Re: New presonus mixer
But then again at 24 or even 16 bit the noise is over a hundred decibels below full scale.
But then again at 24 or even 16 bit the noise is over a hundred decibels below full scale.
New presonus mixer
- Thread starter Douglas R. Allen
- Start date
Re: New presonus mixer
Not to quibble over a few dB but in theory dynamic range is approx 6 db per bit so 16 bit is 96dB and change.
Understanding noise floors in modern digital media can be confusing and yes oversampling with decimation does affect perceived or audible noise.
Luckily we will always have plenty of old fashioned analog noise from mics and preamps to swamp out digital nasties down in the digital dirt.
JR
Not to quibble over a few dB but in theory dynamic range is approx 6 db per bit so 16 bit is 96dB and change.
Understanding noise floors in modern digital media can be confusing and yes oversampling with decimation does affect perceived or audible noise.
Luckily we will always have plenty of old fashioned analog noise from mics and preamps to swamp out digital nasties down in the digital dirt.
JR
Re: New presonus mixer
Another often overlooked benefit of 96kHz is that EQ curves are closer to their analog originals. Most digital mixers draw pretty pictures of the peaking and shelving, but in the 10K to 20K band, the reality is different. Up to 22kHz (for a 44.1kHz Fs), the response curve gets distorted. Exactly how is a more complicated discussion.
By using 96kHz instead, the inaccuracy of the curve is lessened.
This can certainly make a difference in how "English" your EQ sounds at the high end.
Another often overlooked benefit of 96kHz is that EQ curves are closer to their analog originals. Most digital mixers draw pretty pictures of the peaking and shelving, but in the 10K to 20K band, the reality is different. Up to 22kHz (for a 44.1kHz Fs), the response curve gets distorted. Exactly how is a more complicated discussion.
By using 96kHz instead, the inaccuracy of the curve is lessened.
This can certainly make a difference in how "English" your EQ sounds at the high end.
Re: New presonus mixer
Most digital fx (including eq) of today uses oversampling to overcome issues at the extreme points. 4x oversampling is quite common nowdays.
Re: New presonus mixer
Bingo, we have a winner. And you don't need the brick wall filter.
Re: New presonus mixer
Of course, this all assumes proper gain staging, but if you can't do that you have no business spending thousands of dollars on a digital mixer like these in the first place.
Of course, this all assumes proper gain staging, but if you can't do that you have no business spending thousands of dollars on a digital mixer like these in the first place.
Re: New presonus mixer
He does have a gift!
...but i still can't explain how it works like Phil.
JR
He does have a gift!
Re: New presonus mixer
Wonderful question indeed.
I would like to throw in a loose quote from Ivo Mateljan, Arta manual:
"Phase can be estimated up to a quarter of sample frequency"
Uwe
Wonderful question indeed.
I would like to throw in a loose quote from Ivo Mateljan, Arta manual:
"Phase can be estimated up to a quarter of sample frequency"
Uwe
Re: New presonus mixer
Uwe,
Discussions in the context of measurement are a different animal than what Max was asking.
When looking at the phase behavior of a device under test (DUT), the total phase is the combination of the DUT's inherent phase response, and the phase that accrues due to time of flight from the DUT to the measuring transducer.
In the naive case, the arrival time alignment delay of the two signals that are used to calculate the transfer function (i.e. the input stimulus and the measured response) can only be controlled by buffering on the interval of one sample. This means that the phase calculated by the TF could be off due to the limited granularity of the delay time you can pick. Note that this isn't a limitation of the math, but rather of one's ability to pick the alignment delay time for the signals fed to the transfer function. If you could pick the delay time precisely, the phase would be known precisely.
There are more advanced techniques available that allow for inter-sample delay time choices, FWIW.
Uwe,
Discussions in the context of measurement are a different animal than what Max was asking.
When looking at the phase behavior of a device under test (DUT), the total phase is the combination of the DUT's inherent phase response, and the phase that accrues due to time of flight from the DUT to the measuring transducer.
In the naive case, the arrival time alignment delay of the two signals that are used to calculate the transfer function (i.e. the input stimulus and the measured response) can only be controlled by buffering on the interval of one sample. This means that the phase calculated by the TF could be off due to the limited granularity of the delay time you can pick. Note that this isn't a limitation of the math, but rather of one's ability to pick the alignment delay time for the signals fed to the transfer function. If you could pick the delay time precisely, the phase would be known precisely.
There are more advanced techniques available that allow for inter-sample delay time choices, FWIW.
Re: Basic principles of sampling
At the risk of adding noise (dither?) to Phil's very nice explanation, I'll say a little on the subject of reconstruction, which is what happens in the digital-to-analog converter. This is more of a time domain view and does not rely directly on Fourier theory.
Ideal sampling consists of multiplying a continuous waveform by a sequence of (in our case uniformly spaced) impulses. (An impulse, very roughly speaking, is a pulse of infinite amplitude, zero duration, and finite area.) This results in a sequence of impulses that are scaled in energy (area) by the amplitude of the continuous waveform at the time of each impulse.
It turns out that there is an ideal low-pass filter, which we call the reconstruction filter, that does a perfect job of interpolating the original continuous waveform between the sample impulses, provided that there were no frequency components above the Nyquist frequency in the original waveform. This filter has a "brick wall" frequency response and hence a sin(x)/x impulse response, which is of (doubly) infinite duration and therefor physically unrealizable. In practice we approximate this filter, often by using a zero-order hold, or other interpolator, in combination with a realizable low-pass filter. (A zero-order hold replaces each impulse with a pulse whose duration is the sample interval and whose amplitude is the area of the preceding impulse.)
The key here is that the reconstruction filter fills in the spaces between the samples and if Nyquist is satisfied then sufficient information exists in the samples for the filter to do this.
--Frank
At the risk of adding noise (dither?) to Phil's very nice explanation, I'll say a little on the subject of reconstruction, which is what happens in the digital-to-analog converter. This is more of a time domain view and does not rely directly on Fourier theory.
Ideal sampling consists of multiplying a continuous waveform by a sequence of (in our case uniformly spaced) impulses. (An impulse, very roughly speaking, is a pulse of infinite amplitude, zero duration, and finite area.) This results in a sequence of impulses that are scaled in energy (area) by the amplitude of the continuous waveform at the time of each impulse.
It turns out that there is an ideal low-pass filter, which we call the reconstruction filter, that does a perfect job of interpolating the original continuous waveform between the sample impulses, provided that there were no frequency components above the Nyquist frequency in the original waveform. This filter has a "brick wall" frequency response and hence a sin(x)/x impulse response, which is of (doubly) infinite duration and therefor physically unrealizable. In practice we approximate this filter, often by using a zero-order hold, or other interpolator, in combination with a realizable low-pass filter. (A zero-order hold replaces each impulse with a pulse whose duration is the sample interval and whose amplitude is the area of the preceding impulse.)
The key here is that the reconstruction filter fills in the spaces between the samples and if Nyquist is satisfied then sufficient information exists in the samples for the filter to do this.
--Frank
Last edited:
Re: Basic principles of sampling
This is probably way TMI for the OP but comes to mind when talking about starting and stopping samples. I got painfully introduced to this mechanism in one of my early electronics gigs back in the '70s. I was working for a company that performed pitch shift for blind people listening to talking book tape recordings speeded up to 2x normal or more. Very simply the speeded up signal was read into a memory at the fast high pitch rate, then read back out at the slower lower normal pitch rate. Since this was occurring at 2x or more of real time for the recording, redundant data would have to be discarded between samples and then the displayed samples spliced together. Since these samples were not contiguous in time, the audible artifacts created by stopping and stopping these samples (effectively multiplying them) at random times was very objectionably and annoying.
This same mechanism is at play in shunt/switch type noise gates and to a lesser extent when a compressor multiplies a dry signal times a changing gain effectively multiplying the waveform by the gain.
This will hurt you head to think about it too much, but when designing gear it helps to understand where the beeps and farts are coming from... when trying to eliminate them.
JR
This is probably way TMI for the OP but comes to mind when talking about starting and stopping samples. I got painfully introduced to this mechanism in one of my early electronics gigs back in the '70s. I was working for a company that performed pitch shift for blind people listening to talking book tape recordings speeded up to 2x normal or more. Very simply the speeded up signal was read into a memory at the fast high pitch rate, then read back out at the slower lower normal pitch rate. Since this was occurring at 2x or more of real time for the recording, redundant data would have to be discarded between samples and then the displayed samples spliced together. Since these samples were not contiguous in time, the audible artifacts created by stopping and stopping these samples (effectively multiplying them) at random times was very objectionably and annoying.
This same mechanism is at play in shunt/switch type noise gates and to a lesser extent when a compressor multiplies a dry signal times a changing gain effectively multiplying the waveform by the gain.
This will hurt you head to think about it too much, but when designing gear it helps to understand where the beeps and farts are coming from... when trying to eliminate them.
JR
Re: New presonus mixer
Max,
It should be mentioned that quantitized sampling, i.e. discrete amplitude sampling at different bit depths, is not a fundamental requirement of the discrete time sampling process I outlined earlier in the thread. Digital audio is both discrete time and discrete amplitude, but one is not a necessary condition of the other.
Max,
It should be mentioned that quantitized sampling, i.e. discrete amplitude sampling at different bit depths, is not a fundamental requirement of the discrete time sampling process I outlined earlier in the thread. Digital audio is both discrete time and discrete amplitude, but one is not a necessary condition of the other.
Re: New presonus mixer
Sooooo, the conclusion is that a sampling rate of 96KHz is twice as high as a sampling rate of 48KHz.
Any other conclusions?
(Maybe time to split this topic
~;-)~:wink: )
Sooooo, the conclusion is that a sampling rate of 96KHz is twice as high as a sampling rate of 48KHz.
Any other conclusions?
(Maybe time to split this topic
~;-)~:wink: )
Re: New presonus mixer
they said no math Per, so put the calculator away.
lets keep it simple, there is a 4 and an 8 on one and a 9 and a 6 on the other.
48X96 is a sheet of plywood in the US. In inches that is.
Sooooo, the conclusion is that a sampling rate of 96KHz is twice as high as a sampling rate of 48KHz.
Any other conclusions?
(Maybe time to split this topic
they said no math Per, so put the calculator away.
lets keep it simple, there is a 4 and an 8 on one and a 9 and a 6 on the other.
48X96 is a sheet of plywood in the US. In inches that is.
Re: New presonus mixer
Well if we all could agree that bigger numbers are always better, then at least marketing would be happy.
Sent from my DROID RAZR HD
Well if we all could agree that bigger numbers are always better, then at least marketing would be happy.
Sent from my DROID RAZR HD
Re: New presonus mixer
For those wondering what an impulse function is.... Assume that the circuit you are trying to analyze is a big bell. The impulse function is a hammer that you it it with. The impulse response is the way the system rings after you hit it.
By measuring the characteristics of the sound of the bell, you can learn lots about it without having to actually physically measure anything other than how it reacted to being hit.
On the subject of 48 vs 96, I agree. 96 is bigger than 48 I still don't think you can hear the difference in a live console.
For those wondering what an impulse function is.... Assume that the circuit you are trying to analyze is a big bell. The impulse function is a hammer that you it it with. The impulse response is the way the system rings after you hit it.
By measuring the characteristics of the sound of the bell, you can learn lots about it without having to actually physically measure anything other than how it reacted to being hit.
On the subject of 48 vs 96, I agree. 96 is bigger than 48
I still don't think you can hear the difference in a live console.
Re: New presonus mixer
I'll tell my wife that when she brings up my BMI. Comming to think about it, the fast food places would love it!
Re: New presonus mixer
My wife says she misses my 6 pack abs, I tell her I found out at college that a keg is bigger that a 6 pack and that was better.
Bigger it is.
My wife says she misses my 6 pack abs, I tell her I found out at college that a keg is bigger that a 6 pack and that was better.
Re: New presonus mixer
Gosh darnit, and I've been trying to stay fit all my teenage years. Why'd you have to tell me now???
Gosh darnit, and I've been trying to stay fit all my teenage years. Why'd you have to tell me now???
Re: New presonus mixer
So what does a bunch of fat guys hitting their consoles with hammers deduce about the difference in impulse response between 48K and 96K operation?
So what does a bunch of fat guys hitting their consoles with hammers deduce about the difference in impulse response between 48K and 96K operation?