This is why WHF should take measurements

[steve_1979]

I'm not saying that all DAC's will sound identical because they don't. Although IME the vast majority of them have such low destortion that they are audibly transparent and do sound identical. There are several reasons why some DAC's may sound different as a few of them are deliberately 'voiced' to have a certain sound character or the designer may have fudged the design or implimentation of the DAC in some way. Also a badly designed DAC can give off electromagnetic interference which will have a detrimental effect on some amplifiers while other amplifiers will be more immune to this.

In a recent forum post WHF's technical editor Ketan Bharadia suggested to someone that they should spend money on buying a new DAC. Clicky. Due to the vast majority of DAC's having vanishingly low distortion measurements they almost always sound the same. IMO any hifi magazine such as WHF should have a moral obligation to take measurements of DAC's to ascertain whether or not they really do sound different before giving advice to people to spend money on a new one.

Comparing the output of two DAC's is a very simple and cheap exercise to do. All you need to do is plug them into a computer and record the analogue output from each DAC and save it as lossless digital audio files. Then using free software such as Audacity you can analyze the results by performing a null test to check to see if there are any differences between them. The null test creates a third digital audio file from the first two which has only the differences recorded on it and everything else is removed. If you play the null test audio file and there is no sound at all you can be 100% certain that the output from both DAC's will sound identical.

Before giving people advice on which DAC's they should spend their money on WHF should make sure that the advice they give is really true and correct. Anything less from the UK's leading hifi magazine is immoral IMO.

What are your thoughts on this?

 

[steve_1979]

Also, just to add. The same applies to CD player as to DAC's.

 

[Vladimir]

Although I wholeheartedly agree with you Steve, I don't think people care for measurements. Therefore, why waste resources. *scratch_one-s_head*

 

[plastic penguin]

In the real world, people, mainly, want to hear music and not get bogged down with unnecessary garbage.

 

[RobinKidderminster]

+3. Measurements may not tell the whole story and may still rely on interpretation but they could contribute chapter one to the book. Could be applied to most kit.

Car specs dont determine how the car feels to drive but it gives valuable comparison between models and a base line for short listing.

 

[Deleted member 2457]

I think it's lacking something and needs a change so good idea.

 

[rainsoothe]

hm... well actually he was telling the op NOT to switch dac to the one he was considering. And I don't know about measurements and whatnot, all I know is that the Naim Dac V1 sounds WAY better then the Musical Fidelity M1 dac I used to own. I don't know the reason, but I don't really care, since there's nothing I could've done to make the MF sound like the Naim, so why would a measurement be relevant to my sound preference? I mean unless I'm planning on building my own, it really is, imo, a completely irrelevant piece of information.

 

[davedotco]

hm... well actually he was telling the op NOT to switch dac to the one he was considering. And I don't know about measurements and whatnot, all I know is that the Naim Dac V1 sounds WAY better then the Musical Fidelity M1 dac I used to own. I don't know the reason, but I don't really care, since there's nothing I could've done to make the MF sound like the Naim, so why would a measurement be relevant to my sound preference? I mean unless I'm planning on building my own, it really is, imo, a completely irrelevant piece of information.

Naim equipment is voiced to sound like Naim equipment, if you like it, and many do, nothing else sounds the same.

In that respect you are quite correct, measurements would be irrelevant.

The question you should be asking is why the two dacs sound so different, the performance of modern dac chips are virtually identical, so why the big difference?

 

[Al ears]

Whilst I can entertain the idea that things such as DACs that measure the same will sound the same the chance of ever finding two items that measure identically is slim.

Although variations within the actual component is likely to be very slim they do occur. Likewise DAC's are designed to be eminantly 'tuneable'. Built in or user adjustable 'filters' abound (this is how Naim can design a 'house sound' into a DAC). To that end getting to DACs that actually measure the same is not that easy.

I might add that theonly reason I'd change my DAC (the unit itself not the component) would be to get something that might playback higher resolution files.

 

[steve_1979]

hm... well actually he was telling the op NOT to switch dac to the one he was considering.

True. But while he suggested that it isn't worth upgrading to the iRDAC he does says that "iRDAC is better sounding than the rDAC".

It is also suggested that upgrading to the even more expensive Audiolab MDAC would be a worthwhile improvement.

 

[steve_1979]

Whilst I can entertain the idea that things such as DACs that measure the same will sound the same the chance of ever finding two items that measure identically is slim.

While no two DAC's will measure exactly identical most of them will have distortion levels so low that the differences between them are so small that they will sound the same as far as our ears can tell.

This is where the null test comes in handy because it only show the differences between them. You will be able to measure incredibly small differences but if these differences are so small that they sound silent in a null test then the two DAC's are audibly transparent and will sound identical to our ears.

IMO WHF should check to see if DAC's are audibly transparent before claiming that other more expensive DAC's sound better when in reality the majority of them sound identical.

 

[andyjm]

Its actually very difficult to take precise audio measurements, and even harder to interpret them. Rather like giving a man-in-the-street a stethoscope and asking him to diagnose your health problems, having to the right tools isn't enough, you need experience and training in how to use them. There may be exceptions, but Hifi journalists generally don't seem best placed to do this.

The other issue is that even if you knew what you were doing and had the gear, electronic test equipment is generally very narrow in its focus (level, phase, frequency, voltage etc) and it is hard to encapsulate the 'sound' of a piece of equipment in a single number.

It is easy however to establish whether two pieces of equipment are different, as long as you don't care what the difference is. If you record the output of a system, substitute a component, record again and then compare the two recordings, while you won't be able to say what is different, you will at least be able to say if something has changed.

While this sounds simple, there is a bit more to it than that. You need an identical source each time, - these days digital files provide just that. Don't bother trying the test with a turntable. You need to level match and time-align the before and after recordings to produce an accurate difference file. Fortunately there is a software package (for free) 'audiodiffmaker' that does just this.

Even the most hamfisted journo should be able to record a before and after file and run it through diffmaker. Any screw up will make the before and after files more different to each other, producing a null result requires everything to match properly - the test is self policing.

While this won't demonstrate that one amp/streamer/cd player/cable is better than another, it will at least demonstrate that they are different (or not). For the sake of a 15 minute test and a free software download, a journalist could put an end to the endless 'this mains cable/interconnect/speaker cable...' makes the 'blacks blacker, lifts a veil....' (delete as appropriate).

So the question is, why don't they? (to be fair, I think we all know).

 

[spiny norman]

So the question is, why don't they?

 

[Vladimir]

If you owned Haymarket, would you do measurements and DBTs?

 

[Tear Drop]

Before giving people advice on which DAC's they should spend their money on WHF should make sure that the advice they give is really true and correct. Anything less from the UK's leading hifi magazine is immoral IMO.

What are your thoughts on this?

Expecting a magazine like WHF to be anything other than a puffed up advertising leaflet in the world we live in is naive at best. Expecting it to be 'moral' is just a bizarre line of thinking.

 

[spiny norman]

If you owned Haymarket, would you do measurements

Doesn't seem to be a problem for the likes of 'Autocar' and 'What Car?'

and DBTs?

I thought WHSV used to claim it did do blind tests, then it said it didn't, then it did, then of course the Review Editor said he was some kind of omniscient faith-being and had no need for any objective evidence/assessment.

 

[steve_1979]

If you owned Haymarket, would you do measurements and DBTs?

In depth technical measurements: No

Blind A/B, ABX and null tests. Yes

I would also include frequency response (with +/- dB) in the specification section for speakers. But I wouldn't include any graphs or anything like that because most people aren't interested in them.

 

[Vladimir]

In depth technical measurements: No

Blind A/B, ABX and null tests. Yes

I would also include frequency response (with +/- dB) in the specification section for speakers. But I wouldn't include any graphs or anything like that because most people aren't interested in them.

OK. Now take 95% of your monthly income and give it away to charity and live on the remaining 5% for the rest of your life. It's practically the same thing.

 

[Covenanter]

You sell magazines by pandering to people's illusions not by destroying them.

Chris

 

[rainsoothe]

hm... well actually he was telling the op NOT to switch dac to the one he was considering. And I don't know about measurements and whatnot, all I know is that the Naim Dac V1 sounds WAY better then the Musical Fidelity M1 dac I used to own. I don't know the reason, but I don't really care, since there's nothing I could've done to make the MF sound like the Naim, so why would a measurement be relevant to my sound preference? I mean unless I'm planning on building my own, it really is, imo, a completely irrelevant piece of information.

Naim equipment is voiced to sound like Naim equipment, if you like it, and many do, nothing else sounds the same.

In that respect you are quite correct, measurements would be irrelevant.

The question you should be asking is why the two dacs sound so different, the performance of modern dac chips are virtually identical, so why the big difference?

Why? Why should I be asking that question? For me, the only relevant question is HOW the sound of one piece of gear differs from another, and not how the difference is achieved. The rest of the data is for the geek in me, or the engineer in whomever, but from a strictly listening and musical enjoyment perspective, I think the "why" is completely irrelevant.

 

[Vladimir]

Critical thinking is bad for your blood pressure.

 

[relocated]

You sell magazines by pandering to people's illusions not by destroying them.

Chris

Absolutely.

It won't ever happen Steve because there would be far too many items that, for human ears, are indistinguishable. So a twofold collapse, no advertisers and huge numbers of hacked off punters who believed the hype and spent their money needlessly.

It would be the honest and decent thing to do, but it ain't going to happen.

 

[davedotco]

hm... well actually he was telling the op NOT to switch dac to the one he was considering. And I don't know about measurements and whatnot, all I know is that the Naim Dac V1 sounds WAY better then the Musical Fidelity M1 dac I used to own. I don't know the reason, but I don't really care, since there's nothing I could've done to make the MF sound like the Naim, so why would a measurement be relevant to my sound preference? I mean unless I'm planning on building my own, it really is, imo, a completely irrelevant piece of information.

Naim equipment is voiced to sound like Naim equipment, if you like it, and many do, nothing else sounds the same.

In that respect you are quite correct, measurements would be irrelevant.

The question you should be asking is why the two dacs sound so different, the performance of modern dac chips are virtually identical, so why the big difference?

Why? Why should I be asking that question? For me, the only relevant question is HOW the sound of one piece of gear differs from another, and not how the difference is achieved. The rest of the data is for the geek in me, or the engineer in whomever, but from a strictly listening and musical enjoyment perspective, I think the "why" is completely irrelevant.

That is of course the 'correct' answer in moden hi-fi philosophy, but I prefer a different approach.

If two products that should sound effectively identical do not, then I find myself asking why. Generally, I think, it is an attempt by the designer to imprint his, or the brands, character on the sound and this troubles me.

Not, I hope, for any geek-ish reasons but for the simple and quite practical requirement that the system does not favour different types of music. A simple example, a system that excells in terms of PRaT may sound fantastic on most uptempo music, driving a lot of rock and electric jazz along at a pace that sounds involving and exciting. I used to have a system that sounded just like that but, as I was drawn into the world of opera, I found that the system failed to respond to the tempo and mood changes that are so important, everything sounded rushed and the enjoyment was ruined.

I do play a lot of different music, I may 'get into' something and play it a lot but sometimes, as now, I am in a kind of 'scattergun' mode, playing different things. In the last few days this has included some Jamaican Dancehall, mid 60s acoustic jazz, Arizona bar room rock and roll and some more modern 'indie' pop by the likes of Diiv and Imagine Dragons.

Again a simple example, I would not want a setup that gave me extra bass on the reggae, welcome as it might be in that instance, if it was going to smear the bass lines of the jazz which would make some of the stuff I listen to impossible.

Appologies for the length of the reply, but I did not want to contradict your post without trying to explain why I think these differences are more important than a simple matter of taste.

 

[MeanandGreen]

I find it quite intetesting how on this thread the general opinion is that measurements aren't quite so important. That all emphasis shouldn't be on how it measures, but how it sounds.

Yet over on the NAD D3020 thread everyone is banging on about the measurements...

 

[bigboss]

What Hi Fi is not alone in their opinion.

https://www.avforums.com/threads/irdac-vs-rdac-sort-of-review.1807690/