Jacob, I don't know if it's a German thing, but arguing against the use of the term "ears only" is just plain silly. But that has already been pointed out.
Perhaps. They used the term, and based their assumptions on it. Wikipedia tells me that (if I understand it correctly) that Mechanism is to be contrasted with various forms of Vitalism - essentially the belief that what emerges from a complex system is somehow more than what can be produced by its parts. To dismiss the need for double-blind audio tests on the basis of assumptions from philosophy seems to me to show that they have no real scientific argument, so they have to invent a spurious metaphysical argument instead. If they had in mind some other meaning for Mechanism then they should have said so.Jakob2 said:
Most physical scientists and engineers assume that a complex system can be considered to be an arrangement of simpler systems. The behavour of the complex system arises from the behaviour of the simpler systems and their interaction together. When dealing with inanimate objects like audio systems this seems to me to be uncontroversial. Controversy arises when this idea is applied to life, as living systems sometimes appear to have properties which do not arise simply from their parts. Hence some people speak of 'emergent behaviour' (which seems to me to be a description, not an explanation); others prefer various forms of Vitalism - but that then leads us into areas we cannot discuss here.
Their argument would make as much sense if it proceeded thus:
- people who demand DBT-only do so only because they believe in cheese
- not all philosophers believe in cheese, therefore all cheese-based claims must be rejected
- therefore DBT-only must be rejected as being contrary to sound philosophy
Elsewhere they assert that trained listeners can switch off their biases. My understanding is that this is simply untrue. However, I suspect that this claim - although I believe it to be false - lies at the heart of their beliefs about audio testing and all the stuff about Mechanism is merely a smokescreen to lend a spurious philosophical flavour to their assertions. Snake-oil salesmen are doing the same trick when they invoke quantum mechanics (nowadays) or radium (a century ago).
You mean "acting just plain silly" must be a German thing? 🙂
Trying to establish a new term instead of a well defined one and the new one being plain wrong and misleading isn´t silly?
Sorry, but they didn´t dismiss the need for DBs due to philosophical assumptions but instead based on their arguments about the weak points of DBs.
The philosophical part is meant to show that the assertion the DB camp would represent "real science" while the others don´t, isn´t so easy to justify.
"They should have said so" ? Come on, the authors couldn´t have known what someone would find in wikipedia 35 years later. 😉
Their meaning of Mechanism is imho related to the mechanistic view of listeners (you provided an example of this sort of mechanistic view, claiming that a DB must be better because the "sighted bias" was removed).
Aah, that seems to be page two..... 🙂
And in fact they didn´t asserted that, instead they wrote, "... be trained to control their biases...", which is something different.
I agree, that humans are abe to control their biases (at least to a certain degree, as humans aren´t perfect) lieas at the heart of their and _your_ beliefs about audio testing.
As you correctly stated that one can not remove all biases, you have to assume that this control mechanism is possible.
of course an experimenter tries to block out any variable (as possible) and tries to randomize the effects that can´t be blocked, but even that is limited.
I gave a different explanation, maybe you should try a bit harder to control your bias?! 🙂
Nicely put!
Mind you, perhaps we should try cheese based cables, the potato ones seemed to be ok..........
I am not a trained arguer so I cannot control my biases.Jakob2 said:
You don't need a degree in philosophy to understand that a test of hearing should involve, as far as reasonably possible, hearing alone. The problem is that such tests often give inconvenient results, so some means must be found to reject them. Waving philosophy around and arguing that hearing works best when accompanied by other inputs would be regarded as silly in any other field than audio. Of one thing you can be absolutely sure: if DBT confirmed that all cables/amps/capacitors sound different then the true believers would gladly embrace DBT.
Faulty logic. There is a huge difference between a careful experimenter having the ability to design an experiment to reduce or avoid some biases, and a test subject having the ability to control his own biases. If I believe that red amplifiers sound best, then I can still design a test which hides the colour from the test subjects; I probably cannot reduce my own red bias when being a test subject.
If the methods and data are fully disclosed, the bias of the experimenter becomes less important.
Ah, yes, I should have had the caveat of "unchecked biases" make the data extracted unreliable.
(my bold above)
It would seem that CDs (same artist and title) from different pressing plants give different checksums anyway.
In some cases I've had 3 CDs (same artist and title) that give different checksums, and they were error free. So databases that compare rips in this manner are potentially flawed IMHO, as it would appear that differences occur in the pressing operation.
As for transports sounding different, I have compared several CD players (used as transports into the same DAC) and they always sound the same (both to me and my musician son). This was not precise blind testing - just casual swapping around.
Likewise - the transports sound the same as my SqueezeBox touch and my Linux 'bit perfect' MPD player with M-Tech hiface adapter.
There are many CD players out there with poorly designed SPDIF outputs, and these do not provide optimum impedance matching, and this is why cables are perceived as sounding "better" - simply because the cable is masking / correcting any impedance mismatch.
The way I understand this is that it primarily affects the start of a track, the preamble. It should not make any difference in the music part
Jan
There was a paper linked from earlier in this thread which may explain this. Part of CD pressing involves a data conversion (EFM) from 8 bits (music data etc.) to 14 bits (on the disc). A given 8 bit pattern can be coded by several different 14 bit patterns; the appropriate one is chosen to eliminate DC. How this is done depends on the details of the algorithm used in that plant. Hence identical music data can end up as different disc data. At the playing end the music data is recovered exactly, whatever EFM was used.audio_tony said:
Is the checksum based on the 8 bit data or 14 bit data? Does the checksum include metadata - such as data of mastering? (I'm guessing here).
The transfer rate from USB is a lot faster than CD, this is why installations from USB tend to be much faster.
It's nothing to do with CD re-reads (although they will slow the transfer rate down of course).
Consider this: BluRay discs have a much, much higher density than an audio CD, and yet BluRay players are able to transfer much more data in real time with no (visible) errors (unless the disc is really dirty or damaged).
So reading a CD should be (and is!) a "a walk in the park" in relative terms.
The point with the error correction is precisely that you can recover 100% correct audio data bits even when you don't get 100% perfect reads from the disc!
So, from a technical point, it may be nice for EAC to try to read all bits from the disc perfectly, it is unnecessary to get 100% perfect audio data! It's a solution to a non-existing problem.
Jan
There are audio data bits perfectly read from the disc and there are audio data bits that are from the error correction system. Obviously the former is preferred over the latter.
How does one know if the error corrected data is the same, or equal to, the
the actual disc data?
Why would one prefer the bits perfectly read from the disk over the bits coming from the error correction system? I find your statement fascinating.
Data off the disc is the actual file data.
The error correction data is the computed simulation data so that the music actually continues on.
I know that the idea is that the two are supposed to be the same, and my point may be a moot point, but I'd still rather have the actual data.
No, not simulated, it is the actual data. Interpolated errors are exceedingly rare.
- Status
- Not open for further replies.