After LIVE GIG Quantize; Establishing Project tempo, grid & AudioSnap??

I am a little unclear in establishing the tempo and grid (post) live performance.
 
A little background:
I am currently using Cakewalk Sonar 8.5PE in live performance recording.  I purchased X1 and X2, but have not installed them yet; learning curve.  Four times a year I volunteer, track and mix live performances of children (ages 8 to18) at The School of Rock, Devon, PA.  I’ve done this for 2-years.  The track list, below, is typical of a live performance. 
 
The childrens’ timing is not always accurate.  I would like to quantize/fix the timing if I can.  Currently, I’ll split a clip and position it accordingly.  As you know, this approach is very tedious and time consuming.  I’m usually faced with 24-songs and 24-sets of musicians; musicians change after every song.  It’s a pretty crazy environment.  As part of my workflow, I track the show as one project and then break the songs into single projects for mixing.
 
My problem is that I can’t get a handle on establishing each song’s project tempo to set up AudioSnap for quantizing the tracks (all analog) of a mix.  It’s establishing the proper tempo from a live performance that has me flummoxed.  Most tutorials I have seen use either MIDI or audio loops which already have the tempo defined.  I haven’t seen an example/tutorial to quantize the first track of a live (no click) performance.  For some reason I remain in the dark on this.
 
I see the Tempo options within the AudioSnap palette, but it’s not clear to me how I begin to use them; which tracks to start with?  Will this cause the other tracks to change pitch??  Through my ignorance I have totally missed these concepts in the tutorials.  Banging my head against the wall hasn’t helped!
 
I tend to think the tempo would be established by the drums; kick or snare, though, in my case, they are not typically equally spaced/timed sufficiently that I can determine a reliable tempo.  Where do I start, workflow?/
 
My questions are:
 
1.      How would you begin the process of correcting the timing (quantization) given a track list below?
2.      How to establish/define the tempo?
3.      What is a typical workflow regarding a live project?
4.      Can you point me in the right direction of a video/tutorial which specifically discusses live performance mixing?
 
(Note: I do have several of Cakewalk’s videos, but as I mention, all are loops and have already defined the tempo.)
 
DAW Track        Instrument
1                           Kick
2                           Snare
3                           Tom – High
4                           Tom – Median
5                           Tom – Low (Floor)
6                           Overhead – Left
7                           Overhead – Right
8                           Bass
9                           Guitar
10                         Guitar
11                         Guitar – Lead
12                         Guitar – Acoustic
13                         Keyboard 1
14                         Keyboard 2
15                         Vocal – Center (Main)
16                         Vocal - Left
17                         Vocal – Right
18                         Vocal – Keys
 
Thanks very much for your consideration.  Your insights would be greatly appreciated.  This is a problem that I have not been able to resolve since the early days of Sonar.

Personally I would ask myself some critical questions.

• What is the output going to be used for?
• How isolated are the sounds? (i.e. how much bleed is there from one item to the other?)
• Do you have any life outside this project?

If the output is used to give the students a memory of what they did, I wouldn't spend the time to fix timing. You will spend DAYS fixing timing on that much material, and make them think they are better than they are. If it is used to promote the school/fundraise, etc., then fixing the timing may be worth it, but you would only want to spend the time fixing the timing on the tracks needed for that (likely half of the material you recorded if producing a normal length album, probably only one song or part of a few songs for promos.) For that matter, if this were for a promo or album I would try to record in a studio. If budget doesn't allow that, maybe try overdubbing vocals and acoustic guitars? (Even the DI output of an acoustic guitar can catch a lot of stage noise if the guitar's pickup uses a microphone...)


If the sounds aren't isolated well, you are going to run into a huge headache. For instance- if you move the drums forward a bit but don't adjust the vocal mics that have heavy drum bleed, you will cause some issues with clarity/phase.

The third question sounds tongue in cheek, but I'm really serious. If you have any life outside of this project that life will suffer for a while as you try to complete it. If you have people that are expecting everything done perfectly and quickly I would reset their expectations. You are volunteering. That doesn't mean you need to give them your sanity too. 




That all said, if you still want to give it a shot, I would do the following.


1. Start with making things better at the source. If you can take a DI out from an amp it will be much cleaner sounding than a mic in front that is right next to the drums. You may have to tweak with an amp sim for tone, but the added isolation is worth it. If you're not using a drum shield find a way to get one. DI keyboards instead of micing amps as well (or take MIDI from them if you can.)
2. Think about how best to split it up. Doing one project per song might be more work than help. Are the students using the same instruments? Playing similar styles? If you can split into three projects that have similar instrumentation, for instance, instead of one per song that will help you keep the mix organized and will help you mix more efficiently.
3. When it comes time to mix, do one song first. It is likely you will be able to re-use what you did for many of the other songs by using Track Templates, FX chains, etc. At the very least, your settings for one song will be a good starting point for the others.
4. Do yourself a favor and start by setting the low and high filters on everything before you do anything else mix-wise. There can be a lot of bass/rumble in vocal mics, and way too much cymbals/snare sizzle in guitar/bass amp mics, etc. By setting your filters first you save yourself from a lot of headaches later on in the mix process. The classic filters for vinyl mastering were 47 Hz and 12kHz. You probably won't hurt anything by setting all of your filters to default to this for all instruments, then move the low filters for the bass, floor tom, and kick below 40 Hz, and the high filters on the drum overheads back above 12kHz. Your guitars and keys will probably benefit from low filters at 100 or even higher. For vocals, I would start with low filters cutting all below 100 Hz and work your way up. Cutting off the highs above 12kHz will likely help too given cymbal bleed.
5. Similar to using the filters, decide what you do/don't need from each track and remove sections you don't want. You don't need the guitar amp track during sections where the drums are playing loud and there are no guitars playing, for instance. Be careful when cutting out vocal mics, though, as these often catch so much drum bleed that it can be hard to keep the drum sound consistent when cutting out vocal tracks where the vocalists aren't singing. Automation can be your friend here- turning the vocals down a little is more subtle than cutting them out.
6. If you have a drum sample replacement plugin (Drumagog, Slate, etc) use it for the kick, snare, and toms. The likelihood is that the drum tones live are nowhere close to what you'd really like to hear, and this can really tighten up the overall tone.
7. Don't compress too much. Compression with output gain turned up can sound great on many things, but for live recordings with a lot of bleed it just raises the volume of the noise/bleed. So generally I would use limiting to tame peaks a bit, but stay away from compressors with thresholds that are too low.


As for timing.


1. Don't try to quantize things unless you are going to try and add other MIDI tracks or loops later. It's WAY too much work and can kill good vibe, too. In your case I would NOT use AudioSnap.
2. Instead, I would manually fix places where there is an instrument that is obviously off compared to others. 
3. In general, if things are a little off and you want to fix them, but no single instrument is obviously off, move the drums instead of the other instruments. The reason for this is simple. It is much easier to see where the drum transients are, which makes splitting them more accurate, and aligning the drums to where the other instruments sound right is easier than the other way around.
4. When moving drums make sure you move all of them together to avoid phase issues. Clip groups can help with this. Essentially, before you start moving select all the drum clips, right click and choose to group the clips. There is an option in the Preferences that determines whether new groups are created when you split the takes. Make sure this is checked. Then, when you split the drum tracks all of them will split and move together.
5. Remember: "Done and not perfect" is better than "perfect and not done", because "perfect and not done" is neither. Figure out where you want the bar to be, but be realistic. These are live performances from students that are recorded in a less than ideal environment. They won't sound perfect.

Peace,

Tunes

So you are just trying to put all the other tracks in time to one main track?


Id use an over head as a main one to draw transients from

Select Audio Transients from track view - Make sure they all fall where ya want them to and the sensitivity is set where you want through Audio Snap,

Select the Clip

Right Click and select Add Clip to Pool

Then your ready to apply it to the other tracks

Turn on the Audio Transients on the track you want to snap,

Select the clip to be snapped

Right click Clip

Select Quantize to Pool


I AM DOING THIS ALL FROM MEMORY so double check before saving on your experiment copy

digi2ns......this is an interesting approach I would never have thought it.   My first thought would be, too many dynamics going into the overhead mic.  Do you feel this would be a better approach than say, use a kick and snare and add them to the pool?

Anyway, I will try out your recommendation.  Thanks!

Now this way does NOT put things in perfect time to a set ruler or timing, just to the Transients of the track/clip you are putting into the Clip Pool.  So try to pick the one that fits your needs the best

Jim

I just caught your on 8.5, 

Not sure if its the same as X1 and its been along time since 8.5 for me so my memory is off if it is the same.

Thanks JB101 - yes I do have a couple of videos on how to do this on youtube. You just need to extract the tempo map from one of the tracks. I find the kick one of the more reliable tracks to use.

The video for X1 on how to do that is HERE but I see you've bought the SWA Complete X2 video (thanks BTW) and the video on how to extract a tempo map is more detailed in that, including what causes and how to avoid the dreaded "tempo out of range" message.

That is found in this chapter with time in brackets

· Working with Audiosnap (Pt. 2) (7:00:31)
· Preparing clips to avoid "Tempo out of range"
· Editing the clip map
· Extracting tempo from a clip
· Extracting MIDI for drum replacement

HTH

Telstar,

Here is my question - what tempo do you capture the songs at when you first begin??? What is your tempo set at - I'm asking because I have a similar event I'm doing tomorrow.

So for clarity does it matter where or what your tempo is on when you first start to record???

Thanks fastbikerboy

Jim:
 
Typically, where a song has some pickup notes at the beginning, I generally make 1:01:000 on the percieved first beat of the measure those pickup notes are in.
 
Hope that answers your question,
Paul