Become a Site Supporter and Never see Ads again!

Author Topic: Matrix question  (Read 3112 times)

0 Members and 1 Guest are viewing this topic.

Offline Big Muff

  • Trade Count: (3)
  • Taperssection Member
  • ***
  • Posts: 626
  • Gender: Male
  • James Marshall Hendrix 1942 - 1970
Matrix question
« on: July 15, 2018, 08:13:30 PM »

Hi all,

I'm creating a mix of 3.31.85 from the Oade and Miller sources.
My issue is that the tracks are different lengths- meaning I can sync the start of the 2 sources just fine at the start, but about 3 minutes in one gets ahead of the other creating an echo effect.
In the case of Music never stopped; the two sources start off perfectly synced- approximately 3 min in the echo effect - but then it starts to corrrect itself towards the end of the song.

What is a good way to correct this using wavelab 6?

Live rig: DPA 4021 > V3 > iHP-120

Vinyl rig: Music hall 5.1 w/acrylic platter > Ortofon Quintet Black > Pro-Ject Tube-Box II > V3 > UA-5

Offline opsopcopolis

  • Trade Count: (3)
  • Needs to get out more...
  • *****
  • Posts: 2144
Re: Matrix question
« Reply #1 on: July 16, 2018, 06:28:33 AM »
You’re going to have to time correct one of the files. Not sure how to do it 8n wavelab but there are tons of threads about this. Probably been discussed somewhere

Offline nulldogmas

  • Trade Count: (6)
  • Taperssection All-Star
  • ****
  • Posts: 1626
    • How I Escaped My Uncertain Fate
Re: Matrix question
« Reply #2 on: July 16, 2018, 08:31:03 AM »
Yeah, the basic principle is 1) place markers at two clearly defined points (loud clap, snare hit, etc.) near the start and end of each recording, 2) measure the distance between the two for each file, 3) divide one distance by the other, 4) use this ratio with some kind of pitch/tempo filter to either stretch or shrink one file until the two line up.

Though if you're saying that they get out of sync and then go back into sync again, this may be a more complicated problem than just two recorders running at consistent but differing speeds. Still, the above system should work, even if you find you may need to do it in multiple smaller segments rather than just one big stretch.

Offline one8ung

  • Trade Count: (4)
  • Taperssection Member
  • ***
  • Posts: 276
  • Gender: Male
Re: Matrix question
« Reply #3 on: July 16, 2018, 02:15:04 PM »
Mics: Countryman B3 - DPA 4061

Power: CA-UBB - SP-SPSB-10

Recorders: Tascam DR-2d (4x)

Offline morst

  • I think I found an error on the internet; #UnionStrong
  • Trade Count: (2)
  • Needs to get out more...
  • *****
  • Posts: 5950
Re: Matrix question
« Reply #4 on: July 20, 2018, 12:25:12 AM »
Yeah, the basic principle is 1) place markers at two clearly defined points (loud clap, snare hit, etc.) near the start and end of each recording, 2) measure the distance between the two for each file, 3) divide one distance by the other, 4) use this ratio with some kind of pitch/tempo filter to either stretch or shrink one file until the two line up.

Though if you're saying that they get out of sync and then go back into sync again, this may be a more complicated problem than just two recorders running at consistent but differing speeds. Still, the above system should work, even if you find you may need to do it in multiple smaller segments rather than just one big stretch.


Be careful of using an audience clap. Ideally, you would like to synchronize on something close to a stage microphone, or even the break music, in order to get the most accurate time-alignment.


If the sync goes out and comes back in, it could be that they are analog sources with much higher speed variation than we are used to on digital. If that's the case, it will be incredibly difficult to do this without resorting to something like Plangent Process which costs $$ and takes place on 192kHz transfers


Hm, Oade source is digital
https://archive.org/details/gd85-03-31.oade.connor.8244.sbeok.shnf


here's a digital miller source
https://archive.org/details/gd1985-03-31.139080.s2.pcm.sbd.miller.fixed.flac24


I guess watch for dropouts that would cause sync issues. But dropouts would not result in the end of the track coming back into sync.


Are you sure Dan Healy is not just fooling your ears with trippy sound effects?
https://toad.social/@morst spoutible.com/morst post.news/@acffhmorst

Offline Big Muff

  • Trade Count: (3)
  • Taperssection Member
  • ***
  • Posts: 626
  • Gender: Male
  • James Marshall Hendrix 1942 - 1970
Re: Matrix question
« Reply #5 on: July 20, 2018, 07:52:27 AM »

I'm slowly finding my way.
Sadly I'm not a mac person. using only wavelab here.
In wavelab, there are several options/boxes to check when doing a time stretch: "preserve pitch" is obvious, but I haven't dug in to find out what the "DIRAC processor" is.
Or whether or not I should check the "audio quantize" box and adjust the rythym accuracy???
What I have found is working : leaving all those options unchecked and chopping the clip into smaller, roughly 45-1:30sec clips and stretching them each one at a time according to the other source.

I'm certain it isn't Healy's antics :) and more of a warbly reel or an untuned nakamichi that's affected the source along the way.
Using the Oade source as a guide. Given it's a PCM recording and immune to analog mechanisms. I've been stretching the sbd, little by little - a piece a time.

If you want to see the shape of things, open up both Oade and sbd sources of Music never stopped. Out of the 4 tracks I've done now, this one is most warped.
Around 3:30 the sbd takes a big drag and starts to come back and forth around the 5-6 minute mark.
It's what led me to follow nuldogmas suggestions and use the smaller clips.

It's painful compared to the last matrix I made using 2 digital sources (jgb 11.9.91) 
I'm hopeful that the set 2 pcm sbd will be forgiving and downhill from here.


I'd like to know what you all feel is a better listen, with +2db added to the Oade or without?
https://drive.google.com/open?id=1FSJqCw5bBmLESKIubb39LFc2A7hxYmd6



Cheers!
Live rig: DPA 4021 > V3 > iHP-120

Vinyl rig: Music hall 5.1 w/acrylic platter > Ortofon Quintet Black > Pro-Ject Tube-Box II > V3 > UA-5

Offline morst

  • I think I found an error on the internet; #UnionStrong
  • Trade Count: (2)
  • Needs to get out more...
  • *****
  • Posts: 5950
Re: Matrix question
« Reply #6 on: July 21, 2018, 01:22:11 AM »
In wavelab, there are several options/boxes to check when doing a time stretch: "preserve pitch" is obvious,
If your two sources are not at the same speed, they are not at the same pitch. You need pitch correction along with your speed correction because they go hand-in-hand. I would think that you would NOT want to use the "preserve pitch" option.
https://toad.social/@morst spoutible.com/morst post.news/@acffhmorst

Offline nulldogmas

  • Trade Count: (6)
  • Taperssection All-Star
  • ****
  • Posts: 1626
    • How I Escaped My Uncertain Fate
Re: Matrix question
« Reply #7 on: July 21, 2018, 03:04:03 PM »
In wavelab, there are several options/boxes to check when doing a time stretch: "preserve pitch" is obvious,
If your two sources are not at the same speed, they are not at the same pitch. You need pitch correction along with your speed correction because they go hand-in-hand. I would think that you would NOT want to use the "preserve pitch" option.

Seconded.

And yeah, aligning analog sources is a major undertaking. Re-aligning at the start of each song is probably your best bet, but even then you have a lot of work ahead of you. Might be easier to invent a time machine and go back to 1985 with a second digital recorder.

That said, it sounds great so far! I marginally prefer the +2dB version — feels a little warmer and less harsh.

Offline Big Muff

  • Trade Count: (3)
  • Taperssection Member
  • ***
  • Posts: 626
  • Gender: Male
  • James Marshall Hendrix 1942 - 1970
Re: Matrix question
« Reply #8 on: July 29, 2018, 12:50:21 PM »

and done...  http://bt.etree.org/details.php?id=600553

went with a 2.5db boost on the Oade source.
Have to admit, it wasn't an awesome time but had to finish what I started.

« Last Edit: July 29, 2018, 12:54:10 PM by Big Muff »
Live rig: DPA 4021 > V3 > iHP-120

Vinyl rig: Music hall 5.1 w/acrylic platter > Ortofon Quintet Black > Pro-Ject Tube-Box II > V3 > UA-5

Offline yousef

  • Trade Count: (2)
  • Taperssection All-Star
  • ****
  • Posts: 1447
  • Gender: Male
Re: Matrix question
« Reply #9 on: July 29, 2018, 05:46:44 PM »
In wavelab, there are several options/boxes to check when doing a time stretch: "preserve pitch" is obvious,
If your two sources are not at the same speed, they are not at the same pitch. You need pitch correction along with your speed correction because they go hand-in-hand. I would think that you would NOT want to use the "preserve pitch" option.

Seconded.


Thirded.
music>other stuff>ears
my recordings: http://db.etree.org/yousef
http://www.manchestertaper.co.uk
twitter: @manchestertaper

 

RSS | Mobile
Page created in 0.171 seconds with 39 queries.
© 2002-2024 Taperssection.com
Powered by SMF