Welcome Guest ( Log In | Register )

Recording line in with Soundforge, and Audiophile 24/96
post Aug 10 2003, 10:34
Post #1

Group: Members
Posts: 907
Joined: 9-February 02
From: Cheshire, UK
Member No.: 1296

I'm having a problem recording the output from my CD player through the line in of my soundcard, the Midiman Audiophile (Window's XP).

The problem is distortion. I can't seem to find a way to attenuate the line level input in soundforge and this is causing clipping.

I'm beginning to wonder if this can be removed since during normal playback (through PC) I also get audible clipping, even when the line in is attenuated using Midiman's control panel. I suspect the Midiman doesn't like the output levels of my CD player?

I'm quite sure the input signal itself isn't a problem since playback direct through my amplifier reveals no clipping.

Any ideas?

Go to the top of the page
+Quote Post
Start new topic
post Apr 25 2006, 05:39
Post #2

Group: Members
Posts: 2192
Joined: 31-August 05
Member No.: 24222

Most analogue volume controls are passive. Potentiometers are the quintessential, and most common, examples. There are also a variety of resistor ladder designs but those are essentially identical in function to the potentiometer: a passive voltage divider.

There are some active analogue devices such as the voltage control amplifier, a special kind of opamp. Unlike almost every other kind of volume control, a VCA can also amplify the signal, thus this device can be an entire line level preamp. It uses a potentiometer to control the output but the signal does not pass through the pot as it does in a ‘normal' volume control.

A digital volume control operates on the data. It changes the numerical value of samples, thus it loses resolution that cannot be gained back by later amplification. In theory this is not true for analogue volume controls although the distinction is less than absolute.

I'm not sure what your question about the potential difference in clipping between analogue and digital volume controls is about, exactly. My earlier comment about soundcard level controls being digital, such as those in the Audiophile's mixer panel, means that the signal has already clipped (if it is going to) before it gets to the control. Clipping occurs in the ADC. The level controls are in the digital mixing chip, downstream from the ADC. They have no potential to effect clipping on the input; they are only useful when mixing multiple signals in the digital mode.

I don't think ‘headroom' rightfully applies to an ADC. Except for those weird gaming cards, there is always some specific input voltage that produces 0dBfs. Any higher voltage level will always clip; the result will be the same if the input signal is higher by 1mV or by 1V, there is never any non-clipping room above that voltage.

If you look at the operating characteristics curves for transistors or vacuum tubes, most have basically the same shape, only with different numbers defining the particular device. There is generally a significant stretch where the relationship between input and output forms a straight line. At lower or higher levels the relationship becomes radically un-linear.

Reasonably enough, most audio circuits are built around the linear portion of the ‘curve.' If the input signal exceeds circuit specifications (i.e. the linear operating range), the output tends to become rather nonlinear, thus distorted. Somewhere still higher it clips.

It is possible to add some automatic limiting circuitry to prevent clipping, but this is not normal on HiFi because it introduces non-linearities and distortion of its own; actual limiting is left to the fool with his hand on the volume control. Without limiting you get higher quality at the expense of potential danger (clipping, speaker damage, amplifier destruction).

An ADC does have some analogue front-end circuitry. This is more in the nature of impedance matching and buffering; it may have some small + or - amplification but that is not variable. Any considerations towards audio quality will strongly suggests that we want the ADC analogue front-end circuitry always operating on the linear portion of its curve. When the output reaches 0dBfs, and then begins clipping, the input should still be operating completely linearly. The input level that produces clipping in the converter should be lower than the voltage that forces the analogue front-end into non-linear operation, thus well below the clipping level of the analogue front-end. That way there is never any analogue distortion passed on to the digital data.

What an ADC has instead of headroom is dynamic range. For some ways down from 0dBfs its operation is linear and clean. For the Audiophile that is specified as slightly more than 100dB. This is not an exceptional figure, but it is far more than adequate for LPs, cassettes, and quite a few other sources. The peak input can be a fair amount below clipping without ever losing any source resolution. Thus your concern is not "how much headroom does the soundcard have?" but rather "where is 0dBfs relative to my source device, some place convenient or some place inconvenient?"

The dB scale is a relative one. It has no particular values unless it is referenced to something external. For soundcards, as for many other audio devices, there are two main standards. Nominal levels for these two are +4dBu and -10dBV. The "u" and "V" designate some particular meanings, which I believe are u: 1.23V and V: 0.316V. The higher level +4dBu is considered ‘professional' and the other is ‘consumer.'

The real difficulty is that many ‘consumer' devices are constructed without close consideration of the standards. The Audiophile's specs say that its peak analouge input is +2dBV, which is in accordance with the -10dBV standard, I believe. Changing its maximum to a higher voltage could only be done by shifting the entire scale, bring its minimum useable signal level higher also. Then it would not match with other standard devices.

Having read discussions in a number of places over the past four years, I have seen quite a few complaints about inconvenient input levels, when using the Audiophile and a number of other soundcards. A few of these complaints have been the same as in this thread, "my levels are too high for the soundcard," but the great majority have been on the other side " how do I get an adequate input level for good recordings?"

These later are almost always from people who mistakenly believe their peaks should be pushing the red line when in fact they are really doing very well with peaks in the range of -12dBfs to -6dBfs. Regardless, all of these complaints, of which ever stripe, arise because people are attempting to use professional type devices without the full professional equipment regalia. This might be possible if standards were strictly adhered to but I suspect most circuit designers consider themselves artists of some sort and take what ever ‘artistic liberties' they feel inclined towards.

These same discrepancies might well exist with professional equipment -- the actual output levels of source devices may be higher or lower than the standards -- but since most professional setups will include the adequate amenities (mainly mixers), the differences will be take care of in the normal course of set-up for recording, without triggering a second thought.
Go to the top of the page
+Quote Post

Posts in this topic
- evereux   Recording line in with Soundforge   Aug 10 2003, 10:34
- - Pio2001   The line in level is not in Soundforge, it must be...   Aug 10 2003, 18:01
- - evereux   QUOTE If it is reduced, but is still clipped It i...   Aug 10 2003, 19:25
- - Pio2001   I think the solution is to find the line in record...   Aug 10 2003, 21:40
- - ancl   I have an Audiophile 24/96 myself, and as far as i...   Aug 11 2003, 00:01
- - Pio2001   ¿ WTF ?! I'm not going to get rid of my...   Aug 11 2003, 11:40
- - Brajanath   Hi, I'm new here (just started yesterday), an...   Nov 16 2003, 05:38
- - fewtch   QUOTE (Brajanath @ Nov 15 2003, 09:38 PM)My w...   Nov 16 2003, 08:39
- - Pio2001   QUOTE (Brajanath @ Nov 16 2003, 05:38 AM)Befo...   Nov 16 2003, 16:22
- - Brajanath   Your message is perfectly OK. I have not had the ...   Nov 16 2003, 16:56
- - Pio2001   I don't know... The problem with the cassette ...   Nov 16 2003, 17:10
- - Brajanath   QUOTE (fewtch @ Nov 15 2003, 11:39 PM)Not rea...   Nov 16 2003, 17:20
- - fewtch   QUOTE (Brajanath @ Nov 16 2003, 09:20 AM)QUOT...   Nov 16 2003, 17:53
- - JeanLuc   Of course you can adjust input sensitivity with th...   Nov 16 2003, 18:07
- - fewtch   QUOTE (JeanLuc @ Nov 16 2003, 10:07 AM)Of cou...   Nov 16 2003, 18:16
- - JeanLuc   It is the HW mixer which is situated besides the ...   Nov 16 2003, 18:25
- - Pio2001   QUOTE (fewtch @ Nov 16 2003, 05:53 PM)I don...   Nov 16 2003, 18:50
- - ancl   QUOTE (JeanLuc @ Nov 16 2003, 07:25 PM)It is ...   Nov 16 2003, 18:52
- - JeanLuc   QUOTE (ancl @ Nov 16 2003, 05:52 PM)I guess w...   Nov 16 2003, 18:55
- - ancl   QUOTE (JeanLuc @ Nov 16 2003, 07:55 PM)QUOTE ...   Nov 16 2003, 19:01
- - JeanLuc   QUOTE (ancl @ Nov 16 2003, 06:01 PM)You can, ...   Nov 16 2003, 19:09
- - Brajanath   QUOTE (Pio2001 @ Nov 16 2003, 09:50 AM)QUOTE ...   Nov 16 2003, 22:54
- - fewtch   QUOTE (Brajanath @ Nov 16 2003, 02:54 PM)My o...   Nov 16 2003, 23:22
- - Pio2001   QUOTE (JeanLuc @ Nov 16 2003, 07:09 PM)So I j...   Nov 17 2003, 01:46
- - Kos   Thiis is probably very banale for you in here, but...   Apr 21 2006, 09:56
- - AndyH-ha   I have to admit I have not read this very closely,...   Apr 21 2006, 20:20
|- - Pio2001   QUOTE (AndyH-ha @ Apr 21 2006, 09:20...   Apr 23 2006, 03:26
- - AndyH-ha   No that is incorrect. Some of those that use the W...   Apr 23 2006, 10:53
|- - Pio2001   I don't follow you. You are making a distinct...   Apr 24 2006, 19:43
- - AndyH-ha   Most analogue volume controls are passive. Potenti...   Apr 25 2006, 05:39

Reply to this topicStart new topic
1 User(s) are reading this topic (1 Guests and 0 Anonymous Users)
0 Members:


RSS Lo-Fi Version Time is now: 21st April 2014 - 16:11