Xibolba

On 6/14/2020 at 11:46 AM, Base 57 said:

Today I've been recording a bass guitar through adjacent mono inputs (1+2, 3+4,  etc.)  to two mono tracks. I can't duplicate the OP's problem. However this...

confuses me. I have an Orion 32+. CbB displays all of it's inputs as Left, Right or Stereo just as OP described here.

 

There is not another setting as far as I know.  Select a stereo input to record to record a stereo track. Select a mono input to record a mono track.  That's it.

I wish I could be more helpful but I have no idea why you are getting that distortion. All I can do is confirm that it works properly for me. 

Thanks for that feedback - especially that you were recording bass frequencies.  It's been a while I've been avoiding the dual mono scenario since being bitten.  It's worth another go.

I played back the tracks as is - I didn't want to introduce any extraneous processing in the example.  I kept the tracks centered and besides playing both back, I soloed each.  The two tracks are summed to a master buss.  No plugins present.  The distortion artifacts are present in the tracks and not affected by panning.  While this example is setup to demonstrate the distortion artifacts, what I'd normally do is place the mics to pickup different aspects of the sound and then mix the mono tracks together.

I can work around the issue by recording in stereo and then either splitting the track to two mono tracks or send each channel to a different aux track.  Should that really be necessary?  Seems like recording dual mono in adjacent tracks should not result in distortion.

I think I said it in the writeup, but it's not recording two mono tracks that results in distortion, it's only when the two mono tracks are recognised by CbB as elements of a stereo pair.

Am I missing some trick or setting?  Is it a bug?  It's really apparent in the project I shared (link in original post).  I could record at lower bit/sample rates and drop .mp3's here, but as this is an audio fidelity issue it seems better present the best possible wav files.

From outside the box it feels as if different code paths are executed here and one of them isn't behaving well.  Or I'm missing a trick ...

I'm getting distortion when I record (for example) interface channels 1 and 2 as dual mono tracks in CbB.  I do not get this distortion using Reaper or Studio One.  I have found it true with any pair of interface channels seen to CbB as a stereo pair.   Quick example:

In CbB the first to channels of my interface may be addressed as:

• input 1 left mono
• input 1 right mono
• input 1 stereo

If I record two microphones into interface channels 1 and 2 into a CbB stereo track as input 1 stereo I don't get distortion.  

If I record two microphones into interface channels 1 and 2 into a pair of CbB tracks using the sources input1 left and input1 right respectively I do get distortion.

 I'm really hoping it's a trick I'm missing, hence the post.

The test:
This experiment demonstrates the distortion artifacts found when recording low frequency signals in 
adjacent tracks within a stereo pair (by this I mean one microphone in left input 1 and the other in 
right input 1).

An interesting note is that adjacent channels outside  a stereo pair (e.g.: input 1 right and input 3 left) do not have the distortion artifact.

Description of the test bed:
  The source is a upright bass playing pizz (sloppily) through a 2 octave chromatic scale
  The interface is a Focusrite Clarett 8Pre X (Thunderbolt)
  Two SM57 microphones on a stereo bar positioned 12" in front of the source
  One microphone is labled A and the other B
  Microphones are connected to inputs described in scenarios  
  Microphone preamp gain set to 12 o'clock on channels under test
  No parameters changed between tests except as noted in the table below
  Recorded at 96/24
  Exported at 96/24 no dithering
  
General observations;
  No meter registered above -26 dB, (interface, focusrite control, DAW) while recording or during playback

  The quality of the distortion is akin to an extreme EQ boost with a narrow Q where you can hear the filter ringing at resonance.

  The distortion is present in both tracks - it is not an artifact of playing both tracks simultaneously.

  Outside that directly demonstrated by this test, level doesn't seem to change whether or not distortion is present.

  I have reproduced this problem recording amplified electric bass and also drum kit with strong low frequency components (e.g: low toms and bass drum, low power chords on guitar) with a variety of microphones (SM57, RE-20, K-2, SM25, R10, ...).

  In all observed cases the source recorded on both channels was the same (as in the case of this audio example) or similar (as in the case of low toms and bass drum where this is always bleed).

Audio examples:

             Channel
Test      Mic A  Mic B  Result
--------- -----  -----  --------------------
One         1      2    distortion
Two         1      3    no distortion
Three       7      8    distortion
Four        6      8    no distortion
Five        2      3    no distortion

DAWS involved in testing:
   Reaper
   Studio One
   Cakewalk

Hardware involved in testing:
   multiple Presonus interfaces (from FireStudio to StudioLive 16.4.2)
   multiple Focusrite interfaces (from Scarlett/USB to Clarett/Thunderbolt)

The audio is too large to drop here - a link to the writeup, a picture of the recording setup, and a CbB project that demonstrates the audio artifacts may be found here: 

https://drive.google.com/drive/folders/10J5yQVtnrTSXbQws7KwNAfjCttG_5FFm?usp=sharing

Thanks for any info.

On 9/2/2019 at 4:36 PM, mettelus said:

Some of that work was done in a utility called Project Scope a few years ago. The author might be available if you want to add on to it. Aslow3 is another member who did similar work towards translating cwp files to another format and might be another resource for you.

I've looked at the tool and corresponded with the author.  Nice guy and helpful as he could be, but he was similarly hampered by the lack of project file documentation.

As it stands the tool doesn't go far enough to encompass what I'm looking for, and he didn't offer up his source.   

If I had access to an RFC document describing the cakewalk project template I could do this myself. 

I realize that CbB needs the freedom to change the project structure, but so does Microsoft (I'm thinking of the Office suite) where the file formats are documented allowing other tools (e.g.: Open Office) to read/write these files.  Open Source projects are used to chasing this and provide updates as I'm just looking to read.

On the other side of the coin. if it's a feature of Cakewalk there is no need to publish the file structure and the user experience would be seamless across releases.

I would like to see a way to generate a 'project inventory' for offline analysis/documentation of a project.  This would make providing project documentation to a client extremely easy.

If this export could be in one of the delimited formats recognized by spreadsheet programs it would be ideal. 

The sort of information I'm looking for is:

Project:

• Name
• Tempo

Tracks and busses: 

• Name
• Notes
• MSR/Archive/Hidden state
• Trim
• Fader
• Pan
• FX bin
  • On/Off
  • FX name
  • FX parameter/value pairs
• Pro-channel
  • On/Off
  • FX name
  • FX parameter/value pairs
• Track buss routing
• Sends POST and send level

Alternately, if there is a document describing the structure of the CbB project file I'd take a stab at writing it myself.