Richard, if the distortion at low to mid levels is audible, then I'm sorry to say that there must be a problem in the replay chain - not the source material. I have done the exercise sooo many times over the years, where I have thought that a CD was "faulty" or "poor" - but further progressing of refining the overall competence of the playback chain resolved the issues - albums that I could barely think of playing became standard, enjoyable listening fare, and were used as test disks from then on ...
Yeah, that would just be anathema to women, and women make up a large, if not the bulk of my target audience, even if they're not necessarily the ones who will be paying for it, though women have more of their own disposable income than ever before.
Ah, gotcha.
As I said previously, ideally I'd like to sell through Costco. Even rich people have to buy toilet paper. And did you know that Costco is the largest importer of high end French wines? And on their website and in select stores, they sell watches that cost more than the $6,000 I'm setting the ceiling for for the console.
And I think I would also market fairly aggressively to high end interior designers.
Depends what your market is. I'm looking at this as more of a luxury item.
Yeah. The "home entertainment center" isn't quite what I'm aiming for. I want to capitalize more on the "vintage" nature of it. A lot of boomers like myself have some fond memories of those old consoles. And I want to keep with that midcentury modern look, including the old vintage grille cloth.
Something more along the lines of this:
se
Thanks!
Yeah, I think your market would have been limited selling an equipment rack that still needed separate components to function. You're competing on two fronts both of which are pretty well saturated with competitors. For what I'm wanting to do, while the overall market may be smaller, there's just no one out there doing quite the same thing.
se
Well do you have a number for the dynamic range you must have? where must the noise floor be for your ideal system? soory but I cant see any real system being having any noticable differences much beyond 16 bit. Amp errors, speaker errors, room errors and backgound noise all confound the details noticed.
Here's an early sketch of what I have in mind. The drivers on top are coaxials.
se
An externally hosted image should be here but it was not working when we last tested it.
se
Didn't respond to this earlier, thanks for the input, Tom - my comment about this aspect is, yes, in conventional situations that's a very good approach - the Bryston I mention now and again had voltage rails well beyond what typically is required - an easy "fix" for good dynamics. However, I have also done a chip amp with extremely stiff 40V rails, which also provided lesser, but still excellent subjective dynamics, even when running hard up against the rails - an engineering approach will provide an effective solution with an amp that's "only just big enough" ...
I already know that midcentury modern is a very popular design style (and was before Mad Men gave it a nice little boost) and probably will be for some time to come. I'm not terribly intersted in chasing after design styles to try and find the most popular. I want my console to retain the lineage with the vintage consoles that originally inspired it.
se
I understand what You want to do. But the furniture store may have a different view of what sells well with their customers. Can they move enough of them to warrent the space for yours? Just some friendly advice before spending all your hard earned cash.
BTW - it looks almost exactly like one my mother has in her home .... and she is 94 years old. Her's has a LP player, preamp, amp and fm tuner.... plus two 2-way speakers.
THx-RNMarsh
Last edited:
Thanks, but in case you missed it previously, I don't have any intention of selling through furniture stores. Ideally I'd like to sell through Costco. And also market heavily toward high end interior designers.
Yeah? Cool. Does she still use it, and is it tube or solid state do you know?
se
The same applies to Costco. they know what sells and doesnt in their stores. ask them.
She still used it until about 20 years ago...... cant hear any more. She out-lived many features of her body. Was tube electronics and I replaced it all with SS. Just for reliability so i dont have to go there and fix it for her. Garrard Lab 80 TT, Dynaco electonics. Electromagnet speakers changed to a pair of book shelf Dyna A25. All still works fine.
regarding my advice ---
Good judgement comes from experience......
And a lot of that comes from bad judgement.
(W.R.)
-RM
She still used it until about 20 years ago...... cant hear any more. She out-lived many features of her body. Was tube electronics and I replaced it all with SS. Just for reliability so i dont have to go there and fix it for her. Garrard Lab 80 TT, Dynaco electonics. Electromagnet speakers changed to a pair of book shelf Dyna A25. All still works fine.
regarding my advice ---
Good judgement comes from experience......
And a lot of that comes from bad judgement.
(W.R.)
-RM
Last edited:
Looking to see what's selling at a particular point in time and then following their lead is no guarantee of success and ignoring it to do what you want is no guarantee of failure.
I mean, I don't think Apple asked around to see what was the best selling smartphone and said "Let's do something like that."
I'm not comparing myself to Steve Jobs, but I would simply like to do what I have envisioned. Maybe it will end up being a Lisa. Or maybe it will end up being an iPod. But if I don't do what I envision, it wouldn't be something I'd enjoy doing.
Aw, you ditched the field coils? Pity.
Well if this ends up being a bad judgement, I'll have gained some experience, won't I? And that's not entirely a bad thing.
se
How do we lower the thd+N at levels below FS? -- If we just want <0.1% at ANY level down to at least -60dB re 1v (if not the full dynamic range)? This is at the center of why analog continue to sound better on a distortion basis. How many bits are needed for 0.1% thd+N all the way down to -60 and -80dB ?
THx-RNMarsh
Last edited:
No we can't John.
"Discussing" anything without setting the bare goal posts is not a discussion, it's just a lot of noise.
I asked several times and was ignored by the wise men here, who just went on auctioning actually meaningless numbers.
And I thought the questions were basic, but ... The questions were:
1. What is the thershold of hearing distortion as we usually refer to it, as THD and IM? 0.1%? 0.01%? What?
2. What is the level od SPL from which we can talk about "efficient" speakers? 90 dB SPL/2.83V/1m? 95 db SPL/2.83V/1m?
How can we speak about dynamic range when we don't know how much power does that require? And as far as I know, an efficiency difference of 6 dB translates to a hell of a lot of watts, up or down.
How do we lower the thd+N at levels below FS? -- If we just want <0.1% at ANY level down to at least -60dB re 1v (if not the full dynamic range)? This is at the center of why analog continue to sound better on a distortion basis. How many bits are needed for 0.1% thd+N all the way down to -60 and -80dB ?
THx-RNMarsh
Are you joking, is it a plain marketing, or you just have no idea??
Attachments
How do we lower the thd+N at levels below FS? -- If we just want <0.1% at ANY level down to at least -60dB re 1v (if not the full dynamic range)? This is at the center of why analog continue to sound better on a distortion basis. How many bits are needed for 0.1% thd+N all the way down to -60 and -80dB ?
0.1% of -60dB re 1v is 1.0E-6 V (1uV)
0.1% of -80dB re 1v is 1.0E-7 V (0.1uV)
For 16bit and FS of 1V, LSB is 1.526E-5 V
For 20bit and FS of 1V, LSB is 9.537E-7 V
For 24bit and FS of 1V, LSB is 5.960E-8 V
0.1% of -60dB re 2v is 2.0E-6 (2uV)
0.1% of -80dB re 2v is 2.0E-7 (0.2uV)
For 16bit and FS of 2V, LSB is 3.052E-5 V
For 20bit and FS of 2V, LSB is 1.907E-6 V
For 24bit and FS of 2V, LSB is 1.192E-7 V
24bits can cover your extraordinary requirements.
IMO all these is theory.
In practice, even the thermal noise of the analog signal coming into the ADC (thus encoded into the digital stream entered into a DAC) easily reaches the LSB of 20bit/2V FS
George
>Edit. What does this screenshot represents Pavel?
George,
Thank you for this information. I must admit that I don't understand the underlying process here in how the digital bit or word length is changing but do understand your numbers. I am trying to comprehend why changing the output level required truncating the bits to do this, but see even though that is the case that the distortion is still very low even when you do do that. I still don't know the answer as to why it is done this way in the digital domain if that is the result. Would it be an inferior method to doing the attenuation in the analog domain instead of this phenomena in the digital domain or do you just trade one type of noise for another?
ps. I should be asleep right now but just had to go pick up one of the kids in the middle of the night, chewed him out good for that one! Tried to attenuate my anger though, kept the spl level to less than busting his eardrums.
Thank you for this information. I must admit that I don't understand the underlying process here in how the digital bit or word length is changing but do understand your numbers. I am trying to comprehend why changing the output level required truncating the bits to do this, but see even though that is the case that the distortion is still very low even when you do do that. I still don't know the answer as to why it is done this way in the digital domain if that is the result. Would it be an inferior method to doing the attenuation in the analog domain instead of this phenomena in the digital domain or do you just trade one type of noise for another?
ps. I should be asleep right now but just had to go pick up one of the kids in the middle of the night, chewed him out good for that one! Tried to attenuate my anger though, kept the spl level to less than busting his eardrums.
At first, all good parents work nightshifts for to pick-up their teenage offspring when the need arise. If they don’t, they may regret it.
Cool down then and keep on digging.
Next (technical):
I mention two ref voltages because Richard said so. He asked for a ref of 1V and then of the full dynamic range (not proper term) which for a CD player translates to around 2V (This is actually rmsV but I didn’t want to make things too complex )
Envision then a sinusoid with a peak to peak value of 1V(or just a bit lower).
You digitise it with a DAC that works with a ref. Voltage of 1 Volt.
That is to say that the incoming 1Vpp signal is the largest this DAC can handle.
Now the DAC has to slice the 1Vpp amplitude into 2^n (where n=number of bits) slices, for each sample it takes in time. So an 24bit DAC makes 2^24 slices, i.e. 16777216
If you stack all these slices one over the other it makes 1 V, which is the height of the ref . Voltage that this ADC is using.
So each slice has a thickness -an amplitude- of (1/16777216) V, or 59.6 nanoVolts. This is called the resolution of the ADC.
In this –hypothetical-stack of slices, the slice which is on top, is given the name Most Significant Bit The one at the bottom is called the Least Significant Bit, so ADC resolution and amplitude of the LSB have one and the same physical meaning.
Of course, the 1, 2, 3… Least Significant Bits encode the lowest part of the signal.
Now Richard asks for a weak signal (-60dB rel 1V) entering the ADC.
This sinusoid has an amplitude of 1mVpp.
It is short, so as it passed through the 'knifes' of the ADC which have a fixed height, only some will cut it. Thus, it will be sliced into ‘only’ 16777.216 slices.
The 0.1% of the 1mVpp (which is 1nV) that Richard is talking about is almost twice the amplitude of LSBs
If on the other hand Richard is talking about the distortion created by the digitisation of this weak signal, here is what I think:
The precision in height of slicing (which is around ½ LSB and which is close to what is called quantisation error) is -intuitively calculated- 30nanoV and when mathematically derived, it is even less.
So this 30nV is the 1/3 of what the 0.1% of the 1mVpp (which is 1nV) that Richard is talking about.
I hope the numbers are correct
George
Cool down then and keep on digging.
Next (technical):
I mention two ref voltages because Richard said so. He asked for a ref of 1V and then of the full dynamic range (not proper term) which for a CD player translates to around 2V (This is actually rmsV but I didn’t want to make things too complex )
Envision then a sinusoid with a peak to peak value of 1V(or just a bit lower).
You digitise it with a DAC that works with a ref. Voltage of 1 Volt.
That is to say that the incoming 1Vpp signal is the largest this DAC can handle.
Now the DAC has to slice the 1Vpp amplitude into 2^n (where n=number of bits) slices, for each sample it takes in time. So an 24bit DAC makes 2^24 slices, i.e. 16777216
If you stack all these slices one over the other it makes 1 V, which is the height of the ref . Voltage that this ADC is using.
So each slice has a thickness -an amplitude- of (1/16777216) V, or 59.6 nanoVolts. This is called the resolution of the ADC.
In this –hypothetical-stack of slices, the slice which is on top, is given the name Most Significant Bit The one at the bottom is called the Least Significant Bit, so ADC resolution and amplitude of the LSB have one and the same physical meaning.
Of course, the 1, 2, 3… Least Significant Bits encode the lowest part of the signal.
Now Richard asks for a weak signal (-60dB rel 1V) entering the ADC.
This sinusoid has an amplitude of 1mVpp.
It is short, so as it passed through the 'knifes' of the ADC which have a fixed height, only some will cut it. Thus, it will be sliced into ‘only’ 16777.216 slices.
The 0.1% of the 1mVpp (which is 1nV) that Richard is talking about is almost twice the amplitude of LSBs
If on the other hand Richard is talking about the distortion created by the digitisation of this weak signal, here is what I think:
The precision in height of slicing (which is around ½ LSB and which is close to what is called quantisation error) is -intuitively calculated- 30nanoV and when mathematically derived, it is even less.
So this 30nV is the 1/3 of what the 0.1% of the 1mVpp (which is 1nV) that Richard is talking about.
I hope the numbers are correct
George
Last edited:
- Status
- Not open for further replies.