What make up the unit of dB when measuring analog signal?

[waynel]

Your DAC chip need output stage. You can't avoid capacitors at your RCA, XLR analog output. If you want one DAC, you can even trace the circuit. It is not just science, it is engineering.

So you are claiming to understand engineering but don’t know the definition of dB?

1) phase shift is measurable , please show a measurement of a good measuring dac with significant phase shift in the audio band.
2) next ,please show some evidence that this phase shift is audible.

 

[xaviescacs]

Now that the topic has definitely shifted, I'm curious.

Let's say there is some phase shift between voltage and current, arbitrarily significant, how it would affect the performance of an amplifier? And the performance of a speaker?

Disclaimer: It's a sincere and genuine question.

 

[SIY]

Now that the topic has definitely shifted, I'm curious.

Let's say there is some phase shift between voltage and current, arbitrarily significant, how it would affect the performance of an amplifier? And the performance of a speaker?

Disclaimer: It's a sincere and genuine question.

Which voltage and current?

 

[xaviescacs]

Which voltage and current?

hmm... the ones at the output of an hypothetical dac connected to an amplifier for instance

 

[SIY]

hmm... the ones at the output of an hypothetical dac connected to an amplifier for instance

I can’t see how that would happen, at least at audio frequencies. The load is primarily resistive, ditto the source. If the capacitance of the load were high enough to shift current phase wrt voltage, there would be significant frequency response errors.

Don’t try to dig logic out of a badly programmed bot.

 

[NTK]

Now that the topic has definitely shifted, I'm curious.

Let's say there is some phase shift between voltage and current, arbitrarily significant, how it would affect the performance of an amplifier? And the performance of a speaker?

Disclaimer: It's a sincere and genuine question.

Below is from Dr Toole's 1986 paper: Loudspeaker Measurements and Their Relationship to Listener Preferences: Part 2
You can see the phase response of the loudspeakers in the group with fidelity rating 7.5-7.9 were a lot less jagged than the lower rated ones.

 

[solderdude]

hmm... the ones at the output of an hypothetical dac connected to an amplifier for instance

That would mean the input impedance of an amplifier would be complex in nature within the audible band.
Most are resistive in nature or have only mild phase/shift amplitude drop at the extremes (coupling caps, input capacitance).
Speakers and even headphones is another matter.

When the voltage is present and the load is complex it follows that the source has no issues supplying the current, even when not in phase otherwise the voltage would not be there.

 

[xaviescacs]

Don’t try to dig logic out of a badly programmed bot.

Actually the recent requests of information to the bot inspired me to make this question...

Edit: Bots are fine, when you can identify them...

 

[ADU]

If there were any bots here, I think the rest of us bots would probably know and give y'all a heads up... maybe.

 

[mhardy6647]

Science, b**ch!!

Science

xkcd.com

 

[JSmith]

Science, b**ch!!

JSmith

 

[Chrispy]

If there were any bots here, I think the rest of us bots would probably know and give y'all a heads up... maybe.

LOL

 

[GeorgeBynum]

1·10^(-12) W

I enter 1E-12 on my calculator. Fortran's default (output) would be 0.100000000E-11; I may have the number of insignificant zeros incorrect.

 

[Hayabusa]

I enter 1E-12 on my calculator. Fortran's default (output) would be 0.100000000E-11; I may have the number of insignificant zeros incorrect.

1.000000000000E-12 =
0.100000000000E-11 =
0.010000000000E-10 =
0.001000000000E-9 =
0.000100000000E-8 =
0.000010000000E-7 =
0.000001000000E-6 =
0.000000100000E-5 =
0.000000010000E-4 =
0.000000001000E-3 =
0.000000000100E-2 =
0.000000000010E-1 =
0.000000000001E-0 =
0.000000000001

so?

 

[xaviescacs]

My point simply was that one doesn't need to write all those zeros, there are ways to deal with very low or large numbers keeping the length of the numbers to encode them equal.

 

[GeorgeBynum]

so?

My point was just as yours,, but poorly stated. There are an infinite ways to express any number. I elected to add 2 "common" ones, that used in RPN calculators and in the (almost obsolete?) Fortran computer "standard" printer output.

 

[DonH56]

This reminds me of an old, old joke from math and physics courses dealing with statistics, probabilities, uncertainty principle, and the concept of infinity (aleph null or aleph one?):

1 + 1 = 3 for very large values of 1

 

[xaviescacs]

I enter 1E-12 on my calculator. Fortran's default (output) would be 0.100000000E-11; I may have the number of insignificant zeros incorrect.

Curious. I never use the e/E notation by hand, don't like it.

almost obsolete?

Completely obsolete

 

[NTK]

Completely obsolete

From a possibly biased opinion: https://fortran-lang.org/

 

[Blumlein 88]

My point was just as yours,, but poorly stated. There are an infinite ways to express any number. I elected to add 2 "common" ones, that used in RPN calculators and in the (almost obsolete?) Fortran computer "standard" printer output.

As long as those of us who knew and used Fortran live and remember it will never die. So who knows maybe it has 10 more years, maybe 20 years. It was such an elegant, simple, useful way to program a computer for so many things useful in science and engineering.