Taperssection.com
Gear / Technical Help => Post-Processing, Computer / Streaming / Internet Devices & Related Activity => Topic started by: spott on June 18, 2003, 12:33:18 PM
-
I'm sure this has been beat into the ground but I was wondering what the best course of action would be to doing a post show matrix. I'd be happy to compile the infor I get and post it on the site. Damon and I were talking about doing one for the Drive by Truckers last night at the Black Cat but it didn't happen and I left early.
I'd gather that I'd take the DATs (AUD & SBD) and feed them into something like Soundforge or CEP and mix them from there??
Any help or ideas would be appreciated.
Later,
Spott
-
It is easy. Read in both tracks into cool edit pro in the mulktitrack view...time align if need be, and hit mixdown to file. Usually it is easier to time align using cdwave to split the sbd source into tracks!
Daryan
-
Just to further Daryans post, don't bother "stretching" the tracks....it is a pain in the ass, and not easy. I have done a few matrixes (sp?) and the easiest and most efficient way to do it is to cut the SBD wav into tracks, and then use a multitrack program, like Cool Edit to line it up, track by track, increase the volume of the SBD track if you want, then mix it down. Good results are not difficult to come by using this process.
Nick
-
Thanks for the thoughts... will the mixdown function do a 50/50 split or is it adjustable?
Spott
-
it is adjustable...it will normally just lay them equally on top of each other, but you can increase the relative volume of each track.
-
yeah, right clickj on the "V" button and play with the volumes until it sounds good to you. defintely do it track by track then piece it together and mix it down, email me if you have problems
-
I generally decrease the volume of the aud recording by at least 10 dbs. I look for ambient sound from the aud, but not much more.
-
if you're trying to calculate the amount of theoretical delay, i've heard 1 ft (difference between sound sources) equated to 1ms of delay.
good luck!
-
If you want to stretch one file to match the other:
match up a precise start point and end point on each recording and divide the length of one segment by the length of the other (same section of music). You'll find that one is maybe 99.98454% of the other or something like that.
Use wavelab's stretch function (it's the only one I've tried that sounds good). Select the whole file to be shortened and do a time stretch with these settings
It's probably good to set wavelab temp files to 24 or 32 bit as well.
Note that there's no point in using anything other than the ratio box to set the stretch. All the other boxes only appear to be more precise - in fact they don't produce the stretch that they indicate. Rather, they always produce the nearest '3 decimal ratio', so it's better just to set it there so you can see the length of the the file you'll *actually* get. And 3 decimals is it - any more precision ends up getting ignored.
-
Awesome... I will definitely make use of this knowledge if I ever come across a situation where I can do a matrix recording. Thanks!
-
Here's another method to time stretch in wavelab that doesn't involve calculations.
Get your tracks in sync in a montage. Then go to the last few minutes of the set & split them(in the middle/left of the screen hit "edit>split all tracks @ cursor"). Then move the end portion of the track that needs to be stretched so it is again in sync. The gap you have between the track is the difference between the 2 tapes clock speeds, or drift. Move your cursor so it is exactly @ the beg. of the end piece of the track that you intend to stretch. Right click over the longer piece of the track that you intend to stretch & hit "transform>time stretch to cursor". Just make sure "Preserve pitch" is enabled.
-
O-Dog - how'd this solution work out for you? Thinking of doing the same type of mix, would like to know how it turned out. Thanks!
-
Here's another method to time stretch in wavelab that doesn't involve calculations.
Get your tracks in sync in a montage. Then go to the last few minutes of the set & split them(in the middle/left of the screen hit "edit>split all tracks @ cursor"). Then move the end portion of the track that needs to be stretched so it is again in sync. The gap you have between the track is the difference between the 2 tapes clock speeds, or drift. Move your cursor so it is exactly @ the beg. of the end piece of the track that you intend to stretch. Right click over the longer piece of the track that you intend to stretch & hit "transform>time stretch to cursor". Just make sure "Preserve pitch" is enabled.
Korey has definitely found the way as outlined above. I have been doing matrix's as Nick describes above (splitting every 4 minutes or so and aligning). That method works, but is so time consuming. Use Korey's method and it is a snap.
I do have a question for you Korey. After streching, I compare the wave forms, and they do not line up, but they don't sound out of sync.
I unchecked audio quantize after reading the help file. Do you check, or uncheck audio quantize. I will prolly do a test each way, but thought I'd ask.
-
so lost on this :(
-
ok, if yinz could help me out quickly, that would begreat
i have an adobe audition 1.5 question considering a matrix recording:
1. how do i lower/raise the output of one of the stereo tracks ???
-
well, i dontreally wanna boost amplitude since the peaks arealready at 0db, i just want to mix the DAUD/SBD, i want the SBD to be louder in the mix than it is, i would ideally like 60%DAUD/40% SBD
-
Bean - in Audition, click on VIEW > Multitrack View. drag the DAUD source to Track 1 and the DSBD source to Track 2. sync them up.
then, on the left side of track 01, you can adjust the volume with the vertical dB slider (if the vertical slider is not showing, then Right-click on the part in Track 1 that says "V 0" and the vertical slider will appear)... drag it down to around - 6 dB, and listen. adjust that dB slider as necessary until it sounds good, then click on EDIT > MIX DOWN TO FILE > ALL AUDIO CLIPS. then Audition will create the new WAV file (mixed from both sources).
that should set you up nicely.
-
Bean - in Audition, click on VIEW > Multitrack View. drag the DAUD source to Track 1 and the DSBD source to Track 2. sync them up.
then, on the left side of track 01, you can adjust the volume with the vertical dB slider (if the vertical slider is not showing, then Right-click on the part in Track 1 that says "V 0" and the vertical slider will appear)... drag it down to around - 6 dB, and listen. adjust that dB slider as necessary until it sounds good, then click on EDIT > MIX DOWN TO FILE > ALL AUDIO CLIPS. then Audition will create the new WAV file (mixed from both sources).
that should set you up nicely.
you da man brotha, +T 8)
-
I just fionished doing one with wavelab 5, and (thanks Don) it sounds really good.
pm me with address and I will send whoever wants one a copy. This is son volt 22 apr 05
Ray
-
Question about this. I cut the sbd into tracks and loaded the aud and the tracks into CEP how do i go about getting them lined up? I'm confused on that part. Ok i think i have it figured out in wavelab, but i'm wondering about the mix. more board than aud? I don't think the board sounds that great at this show.
-
Question about this. I cut the sbd into tracks and loaded the aud and the tracks into CEP how do i go about getting them lined up? I'm confused on that part. Ok i think i have it figured out in wavelab, but i'm wondering about the mix. more board than aud? I don't think the board sounds that great at this show.
Your ears will tell you. From all accounts I have, there is no formula for the mix, just whatever sounds right. I used about 60% board and 40 aud. I had a great sound tech though. For me both sources sounded pretty good, so i mixed em closer to equal, and it sounds damn good i think.
Ray
-
one more question on this. In wavelab when you create a new audio montage and have all your audio tracks laid out there is a small box on the far right for effects and levels if those levels are changed from 0 how do you move them back to 0 i have been unsuccessful with moving them back
-
one more question on this. In wavelab when you create a new audio montage and have all your audio tracks laid out there is a small box on the far right for effects and levels if those levels are changed from 0 how do you move them back to 0 i have been unsuccessful with moving them back
you mean after you render the tracks??if so, there is no way to move them back, i dont think
Ray
-
this is before you render the tracks. Right to the right of the green tennis ball mute button
-
this is before you render the tracks. Right to the right of the green tennis ball mute button
i always just move them back and forth, no problem.what are you trying to do??
Ray
-
just wondering if there was a way to put them back to 0. I can't make it happen for some reason. I was moving them for different songs in the montage and wanted to move them back to 0 but couldn't
-
just wondering if there was a way to put them back to 0. I can't make it happen for some reason. I was moving them for different songs in the montage and wanted to move them back to 0 but couldn't
thats weird bro.
"????
sorry man+T for troubles.
Ray
-
just wondering if there was a way to put them back to 0. I can't make it happen for some reason. I was moving them for different songs in the montage and wanted to move them back to 0 but couldn't
automation?
-
just wondering if there was a way to put them back to 0. I can't make it happen for some reason. I was moving them for different songs in the montage and wanted to move them back to 0 but couldn't
automation?
??
-
I dunno if this helps but in Sound Forge I've been able to time align SBD and AUD's by resampling one of the sources, then setting the sample rate back to normal.
I find matches at point A and point B on both recordings, usually a snare or kick hit at the ends of either track, then compare the actual number of samples in each clip. Invariably one will have a larger number of samples. So I find the ratio of one to the other and multiply the sample rate times that ratio to find the new sample rate. I've found almost always that my JB3 has an error of something like 2.5 samples/sec at 44.1 so I resample the clip to something like 96007 Hz then just *change the playback rate* (NOT resample) to 96000. Then I resample down to 44.1 or whatever the other clip's rate is. It's worked like magic for me, but your numbers may vary.
Note that the sound forge time stretch does not work well enough. The difference of 1/2 second over 2 hours is too small for it to handle.
Of course now that I have my multitrack portable PC running Sonar 4 on 16-24 tracks this is not an issue. :)
-
I've found almost always that my JB3 has an error of something like 2.5 samples/sec at 44.1
So you run analog-in to the JB3? What was your control device?
-
I've found almost always that my JB3 has an error of something like 2.5 samples/sec at 44.1
So you run analog-in to the JB3? What was your control device?
One matrix I did was set up like so:
SBD: Mono out from the board into MZR-700 MD in mono mode, transferred analog in to my PC thru M-audio Delta 1010LT analog-in.
AUD: Wmod UA-5 optical->JB3
...So I guess I misspoke. The error is probably coming from the MD. Not surprising. But I also noticed a similar (but not exactly the same) discrepancy when doing the same thing with 1 source being a Laptop and the other a JB3. If I had a freaking MD deck with digital out I could make a better evaluation, but I'd rather not use the MD unless I absolutely have to.
Lately I've been bringing my luggable PC and doing a 4-track matrix in Sonar, then just clipping out the stage delay from the AUD side. Much better than having to resample and all.
-
Here's another method to time stretch in wavelab that doesn't involve calculations.
Get your tracks in sync in a montage. Then go to the last few minutes of the set & split them(in the middle/left of the screen hit "edit>split all tracks @ cursor"). Then move the end portion of the track that needs to be stretched so it is again in sync. The gap you have between the track is the difference between the 2 tapes clock speeds, or drift. Move your cursor so it is exactly @ the beg. of the end piece of the track that you intend to stretch. Right click over the longer piece of the track that you intend to stretch & hit "transform>time stretch to cursor". Just make sure "Preserve pitch" is enabled.
holy easter eggs, i just did this for the first time. i was a LONG time cutter, or splitter. dont know why but i resisted learning this until this week, wow.
this is way easier than you think, esp. with all the help from ts.com
one thing i cant figure out, how to grab a block of 24/96 files and have them stretched together, because otherwise i only strech the last hr or so. grouping them, and 'locking' them didnt work properly,