Skip to main content

Topic: Line in overload on soundcards (Read 45469 times) previous topic - next topic

0 Members and 1 Guest are viewing this topic.
  • 2Bdecided
  • [*][*][*][*][*]
  • Developer
Line in overload on soundcards
Reply #25
Sorry to ask again, but can anyone confirm the line-out voltage for digital full scale using the Audiophile 2496?

I'm wondering if the low output on mine is standard (for this card), or a fault.

Cheers,
David.

  • Pio2001
  • [*][*][*][*][*]
  • Global Moderator
Line in overload on soundcards
Reply #26
If you use a multimeter, test at 50 Hz, no more.

Once we used a high-end electronic voltmeter on a signal that was several kHz, and got strange results... We checked all the circuitry until the teacher comes and tells "You won't get any meaningful result from a voltmeter at such frequencies, when measuring alternative voltage, always use an oscilloscope. Alternative voltmeters are designed for 50/60 Hz voltage..."

  • KikeG
  • [*][*][*][*][*]
  • Developer
Line in overload on soundcards
Reply #27
David:

I don't remember the exact figure, but IIRC my Audiophile max. I/O level was similar to yours, it was quite below 2 Vrms. I'm not really sure this 2 Vrms is a true "standard", but IIRC my father's old Denon cd-player has a similar output voltage.
  • Last Edit: 28 November, 2003, 08:33:32 AM by KikeG

  • KikeG
  • [*][*][*][*][*]
  • Developer
Line in overload on soundcards
Reply #28
Quote
In the soundcard challenge, I couldn't tell the difference between the reference file, and the same loopbacked five times in the audiophile 2496 card !
Yes, the mysterious challenger who took the test was me

But, this test was a little bit "unfair" in the first 5-times looped back version, because frequency response differences between all samples had been reduced to minimum via digital processing (not other kind of differences, though). In the next and current version, there was no digital processing applied but dither to the samples, but the number of loopbacks applied was just 3. However, so far nobody has been able to reliability tell the samples from each other in a blind test.

  • 2Bdecided
  • [*][*][*][*][*]
  • Developer
Line in overload on soundcards
Reply #29
Quote
David:

I don't remember the exact figure, but IIRC my Audiophile max. I/O level was similar to yours, it was quite below 2 Vrms. I'm not really sure this 2 Vrms is a true "standard", but IIRC my father's old Denon cd-player has a similar output voltage.


Thanks for confirming this.

It's true that there isn't a "standard", but it seems that, somehow, a standard has emerged.

All the DVD and CD players I could find to test (I found 7), and all the data sheets for current products I could find on the net - every single one of them used or said 2V. The data sheets tend to say just "2V", but testing the output shows this is 2Vrms.

So, whilst the sound card does not break any official standard (because there is none!), it is very unhelpful. As Pio2001 said, it can't record the output of most modern equipment (without an external pre-amp/attenuator), and it can't match the output level of any modern CD or DVD player. This won't matter most of the time, but it's worth knowing.

Cheers,
David.
  • Last Edit: 28 November, 2003, 09:04:02 AM by 2Bdecided

Line in overload on soundcards
Reply #30
1.)The forum suggests the use of a passive attenuator(volume control) with the M-Audio 2496, which may not be such a good idea. The version I had used the Inverting Input of the IC...noise would be a function of the source impedance, hence the floor noise would vary with the position of the potentiometer. (This would also explain why input overload measurements on this card vary with the test setup.)

The better approach would be to place the balance control ahead of a low-noise preamp/buffer eg AD797,SSM2134

2.)Ripping CD music via the Analog outputs is adding more links in the chain then needed, why not use the Digital out or Optical out from the CD player, the 2496 supports this....No more clipping!

3.) In all cases one has to Zero the DC offset on every file.

  • KikeG
  • [*][*][*][*][*]
  • Developer
Line in overload on soundcards
Reply #31
Quote
1.)The forum suggests the use of a passive attenuator(volume control) with the M-Audio 2496, which may not be such a good idea. The version I had used the Inverting Input of the IC...noise would be a function of the source impedance, hence the floor noise would vary with the position of the potentiometer. (This would also explain why input overload measurements on this card vary with the test setup.)
The better approach would be to place the balance control ahead of a low-noise preamp/buffer eg AD797,SSM2134

I doubt using an adequate simple passive voltage divider in order to avoid clipping would degrade notabily input SNR, but I haven't tried. Have you verified this personally via measurements? The use of an active device as input stage makes things much more complex, and I doubt would give any practical benefit.

Quote
2.)Ripping CD music via the Analog outputs is adding more links in the chain then needed, why not use the Digital out or Optical out from the CD player, the 2496 supports this....No more clipping!


We are not talking about ripping here. I believe we are talking about recording and analyzing of the output of a device.

Quote
3.) In all cases one has to Zero the DC offset on every file.


Most cards don't have DC offset problems at the input, since they are AC-coupled (this means that they block the DC at their inputs).

Line in overload on soundcards
Reply #32
Quote
I doubt using an adequate simple passive voltage divider in order to avoid clipping would degrade notabily input SNR, but I haven't tried. Have you verified this personally via measurements? The use of an active device as input stage makes things much more complex, and I doubt would give any practical benefit.


I went through the testing when I first built my Vinyl Ripper ... over 2yrs ago!
I used a Neutrik 3000 series audio test rack for signal and phase measurements; Tektronix 500 rack for Impulse, Tone burst and very low frequency(1-20hz) signals:
CRO Kikusui COR5501

I have 3 vinyl sources, so a switch box/preamp was needed anyway.From memory, the main issue with the M-Audio was anomalies with the subsonics. I am not knocking the product, just putting forward my observation.

Quote
We are not talking about ripping here. I believe we are talking about recording and analyzing of the output of a device


I stand corrected.

Quote
Most cards don't have DC offset problems at the input, since they are AC-coupled (this means that they block the DC at their inputs).


The cards don't have DC offsets?

  • JeanLuc
  • [*][*][*][*][*]
Line in overload on soundcards
Reply #33
DC offset on modern HQ sound cards should be negligible, but still measureable ... the DC offset is introduced after the AC coupling stage mostly due to the pc environment (em radiation cannot always be completely suppressed and might induce DC that leaks over to the card's analog input - not the ideal environment for a high end sound controller)

That's why I use an external ADC to feed digital signals only
The name was Plex The Ripper, not Jack The Ripper

Line in overload on soundcards
Reply #34
The Prodigy 7.1's CODEC features analogue gain control for its inputs, and analogue volume control for the outputs.