the audio was also 192,000khz sample rate, forgot to mention that, adn here are the audio files, the original, the flac, and the decoded from flac. the archive is 7zip, Idk where to upload it so I'll just send it to depositfiles. http://depositfiles.com/files/90anghniw -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.xiph.org/pipermail/flac-dev/attachments/20130415/54572738/attachment.htm
On 15-04-13 17:54, Marcus Johnson wrote:> the audio was also 192,000khz sample rate, forgot to mention that, adn > here are the audio files, the original, the flac, and the decoded from > flac. > > the archive is 7zip, Idk where to upload it so I'll just send it to > depositfiles.As far as I can see, these audio files are lossless and the WAV files are identical when regarding only audio data. When I open both in Audacity and invert one of them, the channels cancel each other exactly, so the audio data must be the same. The checksums don't have to match because WAV is a complicated standard and FLAC might be writing the file a little different (see for example http://xiph.org/flac/faq.html#tools__two_bytes_short) FLAC compresses the audio data losslessly, not the WAV file itself, that's why simple MD5-summing doesn't cut it. The FLAC file is small and encoding is very fast because most of the file is silence, which is very easy to encode.
On 4/15/2013 6:03 PM, Martijn van Beurden wrote:> On 15-04-13 17:54, Marcus Johnson wrote: >> the audio was also 192,000khz sample rate, forgot to mention that, adn >> here are the audio files, the original, the flac, and the decoded from >> flac. >> >> the archive is 7zip, Idk where to upload it so I'll just send it to >> depositfiles. > As far as I can see, these audio files are lossless and the WAV files > are identical when regarding only audio data. When I open both in > Audacity and invert one of them, the channels cancel each other exactly, > so the audio data must be the same.Bit-comparing both WAV streams and the FLAC stream (I did it with foobar2000) does also show that the decoded data is the same. Christoph