Want to merge FLAC audio without re-encoding, better with a GUI Windows App.

[D.Draker]

My theory is that FLAC has a poorly written container that affects the playback because the player has to unpack it to RAM first.

[D.Draker]

Anyways, it's not the topic here, I mean the comparison between the two.

And as for the merging options, I tried MKVmerge suggested by @j7n, it also borked up the timings, just like FFMPEG did.

So it's only somewhat usable if not extracted from MKA. But it's not what the OP wants.

[Klemper]

On 12/14/2024 at 10:59 PM, TSNH said:

Where did you find this info? Compressing audio in a lossless way should be theoretically possible and that's exactly what FLAC is supposed to accomplish. It would have no reason to exist otherwise.

https://www.reddit.com/r/audiophile/comments/tgbhbe/what_has_better_quality_wavs_or_flacs/

FLAC ... "shaves off unnecessary bitrate to result in a smaller file"

How it can be lossless? Please explain.

That's why I want to merge FLAC audio without re-encoding.

[Klemper]

14 hours ago, D.Draker said:

Anyways, it's not the topic here, I mean the comparison between the two.

 

You're welcome to compare, too.

[j7n]

You likely deal with compressed data without realizing it every day. Most programs sent to you from the internet are compressed and can be extracted to their exact original form, otherwise they wouldn't work. When you download a game packed with RAR, all sound in it is compressed in a broadly similar way to FLAC.

Audio data contains redundancy. Neighboring samples are usually similar. This allows a program to analyze them and make a prediction. Imagine you want to record the air temperature every day. It can go between very hot in July to very cold in January. But there is not much difference between consecutive days. So instead of saving every reading in full, you save the difference from the previous day, which is a smaller number. You can still recover the full number exactly with simple arithmetic. This is how all "multimedia" compression with "delta" encoding works. It is in PNG and in 7-Zip.

The differences will be even smaller if you choose to record the temperature once per hour. Then you can build an algorithm to detect a cycle in every 24 units, which now becomes similar to sound with periodic vibrations. Instead of looking only one sample backwards, you can keep a sliding memory window that is a bit longer.

Another aspect that can be exploited in compression is that the entire range of values is not utilized. When the music is fading out, the top bits stay zero the entire time. Back to the temperature analogy, you might choose a scale that encompases the Sahara desert and Antarctica, but those extremes are never reached in practice.

The "FLAC is not lossless" argument has been debated many times, and you can find the discussions on a web search. "Shaves off" is a figure of speech, and can't be analyzed on technical grounds.

[D.Draker]

13 hours ago, j7n said:

You likely deal with compressed data without realizing it every day. Most programs sent to you from the internet are compressed and can be extracted to their exact original form, otherwise they wouldn't work. When you download a game packed with RAR, all sound in it is compressed in a broadly similar way to FLAC.

Audio data contains redundancy. Neighboring samples are usually similar. This allows a program to analyze them and make a prediction. Imagine you want to record the air temperature every day. It can go between very hot in July to very cold in January. But there is not much difference between consecutive days. So instead of saving every reading in full, you save the difference from the previous day, which is a smaller number. You can still recover the full number exactly with simple arithmetic. This is how all "multimedia" compression with "delta" encoding works. It is in PNG and in 7-Zip.

The differences will be even smaller if you choose to record the temperature once per hour. Then you can build an algorithm to detect a cycle in every 24 units, which now becomes similar to sound with periodic vibrations. Instead of looking only one sample backwards, you can keep a sliding memory window that is a bit longer.

Another aspect that can be exploited in compression is that the entire range of values is not utilized. When the music is fading out, the top bits stay zero the entire time. Back to the temperature analogy, you might choose a scale that encompases the Sahara desert and Antarctica, but those extremes are never reached in practice.

The "FLAC is not lossless" argument has been debated many times, and you can find the discussions on a web search. "Shaves off" is a figure of speech, and can't be analyzed on technical grounds.

That's the very definition of a lossy format. xD

[D.Draker]

14 hours ago, Klemper said:

You're welcome to compare, too.

https://audiophiles.co/flac-vs-wav/

Key Features of WAV

1. Uncompressed Audio: WAV files typically store audio data in an uncompressed format, preserving the original audio quality without any loss. This makes WAV ideal for professional audio work where maintaining the highest possible fidelity is crucial.

 

2. PCM Encoding: Most WAV files use Pulse Code Modulation (PCM) encoding, which captures analogue audio signals by sampling them at regular intervals and quantizing the samples to digital values. This results in a faithful digital representation of the original analogue waveform