Hi John,
I apologize, but I trully thougt that the picture will amuse you. Although I tend to agree with Bruno technically speaking (PWM is analog type of modulation just like FM) I also beleive what you said: I made it and I can call it whatever I want. After all, what it would look like if there would be prescribed names for our sons and daughters.
Best regards,
Jaka Racman
I apologize, but I trully thougt that the picture will amuse you. Although I tend to agree with Bruno technically speaking (PWM is analog type of modulation just like FM) I also beleive what you said: I made it and I can call it whatever I want. After all, what it would look like if there would be prescribed names for our sons and daughters.
Best regards,
Jaka Racman
Hah, i saw on the russian forum the post: "I want to buy advanced digital amp UcD tripath"(afraid that it's a result of my efforts for class D reabilitation there ;-). Customers, just a customers, and they tend to know only product name usually.
PS: kids knows, that dogs can do: houseguarding, foodeating, yellowsnowmaking. Other function just not in use.
PS: kids knows, that dogs can do: houseguarding, foodeating, yellowsnowmaking. Other function just not in use.
Putting out the fire with gasoline...
For Bruno: In electronics there is no such thing as a digital implementation of ANYTHING. Not ever, anywhere. All electrical signals are inherently analog quantities. The assignment of digital 'states' is merely a convenience construct for our analysis, so all electronics is analog and digital is in the mind
For JohnW: But electric current consists of a flow of electrons. And electrons are discrete particles. This would make any electrical signal a digital event, but with resolution so fine that we are normally unable to discern it's discrete nature. So we assign an analog continuum to a range of digital events. Thus all electronics is fully digital and analog doesn't exist.
Go make more amps . . .
Regards
JohnH
For Bruno: In electronics there is no such thing as a digital implementation of ANYTHING. Not ever, anywhere. All electrical signals are inherently analog quantities. The assignment of digital 'states' is merely a convenience construct for our analysis, so all electronics is analog and digital is in the mind
For JohnW: But electric current consists of a flow of electrons. And electrons are discrete particles. This would make any electrical signal a digital event, but with resolution so fine that we are normally unable to discern it's discrete nature. So we assign an analog continuum to a range of digital events. Thus all electronics is fully digital and analog doesn't exist.
Go make more amps . . .
Regards
JohnH
I'll drink to that.
Actually I think there is a confusion between the concept of discrete and the concept of symbolic. They are not synonymous. In that regard Bruno and like-minded others have the gist imo.
Perhaps one day it will be commonplace for the presence or absence of a single electronic charge to symbolize a binary one or zero, and our ability to inspect the states of devices like single-electron-transistors will be routine. The ability to recover that symbolic value without ambiguity will permit the circuit to be used as a digital one, but it does not differ in principle from using a latched electromechanical relay for the same symbolic representation.
At least we are further along than the Executive Chairman of H*rm*n, who has been using the term "digital" in a way completely his own, to the utter bafflement of everyone.
Actually I think there is a confusion between the concept of discrete and the concept of symbolic. They are not synonymous. In that regard Bruno and like-minded others have the gist imo.
Perhaps one day it will be commonplace for the presence or absence of a single electronic charge to symbolize a binary one or zero, and our ability to inspect the states of devices like single-electron-transistors will be routine. The ability to recover that symbolic value without ambiguity will permit the circuit to be used as a digital one, but it does not differ in principle from using a latched electromechanical relay for the same symbolic representation.
At least we are further along than the Executive Chairman of H*rm*n, who has been using the term "digital" in a way completely his own, to the utter bafflement of everyone.
Re: Putting out the fire with gasoline...
Bar serious EMI issues, this "something" in this particular case is jitter. Indeed, the CD transport delivers two things. One is digital (bits), another is analogue (timing). Remember that time and hence data rate is an analogue property. Data has no time, just like letters on paper have no time.
A good DAC therefore is one which is capable of fully separating the information from its analogue representation (the SPDIF signal).
As long as the data source dictates the timing of the data, this is not truly possible, because at some time you will have to synchronise the conversion rate with the data rate. If the CD transport is locked to the DAC, the DAC is requesting data at its own pace (next, next, next) and the data itself is the only thing transmitted from the CD transport, not the time.
Of course, the clock circuit (analogue signal!!!) of the DAC has to be good. It wouldn't be the first time that someone makes a CD transport that has a stabler clock than the attached DAC, in which case the DAC would do better heeding the (analogue) time information embedded with the (digital) data.
This does not mean that the concept of digital is not sharply defined (numbers=digital, volts amps and time = analogue), rather that real-life implementations of ADC's and DAC's do not always split the two well enough.
This is correct. In mixed signal applications, analogue effects in the digital circuitry sometimes affects the outcome. The most screaming examples are DACs that sound different when they get their "data" from different CD transports or worse, through different cables. While 'tis clear that the data (that which I call digital) is the same at all time, there is "something" about the signal that also affects the outcome. This "something" is then an analogue trait.John Hope said:
Bar serious EMI issues, this "something" in this particular case is jitter. Indeed, the CD transport delivers two things. One is digital (bits), another is analogue (timing). Remember that time and hence data rate is an analogue property. Data has no time, just like letters on paper have no time.
A good DAC therefore is one which is capable of fully separating the information from its analogue representation (the SPDIF signal).
As long as the data source dictates the timing of the data, this is not truly possible, because at some time you will have to synchronise the conversion rate with the data rate. If the CD transport is locked to the DAC, the DAC is requesting data at its own pace (next, next, next) and the data itself is the only thing transmitted from the CD transport, not the time.
Of course, the clock circuit (analogue signal!!!) of the DAC has to be good. It wouldn't be the first time that someone makes a CD transport that has a stabler clock than the attached DAC, in which case the DAC would do better heeding the (analogue) time information embedded with the (digital) data.
This does not mean that the concept of digital is not sharply defined (numbers=digital, volts amps and time = analogue), rather that real-life implementations of ADC's and DAC's do not always split the two well enough.
bcarso said:
In a digital transmission with a sufficiently wide open eye pattern, the symbolic value too can be recovered fully. As long as you're not converting to or from analogue, there are no analogue effects for the implementation effects to worry about as long as they do not induce data errors.
This is why we record and edit digital audio to computer, and process the digital signal in an environment where we would not like our analogue signal to come even close to.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.