The Basics - Audio
December 31, 2000
Audio - From Camera to FCP
By Ken Stone
This article is being displayed here until the 2-pop Library is restored at the 2-pop web site.
The Basics - Audio
December 31, 2000
Audio - From Camera to FCP
By Ken Stone
In FCP we deal mostly with three different audio sample rates: 32khz, 44.1khz, and 48khz. Most digital camcorders record audio at either 32khz, 48khz, or both. Audio CDs are 44.1khz.
Some DV decks/cameras refer to the audio rate as 12 bit and 16 bit, but FCP/QT refers to the audio rate as 32khz and 48khz. When the deck/camera says 12 or 16 bits on the display, 12 = 32khz and 16 = 48khz. It is very important you know the audio sample rate of the video that you plan to capture BEFORE you capture. If you don't know the audio sample rate, put your DV tape in the camera and press play. Look through the viewfinder it should show you the rate. If you are using a deck, the audio sample rate should be displayed in the deck's information area.
There are two different elements in FCP that need to be set for audio. Capture Prefs, which control the audio capture rate, and Sequence Settings, which control the playback/output audio sample rate from your Timeline.
Before capture, the audio sample rates must be set. The audio sample rate in your Capture Prefs MUST be the same as the audio sample rate at which your source tape was recorded. From the menu click > Edit > Prefs > Capture tab. Click on the AUDIO button to display the 'sound box'. At the top there is a 2 part pull down. Click on it and select SOURCE. Set it to 'DV audio'. In SAMPLE set the rate. You will also want to set 'speakers' to off. If you capture at the wrong rate you can lose audio sync, have audio artifacts, pops, and noise. Once captured at the wrong rate there is not much you can do. You will need to recapture your audio.
Before opening your project or starting a new Sequence you will need to enter an audio sample rate in to your Sequence Preset settings as well. From the menu , select > Edit > Preferences > Sequence Presets > Edit> General tab. Click on the audio button. Set 'Compressor' to NONE and set your audio rate.
The audio sample rate that is set in your Sequence setting will determine the playback/output rate from your TL. If your Sequence audio settings are 32khz then FCP will play the TL at 32khz, if the sequence settings are 48khz then FCP will play the TL at 48khz. Clips that have audio sample rates that do not match the Sequence settings will be re sampled on the fly for playback in real time.
I have mentioned that FCP will re sample audio on the fly. While normally this works it does give the processor an extra chore to do during playback or print to tape. If the load on the processor is too great you will get dropped frames during play back. Using Mixdown Audio, (Sequence > Mixdown Audio) is like rendering. The mixdown audio file is saved in your render folder and used during playback to avoid overloading the processor. In some Applications using 'Mixdown Audio' will replace all current tracks with just two audio tracks, this is NOT true in FCP. When using Mixdown Audio in FCP a rendered file is created and all original audio tracks left are untouched, they remain discrete clips.
We can place a clip with a different audio sample rate then the Sequence audio sample rate into the TL and FCP will re sample on the fly and play back the audio track in real time. While it would be best to work with only one audio sample rate this is not always possible. FCP does have the ability to handle different audio rates. But remember, the less re sampling the better.
AUDIO CDs are 44.1khz. You will need to convert the 44.1khz to either 32khz or 48khz before importing them into FCP. For now (FCP 1.2 and QT 4.1) this is a two step process. There are a number of ways to convert the audio sample rate. Please note that the following procedure for converting CD audio is for Mac OS 9. In OS X the process has become much simpler and can be found at Final Cut Pro in OS X, scroll to the bottom of the article.
Step 1. With FCP open, place the audio CD in your CD player. Click on FILE > IMPORT > 'file'. If 'File' is grayed out make your Browser active. Select and open the CD. Select the track that you want and click the CONVERT button. This will bring up a Save box. Next click on the OPTIONS button. You will notice that you do NOT have the ability to convert your audio tract to 48khz or 32khz. Leave the setting at 44.1khz, 16 bit stereo. Also notice that in this box you have the ability to set IN and out' points to import just the section audio that you want. This will produce smaller files on your HD. Always give yourself big 'handles' (a few seconds of extra footage at the beginning and end of a clip) if you are selecting a portion of a track. When done click OKAY. At the bottom of the Save box you will see SAVE CONVERTED FILE AS. Give your file a name. At the top of the box you will see "Audio CD #". Change this to the location that you want to save your file to and Save. At this point you have not yet converted your audio sample rate.
Step 2. Two options.
a) Open this file in QT Player. I keep an alias of QT Player on my desk top and simply drop the file onto the alias to open it. With your track open in Movie Player, choose EXPORT from the File menu. In the Export box at the bottom set "Export" to 'Sound to AIFF', click the OPTIONS button. set 'compressor' to NONE and set the audio rate to 32khz or 48khz. Click Okay and Save.OR
b) Select the audio file icon in the Browser, and with the Browser active, File > Export > QT. In the Export box at the bottom set "Format" to AIFF, click the OPTIONS button. set 'compressor' to NONE and set the audio rate to 32khz or 48khz. Click Okay and Save.
When done with the conversion you will need to bring the converted audio file back (import) into your Browser. You can now delete the first imported audio clip from your Browser as it points to the unconverted audio file. You may also want to trash the unconverted audio file as well to save HD space.
AUDIO LEVELS. There is a fundamental difference between analog and digital audio levels. We could say: "Analog 0 dB = good, digital 0 dB = bad." This would be for two reasons.
First. With analog audio some over modulation is ok. This is because over modulation of an analog signal is mostly made up of even order harmonics that unless they are very high, don't sound that bad. When a digital signal over modulates you are hearing a clipped square wave which sounds bad, like hearing fingernails on a blackboard.
Second. With digital, a lower recorded audio level of -12 dB or - 16 dB is okay because the noise floor in digital is barely perceptible whereas analog has a rather high noise floor.
Consumer and Prosumer cameras and decks do have audio input meters but it should be noted that these meters are not very accurate, are not calibrated and should only be used as a rough guide. It is possible to add calibrated analog VU meters to your set up for very accurate readings of audio levels. There is no way of changing the dB of the source tape when capturing DV through FireWire.
When recording digital or analog which will be digitized, keep your peak levels below 0 dB to prevent problems when working in FCP.
In FCP once a clip has been capture and is on the TL, double click on the clip to open it in the Viewer select 'audio tab'.
When the audio tab is open in the viewer there is 'dB' display box. This 'dB' display will always show "0". This is not the audio level that the clip was captured at but rather the "0" is simple a 'center' point. As you change the audio level up or down this "0" will change to -x dB or +x dB. The "0" is simple a reference, a starting point. When you have changed the dB setting of a clip and then re open it in the Viewer that change will be shown in the dB box. This facility can also be found in the TL when using rubber bands/overlays.
Once you have captured your audio there is no way from inside FCP to check the audio dB levels of your source material.
In the audio tab there are two adjustment settings. The first is 'Level' which raises or lowers the audio volume. The second is 'Spread' or 'Pan' which for stereo is used to swap left and right channels. For mono or only one audio track set this to '0' to play out of both speakers. These audio adjustments are 'keyframable'.
For additional information on Audio dB levels see Digital Audio in Decibels & Final Cut Pro.
Notes and Conclusions:
1) The higher the sample rate of audio the better the quality, finer grain to the audio. Always record at 48khz when possible.
2) For CAPTURE; the Capture prefs audio sample rate must match the audio sample rate of the source tape.
3) When using Firewire with FCP, plug your external speakers into your deck/camera. This is the proper way to monitor your audio while working or printing to tape. When using Firewire FCP cuts off the audio to your internal Mac speaker.
4) When printing to tape using Firewire the Sequence Settings audio sample rate is the rate that will be recorded to tape. The DV deck/camera is simply making a digital copy to DV tape.
5) For those of you who like to record to tape by hitting the record button on your camera/deck and then playing the TL, remember to use 'Mixdown Audio' first if you have clips with different audio sample rates for best audio quality. This applies to Print to Video and Edit to tape equally. They all benefit from mixdown audio, before a final output. In fact, Print to Video and ETT actually benefit more from Mixdown audio, since they both add an additional audio processing event.
6) When shooting, if you plan to eject your tape and then put it back in your camera for additional recording, record about 10 seconds of extra video before you eject the tape. When you put the tape back into the camera, rewind back to the end of your last shot. This will give you continuous Time Code and will prevent TC problems during Capturing.
7) During Capture make sure you are not capturing any blank video, such as at TC breaks or at the end of the tape where there is no TC. This blank really only effects audio when AutoSyncCompensator is on. Otherwise, blank video at the end of the tape has no deleterious effect on audio. If you have a tape that you need to capture from and there is a TC break you will have to treat this tape as though it were two tapes. In the Log and Capture window, start at the beginning of the tape and give it 'reel' name of "mytape part one". Now log up to but not past the TC break, in fact you will need to stop at least 8 seconds before the break. Now capture. When done manually move the tape past the TC break by 8 seconds. Open Log and Capture and call this 'reel' "mytape part two". Log and capture.
8) It is not advisable to NEST a Sequence who's audio sample rate is different from that of the Sequence being nested into.
9) The 'AutoSyncCompensator' found in the General tab of FCP's Preferences is to be used with the Canon XL-1, and some older GL-1 cameras. When capturing clips with 'ASC' turned on these clips are not effected during capture but are flagged as 'ASC' clips after capture, this flag is stored in the Project file. Each time these clips are played in FCP the flag is read by FCP and the audio is re sampled to 32k or 48k exactly. If you use 'Mixdown Audio' these clips they will be re sampled and stored as a rendered files eliminating the need to resample on the fly.
If you wish to remove (disable) the 'ASC' flag from captured clips (having it turned on by mistake for example) do the following: delete the 'ASC-on' clips from your browser, save your project, quit FCP (your 'ASC-on' clip would still reside in the UNDO queue stored in RAM and quitting FCP empties ram), relaunch FCP, turn off 'ASC', then re import or re link your clip.
10) If the volume level is too low in FCP and raising the dB to 10 dB or 12 dB is not enough then 'copy and paste' the audio track to the next audio track down. Now you will have two sets of audio tracks and FCP will play them both, boosting the volume. You may need to pull down the dB settings in the audio tab of the Viewer for these tracks if the volume is now too high.
11) Capturing audio from Ch 3 & 4. The camera/deck has to be in play/pause position for the QT settings to give you the input option of second two channels. Put your deck in pause; Menu > Edit > Prefs > Capture Tab > click on the Audio button > select 'Source' under 'Sample' you will then have a pull down menu choice of first two channels or second two channels.
12) If you are shooting DV you are dealing with a soundtrack area that is smaller than a human hair, Always run a mono mic source into both channels using a Y splitter or a professional audio mixer with stereo or dual mono output. By running your mono mic source to both channels, you have cheap insurance that you will have a backup, secondary channel of audio in case you have a tape or head problem with the first track.
13) Before you import MP3 audio files in FCP open them in QT. Export as AIFF at 32k or 48k, then import.
14) Audio Transitions. The '0db cross fade' filter works by dropping the volume at the edit point. This effect is used when the ambient or background sound is different for the two clips or if there is very little ambient sound. It is also useful when cutting dialog. The '+3db cross fade' maintains a constant overall volume crossing from one clip to the next. This cross fade is used when the background sound is the same and provides a smoother transition. For information on using the Audio Filters in FCP see A Beginner's Guide to Final Cut Pro's Audio Filters.
15) Always plan out your audio and know your audio sample rates before you start to work.
Enjoy,
--ken
