Subject:Proper order in which to edit

Posted by: Hummingbird

Date:2/11/2003 1:38:03 PM

I record two worship services every Sunday morning on a minidisk recorder and transfer
the musical selections to Sound Forge. There I remove the DC offset and trim the
beginnings and endings to about one second each. I save each selection as a wav file.
When a piece comes off especially well, I edit it further The object is to eventually
accumulate about 60 minutes of really good material, pay the royalties, and burn CD's to
be: 1.) given to prospective new members 2.) given to shut-ins, and 3.) sold roughly at
cost to pay for those we give away.

For the pieces I want to edit, I need to do the following things and wonder what the best
order to do these things might be?

1. Fade in and out
2. Normalize (typically to -.50db)
3. Sometimes edit EQ (passing car sub-woofers occasionally creep in)
4. Apply an impulse file using Acoustic Mirror (our space is pretty 'dead.')

Additionally, some pieces are especially soft and when brought up with normalization the
background 'hiss' comes up as well. I can remove it with noise reduction, but am
wondering if I should do so before normalizing, or after? I don't really hear it until after
normalization is applied.

I would be grateful for suggestions. Thanks - Hummingbird

Subject:RE: Proper order in which to edit

Reply by: Chienworks

Date:2/11/2003 3:53:25 PM

The order i'd use is:

1 - Noise Reduction
2 - Impulse / Reverb
3 - EQ
4 - Compress
5 - Normalize
6 - Fade

If the initial signal is very quiet you might want to normalize to a level like -6dB first just so that there's sufficient data to hear and to work with. On the other hand, if it's so quiet that it's hard to hear or work with, you'd probably be better off bringing it in to SoundForge again at a higher level to begin with.

Noise reduction works best before you've altered the signal so that it can detect the original noise. Impulse and eq can both change the noise and make it harder for Noise Reduction to do it's job. All of these can alter the overal volume level, so it's best to normalize after these steps to make sure you don't clip or have to normalize a second time. Less processing is better!

Subject:RE: Proper order in which to edit

Reply by: Hummingbird

Date:2/11/2003 4:51:00 PM

Chienworks - Thanks for your reply. Since I save the files prior to editing, it will be easy enough to do it your way and see how it works. The only question I have regards using compression, which is a process I didn't mention on my list. I've always thought compression was best-suited for 'pop' music that needed to be pretty much at one dynamic, or needed to accentuate a bass line or kick-drum. Everything I deal with is 'classically' oriented with extreme dynamic changes being the norm. How will compression affect this style of music? Since I haven't used it, I would be interested in its positive contribution to the finished product.

Thanks - Hummingbird

Subject:RE: Proper order in which to edit

Reply by: Chienworks

Date:2/11/2003 6:39:23 PM

Hummingbird, almost any recording can benefit from compression. A live performance may have a vast dynamic range. In my live theatre work i've measured at least 40dB difference between spoken lines and full cast musical numbers. This is acceptable in live situations because the human ear responds to a wide dynamic range and adjusts accordingly, and the brain also uses visual cues to help handle the volume range. Audio recordings suffer from two problems. Early analog equipment couldn't handle the wide dynamic range; loud sounds could distort and quiet sounds could get lost in the medium's noise level. Digital recordings aren't quite so limited, but the equipment still responds less faithfully than the ear does. The other problem is that when someone listens to a recording they are separated from the original performance and unprepared for large volume variations. It can be uncomfortable and annoying to have the same dynamic range sitting in your living room as in the theatre.

Classical music definately benefits from a wider dynamic range than pop, but compression at a lower level can help it sound more pleasing. The single biggest thing i use compression for in live recording is to automatically remove the stray huge peaks that would otherwise prevent normalization from bringing the rest of the signal up to a more usable level. The compression i apply to a classical or theatrical recording changes very little of the recording; it only eliminates the extremes.

Subject:RE: Proper order in which to edit

Reply by: musicvid10

Date:2/12/2003 12:45:24 AM

Just wanted to add a starting point to Chien's excellent (as always) advice. I like to start conservatively and back off when things start to sound unnatural or the peaks get muddy. A good starting point is the "Medium Compression" preset in the Wave Hammer plugin. You can then adjust the compression to suit your needs. Be careful not to set the threshold too low or you will amplify the noise floor.

When you're done, run the results by another musician. Two sets of ears are always better than one!

Subject:RE: Proper order in which to edit

Reply by: Hummingbird

Date:2/12/2003 8:04:49 AM

Chienworks - Thanks for the explanation of compression. Since I've not used it before, it'll give me a new 'toy' to play with. I especially appreciate you providing the theater illustration which helps put your information in proper perspective.

Musicvid - I also appreciate your advice and certainly agree that using other ears (provided they are qualified) is valid advice. I apply the same when composing. The creative process is often so personal that it's hard to step away and be objective.

Much obliged, gentlemen. - Hummingbird

Subject:RE: Proper order in which to edit

Reply by: drbam

Date:2/12/2003 8:51:46 AM

<< I also appreciate your advice and certainly agree that using other ears (provided they are qualified) is valid advice.>>

In my experience, I find that the more "unqualified" ears regularly offer extremely valuable advice, suggestions, or comments. Of course feedback from other musicians and eventually a good mastering engineer are indispensable. ;-)

drbam

Subject:RE: Proper order in which to edit

Reply by: musicvid10

Date:2/12/2003 11:53:24 AM

*I find that the more "unqualified" ears regularly offer extremely valuable advice, suggestions, or comments.*

Yeah, like the time I was roundly chastised for playing the "wrong" notes ("C" natural) in the last section of the Pachelbel Canon. :?)

hehehe

Subject:RE: Proper order in which to edit

Reply by: drbam

Date:2/12/2003 5:49:34 PM

I suppose I'd should have qualified my comment by noting that one does need the ability to determine which "unqualified" comments might be indeed useful! ;-D

drbam

Subject:RE: Proper order in which to edit

Reply by: Kowlinga

Date:2/13/2003 7:33:52 PM

Of course that C natural works just fine because it gives you the 7th which leads so beautifully to the key of G. Then you have your audience amazed at Pachelbel's Canon in G.

Subject:RE: Proper order in which to edit

Reply by: pubcrawl69

Date:3/28/2004 5:36:41 AM

i hope you won't consider this "spam"; -----

As a "newbie", I just wanna' say, "WOW" ---- you folks share an AMAZING amount of

priceless information! Thank You! I feel like I've died and gone to Soundforge Heaven!

Subject:RE: Proper order in which to edit

Reply by: Rednroll

Date:3/28/2004 11:02:00 AM

I agree with about 99.9% of Chienworks advice. And usually, I recommend normalization as one of the last steps of processing, but in your case, it might be a good idea to normalize before the noise reduction. I've ran into this type of work quite a few times, and what I found is if I use the noise reduction first, then normalize, when I normalize it brings up a lot of the noise floor that I didn't hear before. So if you normalize first, this brings up the level of the noise floor, giving it more resolution to take a noise print within the noise reduction plugin. It also, allows you to more accurately adjust the "reduce noise amount", more accurately because you're listening too, what will be closer to the final overall level. Otherwise, there may be a lot of noise hiding in that -90dB to -60dB level area, which your ears have a hard time hearing and by normalizing first it may put that once unheard noise into the decibel level that can be heard. Rather than normalizing after noise reduction, and having the level raise the noise you previously didn't hear, into a level you now can.....Then you're saying to yourself, do I use noise reduction again? I thought I already removed the noise?

My order would go more like this:

1-Normalize
2-Noise Reduction
3-EQ
4-Reverb
5-Compress
6-Fade

Subject:RE: Proper order in which to edit

Reply by: rraud

Date:3/29/2004 11:01:47 AM

Just to clarify the above advice regarding the Noise Reduction process. NR will not remove much, if any unwanted noise from a passing car. It does however work wonders on constant "din" type noise such as HVAC and distant traffic.
Also, if your transfering from MD in the analog domain. Set your record levels as high as possible while maintaining enough headroom to avoid any clipping. This appies to the intial MD recording as well.