Taperssection.com
Gear / Technical Help => Post-Processing, Computer / Streaming / Internet Devices & Related Activity => Topic started by: Kush on April 09, 2009, 11:24:05 PM
-
I finally ran 24 bit 48 kHz and I need some help with my post-processing work-flow. Here's what I've done so far for the first set:
Open the .wav file in CD Wave Version 1.94.7 beta
Trim excess of the beginning and end of set by breaking it up into 3 tracks. Track 1 is the excess before the band takes the stage. Track 2 is the entire first set. Track 3 is the excess after the set is completed. Save tracks in Alternate 24 bit version.
Open the entire first set in Audacity Version 1.2.6
Import audio as 48 kHz and 32 bit realm. Add 3 second fade-in to the beginning of the set and add 3 second fade-out to the end of the first set. Split into stereo track. Amplify the left channel to peak at -0.02 dB. Amplify the right channel to peak at -0.2 dB. Then I export as Wav Uncompressed Wav, Signed 24 bit PCM.
When do I dither and resample and what should be done first? After tracking in CD Wav? Should all this be done in one step in Audacity including what's mentioned above?
-
Here is my 24 bit workflow:
1. Edit (fades, Eq, etc) in your favorite wav editor
2. Track in CDwav
3. Convert 24 bit files to 16/44.1 with r8brain (using batch convert).
4. Convert 24 and 16bit filesets to FLAC with Traders Little Helper. Also create FFP files with TLH.
5. Create Torrert files with uTorrent Seed both 24 and 16bit FLAC filesets.
-
add gain>resample>dither>track>flac
-
Audacity will resample and dither during the export process. Check the stickied Readme1st thread for an Audacity workflow and configuration thread.
-
I'm not saying this way is the best way but it works for me.
I just open the file in Audacity,
trim off the junk at the beginning and end (I always save the original files unaltered),
add a fade in and fade out to the beginning and end if it needs them,
start a label track and add a label the beginning of the first cut,
add a label at reasonable points between songs (go in order from beginning to end, if you go in any other order and then remove a section the labels will be in the wrong places, I cut down long silences between songs if they loose continuity),
then either normalize or amplify the whole file as needed (rarely any other modifications like EQ),
then once all the labels are set choose "export multiple" and go away while Audacity does its thing.
You may want to get fancy and add crossfades etc, as needed in places. The presets for Audacity will save the individual tracks as 44.1/16 unless you tell it otherwise. There is a workflow for Audacity somewhere on this site that gets into the details of the settings but this will work straight out of the box.
-
My 24-bit workflow:
- Load 24-bit WAV into Audacity (I'm using 1.3.7).
- Insure that Audacity is measuring time in CD frames (75 FPS - right-click any one of the Selection Start, End/Length, or Audio Position counters to set this).
- Make sure that the Snap To box is checked. Now, selections will be made on frame boundaries, which is important if you're going to downsample to 44.1 kHz for CD-DA. Since both 48000 and 44100 are evenly divisible by 75, this will prevent SBEs.
- If I'm using my SP-TFB-2 mics, split the stereo track into separate left and right tracks, and boost the left by +3 dB. Then, merge it back into a stereo track. (Another mismatch scenario involves my AKGs going through my 0404 used a preamp, where I have independent level controls. I'll first use the fader to figure out how much boost to use in this case, then cancel the fader and apply the boost.)
- Cut out excess before the performance starts - leave about 10 seconds as a "leader."
- Apply fades, cuts, etc. Drop labels where you want track boundaries. NOTE: At the end of the last track, UNCHECK the Snap To box, otherwise you won't be able to cut out a last little "click" at the end of the recording.
- If I'm using the LS-10 internal mics, apply EQ to reverse the baked-in bass rolloff. Yes, this works very well.
- If necessary, apply a hard limiter to applause. You need to download and install the LADSPA plugin package to get the hard limiter.
- Normalize.
- Use "Export Multiple" to FLAC (I save in 24-bit FLAC, always). Audacity will name the files using the labels that you added. This is why I leave a 10-second "leader" at the beginning of the recording.
- Load the FLACs into Foobar2000 for tagging and ReplayGain scanning.
- If I'm going to burn a CD, I bring up a command line on my Linux server, and use a Makefile which calls SOX to convert the sampling rate, dither to 16 bits, and output 16-bit WAVs. This makes it easy to run four simultaneous jobs to take advantage of the server's quad-core CPU.
-
I use Gold Wave:
Combine Files from Edirol to a Single Wave File
Trim Beginning and End
Add Cue Points and Name Tracks Per Personal Protocol/Mask
Locate Blips, Handclaps and Drunks and Process Them Down, Either by Zapping or Heavy Compression.
Do the Same for Cymbals, Piano, High Vocals and Compress Them Now to Slightly Above Music Level
Add any EQ Boost or Roll Off
Volume Boost to Slightly Below 0db
Final Review for Other Tweeks--Zaps and Compression Mainly.
Convert to 16b
Split and Save Tracks.
I Save Raw Feed for about a Month, and after listening to concert a couple times, I delete it and use the modified files.