Fake Flac? I.e fake lossless.

[zoomo]

Downloaded BRG 2 from the full list site and found a folder containing each track as flac, and all the same files as mp3, after listening to them both I couldn't hear much difference between the flac and the mp3, I performed a spectrum analysis using Audacity and it reports the same results for both flac and mp3, the result showed what you'd expect from a flac file, but to have a similar result for the mp3 was odd, so going by the details showing while hovering over the flac files indicates that replaygain was used encoded by FLAC v1. 1.4a with FLAC frontend v1. 7.1

The plot spectrum analysis results as mentioned earlier for both the flac and mp3 were approximately 22000Hz.

Question is, would you say that the flacs are more likely to be fake in that it was no doubt the mp3 files that were used to make these flac files which would mean that they aren't true lossless files, so you'd be best to keep the mp3 files and ditch the flac, after all they both come to the same analysis.

With regards to the mp3 files they were encoded with Lame Front-End 1.8 using pazera software.

[zoomo]

Testing track 1 Funky Captions using Spectro shows the average bitrate as 272 kbps, the same track in flac reported a average bitrate of 925 kbps, does this not show that the flac file is a higher quality than the mp3 at 272 kbps, or is that of no importance?

[Retronic]

I'm pretty positive this is my own (old) rip in FLAC for comparison perhaps:

mediafire.com/file/esu8laov8eqq4j9/BRG_02_-_Contemporary_Orchestral.rar/file

[zoomo]

I'm pretty positive this is my own (old) rip in FLAC for comparison perhaps:

mediafire.com/file/esu8laov8eqq4j9/BRG_02_-_Contemporary_Orchestral.rar/file

Cheers, downloading it now, will let u know later.

[zoomo]

I'm pretty positive this is my own (old) rip in FLAC for comparison perhaps:

mediafire.com/file/esu8laov8eqq4j9/BRG_02_-_Contemporary_Orchestral.rar/file

average bitrate 839 kbps.
Funky Captions reports a cut-off at 19.8KHz

[Porn Library]

This is why I like and will always stick to WAV. Regardless of what people think or say, FLAC files are not always lossless. Getting a true lossless FLAC is very tricky and it all depends on the recording software being used. The free recording software out there does not make lossless FLACS I found out not too long ago. You will be losing something.

Also, anytime someone converts a WAV file or any other lossless format to FLAC, you just killed it. It more or less becomes a 320k mp3. So that's why I never rip anything in FLAC that I transfer. Not to mention, the format is not easy to make CD's with (and i'm one of those people who still listens to music on home stereo 80% of the time). So just thought i'd reply to this with my thoughts and info

[Andrew]

Hi Porn Library,

A FLAC file can be thought of as a compressed WAV file, similar to a ZIP file.  The size is made smaller and the contents are not changed once unzipped.

You can test this yourself by doing the following:
- Compress a WAV file to FLAC
- Decompress the FLAC file back to a WAV file
- Compare the original WAV file to the decompressed one
They should be the same bit for bit

WAV files can also be faked.  I can fake one by taking an mp3 and converting it to a WAV file.

[Metacity]

I can vouch for the "safety" of the FLAC format as well. I have on several occasions made FLAC copies of WAVs in my collection to fit them under file size limits, and there was no difference at all in the audio or in a spectrum analysis.

That being said, I can easily see "degradation" happening if the software involved isn't quite up to par, so to speak. It might be worth trying several different programs and seeing if it makes a difference in the end result? (As I'm on Linux, I prefer this one myself.)

[Uvuzau]

Maybe a bit off topic, but for me is bigger problem when people rip their records very quiet.
I guess main cause of that is recording audio signal before the amplifier, so "ripper" listens normal loudness while recorded audio is very quiet.
In the end for me, it doesnt even matter is it flac, wav or mp3 when rip is very low in volume.

Retronic for example is doing great rips when it comes to loudness, nothing compressed or limited, just pure recordings with correct gain.

[Retronic]

Mine were quiet for a long time- I think it was simply because I didn't have Audacity at 100% recording level!

It's trial and error I guess.

[Andrew]

I've found records themselves can be all over the place in volume.
I'm guilty of not correcting for the differences afterwards when processing them.

[Silverrider85]

I'm ripping with audacity and not correcting the sound because i simply don't have the knowledge to do so. The only reason I rip in FLAC is because I think it's the most used format here with the best quality, correct me if i'm wrong. Don't think my rips are too quiet either, but i never compared to others ... interesting discussion for sure!

[Uvuzau]

There is no need for afterwards audio level editing, its just matter of finding the loudest part of song or whole record and compare is loudness somewhere near lets say CD release or some other well ripped record.
That way loudest part/song will be up to "industry standard", so you dont have to turn your speakers up
Its much easier with songs that have drums, not so easy with orchestral suspenses for example or similar stuff with huge dynamic differences to define the loudest part and set the recording levels accordingly.

Edit: One example from few days ago, i was looking for upgrade of Bruton BRI 16, which is very low in gain, that low that i cant even define is it really flac or no, im not sure who ripped it originally.
John_Fred helped me with sharing some rip of it but it was same as one i had, very very low in volume.

[Greta]

@ Uvuazu - I can try and see if my copy of BRI 16 is any better. I'll up it in a few days ok?

[Greta]

Going back to the "flac is not 100% lossless"....Porn Library....you killed me. I simply DON'T WANT to believe it.
I remember well, somebody, many time ago, in a previous discussion about flacs, posted a very tech experiment about conversion from wav to flac and then back to wav, demonstrating there was no loss of a single bit. (checksum files, or something like that, were involved too)
Does anybody recall that article?