Low quality media perhaps?

[WhiteFelineWonder]

Lately, I've been thinking about how it's possible to make audio sound really low quality, like something you'd hear in Flash. Did a little bit of searching and in Audacity, it's to do with the exporting options. You'd set the bit rate mode to average and get the quality down to 8 or 16 kilobytes per second to get that really compressed sound to it. This in junction with me experimenting with some conversion options in VLC, and thinking about the typical file size I would get for taking a photo of a drawing I made made me curious.

I feel like there's still a place for utilizing low quality audio and video. Part of me just thinks that even with the luxury that is very high quality media that's at 60 frames a second and 1080p resolution and all that, maybe some of us just want to be a bit primitive or retro with the stuff that we film or make or sing to. Those old cameras that you'd find on tech from the 2000s or even 2010s? The DSis or Blackberries, even old physical handheld cameras that had those tiny screens? I'd say there's still a place for them. Granted, some old tech might be a bit hard to find and have it working with modern technology given they're from an era that's quite far back.

Maybe it's just me reflecting on how far we've come with technology, and getting nostalgic for how audio can sound in SWF files and even in video games, but I'm curious, let me know what you guys think. Maybe there can be a sort of casual renaissance with low quality stuff. Maybe low quality media could spark some new ideas.

[BlazingCobaltX]

I had this thought for a while too, when YouTube was borking out on my PC and wouldn't load unless I played things at 240p. Though it was annoying at the beginning, gradually I stopped minding the lower quality video. Eventually I even felt that most videos didn't need to be above 720p.

...Then YouTube started being normal again and I went back to 1080p videos. Unfortunately the eye just gets used very quickly to higher fidelity images and video, but this episode did make me question whether we need 4K or lossless FLAC for every single thing. For me, this discussion also has to do with file storage and preventing a file size creep because everyone wants to output their work in the highest possible quality.

Recently, this Dutch artist recorded a music video with the 3DS, along with some clever editing. I think it looks neat.

[WhiteFelineWonder]

Not only does this sound awesome, but the way the footage looks would be like something I would find in around 2012, maybe 2013 or maybe even before 2012. Those early 2010s vibes are quite strong with this and I think that's pretty neat! :D

The fact this was done with a 3DS makes this seem quite humble or modest in a positive way I will say.

[xwindows]

I feel like there's still a place for utilizing low quality audio and video.

I agree, especially the video part-- which I have already written about in length here at MelonLand; so I wouldn't repeat that part here.

Did a little bit of searching and in Audacity, it's to do with the exporting options. You'd set the bit rate mode to average and get the quality down to 8 or 16 kilobytes per second to get that really compressed sound to it.

There are actually more to this other than just bitrate of lossy codec output...

The most basic variables, which exist in every digital audio types are:

• Sampling rate: 44100 Hz (or 48000 Hz, depending on media format) are "ideal" quality (1), but things hasn't always been that way.
  
  Copper-wired landline telephone network runs at 8000 Hz; if you're old enough to have used them in actual form (2), you would remember its stripped-down sound where anything above speech frequency were nerfed (sometimes including the wisp of "-esss" in your speech too); the earliest PC sound card also ran at the same frequency. Middle-ground sampling rate like 22050 Hz could be heard in many CD-based videogame audio in late-90s to early 2000s; treble would sound a bit muted in those, but they were otherwise serviceable and people would not chalk it as "low quality" at first glance.
  
• Channels count: Stereo audio is at least minimum these days; but if you are old enough, you might remember the time when TV broadcasts were mono by default, and you might remember AM broadcast radios as well (which were and still are often mono). Of course, phone calls were (and still are) mono by default.
  
  In my experience, switching to mono would not affect the apparent quality you'd experience that much, as long as you were listening to it on-speaker. When using headphones/earbuds, stereo and mono can make worlds of differences: like feeling as if you were sitting in a wide concert hall vs. lying confined in a coffin while listening to the same tune.
  

^ Both could be set on Audacity's individual track (there is also a project-wide as well as global preference for sampling rate); they aren't set on exporting process.

In uncompressed audio (like WAV or AU file), there is another variable:

• Bit depth: 16-bit is gold standard. (1) The most common ones are 16-bit and 8-bit; but 16-bit wasn't always available for consumer. However, the sound of reduced bit depth would not be very apparent unless you know what to listen for (basically, it disappears some quieter components of your music, and introduce subtle "hiss" (3)).
  
  Audio CD uses 16-bit audio, and it has been the first mass-market format of 16-bit digital audio. Cassette tapes, while analog, have best-case hiss noise floor equivalent to 6-bit digital audio. The most-advanced copper-based landline telephone exchanges use 8-bit-with-cheat (4) bit depth for the audio. Many CD-based videogames in late-1990s to early-2000s used 8-bit uncompressed audio in some or all parts of game audio.
  

^ Audacity doesn't support 8-bit uncompressed audio for some reason, so you can't use Audacity alone in experimenting in this one. (5)

Back in Microsoft PowerPoint 2002-2003 days, I used embed uncompressed .wav music in slideshow by converting it (using Microsoft Sound Recorder) to 8-bit mono first-- quartering the size, to avoid making the slide file too huge.

For lossy-compressed audio (like ADPCM (6), MP3, Ogg/Vorbis, MP2, AAC, Opus), there are other different variables:

• Audio bitrate: low bitrate means more codec-introduced compression artifacts; but note that this also play hand-in-and with the sampling rate you use too. In MP3, encoding music with 44100 Hz sampling rate with 128 kbits/sec bit rate is usually considered a minimum decency; but if you like being sacrilegious, you might get away doing so at 96 kbits/sec if you use good encoder and tune it carefully.
  
  If you like cringe however, force your MP3 encoder to encode your music at 48 kbits/sec (or 32 kbits/sec) bit rate while maintaining 44100 sampling rate...
  
• Audio codec: at high bitrate, all codecs sound similar, but at low bitrates, audio encoded by each different lossy codec sound different.
  
  You're probably be most familiar with MP3 codec; but it's useful to know that other exist too, like the royalty-free Ogg/Vorbis-- which have great support for browsers and hardware player for a very long time now, and is said to have better low-bitrate sound than MP3.
  

and getting nostalgic for how audio can sound in SWF files

That is a mix of reduced sampling rate, mono audio, and MP3 lossy encoding in a relatively-low bit rate: mono audio with 22050 Hz (sometimes 11025 Hz) sampling-rate encoded as MP3 at 32-64 kbits/sec should bring your tune back to that time.


(1) Beyond 44100 Hz 16-bit no one actually hear any improvement in proper benchmark settings. Some press dubbed 96000/192000 Hz 24-bit consumer music releases as "The emperor's new sampling rate".

(2) As opposed to a form emulated from VoIP/fiber-phone using local ATA (Analog Telephone Adapter) box, which aren't confined to this limit anymore; but introduce its own frequency/phase-butchering, and latency/jitter issue which bork many attempts to run modem connection through.

(3) See this video at 00:10:32.

(4) Well, I'm not going to explain here how u-Law and A-Law encoding works; if you wanna know, you know where to read them. While using 8-bit data, these cheats allow their sound to have apparent bit depth of 14-bit in many kind of uses.

(5) Dunno on your version of Audacity, but on mine (2.0.1), even if I went full-tinkerer mode and set "Custom FFmpeg Export" type on exporting, I would still not be allowed to export WAV file with 8-bit PCM audio despite FFmpeg itself is fully capable of doing so; because of some... unfortunate assumptions on Audacity's part when invoking FFmpeg library.

(6) ADPCM encodings, usually IMA-ADPCM (but there is Microsoft ADPCM too) are the odd ducks; technically, this is considered to be in the area of lossy compressed audio codec, but it is 4-bit PCM audio with simple time-based cheats (as opposed to sample-value-based cheats like ones described in (4)) and it has no bitrate adjustment. This is used in audio recording function of many feature phones; can sound surprisingly decent-- thus okay for many voice recording uses.