Colophon

Here are my production notes and references for stuff I found useful while creating the podcast and website for Player/Missile.

Theme Song

Software

Podcast Production

Audacity Workflow

  • Record @ 44100Hz, mono, 32-bit float

    • separate tracks for each segment

    • have a piece of paper, write down edit points as they happen!

    • don’t delete the bad section at the edit point, zero it out so that edit points written down for later times are still valid

  • Amplify -5db (make slightly quieter to remove more clicks but still keep softer vocalizations)

  • Click Removal (New!)

  • Truncate Silence

  • Noise Reduction (make sure to save a track with noise!)

  • Compress

  • Post

    • add Odin’s Atari

    • record teaser

    • add bumpers for each segment

    • add morse code for interview subjects

    • re-align tracks

    • downmix tracks to 64kbit mono mp3

Audacity Reference

  • ARGH! Note to self: save the project and operate on a new copy before doing any of these operations! Undo works to restore bad operations, but I don’t think things can be undone across saves.

  • Noise Reduction:

    • Current:

      • Noise reduction: 20dB

      • Sensitivity: 1.5

      • Frequency smoothing: 0

  • Noise Removal (old):

    • Maybe try:

      • Noise reduction: 25dB

      • Sensitivity: 10

      • Frequency smoothing: 500 Hz

      • Attack/decay time: 0.5 s

    • Current:

      • Noise reduction: 25dB

      • Sensitivity: .12

      • Frequency smoothing: 500 Hz

      • Attack/decay time: 0.25 s

    • Previous:

      • Noise reduction: 18dB

      • Sensitivity: 1.56

    • Previously previous:

      • Noise reduction: 12dB

      • Sensitivity: 1.56

  • Compressor:

    • Current:

      • Threshhold: -20dB

      • Noise Floor: -30db

      • Ratio: 4:1

      • Compress based on Peaks

    • Previous:

      • Threshhold: -26dB

      • Noise Floor: -35db

      • Ratio: 8:1

      • Compress based on Peaks

    • Previously previous:

      • Threshhold: -30dB

      • Noise Floor: -35db

      • Ratio: 8:1

      • Compress based on Peaks

  • Click removal

    • Current:

      • Threshhold: 157

      • Max spike width: 31

  • Align tracks

    • Edit -> Select -> All

    • Tracks -> Align Tracks -> Align End to End

    • Manually (from http://forum.audacityteam.org/viewtopic.php?f=16&t=64180)

      • Double click on the top track to select it.

      • K key (moves cursor to end of the track)

      • ENTER (deselects the track)

      • Down cursor key (moves focus to second track)

      • ENTER key (Selects the second track)

      • Alt + T (Tracks menu)

      • A (Align Tracks)

      • C (Align with Cursor)

  • Truncate Silence

    • Current

      • Level: -25db

      • Duration: .15s

      • Truncate Detected Silence

      • Truncate to: .4s

    • Previous

      • Min silence duration: 1200 ms

      • Max silence duration: 50000 ms

      • Silence compression: 50:1

      • Threshold for silence: -20 dB

  • GLAME Highpass Filter (not Butterworth)

    • for use with the headset mic that doesn’t center on zero

    • freq: 100Hz

  • Mix and render

    • mix and render to new track: Ctrl + Shift + M

  • blooper tone

    • 1000 Hz, .4 seconds

  • quindar tones

  • SAM TTS (and phonetic reference)

    • atari800 -basic sam.atr

    • RUN “D:SAYIT.BAS”

    • S

    • KRAYKIY6

Interviews

  • Recording using line in

    • Cell phone headphone out to line in

    • ATI SB mixer: Line 0, input volume about 50%

    • Audacity: input device: HDA ATI SB: Line 0, input volume, 90%

      • using default input device, sysdefault: Line 0, results in 10 Hz clicking

    • after suspend/resume, the line input sometimes doesn’t work. Using arecord -vv /dev/null seems to force it to work again (ref)

  • Recording with Skype

    • use a call recorder

    • backup:

      • headphones out to line in on 2nd machine; audacity recording

      • digital recorder to record my side of the conversation

    • on linux

      • Using https://launchpad.net/audio-recorder it is possible to record Skype. It doesn’t record in separate tracks by default, but seems like it can be mostly made to work by using the audio settings dialog when the recorder is active. When it is active, the Recording tab is populated with sliders, so you can adjust the mic to the L track and the playback to the R track. However, there does seem to be some bleedover of the audio to the mic.

  • Recording from Google Hangouts

atari800 Emulator

  • Alt+W starts/stops recording, saved in current directory, named atari[0-9]{3}.wav

  • F10 captures screenshot in current directory, named atari[0-9]{3}.png

Microphone

  • Samson C01U microphone

    • Wasn’t recognized with audacity in linux until after I ran audacity in WinXP in VirtualBox, but after that it worked fine in linux

Website

CSS

This is the best I can do so far:

td.rotate {
  /* Something you can count on */
  height: 130px;
  white-space: nowrap;
  text-align: center;
  vertical-align: middle;
}

td.rotate > div {
  transform:
    /* Magic Numbers */
/*    translate(25px, 51px)*/
    translate(10px, 55px)
    /* 45 is really 360 - 45 */
    rotate(315deg);
  width: 0px;
}
td.rotate > div > span {
  border-bottom: 1px solid;
  padding: 5px 10px;
}

using HTML like: <td class=”rotate”><div><span>1979</span></div></td>

Magazine Listings

Using archive.org’s advanced search, choose json format (with indented response), and from the fields list select date, identifier and title. Sort date ascending, choose a really big number for number of results, and save the resulting json file.

I can’t find the advanced search on the new interface of the Internet Archive, but you can still get to it directly at https://archive.org/advancedsearch.php

Multimedia

Audio

  • Convert .wav to .m4a for ringtones: faac spelunker.wav -o spelunker.m4a

Video

  • Extract audio from video files: avconv -i source.flv -vn -c:a copy dest.mp3

  • Extract audio from video files and re-encode: avconv -i source.mkv -vn -acodec flac dest.flac

  • Extract 8 second subset of video file starting at 805 seconds from the start: avconv -ss 805 -t 8 -i Fletch.mkv -vn -acodec flac charcoal.flac

Images

  • scale images using a script: convert -resize "25%" 20150218_1*jpg which will create multiple output images using suffixes based on the last filename in the input list

Speech Synthesis (Robot Overlord voice)

Morse Code

PDF Viewing/Editing

  • Turn off Adobe Acrobat’s PDF/A viewing to enable File -> Save As -> Optimize PDF

    • Preferences -> Documents: View Documents in PDF/A Mode: Never