It does not require an expert to answer the question. If your website description of no filters is accurate, then (as I said in post 219) any signal with components above half the sample frequency will generate an alias. If this does not happen then you do have a filter.Charles said:
The alias will be at a frequency of fsig - fs, and at about the same amplitude as the signal would have been were it not for aliasing. This means that aliasing is worse than the images produced by a filterless NOS DAC, as images are at high frequencies where audibility will be poor, while aliases are at lower frequencies.
At 96kHz sampling aliasing might not be too much of a problem as there won't be much music above 48kHz, but at 44.1 or 48kHz it will be unless the studio has poor bandwidth.
Are you?Charles said:
Last edited:
What does the fact that you didn't answer it tell us ?
But one point for action
Any others ?
Charles
What does the fact that you didn't answer it tell us ?
But one point for action
Any others ?
Charles
you need an anti-aliasing filter otherwise your board is unusable. Any HF noise or signal content above fs/2 will be folded back into the normal audio spectrum.
Can I at least expect from you to read my question correctly ?
Here it is again:
What do I have to connect to my SAR ADC's input in order to get really nasty alias-distortion, even if I sample at 192kHz ?
Now don't hide in textbooks, come with your experience !
Okay, you probably cannot have the experience, then come with your engineering !!
Charles
Here it is again:
What do I have to connect to my SAR ADC's input in order to get really nasty alias-distortion, even if I sample at 192kHz ?
Now don't hide in textbooks, come with your experience !
Okay, you probably cannot have the experience, then come with your engineering !!
Charles
Last edited:
Can I at least expect from you to read my question correctly ?
Here it is again:
What do I have to connect to my SAR ADC's input in order to get really nasty alias-distortion, even if I sample at 192kHz ?
Now don't hide in textbooks, come with your experience !
Okay, you probably cannot have the experience, then come with your engineering !!
Charles
it's a silly question because it depends on what you mean by nasty.
good engineering practice dictates that you can't assume the nature of the input signal to the adc and that any aliasing artefacts are not desirable.
Last edited:
right, so you made an ADC that is only able to sample at 192khz if you want to avoid aliasing being folded down into useful bandwidth?
well thats really useful.... but then you are stuck in a spot, since I presume you dont like resampling either?
just like the little player thats only ~16bit with the volume turned all the way up
the products are toys that you should have just kept as personal experiments
well thats really useful.... but then you are stuck in a spot, since I presume you dont like resampling either?
just like the little player thats only ~16bit with the volume turned all the way up
the products are toys that you should have just kept as personal experiments
Last edited:
That you did not ask the question you thought you were asking? I gave the correct answer to the question you wrote, but perhaps not the correct answer to the question you thought?Charles said:
Or is this just a game of riddles to divert attention from your claim of an ADC which lacks, and does not need, anti-aliasing filters even when run at 44.1kHz? The statements on your website advertising this device appear to show a lack of understanding of how sampling works. This would be a concern to any potential customer who does understand sampling.
Please do not try to erect an artificial barrier between textbooks, engineering and experience. Everything we experience is done so in the context of our understanding, which may be true or false or a mixture. If your experience suggests that the Shannon sampling theorem is incorrect then either you deserve a Nobel Prize or the mental model by which you interpret your experience is incorrect and misleading you.
Im not an expert in the field of anti aliasing filters, but I assume that to make an anti aliasing filter on the input of an ADC you would need either multiple op-amp filters to create a very sharp HF roll off at half or just below the sampling frequency, or a complex passive network to do the same job.
Many audiophiles seem to subscribe to the theory that less is more. The audio signal traveling through less components will result in a more pure sound. I personally find this to be the case - use what you need, not more than you need and also not less than you need.
Now, If this ADC has no anti aliasing filter, AND the user samples at say 88.2Khz, and say is archiving an LP record, then is the end user going to get any aliasing?
I suggest that they will not, because most LP's don't have audio above 40khz.
But, because the audio passed though less components before the AD conversion, there might be subjectively better sound from using this ADC. I think this is possible.
I also think that if making a live recording at 96khz or greater, that there would be little chance of aliasing occurring.
To me this sounds like a unique product, which when used in an educated way, by an educated person could result in better sound than a standard ADC with a brick wall filter.
Am I wrong about any of what I wrote above?
Many audiophiles seem to subscribe to the theory that less is more. The audio signal traveling through less components will result in a more pure sound. I personally find this to be the case - use what you need, not more than you need and also not less than you need.
Now, If this ADC has no anti aliasing filter, AND the user samples at say 88.2Khz, and say is archiving an LP record, then is the end user going to get any aliasing?
I suggest that they will not, because most LP's don't have audio above 40khz.
But, because the audio passed though less components before the AD conversion, there might be subjectively better sound from using this ADC. I think this is possible.
I also think that if making a live recording at 96khz or greater, that there would be little chance of aliasing occurring.
To me this sounds like a unique product, which when used in an educated way, by an educated person could result in better sound than a standard ADC with a brick wall filter.
Am I wrong about any of what I wrote above?
Last edited:
but you could be making the sound worse from aliasing artifacts and is an example of where less is worse !!
Higher sampling rates means the aliasing filters can be simpler but that doesn't mean they are not needed. Without aliasing filters any HF noise entering the signal chain can be folded back down into the audio band and corrupt the recording. It's just not done !!
Firstly I don't own the product nor make the product so don't get upset with me.
When archiving an LP, what sort of HF noise gets into the signal chain?
If I put my turntable on the scope and check for HF noise and cant see any, would it be safe to use this ADC?
Same thing in a studio, if I put my scope, or spectrum analyser on the mic-pre, and couldn't see any HF noise, would it be safe to use this ADC?
When archiving an LP, what sort of HF noise gets into the signal chain?
If I put my turntable on the scope and check for HF noise and cant see any, would it be safe to use this ADC?
Same thing in a studio, if I put my scope, or spectrum analyser on the mic-pre, and couldn't see any HF noise, would it be safe to use this ADC?
Thinking about this some more, if you can only hear a tiny hiss from your speaker when the phono amp is on, then this means the hiss is at very low amplitude, any noise above the range of hearing is also likely to be at very low amplitude, even if this low amplitude HF noise was to create an alias, it also would only be at very low amplitude, and come out as low level "noise", I would not expect this to sound like a "birdie"
I don't see the problem!!
I don't see the problem!!
Last edited:
The tracing noise along the groove, ticks, pops, any minor mistracking... if you use a spectrum analyzer to look at the HF content of a phono output (a scope is not the appropriate instrument), it's pretty significant.
I did actually say spectrum analyser when referring to the mic pre. I just omitted to say so with reference to the phono amp. My bad.
And for archiving an LP I would presume that a person would be locating a pristine copy, or de-clicking the track after the recording anyway.
And even if a person was to leave the "pop" on the recording the alias would only be of the "pop". In real life, I can't see the issue.
Last edited:
pops and clicks have wide bandwidth well above the audible band !
you need to de-click the track before you sample it with an A-D and that means you need a filter to do this !!
Hi Trevor,
thank you for your suggestion, but I think I have shown with the simple scopeshots on the Creation ADC website The Altmann Creation ADC, that a click, containing high bandwidth does indeed cause distotion when recorded with a sigma-delta ADC, but does not cause distortion when recorded with my ADC.
If you take a look at the scopeshots on my website, you will see.
A high frequency click or pop is not a problem for a filterless true ADC, so my initial question remains:
What is the only thing that causes really nasty and audible alias distortion when recording with a filterless SAR ADC even at 192kHz sample rate ?
Charles
Hi Trevor,
thank you for your suggestion, but I think I have shown with the simple scopeshots on the Creation ADC website The Altmann Creation ADC, that a click, containing high bandwidth does indeed cause distotion when recorded with a sigma-delta ADC, but does not cause distortion when recorded with my ADC.
If you take a look at the scopeshots on my website, you will see.
A high frequency click or pop is not a problem for a filterless true ADC, so my initial question remains:
What is the only thing that causes really nasty and audible alias distortion when recording with a filterless SAR ADC even at 192kHz sample rate ?
Charles
You need to understand Fourier transform theory otherwise your arguments are totally unsound and are derived from false premises.
Also there are two pictures of oscilloscope displays and I can clearly see differences between the two so I don't know what you are try to demonstrate. Perhaps you should test your AD converter with a 20 KHz squarewave and see what happens
You need to understand Fourier transform theory otherwise your arguments are totally unsound and are derived from false premises.
Also there are two pictures of oscilloscope displays and I can clearly see differences between the two so I don't know what you are try to demonstrate. Perhaps you should test your AD converter with a 20 KHz squarewave and see what happens
Hi Trevor,
I have shown that a sigma-delta ADC causes distortion when fed with a HF-click and that a zero-filter SAR ADC (i.e. my Creation ADC) does not create distortion, btw. the test wav I used is available for download on my website. So this is something everybody can try out himself and I need not add anything to pure facts, as they stand for themselves
In the examples on my websites I have chosen this mixed signal which contains a square step added to a sine wave, so that you can clearly see, that the step as well as the sine contains no distortion when recorded with my Creation ADC, but does get corrupted when recorded with any sigma-delta ADC
But your idea of sampling a 20kHz square wave is good. I have the DAC that can play back a true square-wave and I have the ADC that can sample a true square wave, and the output will again be a pure square wave.
Try this with a sigma-delta ADC/DAC and you will get a huge amount of distortion. To say that this distortion can be explained by the theory of Fourier transformation will not remove the distortion.
The definition of distortion is that the output is different than the input.
Charles
Charles,
No, you haven't. You have shown that your creation doesn't seem to cause such gross distortion that it is visible with the bare eye on a waveform display on a scope. It is a totally meaningless measurement, unless your distortion is really gross (as in several %), and as long as the difference you are trying to show is in the amplitude (and not frequency/temporal) domain.
Could you show some spectrum plots instead - at last that would have *some* relevance?
No, you haven't. You have shown that your creation doesn't seem to cause such gross distortion that it is visible with the bare eye on a waveform display on a scope. It is a totally meaningless measurement, unless your distortion is really gross (as in several %), and as long as the difference you are trying to show is in the amplitude (and not frequency/temporal) domain.
Could you show some spectrum plots instead - at last that would have *some* relevance?
Hi Julf,
then I have at least shown that really gross distortion does appear with a sigma-delta ADC, and you can do the same test if you have an oscilloscope.
Charles
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.