Vikash said:
janneman said:
Couldn't agree more. There's a huge amount of non-science and BS in the audio world, and CD players are one of the worst (perhaps second only to cables).
I saw a study a few years ago demonstrating that the jitter produced by players was well below what is audible to human ears. When you consider that even cheap CD players can produce a 20-20k signal that is flatter and with massively less distortion than any speaker system, you do begin to question why high-end players cost so much.
Where there are (audible) differences in sound, it's usually due to the designer modifying the frequency response - a little lift in the higher frequencies is quite common.
Another technique I've seen is to use a valve based pre-amp, effectively adding distortion to the output signal.
There's nothing wrong with doing these things, as long as you accept they are techniques being used to modify the recorded material, in a way that the designer believed would make the resulting sound more pleasing. I.e. there's no magic involved.
IRC The magazine 'HiFi World' shows frequency response and distortion data in its CD player reviews.
sploo said:
HiFi World was a good mag. with good DIY content.
Now it has gone the way of all the others !
Biased reviews, boring repeats or upgrades to previous projects etc.
The bias towards the CD63 DP is typical. (Understandable, I suppose, he got it free!)
This months is no different.
A 'tube clock' review which contains pseudo-technical nonsense, an 'entry-level' amp for 2500, an 'entry-level' turntable for 1500 and part 10 of designing speakers.
Andy
SY said:
Good point. Certainly, a noisy transformer stage would be likely to cause (audible) problems.
pinkmouse said:
*LOL* Actually, I've just checked my notes (forgive me... it was a long time ago) and the valve unit was actually a DAC design - but hey, you get a DAC in almost all CD players!
poynton said:
That is a great shame. The problem of course, is that big statements sell, and you also need advertisers to fund a mag. Explaining to your readers that these five different cables are all passive components, with resistance, inductance and capacitance, and therefore aren't going to make an audible difference, well, it don't get you advertising!
The Audio Critic is well regarded, though online only. http://www.theaudiocritic.com/
planet10 said:
Why? Because they approach things from a scientific point of view; actually measuring frequency responses etc. etc, instead of saying how a particular piece of kit sounds 'delicate with airy mids and a fruity aftertaste'?
They're certainly different to most of the audio press, but I'd place far more weight on a set of calibrated measurements than some bloke testing some product in some unknown environment, who then writes an essay on what he thought he heard on that particular day.
I accept that nothing in the world is black-and-white, and that everyone has an agenda, but I'd definitely lean towards their philosophy over most other mags.
sploo said:
Their science is so biased, limited, & closed minded as to be just as bad as the science of the listen only, full-on snake-oil camp. The analogy of "if the only tool you have is a hammer, then you treat everything like a nail" is highly applicable. They are just as biased but wrap themselves up in a far too limited veil of the apperance of science.
dave
planet10 said:
I guess we'll just have to agree to disagree. I have a scientific background, and work in an electronic engineering environment, and their measurement based approach matches what I see every day.
Over the years, I've read more and more papers on human perception (and its fallability) and papers on rigorous blind testing, that have shown so many audiophile (phool?) myths to be just that.
If I tell you that a speaker has 'great bass', you have no idea of my perception of great bass. It could be that I've spent my life listening to small bookshelf speakers. If you usually run towers with a sub, you're going to have a very different perception of what 'great bass' means. If I tell you that the -3dB point is 25Hz in room, it's far more clear (assuming I give a brief description of the room).
I fully accept that overzealous science is just as bad as overzealous audio 'quackery', but I'd prefer an argument backed up with data, rather than one based on someone's perception.
I do agree that boiling it all down to science makes things a little dry (no pun intended), but I doubt NASA designed the shuttle based on what 'sounds' right, and I wouldn't buy gear based on the same level of review.
sploo said:
As do i
Fallibility as well as the unkanny ability to also be able to pick out things that can be measured but had never been before.
I've yet to see a paper with rigorous blind tests. I'd like to see some. Certainly all the ABX tests being touted are useless because the method is statistically & methologically flawed.
I'll agree with you. When most people talk good bass, they are referring to 50-150 Hz.
2 ends of a spectrum. Far right vrs far left. Incomplete or irrevent data can be as mis-leading as someone's hand waving.
I like to be somewhere in the middle, but i do know that in the end all that is important is that a person can sit down and be imotionally drawn into the musical experience.
But you can be sure that they explored ideas and intuition and then developed ways of exploring that... otherwise we would still be on the ground. That is analogous to hearing something and then digging in to see how to detect that with measures (or find it was illusion). There is no question that our current measurement technology is very primitive (and in a lot of cases seen thru rose-coloured glasses -- ie our ear-brain works in the time domain, but we mostly test in the frequency domain)
dave
Start with Lipshitz and Vanderkooy, move on to Toole, and throw in some Olive to make a perfect Canadian martini of rigorous and well-controlled testing, included the PROPER statement of hypothesis (which is where your usual argument falls down- you like to restate the hypothesis and then triumphantly show that their tests don't answer the restated hypo).
I wish I had been able to get to England to give my talk on this; it's a very common mistake and usually made by pretty smart people. Ah, well, one more reason you'll have to figure out a way to get down here for next year's Burning Amp meeting.
planet10 said:
Well... possibly. I'm sure simulations of the human auditory system could be improved. Could you give examples of the sort of things you mean?
I've seen one designer note: "If it's audible, it's measurable. Just because it's measurable, doesn't mean it's audible". Not saying I completely agree with it, but it gets good point across.
planet10 said:
Wow! Well, that's surprised me. Could you elaborate? That is quite a sweeping statement.
planet10 said:
Certainly. You always have to have a level of faith in the data being presented, and the unbiased nature of the presenter. However, honest data is always better than an honest perception.
planet10 said:
Agreed. At the end of the day, if you enjoy it - it's good enough. I just have a bit of a chip on my shoulder when people make extravagent claims about gear, especially when it doesn't match the science (and in some cases I've seen, is technically impossible).
SY said:
Got a 1st name for Olive. I've seen quite a few L&V papers but none with double blind tests, a general comsumption white paper on a test Toole did, but with little that was concrete.
dave
sploo said:
Not any links, but lots of instances scattered over the last 30 years... even a few posted amoungst the threads here
the 1st bit should be "If it's audible, it's measurable, but we may well not have figured out how to measure it yet"
The test is statistically unable to assert what is often asserted (see SY's comment at the end), the apparatus used in the test can easily add enuff degradation to spoil the results, and most importantly, the nature of the test changes the way people listen. Instead of kicking back and absorb the musical gestalt, you concentrate on listening for identifiabe changes... ie the inherent conditions of the test are not the same as we listen to music.
Carefully selected, scaled, presented data can easily be as deceptive & slippery as a "hi-end audiopile" reviewer's flowery descriptors. I don't believe either at face value.
A standard for instance is a single number THD... almost totally meaningless without seeing the spectrum.
One of the reasons i consider being called an audio-phile an insult... people like that have twisted the original meaning. I've run into lots of those, and have just learned to ignore them. The Audio Critic guy is just as bad, only he is at the complete opposite end of the spectrum -- he gives engineers a bad name.
dave
This is all getting way off-topic. But I'd like to add my 2 cent's worth. All the guys that are saying they are for science are forgetting there is very little science that can predict how well an amplifier will sound AFAIK.
So I have always cared very little for measurements. Does lower THD mean better sound. No. Does the greater channel seperation, greater dynamic range and the lower THD of a CD player mean that a CD player sounds better than a turntable? Not necessarily.
Does more feedback mean better sound. Not necessarily. Does zero feedback mean better sound. Not necessarily.
So the science is only a tool...just as listening is only a tool. And using both is I think by far the best way to develop an amp. But measurements as a reviewing tool I have always found pretty pointless. The amp has been developed and the only thing that matters now is what does it sound like. What is the purpose of showing measurements in a review if it says nothing about the way a product sounds in a certain system? My guess is: That the reviewers somehow want to come across as learned thereby "adding value".
And even saying how something sounds can be pointless unless the reader understands in what context an amp will sound good. Because a 1 watt amp simply does not work with a 82dB sensitive speaker. And a big *** amplifier with 100 parallel output transistors often sounds worse on a 108dB efficient loudspeaker than a 500mW amp with one output device.
Why are diy'ers reading hifi magazines anyway? Most good sounding systems are about synergy and as far as I'm concerned that is where art and emotion take a lead and science takes a supporting role.
So I have always cared very little for measurements. Does lower THD mean better sound. No. Does the greater channel seperation, greater dynamic range and the lower THD of a CD player mean that a CD player sounds better than a turntable? Not necessarily.
Does more feedback mean better sound. Not necessarily. Does zero feedback mean better sound. Not necessarily.
So the science is only a tool...just as listening is only a tool. And using both is I think by far the best way to develop an amp. But measurements as a reviewing tool I have always found pretty pointless. The amp has been developed and the only thing that matters now is what does it sound like. What is the purpose of showing measurements in a review if it says nothing about the way a product sounds in a certain system? My guess is: That the reviewers somehow want to come across as learned thereby "adding value".
And even saying how something sounds can be pointless unless the reader understands in what context an amp will sound good. Because a 1 watt amp simply does not work with a 82dB sensitive speaker. And a big *** amplifier with 100 parallel output transistors often sounds worse on a 108dB efficient loudspeaker than a 500mW amp with one output device.
Why are diy'ers reading hifi magazines anyway? Most good sounding systems are about synergy and as far as I'm concerned that is where art and emotion take a lead and science takes a supporting role.
I'm saddened to see even this thread get swept away by "subjectivism vs objectivism" as does every other thread on this otherwise great forum. 
Is every mod posting on this thread now too??
I'll bring my non-measured CD player to the next meet and the scientists can do a double blind test (maybe when I'm off having a beer or something) and see if they can tell the difference between onboard laptop sound and excellent CD player. I wonder if they can 🙄
Is every mod posting on this thread now too??
I'll bring my non-measured CD player to the next meet and the scientists can do a double blind test (maybe when I'm off having a beer or something) and see if they can tell the difference between onboard laptop sound and excellent CD player. I wonder if they can 🙄
SY said:
I'll have to dig into my AES journals to see if i can find some L&V stuff... so far google has only found one paper that i've been able to download and that was their critique of DSD.
Thanx for the other info.
dave
SimontY said:
I always argue for the middle... all those on the far ends are out-to-lunch IMHO (and that could probably be applied to all facets of life) 🙂
dave
- Status
- Not open for further replies.