• SONAR
  • Should Clipping Be More Pronounced With Different Audio Interface
2015/02/03 21:26:59
cpkoch
I recently replaced a TASCM US 122  with a  Steinberg UR22 Audio Interface.  With the US 122 I seldom (if ever) noted a sharp nasty clipping sound.  Others who had listened to my stuff mentioned clipping but I chalked it up to their wanting something closer to perfection in the production.  However now, using the Steinberg unit, the clipping is, to me, very pronounced and irritating.  Is this to be expected?  
2015/02/03 21:32:06
AT
Not unless you are pushing the unit.  Try turning down the input of the unit when recording.  If it does it on playback of commercial stuff, then something is wrong w/ the output.
 
With low-cost interfaces it is too easy to clip.  Digital doesn't have the noise floor of analog, there is no reason to try to capture at high volume.  -12 dB is fine, even peaking to -6.
 
@
2015/02/03 21:46:17
cclarry
AT is correct.  Most hardware units are designed to do their best work
on an input signal that is at -18 db +/-  RMS.  This is what is referred to as the "Sweetspot"
for the hardware.  So, as AT stated, -6 db PEAK is a good place to be...it leaves you "headroom".

Audio Interfaces have varying types of Preamps, and,
consequently, their gain and headroom vary.  But as long is you're not
"Peaking in the red" then you should be in good shape on recording.  
Once the signal is recorded, how (and what) you use to process the signal, and your
gain staging, both pre and post processing, can impact "clipping" if you are over driving the signal on
output.  That's why it's very important to use your meters, and even
a 3rd party metering plugin if you are not confident in the meters in
your DAW or plugins.
2015/02/03 21:50:54
mettelus
+1 to AT... it is easy to clip some interfaces if trying to run them too hot. When tracking it is best to insert an audio track and leave all settings on that track at unity. Arm the track and adjust gain on the interface to peak barely onto the orange (in SONAR) at most (never red). If there is a performance part that is noticeably louder, that is a good time to test it (loud vocal passage or full guitar chords).
 
As long as an interface has good signal-to-noise ratio, there is a lot of leeway to bring that signal up and eliminate noise once recorded, but if a track is clipped during recording (i.e. SONAR records the raw audio with a flat top) there is very little to recover this situation other than to re-record it.
 
Do the wave forms of the tracks you are referring to have flat tops and bottoms? This is a key indicator of clipping regardless of what the interface is telling you.
2015/02/03 22:08:46
swamptooth
Right to all of the above, and you have to remember as well that clipping does not happen in Sonar, it happens in the interface.  There is no such thing as clipping in software, it's all math.  Clipping happens at the analog level.
With the steiny interface make sure +48v is off and guitars or other instruments run through the line inputs in the back of the unit.
2015/02/03 22:26:50
noynekker
swamptooth
Right to all of the above, and you have to remember as well that clipping does not happen in Sonar, it happens in the interface.  There is no such thing as clipping in software, it's all math.  Clipping happens at the analog level.
With the steiny interface make sure +48v is off and guitars or other instruments run through the line inputs in the back of the unit.


So what's the purpose of all the red clipping lights, do we pay no attention to them ?
Does it only matter when recording from the interface ?
ie. on mixdown . . . if the track is clipping, but the bus is not clipping, will there be no distortion at the interface ?
I always thought that gain staging throughout the signal chain had some importance.
2015/02/03 22:54:51
swamptooth
noynekker
So what's the purpose of all the red clipping lights, do we pay no attention to them ?
Does it only matter when recording from the interface ?
ie. on mixdown . . . if the track is clipping, but the bus is not clipping, will there be no distortion at the interface ?
I always thought that gain staging throughout the signal chain had some importance.

On the interface, yes  - in the daw, it depends on what you want to do.  Dan Worrall has a short video illustrating this in a pretty extreme way here.  I think a lot of it has to do with familiarity, and setting up guideposts.  Most daws as a rule, including sonar, don't define 0db according to the aes-17 standard which gives you a bit more leeway on output, but not input. 
2015/02/03 23:10:28
JonD
 
A quick google-search brings up more than a few reports of the UR22 being too hot (clipping).  You may want to research this and determine under what circumstances and/or if it's fixable.
2015/02/03 23:16:58
noynekker
Thanks swamptooth, an interesting video . . .
- no distortion in the DAW because it uses floating point maths
- the signal only distorted when converted to integer format to send to the audio interface
- not possible to clip a modern DAW's mix bus internally . . . consider it to have infinite headroom.
 
Lots to think about here, seems I really need to pay closer attention to my soundcard mixer's I/O settings.
 
As to the OP's original question, is it possible that the quality of the newer soundcards analog - - -> digital conversion is of a lower grade, causing the more noticeable distortion. What else could it be, unless the I/O settings have been changed in the new set-up ?
2015/02/03 23:20:34
JonD
noynekker, to answer your question, see my post above.
 
© 2026 APG vNext Commercial Version 5.1

Use My Existing Forum Account

Use My Social Media Account