Not only Michael Fremer could not get enough of it. I have another magazine (the magazine was a gift, I don't buy hifi magazines anymore) where the reviewer had to redefine his audio universe because of that CD player.
[EDIT] It was Roy Gregory (the other reviewer) and he rated it, together with the Lyra Connoisseur 4.0 as the best products he ever reviewed.
There's still a great deal about listening that we still do not fully understand.
http://www.livescience.com/health/080818-seeing-sound.html
Maybe this explains some?
bandwidth? THD above 20khz? digital sampling rates? feedback? other?
Maybe we should spend less time at the testbench and rely more on what our mind is telling us.
The goal, IMO, should be the listening experience, not the specs game.
http://www.livescience.com/health/080818-seeing-sound.html
Maybe this explains some?
bandwidth? THD above 20khz? digital sampling rates? feedback? other?
Maybe we should spend less time at the testbench and rely more on what our mind is telling us.
The goal, IMO, should be the listening experience, not the specs game.
Back to the challenge of layout and parts quality. What do I mean by parts quality? Do I mean high prices, pretty finish, both?
Think about faked products that look the same for all intents and purposes. Is a fancy watch that is counterfeited that same, for all practical purposes as an original?
Is all chicken that is to be cooked, really the same quality, and therefore taste?
Is all fruit at the supermarket the same as if you could pick it from a tree or buy it from the local farmer?
For all intents and purposes, the watch, the chicken, and the fruit look and appear to be the same. At least until you use the watch for awhile, or eat the prepared chicken or fruit.
Now, if you have no appreciation of these differences, then the minute differences that I work with will be of no use to you. For you then: 'Parts is parts'. It is the same for chicken or audio.
Now what are the differences in audio parts, including caps, connectors, and wiring?
Think about faked products that look the same for all intents and purposes. Is a fancy watch that is counterfeited that same, for all practical purposes as an original?
Is all chicken that is to be cooked, really the same quality, and therefore taste?
Is all fruit at the supermarket the same as if you could pick it from a tree or buy it from the local farmer?
For all intents and purposes, the watch, the chicken, and the fruit look and appear to be the same. At least until you use the watch for awhile, or eat the prepared chicken or fruit.
Now, if you have no appreciation of these differences, then the minute differences that I work with will be of no use to you. For you then: 'Parts is parts'. It is the same for chicken or audio.
Now what are the differences in audio parts, including caps, connectors, and wiring?
1audio wrote
Kendall Castor-Perry's work is a reminder of two things:
1) Manufacturer's SPICE models range in usefulness from reasonably useful to downright misleading
2) Simulations that account for only a portion of the real physical circuit under consideration are no more than pointers to better performance.
This particular body of work I recommend because it can be replicated by anyone with the time and inclination because the essential tool is free. (A hearty cheer for Linear Technology - and no, I don't work for them!)
To get simulation results that translate directly (or so damn close as not to matter) to reality then expensive/proprietary simulation tools are needed. This is well outside the scope of DIYers and all but the seriously well-funded commercial operations. I can't imagine many audio designers are going to put down $30k for a field solver.
But we can deduce a few really useful points:
1)Everything else being equal, the decoupling cap in the smallest package is the best - so surface mount 1608 package, which is the smallest package that can be comfortably/repeatably hand soldered/reworked.
2)Minimize trace inductance by using PWR/GND plane pairs, short PWR/GND traces and minimize the number of vias in the PWR/GND traces
3)Use a good dielectric (NPO/COG ceramic or PPS film)
These points are realizable by the hobbyist - not everyone has a sub-1608 package SMT assembly plant in their garage...
For those you just want a paint by numbers answer:
Murata GRM31C5C1E104JA01L .1UF 25V 5% C0G 1206/3216
Panasonic ECH-U1C104JB5 .1UF 16V PPS FILM 1210/3225
These are available in ten-up quantities from the likes of Digikey for a little over a dollar a piece. Quite reasonable compared so some pricing that has been discussed earlier on.
NOTE/WARNING/ALERT : these are surface mount parts -
Kendall Castor-Perry's work is a reminder of two things:
1) Manufacturer's SPICE models range in usefulness from reasonably useful to downright misleading
2) Simulations that account for only a portion of the real physical circuit under consideration are no more than pointers to better performance.
This particular body of work I recommend because it can be replicated by anyone with the time and inclination because the essential tool is free. (A hearty cheer for Linear Technology - and no, I don't work for them!)
To get simulation results that translate directly (or so damn close as not to matter) to reality then expensive/proprietary simulation tools are needed. This is well outside the scope of DIYers and all but the seriously well-funded commercial operations. I can't imagine many audio designers are going to put down $30k for a field solver.
But we can deduce a few really useful points:
1)Everything else being equal, the decoupling cap in the smallest package is the best - so surface mount 1608 package, which is the smallest package that can be comfortably/repeatably hand soldered/reworked.
2)Minimize trace inductance by using PWR/GND plane pairs, short PWR/GND traces and minimize the number of vias in the PWR/GND traces
3)Use a good dielectric (NPO/COG ceramic or PPS film)
These points are realizable by the hobbyist - not everyone has a sub-1608 package SMT assembly plant in their garage...
For those you just want a paint by numbers answer:
Murata GRM31C5C1E104JA01L .1UF 25V 5% C0G 1206/3216
Panasonic ECH-U1C104JB5 .1UF 16V PPS FILM 1210/3225
These are available in ten-up quantities from the likes of Digikey for a little over a dollar a piece. Quite reasonable compared so some pricing that has been discussed earlier on.
NOTE/WARNING/ALERT : these are surface mount parts -
On parts.
Pride of ownership is a powerful reason for choosing things, HiFi included.
The visual aspect is most important. The eyes have it.
Minute differences in sound quality between parts are swamped by these factors.
Choosing good looking components is good thinking.
It is no accident that expensive parts generally look good.
Pride of ownership is a powerful reason for choosing things, HiFi included.
The visual aspect is most important. The eyes have it.
Minute differences in sound quality between parts are swamped by these factors.
Choosing good looking components is good thinking.
It is no accident that expensive parts generally look good.
VivaVee said:
1. i really can confirm this from my own spicy experiences
from useless - to be almost identical to datasheet curves
now, datasheet curves are not even those the same as your circuit
2. a portion of the real physical circuit
won't do! .... even how good your component spice models are
- ideal voltage/current sources
- ideal rails ... it really is no rials at all, just ideal network nodes
I want to know this, i have wondered to ask about for a long time:
Are there simulation softwares
where you can specify inductance, resistance and capacitance of your PCB tracks?
By for example enter the dimensions, the geometry & length of your tracks between you components.
Of course i can enter small inductances, resistances & capacitances LCR of every track in my simulated circuit.
But what a job
Better to specify a few types of tracks, by copper dimensions,
and put these between components.
Power supply instead of ideal voltage source. This I have solved already.
Not difficult at all
See my attachment!!!!
Lineup Audio Simulations issues forum
Attachments
lineup said:
I want to know this, i have wondered to ask about for a long time:
Are there simulation softwares
where you can specify inductance, resistance and capacitance of your PCB tracks?
By for example enter the dimensions, the geometry & length of your tracks between you components.
First I can't take credit for the quote even if it represents my thoughts as well. Second I believe both Mentor's and Cadences premium product would allow you to build the complete model of the the layout and see what it does, necessary for the EMI/RFI compliance. But those are $50K to $100K per seat plus a year of experience to get value from them. And then all the parts that aren't fully modeled in the way you use them will leave your efforts hanging.
Better to get or make the instrumentation and test setups to see whats happening. I suspect that finding and "fixing" the details, like layout issues, tuning cap resonances etc. will make a big difference and be an enormous amount of work.
If all you are looking for is the linear response of the layout and circuit, you don't have to spend that much for software. At work we use a version of Eagleware Genesys that will do the entire linear simulation. From DC to 10+ GHz, if you like. I've used that software to simulate bypass schemes, including the PC traces, quite successfully - the measured performance of the circuit on the network analyzer was virtually identical to simulation.
There are other RF simulators that do much the same for less than a $1000. You could probably do similar simulations using RFSim99 if you wanted to put in the effort. You might have to use some other free software along with it to get the simulations you want. But, you'd be surprised what you can find on the web if you look. There's the free student versions software from Ansoft, for example.
There are other RF simulators that do much the same for less than a $1000. You could probably do similar simulations using RFSim99 if you wanted to put in the effort. You might have to use some other free software along with it to get the simulations you want. But, you'd be surprised what you can find on the web if you look. There's the free student versions software from Ansoft, for example.
Bas Horneman said:
If one does the right thing in the first place..which is to consider the reports on 'audio quality perceived' to be authentic, then that beggars the question of whether audio measurement (as it stands today) is all it's cracked up to be, when considered in comparison to the human ear's capacities and expectations of what a music signal should be.
This thread and the argument would not be here, in existence, at all..period..if measurement itself...knew what to look for.
Read that last sentence again, and remember it, forever.
To think that the persons in high end audio are psychotic, if one does not hear these given qualities, that is the real psychosis. such a thing is more a reflection of the personality making the accusations.
And more than one person in this thread is guilty of that.
I except Scott from that
, as he is decent enough to speak on such points himself..and to remember it as a reality he understands.The reality is that only psychotics themselves say such things as ' these high end audio guys must be liars as I don't hear it and I don't understand it'. That is a vocalized reflection of a limited intellect and personality. Nothing more.
- Status
- Not open for further replies.