, 16-bit processing was very common in the past.  Most newer software operates 32-bit fixed, 32-bit floating point, or occasionally at 64-bits.

Operating systems also insert SRC processing to match up all of the sample rates.  Players such as JRiver, bypass the SRC and use high-resolution gain controls with dithering.
Title: DAC IV stages
Post by: Arnold B. Krueger on 22 February, 2013, 05:01:25 PM
But let me bring this thread back to the question of detecting D/A converter differences in an ABX test.  I believe the SNR performance of many common audio products, combined with the ubiquitous use of generous amounts of digital volume control is sufficient to expose detectable differences in an ABX test.  This does not mean that the general public is dissatisfied with these devices, and it does not imply that the converter noise is objectionable, but it does indicate that converter differences should be detectable in an ABX test.  In fact, this has been my experience.

In the past, I posted the results of 2 ABX tests I conducted here at Benchmark:

[a href='index.php?act=findpost&pid=0']ABX test - DAC1 vs MacBookPro via headphones[/a]
[a href='index.php?act=findpost&pid=181']ABX test - DAC1 vs MacBookPro via speakers[/a]


I did a little research into this topic and found that there seems to be  a general agreement that the Macbook uses an ALC888 sound chip.


If thats true, then I suppose I was right to be skeptical about the above claims given that the ALC888 has analog volume control.


The block diagram of the ALC 888 can be found in Figure 1, page 3 of ftp://209.222.7.36/pc/audio/ALC861_DataSheet_1.3.pdf (http://ftp://209.222.7.36/pc/audio/ALC861_DataSheet_1.3.pdf)

I read it as saying that only the analog inputs of the ALC 888 have an analog volume control.  It appears to me that the path from the DAC to its output terminal ha no analog volume control.  Only muting (off/on analog swtich) seems to be shown.
Title: DAC IV stages
Post by: greynol on 22 February, 2013, 05:05:55 PM
Yes, 16-bit processing was very common in the past.

Right, though there have been free or at least extremely inexpensive solutions that did not involve buying overkill hardware for over a decade now.
Title: DAC IV stages
Post by: John_Siau on 22 February, 2013, 05:06:01 PM
Most measurements of the performance of an analog pot are made under unrealistic assumptions. They presume that everything following the pot is practically noiseless which is most definitely the the case in the real world.

Slide 13 in http://www.esstech.com/PDF/digital-vs-anal...ume-control.pdf (http://www.esstech.com/PDF/digital-vs-analog-volume-control.pdf) is an example of this unrealistic assumption. In fact the equipment following the pot has far more noise than is shown in the right hand FFT plot.  The actual system noise in the equipment following the pot may be the same or even 20 dB higher than the left hand plot!

We fully agree on this. 

Low impedances and/or high signal levels are required to achieve low noise.  An active variable gain stage can be constructed from and opamp and a linear pot.  This configuration can provide 20 dB adjustment while maintaing a 110 to 120 dB SNR.  It takes a low-noise opamp and low resistor values.  We do this in many of our products.  To your point, this isn't going to happen in a low-power device.
Title: DAC IV stages
Post by: saratoga on 22 February, 2013, 10:02:48 PM
Quote

If thats true, then I suppose I was right to be skeptical about the above claims given that the ALC888 has analog volume control.


The block diagram of the ALC 888 can be found in Figure 1, page 3 of ftp://209.222.7.36/pc/audio/ALC861_DataSheet_1.3.pdf (http://ftp://209.222.7.36/pc/audio/ALC861_DataSheet_1.3.pdf)

I read it as saying that only the analog inputs of the ALC 888 have an analog volume control.  It appears to me that the path from the DAC to its output terminal ha no analog volume control.  Only muting (off/on analog swtich) seems to be shown.


I can't use that link, but it says "ALC861".  Using this link:

http://www.hardwaresecrets.com/datasheets/...taSheet_1.1.pdf (http://www.hardwaresecrets.com/datasheets/ALC861-VD-GR_DataSheet_1.1.pdf)

The ALC861 does have analog volume control.

Looking at this link:

http://www.hardwaresecrets.com/datasheets/ALC888_1-0.pdf (http://www.hardwaresecrets.com/datasheets/ALC888_1-0.pdf)

There is a viable gain amp for the ALC888. 

I think digital gain control is very uncommon for PCs as well.  The intel AC97 spec called for analog control.  I believe the newer HD audio spec does as well.

I'm less familiar with PCs though.

But if we start with a laptop that has a 95 dB SNR and use the digital volume control on iTunes or Windows Media Player or the OS then the situation is very different.  iTunes and WMP do not control the gain of an amplifier following the internal DAC, they control the gain in software (digital gain control).


Is that actually true?  Seems like such a dumb design choice I doubt it, but I don't know much about how the guts of Windows work.

Yes, the gain control is software-based DSP.

Software-based digital gain control is easy to implement on a fast processor.  A stereo gain-control function places almost no load on the CPU.[/quote]

Yes but it puts a lot of strain on the DAC by reducing the SNR for no reason at all.  It would be very simple to implement a more intelligent mixer that simply scaled the analog volume to maximize SNR.  Do you have some documentation indicating that MS or Apple chose not to implement such an obvious feature?   


Title: DAC IV stages
Post by: Arnold B. Krueger on 23 February, 2013, 11:11:36 AM
Looking at this link:

http://www.hardwaresecrets.com/datasheets/ALC888_1-0.pdf (http://www.hardwaresecrets.com/datasheets/ALC888_1-0.pdf)

There is a viable gain amp for the ALC888. 

I think digital gain control is very uncommon for PCs as well.  The intel AC97 spec called for analog control.  I believe the newer HD audio spec does as well.

I'm less familiar with PCs though.


My ALC 861 link was obtained by following an ALC 888 link on the Relatek web site. Sight!

I agree that the proper ALC 888 block diagram shows an analog volume control on its output.

Therefore it seems improper to unconditionally attribute any possible misbehavior of a device containing the ALC 888 on a digital volume control.
Title: DAC IV stages
Post by: John_Siau on 27 February, 2013, 09:10:26 AM
I agree that the proper ALC 888 block diagram shows an analog volume control on its output.

Therefore it seems improper to unconditionally attribute any possible misbehavior of a device containing the ALC 888 on a digital volume control.

The ALC 888 block diagram may show an analog volume control, but this does not necessarily mean that audio applications have access to this feature.  In our tests, the iTunes volume control was used to set the level at -20 dB when playing from the Macbook's headphone output.


iTunes has a digital volume control.  Prior to iTunes 7.X the volume control was 16-bits.  With iTunes 7.X and up, the volume control is 24-bits dithered.  Starting with iTunes 7.X, iTunes establishes a 24-bit connection to core audio.  Prior to 7.X this connection was 16-bits.  More information is available here:

http://benchmarkmedia.com/wiki/index.php/I...c_-_Setup_Guide (http://benchmarkmedia.com/wiki/index.php/ITunes-QuickTime_for_Mac_-_Setup_Guide)

The Mac's audio path looks like this when playing audio with iTunes, and is very similar when using most other audio applications:

1) iTunes applies a digital volume control (16-bit truncating before version 7.X, 24-bit dithered starting with 7.X)
2) If the sample rate of the file being played does not match the sample rate set in the audio-midi control panel, then sample rate conversion is applied to the output of iTunes.
3)  After sample rate conversion, the audio is mixed with other audio sources (midi-generated audio, microphone input, and other wave inputs).
4) The digitally mixed audio then reaches the Mac's master gain control (another digital gain control). 
5) Audio is then sent to the output device.

In the case of digital interfaces, (USB and optical are both available on the Macbook), there is no opportunity for an analog gain control.  An analysis of the data shows that digital gain control is active when feeding digital outputs.  In some cases, this digital gain control is undithered 16-bit (older versions of iTunes).  In other cases, it is undithered 24-bit (many other audio applications), or dithered 24-bit (some newer audio applications including iTunes 7.X and up).  The bit depth delivered to the output is a function of the playback hardware, the bit-depth of the source, the version of iTunes, the OS version, and certain settings in the audio-midi control panel.  It is often very hard to acieve a transparent data path.  For this reason, specialized media players have been developed that bypass much of the audio processing, while taking control of the system (audio-midi) settings.  The latest of these players is the JRiver Media Center for Mac (released this week).  JRiver turns off mixing, SRC, and other processing, to deliver bit-accurate transmission of the audio to the digital outputs.

In the tests we ran (Macbook vs. DAC1), the gain control was 16-bit undithered due to the fact that iTunes 7.X was not available at the time.  The iTunes volume control was used in the test.  The existence of 16-bit truncation was verified with an AP System 2, and can be heard in the fade-out at the end of the samples I posted on this forum.

While some on this forum have pointed out the fact that MP3 players often use variable gain amplifiers as a primary gain control, the situation is different in computers.  I had assumed that MP3 players followed the same topology used in computers, but there are some differences.

The need to run a variety of audio producing applications simultaneously demands that a computer use digital volume controls, digital mixing, and sample-rate conversion.

The Windows audio system is very similar to the topology of the Mac audio system, and I will not repeat the overview here.
Title: DAC IV stages
Post by: Arnold B. Krueger on 27 February, 2013, 10:30:59 AM
In the tests we ran (Macbook vs. DAC1), the gain control was 16-bit undithered due to the fact that iTunes 7.X was not available at the time.  The iTunes volume control was used in the test.  The existence of 16-bit truncation was verified with an AP System 2, and can be heard in the fade-out at the end of the samples I posted on this forum.


Can this be summarized as saying that those tests wouldn't produce the same results if repeated today using up-to-date iTunes software?
Title: DAC IV stages
Post by: Nessuno on 28 February, 2013, 07:57:22 AM
While some on this forum have pointed out the fact that MP3 players often use variable gain amplifiers as a primary gain control, the situation is different in computers.  I had assumed that MP3 players followed the same topology used in computers, but there are some differences.

The need to run a variety of audio producing applications simultaneously demands that a computer use digital volume controls, digital mixing, and sample-rate conversion.

Which happens to be the same working condition of "smart" devices. On an iPhone, for example, I verified you can even listen to the audio from a music player application in the background of a phone call if both are routed to the headphone out and in this case both are controlled by the same (digital or analog?) master volume.
Title: DAC IV stages
Post by: John_Siau on 04 March, 2013, 11:28:58 AM
In the tests we ran (Macbook vs. DAC1), the gain control was 16-bit undithered due to the fact that iTunes 7.X was not available at the time.  The iTunes volume control was used in the test.  The existence of 16-bit truncation was verified with an AP System 2, and can be heard in the fade-out at the end of the samples I posted on this forum.


Can this be summarized as saying that those tests wouldn't produce the same results if repeated today using up-to-date iTunes software?

The results may be different.  The truncation artifact would be eliminated if it was caused by iTunes.  The operating system may also need to be updated to eliminate the truncation.  These changes should eliminate the truncation artifact (one of two artifacts that made ABX detection easy).  The second artifact (noise) probably would not change.

One could argue about which artifact is more objectionable, but I have no intention of doing that here.  ABX tests are intended to detect audible differences and do not tell us anything about which "sounds better" when a difference is detected.
Title: DAC IV stages
Post by: Soap on 04 March, 2013, 07:11:34 PM
In the tests we ran (Macbook vs. DAC1), the gain control was 16-bit undithered due to the fact that iTunes 7.X was not available at the time.  The iTunes volume control was used in the test.  The existence of 16-bit truncation was verified with an AP System 2, and can be heard in the fade-out at the end of the samples I posted on this forum.


Can this be summarized as saying that those tests wouldn't produce the same results if repeated today using up-to-date iTunes software?

The results may be different.  The truncation artifact would be eliminated if it was caused by iTunes.  The operating system may also need to be updated to eliminate the truncation.  These changes should eliminate the truncation artifact (one of two artifacts that made ABX detection easy).  The second artifact (noise) probably would not change.


So the solution to a piece of software and/or an OS which implemented a mixer so poorly it truncates is to attach an expensive external DAC?  So instead of just using free software which does its job competently or using an analog volume control (the MacBook Air tests at ~104dB of SNR, leaving plenty of room for other options) for $20 one should throw out the baby and buy your product?
Title: DAC IV stages
Post by: Mach-X on 05 March, 2013, 05:47:13 AM
Rather ironic that everybody is rather correct. The clip+ when the rockbox firmware is installed has near infinite volume settings, that attenuate very smoothly in nature, that would suggest to me, an analog gain adjustment.

However my Galaxy cellphone with wolfson wm8994 has both digital and analog adjustment. With stock firmware the analog is locked (probably for liability reasons as the amp is quite a beast), while the digital gain is adjusted in 8 steps. With the 'voodoo' software modification, the analog amp is unlocked, and the digital amp gets locked to max signal, theoretically improving SNR, as well as EMI, as in most cases with full bit depth, you are using much less gain on the analog amp

Really cheap pmps seem to use digital only with a fixed analog gain. So it would seem both implementations are used.