For example, I have two video files of the same movie:
m2ts with LPCM 2.0 audio
mkv with FLAC 2.0 audio
If I convert the first one to ac3 2.0 640 and the second one to ac3 2.0 640 ¿Do I have the same ac3 audio file? ¿exactly the same quality? ¿100%?
+ Reply to Thread
Results 1 to 18 of 18
FLAC is losslessly compressed digital audio. If you had a FLAC audio file that was created from an LPCM file and decompressed the FLAC file to LPCM, then the LPCM audio in the new file would be identical to that in the original LPCM audio file from which it was created
However, unless you know that the exact same source was used for both your LPCM file and FLAC file, and know everything that was done to them before you obtained them, you should not assume that the audio on the FLAC file will be identical to that on the LPCM file once the FLAC file is decompressed.Ignore list: hello_hello, tried, TechLord
FLAC will provide 100% same (bit exact) version of source as PCM. It works functionally same as Zip or similar compression.
LPCM 2.0 bit rate: 1536 kb/s
FLAC 2.0 bit rate: 259 kb/s
The LPCM is 1536 kilobits per second, but the FLAC is 259. Will I have the same 640kb/s ac3 audio file converting 256kb/s FLAC audio as I would with 1536kb/s LPCM audio?
You may not realize this, but unlike LPCM, FLAC doesn't have a constant bitrate. It varies, sometimes wildly.
So lpcm = 1536, all the time for that set of bitdepth & samplerate & channels. But your flac, which you say is 259, is only AVERAGING 259 (which btw sounds way too low to be true - flac usually has a compression of ~2.39:1). Some moments it will be less, other moments it will be more. In fact with flac, it is even possible for certain instantaneous bitrates to spike HIGHER than the lpcm rate.
Regardless of the flac bitrate, recompression to another format is NEVER done to the flac file directly, but is done to the intermediate, decompressed lpcm version that was its source, just like u_q said. So the bitrate of the flac actually has no bearing in these equations.
Again, *IF* your source lpcm in the one file is identical to the decompressed lpcm intermediate from the flac file, AND you perform the same recompression application on both, and have resultant files which are identical in bitrate, those SHOULD be equivalent in quality. Not identical, as lossy compression does have some variability (uniqueness, individuality) per pass, and so may not have identical hashes, but overall they can be considered as identical.
mkvextract to extract the flac audio from MKV file and the result is:
The extension is .??? audacity can’t recognize the file. Should I change the file extension or import the file in any other way?
I decided to open eac3to and import the mkv file with flac audio and the program recognized it. Would it be possible to decompress the flac with eac3to?
I've finally done it with eac3to, I haven't been able to do it with Audacity. Once I fed eac3to with the mkv file I used this:
The result is the same bitrate as the original LPCM and identical bit of depth. Thank you very much to everyone, I had no idea that FLACs were compressed audios.
Just one more thing, what can I do if the original instead of being LPCM is DTS-HD?
Suppose I have another MKV file with FLAC, but this time the original audio is:
2.0 / 48 kHz / 2095 kbps / 24-bit (DTS Core: 2.0 / 48 kHz / 1509 kbps / 24-bit)
Would it be possible to extract DTS Core from FLAC?
That's not how it works.
Anytime you compress something, whether it's lossless or lossy, it is using as its source an uncompressed copy (might be a file or a decompressed intermediate or a pipe).
If your "source" was intended to be DTS-MA, prior to compression to flac it would first decompress the dts to lpcm and then use that as input into the flac. ALWAYS.
This is what I did today:
First, Flac to wav. The original bluray audio is DTS-HD instead of LPCM.
DTS-HD Master Audio English 1554 kbps 2.0 / 48 kHz / 1554 kbps / 16-bit (DTS Core: 2.0 / 48 kHz / 1509 kbps / 16-bit)
The resulting wav doesn't seem to have the same bit rate:
WAV, 2.0 channels, 1:26:10, 16 bits, 1536kbps, 48kHz
Anyway, I fed eac3to with the wav audio:
Creating file "C:\Users\AD\Desktop\video.mkv_2eng.wav_.L.wav "...
Creating file "C:\Users\AD\Desktop\video.mkv_2eng.wav_.R.wav "...
The original audio track has a constant bit depth of 16 bits.
Encoding DTS <768kbps> with Surcode...
Surcode DTS Encoder doesn't seem to be installed. <ERROR>
1.Why this difference in bit rate? It doesn't exactly match either the DTS-HD or the DTS Core, only the bit of depth.
2.Why eac3to separates L and R tracks and tries to encode at 768 kbps if the wav is 1536kbps.
3.If I extract DTS -core from DTS-HD I do not receive any error message, should I install codecs to get the DTS -core from FLAC?
RAW data are easy to calculate - sample rate * number of channels * bitdepth (assumption each channel has same bitdepth) equal your RAW bitrate, in real numbers this look like this: sample rate 48000 * 2 channels * 16 bit per channel = 1,536,000 bits per second.
Some tools may report incorrectly parameters. Anything above 1536kbps may be overhead.
Additional remark so called DTS (core) is a lossy audio compression, 768kbps DTS exist thus software may try to reencode your source to 768kbps DTS core (do not forget that whenever DTS 768 or 1536kbps is mentioned then we should talk about 5.1 audio i.e DTS compression ratio 6 or 12 to 1, for 2.0 audio 768kbps DTS provide only 2 to 1 compression)
Anyway: keep in mind that Surcode is outdated and obsolete software, designed when DVD-Video "ruled" (so to speak).
eac3to should have been updated to use ffdcaenc-2 (freeware) or/and the DTS-HD Master Audio Suite, but sadly the author (madshi) doesn't care......users on my IgnoreList: 161 names thus far, but featuring DB83 and manono.
Thank you very much. I'll try with ffdcaenc-2 as soon as possible.
users on my IgnoreList: 161 names thus far, but featuring DB83 and manono.