What is wrong with all those reviewers? (well, maybe with some reviewers...)

[FennerMachine]

Different species ears are tuned to various frequencies ranges hence how frogs of one type can hear there own calls over others.

We know the frequency range of the human ear but do you actually know how sensitive our ears are? Interesting to research that.

 

[John Duncan]

See my reference to Bayesian inferences as well and/or a transcript (if one exists) of Professor Ray Dolan's talk as part of this past week's 2012 Alan Turing Lectures.

Tempting, but I'm going to listen to some Peter Bradley Adams instead.

 

[the record spot]

It would differ from one age category to the next; human hearing for the most part loses its sensitivity over time, so you want to get in there early if you want to hear how great a system sounds!

 

[manicm]

Ha ha ha Record Spot - humans also lose their sense of taste in old age, so nothing new there . And I'll bet Andrew is older than me, not sure about Clare (she always looks 25 ) and the entire staff of the other UK rags are twice my age.

Tell them they're not fit to run hifi mags anymore. Tell Neil Young he's talking utter nonsense (btw Linn's MD visited him at his home recently).

 

[the record spot]

See my reference to Bayesian inferences as well and/or a transcript (if one exists) of Professor Ray Dolan's talk as part of this past week's 2012 Alan Turing Lectures.

Tempting, but I'm going to listen to some Peter Bradley Adams instead.

Fair dos JD. It's Friday night after all. Here's a link anyway!

http://conferences.theiet.org/turing/

 

[Alec]

...human hearing for the most part loses its sensitivity over time...

This is true, it just is.

And Neil Young has always at least recorded utter nonsense.

 

[the record spot]

Ha ha ha Record Spot - humans also lose their sense of taste in old age, so nothing new there . And I'll bet Andrew is older than me, not sure about Clare (she always looks 25 ) and the entire staff of the other UK rags are twice my age.

Tell them they're not fit to run hifi mags anymore. Tell Neil Young he's talking utter nonsense (btw Linn's MD visited him at his home recently).

You'll probably need to explain your logic more here manic - what are you trying to suggest?

EDIT: More to the point and probably more accurately, what on earth are you talking about...? There's nothing new in what I'm saying; ask any Audiologist in the NHS what happens to human hearing capabilities with ageing.

This is nothing to do with running a magazine, more down to the claims people can make around what they hear, what they think they hear, perception, suggestion and the way the brain deals with all of that.

For pete's sake, it's just music on a stereo!

 

[fr0g]

...human hearing for the most part loses its sensitivity over time...

This is true, it just is.

And Neil Young has always at least recorded utter nonsense.

Actually I like Neil Young's music. It's when he stops playing music and starts preaching nonsense he loses me.

 

[steve_1979]

Did you ABX them?

Sorry everyone, I just couldn't resist it.

Hehehe.

Sorry. I congratulate myself a little every time I get to reply to anyone with that link. Sad innit? And possibly against the rules? Sorry.

Even though I knew what was coming before I clicked the link it still makes me laugh every time.

 

[SteveR750]

Hey whathifi, please review musical fidelitys m1 dac

Why? If you have it and like it, why do you need someone elses opinion? Would you change it if it got 3 stars or less?

Save their effort to review stuff you haven't yet got but might want.

 

[oldric_naubhoff]

Quick - without prejudice - question: have you heard the DACs under review?

no, I didn't Clare. unfortunately I don't have an easy access to everything that pops out on the hi-fi market. I wish I could hear them. however, that's not the point. I'm not claiming (as per my firs post) that all those devices sound the same. the point is the differences emerge not from variations in freq response because freq response of digital sources is essentially the same.

 

[oldric_naubhoff]

link to test?

don't think one exists yet. but even is there was you'd probably wouldn't be able to understand the language (obviously not English).

 

[oldric_naubhoff]

A manufacturer can make their DAC sound any way they want to, so the FR graph for any of the DACs may not necessarily be 'ruler flat'.

I yet have to see one that isn't. at least those which are claimed to be called hi-fi products are flat.

 

[oldric_naubhoff]

How are frequencies tested? I don't know.

Just a constant tone at a set frequency?

Then another tone?

Or an even increase in tones from 20Hz to 20Khz?

Or a spread of tones randomly generated?

Surely different speakers, amps, and even DACS will respond differently to rapid changes?

Making a 1KHz tone is one thing,

Changing from 500Hz to 1Khz to 2Khz and so on rapidly and randomly within a second or two will show up the strength and weaknesses of a DAC.

Oh, wait, that's called MUSIC!

Scientific testing has its place, but I don't think that scientific measurements are up to the standard of our ears and PERCEPTION of sound just yet!

usually you'd be feeding a sweep or white noise, but I presume sweep is more common. and then you see what you're getting on the other side.

I also note your point that devices should play music. that's what they are for in the end, aren't they? definitely not to play isolated sinewaves or squarewaves. but if a device is not characterised with a flat freq response it will not be able to reproduce transients and harmonics correctly. it will not be able to follow source material "to the note", so to speak.

 

[Anonymous]

the point is the differences emerge not from variations in freq response because freq response of digital sources is essentially the same.

methinks you need to retake electrical engineering 101. A frequency response graph over the audible frequencies is exactly the kind of test that will highlight differences in how a DAC sounds.

And, more pedantically, you can't measure the FR of a source. As the name implies, it's a response test, in other words you control the input and you measure the output. When testing a dac, you feed it a digital input and you measure the analog output. There is no "digital source" to be measured.

usually you'd be feeding a sweep or white noise, but I presume sweep is more common. and then you see what you're getting on the other side.

There are two landmark tests for an all-analog ciruit: impulse response and noise shape. The sweep is interesting from an audio standpoint, but not when measuring because it's very hard to characterise: analog circuits include delay elements, which means that if you spot an anomaly in the response you can't tell whether it's caused by the current frequency or an echo of one of the previous frequencies.

 

[BenLaw]

A manufacturer can make their DAC sound any way they want to, so the FR graph for any of the DACs may not necessarily be 'ruler flat'.

I yet have to see one that isn't. at least those which are claimed to be called hi-fi products are flat.

Figure 3 on this Stereophile review shows a non-flat frequency response (using one particular filter):

http://www.stereophile.com/content/bricasti-design-m1-da-converter-measurements

 

[oldric_naubhoff]

Figure 3 on this Stereophile review shows a non-flat frequency response (using one particular filter):

http://www.stereophile.com/content/bricasti-design-m1-da-converter-measurements

ah yes. I actually know this graph. note that it's a special case, i.e. different filter. other filetrs are flat. and more importantly, note the scale! 0.25dB variation. AFAIK 0.25dB is the limit for a transient recognition. so not really relevant variation in this context, is it?

 

[lindsayt]

Figure 3 on this Stereophile review shows a non-flat frequency response (using one particular filter):

http://www.stereophile.com/content/bricasti-design-m1-da-converter-measurements

LOL. Variations of +0db, -0.25db between 20hz and 20khz. Now compare that to any speaker frequency response.

You'd be likely to get more variations in frequency response by slouching in your listening chair instead of sitting up straight.

 

[oldric_naubhoff]

A frequency response graph over the audible frequencies is exactly the kind of test that will highlight differences in how a DAC sounds.

I'm a little bit lost here. do you mean DAC as a product for hi-fi consumers or DAC as a chip meant to be put into a PCB? if the latter then I agree. there's a reason for all this post processing/ filtering affair. but DAC as consumer products have essentially flat response. you could follow Ben Law's link to see some test results on Stereophile web page if you don't believe me.

And, more pedantically, you can't measure the FR of a source. As the name implies, it's a response test, in other words you control the input and you measure the output. When testing a dac, you feed it a digital input and you measure the analog output. There is no "digital source" to be measured.

yet again I think you misunderstood me. term "source" refers to a device feeding signal into an amp (pre or integrated). and since DACs deal with digital input signal hence term "digital source". I've always thought this term was well accepted and well understood in hi-fi circles to group-name devices making use of digital source signal. and to distinguish them from "analog sources". which obviously deal with analog source signal. however, I admit I'm just a mere hi-fi enthusiast and could be wrong. if you feel this term is illogical maybe you have a better suggestion?

There are two landmark tests for an all-analog ciruit: impulse response and noise shape. The sweep is interesting from an audio standpoint, but not when measuring because it's very hard to characterise: analog circuits include delay elements, which means that if you spot an anomaly in the response you can't tell whether it's caused by the current frequency or an echo of one of the previous frequencies.

interesting and thanks for sharing. so here you have it. you can have two devices measuring very similarly in response domain and yet they may have different "sonic signatures". or am I missing something?