Interesting...

[darrenyeats]

In my personal experience, digital volume control has been more transparent than physical attenuation - when the gain structure has been correct.

The natural gain structure MUST be correct for digital volume control to work properly. Otherwise, excessive digital attenuation becomes necessary and an audible loss of resolution occurs IMO. (The loss is actually around the DAC-pre's analogue resolution, but anyway ...)

 

[Julf]

There is a middle ground if you take the absolutism out of science.

There is middle ground if you take out the pseudo-science from the claims defending a subjective opinion, and recognize subjective opinions for what they are.

 

[merlin]

There is a middle ground if you take the absolutism out of science.

What you mean is "if you accept that science doesn't know it all".

Sadly science functions on the basis that it works with a credible theory until another credible theory comes along. There is a difference between credible theory and what is put forward in most cases sadly.

 

[Julf]

People relying on science naturally argue their point from a scientific POV and require scientific evidence to counter their stance.

People relying on science naturally argue their point from a scientific POV and require scientific evidence for claims that run against established scientific knowledge.

People relying on science naturally acknowledge subjective opinions for what they are, and don't require any evidence as long as the subjective opinions are given as such, instead of supported by pseudoscience and muddled misunderstanding of technology.

 

[ssimon]

It's a topic well worth users of digital music to debate, whether it is better to use a pre amp, or rely on digital attenuation.
Sometimes a debate can become difficult when it becomes incomprehensible.
[QUOTE
The natural gain structure MUST be correct for digital volume control to work properly.[/QUOTE]

No doubt somebody will understand it.

 

[Steven Toy]

Spoil-sport. We want our fun.



Yes it's down to implementation and like everything in life there is good, bad and meh.

I think Steve is still using a level 95 turbo fusion powered MDAC so I would expect that to be a superb design. Likely the results are down to expectation bias, probably from reading dodgy reviews

Regarding my own experience you are one of three people to have heard my system both with and without the pre. St the time you acknowledged the effect of the pre (and its analogue attenuation.)

 

[Julf]

Regarding my own experience you are one of three people to have heard my system both with and without the pre. St the time you acknowledged the effect of the pre (and its analogue attenuation.)

So how do you know what part of the difference is due to the pre itself, and what part is due to the analog attenuation?

 

[Steven Toy]

So how do you know what part of the difference is due to the pre itself, and what part is due to the analog attenuation?

I couldn't draw a line down the middle like that. Obviously both are having an effect.

 

[Julf]

I couldn't draw a line down the middle like that. Obviously both are having an effect.

"Obviously"? Couldn't it be that the audible difference you hear is completely caused by the pre, independently of the analog attenuation?

 

[Whaleblue]

Surely attenuation per se is the bigger issue. What I mean is that the difference in both a human's perceived frequency curve, and characteristics of speakers changing frequency response, at different volumes, is going to be far more significant than digital or analogue circuitry attenuation introduces?

 

[merlin]

"Obviously"? Couldn't it be that the audible difference you hear is completely caused by the pre, independently of the analog attenuation?

That is undoubtedly the most likely explanation.

I for instance use a tube pre for the very reason that it ADD's something to the system that I personally value.

I'm in no doubt that it's more coloured than a straight line with gain. It's more entertaining though.

 

[Paul R]

"he is wrong" is just as sweeping a statement as the very one's you have issue with unless specifically evidenced.

You paraphrased Steven's point as,

It's just an article that Steve picked to help him illustrate a point. Yes, there isn't concrete reference to theory and the author is somewhat divisive but Steve's point (I think) is that there are few available good digital attenuation devices in built into things like DACs, so getting hung up on the author or the article is perhaps missing the point

And if that is his point, then he is wrong, as I said. It would be rare for a modern DAC to have a poor digital attenuation implementation. Steven's DAC contains an ESS Sabre chip,(they seem to be ubiquitous), and that does not have a poor attenuator.

I agree with you RE the writer. I don't think that his theorising is credible but the GENERAL point was that in the main, analogue attenuation seems to sound better to some (myself included). As stated (please re-read) there are of course exceptions where digital attenuation is done properly.

It's a non-point that some people believe something.

No-one is arguing that properly configured, digital attenuation is largely inaudible until at least -30dB is reached but it is simply foolish to poo-hoo the fact that some digital attenuation devices (precisely because they are not properly implemented) do in fact degrade SQ quite noticeably, therefore quite measurably too.

Name some of these poor attenuation implementations, and can you show the measurements?

Why do so many threads on PFM these days simply seem to degenerate into pointless circular arguments usually involving subjectivist V's objectivist with no middle ground these days? Very few threads, this one included, seem to allow for genuine grown up debate without degenerating into meaningless squabbling.

This isn't 'subjectivist' vs 'objectivist'.

I formed my opinion of the digital attenuator in a Sabre by listening, and, unlike Steven, I've also done some listening at the extremes of level and attenuation, just to see how well it all works.

Paul

 

[maxflinn]

It's true. People relying on science naturally argue their point from a scientific POV and require scientific evidence to counter their stance.

It's generally not possible to provide such evidence, and so people with a genuine love of music and a passion for it's replay are left bashing their heads against a brick wall.

I can't see how there can be a middle ground when it comes to science.

Are you suggesting that only those that don't care for science have a genuine love of music and a passion for it's replay?

 

[adamdea]

It's a topic well worth users of digital music to debate, whether it is better to use a pre amp, or rely on digital attenuation.
Sometimes a debate can become difficult when it becomes incomprehensible.
[QUOTE
The natural gain structure MUST be correct for digital volume control to work properly.

No doubt somebody will understand it.[/QUOTE]
The point is not too complicated. A power amp generally has fixed gain. This is not the same as the power spec. All amps with the same gain will (up to the point where they start to run out of power) amplify the same input signal by the same amount.
Active pres also tend to have gain, whereas a passive pre will generally only reduce the output of the dac (or other input).

If you have a lot of gain in the system, then a full (2v plus) dac output will map to deafening output. If your actual listening level is sane then you will need lots of attenuation. This is silly- it's a bit like having your foot permanently down on the accelerator so that you need to much harder on the brakes.

This is silly whether you use digital attenuation or analog attenuation, but the silliness of it with analog attenuation doesn't tend to bother the sort of people who prefer analog attenuation, whereas people tend to make a lot of fuss about it in relation to digital attenuation. Either way it makes more sense to reduce the gain after the volume control either in the dac output, the pre amp or in the power amp.

In many ways it makes most sense to reduce the gain at the end. This is because it will amplify everything upstream (eg, say, noise in the Dacs output stage or picked up the cable between the dac and power amp).

Putting this together.... It is only because of high (and probably unnecessary) levels of gain after the dac that you need to have high levels of digital attenuation. If there is less gain after the dac then it can be run at a higher level.

This is common sense even if (which is probably the case) there is nothing wrong with digital volume controls. (think about the analog noise amplification).

 

[zarniwoop]

If this is so the dCS Vivaldi stack at £67,000 is also poorly designed.

It could be.

A more cynical view would be that the demo was conducted by someone who has an expensive pre-amp to sell, and therefore a preference for an outcome that does not render that redundant. Unless you dismiss entirely the power of suggestion then the way in which the demo was conducted, and the words used, could have influenced the result.

I'm not saying this is what happened - just offering a plausible alternative explanation.

 

[Joe P]

Max,

Are you suggesting that only those that don't care for science have a genuine love of music and a passion for it's replay?

I think what merlin is saying is that people with a scientific outlook view the world differently from blokes off the street, unless those blokes happen to be scientists.

Actually, I have no idea if this is what merlin is suggesting, but it gave me an opportunity to post a comic that's funny because it's true.

Joe

 

[Steven Toy]

It could be.

A more cynical view would be that the demo was conducted by someone who has an expensive pre-amp to sell, and therefore a preference for an outcome that does not render that redundant. Unless you dismiss entirely the power of suggestion then the way in which the demo was conducted, and the words used, could have influenced the result.

I'm not saying this is what happened - just offering a plausible alternative explanation.

That clutches at straws a bit.

 

[adamdea]

Surely attenuation per se is the bigger issue. What I mean is that the difference in both a human's perceived frequency curve, and characteristics of speakers changing frequency response, at different volumes, is going to be far more significant than digital or analogue circuitry attenuation introduces?

Signs of thought. Have him shot.

 

[zarniwoop]

That clutches at straws a bit.

Really? What about the infamous toe-tapping, head-bobbing audio salesmen? Are you familiar with the work of Derren Brown?

 

[darrenyeats]

Actually, I have no idea if this is what merlin is suggesting, but it gave me an opportunity to post a comic that's funny because it's true.

I like that.