For ROH covent garden I chose the balcony in preference as you can look into the orchestra pit and get the direct sound rather than the reinforced. The imaging is very odd from the line arrays up the side.
I don't ASK for eyes only and even said I consider-it as "stupid".
But, when i want to know "what sound better" (for me ;-), I use my ears.
Imagine an amplifier that produce distortions in the opposite of those of your speaker. The measurements of this amplifier will show worse performance, the result in your system will sound better.
So I use measurements when I develop, to help-me to go in the right direction, and my ears to figure out if it sound good (or not). You know, those subtle differences that are not unlighted by measurements.And if i have to choose between the two, I will privilegiate my "ears".
I understand your position, but designing audio circuits is something different from "operational". And near each operational IC I tried use to sound a little different from others, even if their datasheet's or measured performances are very close.
Of course, it depends from the application, implementation, environnement etc.
-RM
Something i havent heard here is how channel to channel separation affects the sound between comparable systems.
What is the affect of differing separation or isolation on the sound?
I have my opinion about it but like to hear what others think?
It is an often over-looked parameter in the audible differences between products and entire system.
THx-RNMarsh
you are never too old to try something stupid.
Last edited:
For those interested, my take of Scott's balanced phono with 60dB gain at 1kHz.
http://www.diyaudio.com/forums/anal...ential-single-phono-preamp-6.html#post5420830
Patrick
http://www.diyaudio.com/forums/anal...ential-single-phono-preamp-6.html#post5420830
Patrick
Not that I am aware of. Audio reproduction is a pretty specialized topic that mainstream academia does not appear to be interested in. Why would they?
My DAC and headphone amp experiments showed that channel separation matters a lot. I had one integrated HPA and two mono block HPAs. The mono blocks clearly showed that I had work to do on the integrated unit. Up until that point I didn't realize there was a problem. Didn't think to look closely, actually. Turned out the most of the inter-channel coupling was through the power supply rails. All were decoupled with at each IC with tantalums and ceramics. It took some big electrolytics and some films before I was happy with channel separation. Lesson learned: amplifier stability is one thing, and channel separation is another as far as power conditioning goes.
Ended up with the integrated unit much better and I think quite acceptable to me, but still not quite as good as mono blocks. At least the difference is very small now.
Also noticed that ESS seems to recommend isolated AVCC power supplies for each DAC channel, and preference for single opamps over duals in the output stages (at least where channel separation could be affected).
Bottom line for me: Good channel separation makes the stereo listening experience much more enjoyable.
Last edited:
Now your missing a little humor to remain combative. I test optics eyes only which mean I have to peek.
I know what MTF is I took two graduate courses in image processing. I thought it was wonderful how much the same Fourier math can serve you in a totally different field. My teacher was one of the guys that helped fix the Hubble data before they put in the new correction mirror. You have to wonder how that happened. It's as bad as the guy that left the empty Heineken can inside the LHC.
I personally don't think we need to be experts in quantum mechanics to tackle the audio problem. But at the other extreme there are bad assumptions made, for legitimate reasons, where overly simplistic theories are doggedly clutched. You mention linearity simplifications and that is one I come across all the time, especially in the context of NFB. It is not practical to solve a complex non-linear system by calculus; it can be simulated but then you need to know the limitations of the simulation. When simplistic, linear models of op-amp circuits are used as an argument that a real circuit is perfect and the disappointed listener is at fault for having a distortion dependency, like some addict, then we are in trouble.
IMO engineering is not about force-fitting reality to standard equations from textbooks. It is about seeking out that which has not been included in the equations or models that will explain observations. Or maybe some would call that science; whatever.
Another example comes to mind about the first CD players in the early 1980s. Those around at the time will recall how Philips and Sony, and others, boasted "the perfect sound" based on engineering theory. Well, that was a croc because their engineering theory at the time was not fit for purpose. Even back then there was a media war between those who thought their Dual 505 walked all over a CD player and those who could not accept that a digital technology could be anything but perfect.
If the repeatable observation does not match the theory, first challenge the theory.
Last edited:
Am I the only person who reads speculative, suggestive, and completely undefended comments (and I'm not just picking on you, Richard), like Jeremy Clarkson introducing The Stig?
It's the only logical way I can interpret them. Not that channel separation isn't important, but at what level? Does -100 dB cross talk do it? 0 dB is obviously mono!
Similarly other stupendous claims that would require spot sensitivities in hearing well below the Brownian noise floor (much less any sort of thermal gradient) of air movement in a room at RT. As nicely summarized here: acoustics - How loud is the thermal motion of air molecules? - Physics Stack Exchange
Give me a break.
Try this one :
http://www.diyaudio.com/forums/anal...ntial-single-phono-preamp-30.html#post5420830
Failing that it is post #292 & #298 of your phono thread :
A simplified universal differential or single ended phono preamp
Patrick
Maybe it would help to provide some context to say that some of the test music I listen to is not highly compressed and mostly stays away from FS. There is also some quiet space in the music to listen to decaying cymbal tails and various other little details typically pretty obscured on many systems (the decaying rattles of the snare wires on the bottom head of the snare drum, say). It helps to be able to distinguish some of them if they aren't masking each other. Channel separation helps a lot when they occupy the same frequency range.
EDIT: Besides a system that can reproduce small details, ear training and practice may be needed to hear (notice) things that go by fast and are small. Perhaps something like recognizing words spoken very quickly with an accent. If on is familiar with it, it can be easy to understand. OTOH, if unfamiliar it can all sound like gibberish or simply noise. Or maybe like hearing all the letters in Morse code at 40 WPM. Was that dit-dah-dah-dit, or dah-dit-dah-dit, or what? Too late, its long gone.
EDIT: Besides a system that can reproduce small details, ear training and practice may be needed to hear (notice) things that go by fast and are small. Perhaps something like recognizing words spoken very quickly with an accent. If on is familiar with it, it can be easy to understand. OTOH, if unfamiliar it can all sound like gibberish or simply noise. Or maybe like hearing all the letters in Morse code at 40 WPM. Was that dit-dah-dah-dit, or dah-dit-dah-dit, or what? Too late, its long gone.
Last edited:
One a first order basis, reduction in channel to channel isolation (cross-talk, also) reduces or narrows the "stage" width. And, it tends towards monophonic in character. Some information is lost as well, because of some phase cancellations occurring as you blend the two channels together.
Again, it can be system specific in some cases.... wiring layouts etc. So, you cannot get the full picture unless you measure the entire system you are listening to.
This heard affect of reduced cross-talk figures is something we cannot measure directly as to this effect on the sound we hear but it is very real.
THx-RNMarsh
Last edited:
It's a nonsensical component. Fermi velocity (Fermi energy) is defined at 0 K as the difference between the energy of between the highest occupied electron state and the lowest non-interacting electron state, which for a metal is the bottom of the conduction band, where we'd define the electron's kinetic energy to be zero. (notice METAL: this requires degeneracy, therefore in context of an insulator, i.e. BeO, it doesn't even make sense).
If you're talking Fermi level, then that's all well and good. We're dealing with a wide bandgap material at maybe 500 K? Not gonna be too many CB electrons as result. BeO has a bulk bandgap of 10.6 eV, but this becomes messy when dealing with grain boundaries and the huge number of surface defect states in a powdered material. When does a semiconductor become an insulator and vice versa? Depends on who you ask, but 10.6 eV is a big bandgap; Si is 1.12 eV and GaN, used for high power switching is 3.4 eV, meaning BeO cannot be used for current conduction in what ultimately is a 0.025 ohm resistor or thereabouts. There's just not the free electrons around to make a lick of difference. In sum BeO is NOT going to be a good material to design a low-impedance resistor, at least certainly not for playing part in current conduction.
BUT! All this is generally nonsense when looked at more carefully anyhow. We're dealing with a bulk materials in an ensemble, i.e. messy and a world away from needing QM to explain. So let's get down to something far more reasonable, how one constructs a resistor. A look at how resistors are made show that no one in any mind (right or wrong) will use BeO for anything but the core (for WW and film) or the resistive admixture in a bulk carbon resistor, BeO would make a pretty decent, albeit expensive choice for core given it's excellent temperature properties.There's mention of carbon/graphene in the item description: news flash, bulk graphite has microscopic amounts of graphene and has for eternity. This is nothing new, but makes for great woo to people who don't know better. Your pencil has graphene in it! Such an amazing quantum device!
So to a bulk, we've got either a carbon comp or carbon film resistor using BeO for the ceramic portion of the construction. There's nothing about either large (and by large I mean micron-sized or bigger to be conservative, probably >10 nm and I'm safe) graphite particles nor BeO that makes it an exotic material.
Of course the claims are made that measurements won't be able to discern this noise advantage, which means that the effect isn't actually there. Funny!
I disagree, their assumptions and the framing of the problem were flawed as well as the limits of the available technology. I remind you that there is still a quite vociferous group that thinks 1980's multi-bit DAC's in NOS mode sound wonderful.
The basic theories far predate the CD.
Don't you noticed i wrote: "I don't ASK for eyes only and even said I consider-it as "stupid" instead of "ears only" ?Now your missing a little humor to remain combative. I test optics eyes only which mean I have to peek.
I suggest we segue from the Bybee stuff, it can be a real thread killer. Your points are correct but I've lost count of how many times they have been made. The claims violate some very basic first principles and would require a ground up rewrite of physics as we know it (not a correction).
Last edited:
Very tricky to measure with a mic I would think, most of the effects are going to be in the room but would probably be averaged to a degree between the two ears (in the brain also
)- Status
- Not open for further replies.