Welcome Guest ( Log In | Register )

Line in overload on soundcards, M-Audio etc unable to record properly ?
post Aug 17 2003, 20:13
Post #1


Group: Super Moderator
Posts: 3936
Joined: 29-September 01
Member No.: 73

Here's a conversation between Patsoe and me started in private messages, about the analog line in of soundcards. It seems that it's not possible to record properly many analog sources, because the recording level is clipped no matter what.

QUOTE (Patsoe @ Private message)
QUOTE (Pio2001 @ Private Message)
[About Marian Marc 2]I couldn't record vinyls since the input is overloaded by my vinyl output level. Note that the M-Audio Audiophile is worse : not only the input is overloaded, but there is not even a volume control for analog recording !!! Clip or die !

I've been thinking about this again. How can it be? Marian lists a maximum of +8dBU for the input in the manual, and the Audiophile should actually do fine too, with a +2 maximum (fine for most consumer line-outs). Or?
QUOTE (Pio2001 @ Private message)
We have no more infos about line in level behaviour, compared with the Audigy, for example. This is a big issue IMHO, becase we're advertising soundcards that are unable of proper analog recording.

Shouldn't comparison of the line-ins the way Digit-Life do it be fine? They play back from their Lynx Two with all cards, so this makes for a reference by which we should be able to say something, I'd think.

Here's what I can say about the marian Marc 2 Soundcard :

Max input : +8dBu

I took a track that I've got on CD and LP (45 RPM). I played the CD (an MPC file) throught the digital output, bit exact, no DSP, to the Sony DTC 55ES :
Impedance : 470 Ohm, Rated output : -4dBs. Load : Arcam Diva ampli + Marc 2 analog input + External equalizer (2 tape outs + monitoring). This stands for an analog source that should be calibrated to a correct line out level.

I recorded it with the Marian Marc 2 analog input, line in level at 0 dB, then I also recorded the vinyl, without changing the recording level.

Here are the results :

Original :

Copy of the Sony Playback :

A "statistics" tool run on 14 seconds of audio shows an RMS level 1 dB lower than the original. Thus the output of the Sony is well tuned with the input of the Marian. The 1 dB loss might even just come from the triple load.

Copy of the vinyl (phono tape out) at the same recording level :

The same analysis (ran on a selection that is not clipped) shows a level 10.4 dB higher than the Sony recording (same load). Thus if the Sony has a rated output of -4 dBs, the Phono out is +6.4 dBs.
The cartridge is a Stanton Trackmaster EL (rated 5.2 mV at 5.5 cm/s), the preampli an Arcam Diva A85 (phono input sensitivity 2.5 mV). Thus the +10.4 dB comes from the cartridge output, that is 5.2 mV instead of 2.5 (+6.4 dB) and that plays the 45 RPM at about 55 cm/s (ten times faster than for the 5.2 mV measurment).

In addition, the peaks are completely clipped.

Lowering the recording level leads to this :

Completely unuseful, the clipping is still there, but lower. The recording level is just a digital process applied to the line in digital stream. We could record at full volume and adjust the level later in the wave editor, the result would be the same.

It seems that Midiman understood this, since they don't feature any recording level slider in the Audiophile 2496 soundcard.

The clipping occurs at -0.5 dB. Granted that the recording level resulted in a file 1 dB lower than the original, it means that the analog input of this card has no headroom at all. It can record a line out signal that is equal to the Sony one (-4 dBu), and it will clip exactly at the same level as the Sony playing a clipped file (0.5 dB higher, to be precise).
If a CD Player has a line output that is 1 dB higher than the Sony, it can't be recorded with this soundcard ! The input is overloaded.

The phono output level is 10 dB higher, and neither the Arcam Diva A85 inputs, nor the Cyrus One inputs, nor even the analog inputs of the Sony DAT are overloaded by it ! They handle it perfectly.

The solution seems simple : asking soundcards manufacturers to add an analog volume control for the line input before the ADC in their soundcards. Otherwise, the line input feature is quite unuseful.
Go to the top of the page
+Quote Post
Start new topic
post Aug 18 2003, 12:05
Post #2


Group: Super Moderator
Posts: 3936
Joined: 29-September 01
Member No.: 73

Sorry, deaf, I didn't understand what you mean. Can you give an example based on a CD player line-out -> Soundcard line-in, for example ? I'm a bit confused.

QUOTE (deaf @ Aug 18 2003, 05:42 AM)
When you place an attenuator in front of the line in ADC,  it will degrade the noise performance of the ADC.

How does an attenuator introduces noise ? Can you explain, please ?

QUOTE (deaf @ Aug 18 2003, 05:42 AM)
Then the best is to not have the attennuator on the soundcard, so when using "passive" sources you get no signal attenuation and the lowest noise added.

How do avoid clipping without attenuating ?

QUOTE (deaf @ Aug 18 2003, 05:42 AM)
When you have an "active", amplified source then you can add a passive attennuator if the amplifier does not have one to lower the signal amlitude to avoid clipping. A 24bit ADC has  a dynamic range sufficient not to degrade the signal from the amplified soure.

What is an "active" and a "passive" source ?

QUOTE (Patsoe @ Aug 18 2003, 09:32 AM)
Gaining/Volume is made by the AD/DA converter on the card.

Stupid ADC mad.gif !

This post has been edited by Pio2001: Aug 18 2003, 12:07
Go to the top of the page
+Quote Post

Posts in this topic
- Pio2001   Line in overload on soundcards   Aug 17 2003, 20:13
- - DigitalMan   QUOTE (Pio2001 @ Aug 17 2003, 11:13 AM)The so...   Aug 17 2003, 20:33
- - Pio2001   How do you know that your line in level was analog...   Aug 17 2003, 21:15
- - picmixer   I wonder if this by any chance could be related to...   Aug 17 2003, 23:48
- - deaf   When you place an attenuator in front of the line ...   Aug 18 2003, 03:42
- - Patsoe   QUOTE (DigitalMan @ Aug 17 2003, 08:33 PM)Is ...   Aug 18 2003, 07:32
- - Pio2001   Sorry, deaf, I didn't understand what you mean...   Aug 18 2003, 12:05
- - 2Bdecided   Pio2001, I think you're being overly critical...   Aug 18 2003, 13:02
- - Patsoe   I guess I agree with David. About the second part ...   Aug 18 2003, 16:00
- - deaf   Let me try to explain again. Basically, we want to...   Aug 18 2003, 18:19
- - Pio2001   If the maximum output level of CD Players (and sou...   Aug 18 2003, 20:42
- - 2Bdecided   deaf, your distinction between an active and pass...   Aug 19 2003, 10:46
- - Pio2001   I wonder if other soundcards behave the same.   Aug 19 2003, 11:22
- - deaf   2Bdecided, You are almost correct. It is hard to ...   Aug 19 2003, 13:19
- - Mac   QUOTE (2Bdecided @ Aug 19 2003, 09:46 AM)The ...   Aug 19 2003, 15:12
- - 2Bdecided   Mac, I (and the others, I think) are talking abou...   Aug 19 2003, 16:29
- - Pio2001   Your card is designed for recording vinyls directl...   Aug 19 2003, 20:10
- - Mac   Yay, I'm not stoopid I've not noticed li...   Aug 19 2003, 20:53
- - Patsoe   OK - summarizing: for highest compatibility, the r...   Aug 22 2003, 00:57
- - Pio2001   I don't know what to say. The cards used for p...   Aug 22 2003, 12:06
- - lucpes   Had the same problems. Solved by using some trim p...   Aug 22 2003, 12:31
- - Patsoe   QUOTE (Pio2001 @ Aug 22 2003, 12:06 PM)I don...   Aug 22 2003, 14:51
- - Pio2001   For "sound quality", it is very difficul...   Aug 22 2003, 18:36
- - 2Bdecided   Sorry to bring this thread "back from the dea...   Nov 21 2003, 15:04
- - spoon   Technically they shouldn't call it line in and...   Nov 22 2003, 12:04
- - 2Bdecided   Sorry to ask again, but can anyone confirm the lin...   Nov 28 2003, 11:49
- - Pio2001   If you use a multimeter, test at 50 Hz, no more. ...   Nov 28 2003, 12:22
- - KikeG   David: I don't remember the exact figure, but...   Nov 28 2003, 14:30
- - KikeG   QUOTE (Pio2001 @ Aug 22 2003, 06:36 PM)In the...   Nov 28 2003, 14:43
- - 2Bdecided   QUOTE (KikeG @ Nov 28 2003, 01:30 PM)David: ...   Nov 28 2003, 15:03
- - rhadinocentrus   1.)The forum suggests the use of a passive attenua...   Nov 30 2003, 07:54
- - KikeG   QUOTE (rhadinocentrus @ Nov 30 2003, 07:54 AM...   Nov 30 2003, 20:35
- - rhadinocentrus   QUOTE I doubt using an adequate simple passive vol...   Dec 1 2003, 22:24
- - JeanLuc   DC offset on modern HQ sound cards should be negli...   Dec 1 2003, 22:33
- - fallen_angel   The Prodigy 7.1's CODEC features analogue gain...   Dec 28 2003, 23:39

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: 25th November 2015 - 04:57