Advice on mastering stealthed Jazz show needed!

[Nick in Edinboro]

I stealthed a Wynton Marsalis concert at my University (thanks to Heath for letting me borrow mics).

The situation is that I was center, about 2/3's back in the venue.  The band, while playing mic'ed instruments, generally was playing unamplified.  As such the levels rarely even breach -20db in CEP2 during the music but are nice and HOT during the applause.

When I bump the volume up using the amplify filter (can't use normalize obviously) the music sounds fantastic, little hiss from being amplified or compression but again the clapping is obviously smashed well above the -0db mark.

Since people clap after solo's, after Wynton says something, and not only after songs it would be impossible to isolate the high level applause from the low level music.

Anyone have any ideas of what to attempt?  It's a shame as the actual music is breathtaking but as it is now would be unlistenable and I can think of a decent way to master this into something better.   

Edit:  Oops, meant to edit my own post, but did this one by accident.  Just leaving a note so you don't think I'm messing with your post intentionally when you see the "Last edited by Brian Skalinder...".

[blu666z]

Sounds like a job for the Waves Plugins.  L2 Limiter.

-Kevin

[Nick in Edinboro]

Ahh, I have waves (but little clue as to how to use it).  L2 Limiter eh?  I will look into that plugin, thank you sir!  Exactly the kinda response I was hoping for.  Big +T for the fast reply.

[zowie]

Apply a high ratio compressor/limiter with threshold set at the peak of the music level, fairly fast attack and medium decay.
Send me a PM if you want me to play around with it for you.

[Brian Skalinder]

Edit: Uhmmm...what the other guys said.  Too slow on the draw composing my reply! 

Compression is your friend.  It should allow you to get the applause no louder than the loudest music section which should make for tolerable listening.

[1] Find the loudest music peak
[2] Amplify | Dynamics Processing
[3] Set your threshold to the loudest music peak
[4] Set your compression ratio (in this case, compress the snot out of everything with a high ration, > 20:1 - I find this accomplishes much the same effect but sounds better than hard limiting, which is another function in CEP that IMO sounds like shit)
[5] Amplify as normal to get levels up.

So, for example, if the music peaks at -20dB, you'd go into Dynamics Processing and select Compress from the dropdown, set your ratio to 10:1, and set the threshold at -20dB.  In the next row, select the Flat dropdown.  Done.  It'll read something like Compress 20:1 above -20dB, Flat below -20dB.  Then amplify the whole thing.

Play with different ratios / attack / decay to find what sounds best to your ears. 

[Nick in Edinboro]

You guys really do rock.  I've had this sitting on my hard drive for weeks now untouched as I envisioned it being impossible to correct given my knowledge of my software.

I owe it to myself to educate/understand these basic audio tools more (compression, limiting, and the like).  They probably could've come in handy in the past and definately will in the future.  I've only learned "normalizing" to this day and still rarely use it as anything I generally tape has hot levels already.  This situation was a rare circumstance!

I'll play around with everyones advice here and see what I think sounds best.  Zowie, if I can't seem to get desired results given my attempt I might just send you some flacs!  If you'd like it just to try anyways or are interested in a copy after I get it listenable just let me know!

Thanks again to everyone, yer so smrt!
Nick (thought it couldn't be done)

[TenoRichards]

Great advice from you all. I have basically the same problem on a master of opera. Venue is unamplified and applause just sends tthe hing over the edge. Rarely did I clip during the applause, but the contrast in dynamics is too vast. So, anyone walk me thru this in SoundForge 7? I'd sure appreciate it.

In setting levels in acoustical environments, what is the best tact to take? I KNOW the applause is gonna be loud. What do I DO about it? Do I set my gain to the loudest MUSICAL portions and not worry about the clipping that will occur during applause, or just keep the levels down and boost it in post production?

The original poster said that "I'd obviously not normalize." Esplain Lucy. Why is this obvious?

TIA
Andrew

[pfife]

Great advice from you all. I have basically the same problem on a master of opera. Venue is unamplified and applause just sends tthe hing over the edge. Rarely did I clip during the applause, but the contrast in dynamics is too vast. So, anyone walk me thru this in SoundForge 7? I'd sure appreciate it.

Do you have the waves plugs?  I might be wrong, cause I only have SF6.0, but I don't think that has a compressor.  There are other options, but waves is the best.  If you don't, do yourself a favor and grab them!

In setting levels in acoustical environments, what is the best tact to take? I KNOW the applause is gonna be loud. What do I DO about it? Do I set my gain to the loudest MUSICAL portions and not worry about the clipping that will occur during applause, or just keep the levels down and boost it in post production?

This is exactly what I do, but I usually leave myself some headroom, because what I have noticed is that acoustic music also seems to be the most dynamic music to record... so, right about the time I get my levels set, and they are hot, the artist will raise the volume of their voice or something, and peak the shit out of my levels.

Its really a decision between the following:
1) running "not so hot" - which has the negative of not using the full dynamic range of your rig  -or-
2) running hot, and peaking during loud parts - which to me can come close to ruining a recording.  Most of the time the louder parts are loud for a reason, and you don't want them to be clipping all over!

To me the choice is clear...

[hexyjones]

Just do the opposite of what you started with...

Reduce the applause...(a few ways to do that...)

then normalize...

Again gang...dont sit too deep in the center of venues and gigs like this - the applause will be overpowering...

Opt for a closer side seat and you will have a more even recording...I know it doesnt seem right...

Someday I'll do a sampler of recordings I made at PSU in Eisenhower Aud - Metheny - Al DiMeola - Betty Carter...all different seats...

[TenoRichards]

thank guys. I'll keep working on it. appreciate the help. The help given in SF7 is minimal in how to acquire plug-ins. It has something called Wave Hammer, which seems to be a volume maximizer and a compressor. I'll continue to research about this.

While I'm thinking of it: Anyone want to suggest publications on doing Mixing. (dying for the compilation SirChek is starting) I really need to start from the ground up.

your favourite opera singing n00b
A
p.s. WHY should I care if the recording clips out during applause? Aside from the whole ambiance thing, the thing I'm wanting is the music, not applause... Why should I not let it clip then just edit it out, or compress the clipped applause, which happens in opera every 15-20 mins or so?)

[BC]

In soundforge 6.0 I have typically used Effects>Graphic Dynamics (=compression) in this situation.

When editing material like this I just peak out the music, which results in the applause being peaked out as well as squashed (and sounding kind of funny) but I don't mind that so much since it is just clapping and crowd noise.

Lately I have been using the waves L2 ultramaximizer plug-in in SF 6.0, just set the threshold about to where the music peaks, it will do the rest for you. I think this has a nice algorithm, the volume increase sounds very transparent and natural to me. I think I typically set the threshold about 2dB above the music peak which results in the peak ending up at -1 to -2dB or so. You can play around with the threshold to see where you like it best.

Good luck!
Ben

[pfife]

In soundforge 6.0 I have typically used Effects>Graphic Dynamics (=compression) in this situation.

When editing material like this I just peak out the music, which results in the applause being peaked out as well as squashed (and sounding kind of funny) but I don't mind that so much since it is just clapping and crowd noise.

Lately I have been using the waves L2 ultramaximizer plug-in in SF 6.0, just set the threshold about to where the music peaks, it will do the rest for you. I think this has a nice algorithm, the volume increase sounds very transparent and natural to me. I think I typically set the threshold about 2dB above the music peak which results in the peak ending up at -1 to -2dB or so. You can play around with the threshold to see where you like it best.

Good luck!
Ben

Yup... I think the Ultramaximizer rules too.

[TenoRichards]

Yup... I think the Ultramaximizer rules too.

so, how much is the L2 (or now, the L3) gonna run me?

[pfife]

They are quite expensive... its the Waves bundle...  I am not really sure how much they are though.