You can add location information to your Tweets, such as your city or precise location, from the web and via third-party applications. You always have the option to delete your Tweet location history. Learn more
Looks like most of the issue here was due to the original being very hot (in the stream it would've been lower). Redoing the test with volume reduction...
Here we go. libfdk_aac is arguably better than lame (it runs on 16-bit int math so it can't handle overshoots well at all, after lowering the gain it's fine), ffmpeg is *terrible* in comparison to both. https://mrcn.st/t/lame_vs_fdk_vs_ffmpeg_delta_lower.wav …
Substracting resulting encode of two psycho acoustic codecs is not a proper way to compare audio signal, but yes, the main difference in the high frequency, and I agree with you, I can hear high-pitched 'hiss' too.
Check how @SoundExpertOrg does testing:
http://soundexpert.org/articles/-/blogs/visualization-of-distortion …
I know, but in this case it helps make the difference which I can already hear in an A/B test anyway completely obvious. Here's a side by side comparison without subtracting. The bonus two double encodes at the end are what I was comparing in the stream. https://mrcn.st/t/orig_lame_fdk_ffmpeg_ffmpeg2fdk_ffmpeg2ffmpeg.wav …
Can anyone remind me why we collectively gave up on ogg vorbis? It's because of high bandwidth availability favouring lossless like flac for audiophiles and the huge availability of mp3 hw decoders for avg users (apple/ms specific formats aside, still shivering thinking at wma)?
Because Opus is better. But it's not well supported yet, so we still need AAC for compatibility.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.
