Jump to content
TrailerParkDarling

Sky Ferreira

Recommended Posts

14 minutes ago, rollwithme said:

Is this real or a demo? The song came out after that date so I’m confused

Sky Ferreira - Dont Forget 7.8MIX 24bit 1.28.22.mp3

sounds the exact same to me i have to listen to this and the released one side by side to see if they have any differences 

Share this post


Link to post
Share on other sites
1 hour ago, AtomicMess said:


I could if you'd like me to, BUT, I wouldn't be able to be the 'seal of certainty' on legitness. Honestly only the folks who leaked them would be able to do that.

Putting a spoiler tag because this will be long, but for the girlies who care (and for those who want to learn), here's a mini-class on reading spectrums as a mark of quality and a shitty way of how we get to "upscales".

 

  Hide contents

 

Don't take this as gospel, but rather just a general quick-pass on the topic.
 

This is the spectrum for an OG file; I've got the CD, ripped in lossless. (90s Dance Banger; doesn't exist in the digital era, RIP) This would be our OG/WAV. Notice how the full spectrum here is populated with data. In terms of hearing frequencies, most humans top out at around 20khz, however some folks (probably not you, you're not special, probably not myself either) can hear a little higher, but not much.

All the sonic data is here. Now, notice at the start of the track, the spectrum isn't really "full" - this particular song starts out quiet, fades in, and gradually adds instrumentation and vocals, which explains why there's not a lot of sonic data at first, and then suddenly, there is. In short - there's some variables here that would affect whether a song or a part of a song takes up the "full" spectrum, but overall, a fully lossless/mastered track should look something like this, or have parts of the song that take up the full spectrum of sound, like this.

 

C9qrUCp.png


Another example of a mastered/finished track, also ripped from a CD, looks like this:

 

https://i.imgur.com/am1I6O6.png

 

Notice that the spectrum doesn't go quite to the top @ 22khz, but it comes close. It all depends on the track mastering and instrumentation. (Sidenote - when you see things like 44.1khz or 48khz on WAV or really ANY audio files - that's taking into account stereo. 44.1kz/2 = 22khz per left/right channel. 48khz/2 - 24khz per channel)

Now, if I take that first track, and I convert it to 128k, we immediately look like this:

https://i.imgur.com/IcAYN7O.png

 

 

Bam - all that higher frequency data? Bye. That sharp line of sonic demarcation right above 15khz is the canary that says "You've been compressed!", especially for 128k mp3.

 

The algorithm for encoding, in most standard encoders without any fiddling, starts to roll off frequencies over certain points. Take this with a grain of salt, but in general, these are the "cut off" points, so to speak, for the mp3 encoder.

320kbps - 19.5kHz
192kbps - 18kHz
160kbps - 17kHz
128kbps - 16kHz 

Sonic data above these lines is either discarded entirely, or heavily compressed, at these bitrates. VBR (variable bit rate) encoding changes things a bit, but I'm not getting into that.

Here are all the screenshots of this same song/file, at various different bitrates, just to give a strong visual of what compression does to a song at different levels (raw/og, 128k, 160k, 192k, 224k, 256k, 320k): https://imgur.com/a/xX4emZt

 

If you scroll through, you can see how at each level, more and more data is kept at each bitrate/encode level, however, even at 320k, there's still a substantial amount of sonics that are missing at the 20khz line and above. But, given that human hearing tops out at this level, this is why files at 320k are considered by many to be lossless/OG, because functionally, they are (technically they're def not). Hearing encoding artifacts and differences at this level is something only very, very good ears will notice, AND, you'd need to hear the uncompressed OG, to really be able to make the comparison to tell that something was amiss to start with.

 

Now, this is where this gets interesting, and this is where people start to point to the word/phrase "upscale", and how we get here.

I took the 128k mp3 file, which shows up as missing/having compressed allll that sonic data above ~16khz, and loaded it into Ableton (Audio/DAW program).

Then, I re-exported it, without any changes, back out to a raw WAV file.

Now the spectrum looks like this:

 

https://i.imgur.com/qDSEgXo.png


Now, if I listen to this, it's still obvious its still a 128k source file, regardless of the fact that the spectrum shows otherwise. The hi-hats are crunchy, the kick drums sound "sharp", the synths are warbled on some notes. Even though it looks like there's plenty of "good" high frequency data here in the spectrum, it's a lovely lie (song title pun here :oop2:)

People who are far more talented than I though, would be able to take these low quality, 128k encodes/sources, and "clean them up", so to speak, with various audio tools and plugins. They can trim back some of that sharp/crunchy crunch on the hi-hats and kick drums, they can play with the EQ to make the vocals sound less muddy, etc, and then afterwards, export the track back out. There's a few out-of-the-box plugins (iZotopes 'UnChirp') that claim to be able to do just this, however, from experience, they're far from perfect.

Spectrum wise, they'll look like they're lossless/OG, or even just "higher quality", like the above here, and sonic wise, if cleaned up well enough, they might pass the "ear test" on not having any identifiable compression artifacts, but at the end of the day, they were still from a compressed source, original sonic data is missing, and the acoustics are all synthetic, in a nutshell.

So in relation to the original question - while I, or anyone else who understands a little bit about all of this, could check files for "legitness" on if they were OG or lossless, it gets a bit mucky. It's VERY easy and obvious to see if a file was/is compressed audio, because you see very sharp lines on a spectrum analyzer at the 15/16/17khz ranges where shit above is cut-off and discarded. You're not escaping that. But when it comes to 320k files, or WAV/FLAC/AIFF/other lossless formats, the spectrum sometimes can't *wholly* be trusted, because running files through professional audio DAWs and tinkering with them can produce back out a file that *looks* like it's OG/Lossless, but still actually came from a very compressed source (aka resampled/upscaled).


(And, playing them on any kind of high end system will still flesh out artifacts and chirps in the sound, usually)


Anyone who gets this far can clown me now for having no life :flop::flop::flop::flop::flop:

 

 

 

 

 

Thank you for this post! It's something I've been interested for a very long time and couldn't manage to find a very good source\explanation that wasn't fully tech\expert and you did a perfect job with the visual examples. This should be fixed on top of the Audio section of this place.

 

-

 

Oh wow we have the whole Maso Album now, just missing from the title track (which i'm not even sure if it was confirmed to be a Title Track? the "Hold my tongue until I choke" snippet? It's so good tho)

Share this post


Link to post
Share on other sites
40 minutes ago, iamjocelyn said:

sounds the exact same to me i have to listen to this and the released one side by side to see if they have any differences 

 

everything from the folder is real, the DF file is apparently just unmastered

 

so glad it all finally leaked now we can put this woman to rest :deadbanana:

Share this post


Link to post
Share on other sites
4 minutes ago, a wild one said:

 

everything from the folder is real, the DF file is apparently just unmastered

 

so glad it all finally leaked now we can put this woman to rest :deadbanana:

what folder luv :assad:

Share this post


Link to post
Share on other sites
24 minutes ago, iamjocelyn said:

what folder luv :assad:

 

originally it was a dropbox folder that had all the songs that leaked today + ntmt songs (its now replaced with a virus:flop:)

 

its basically the set of songs she performed this year, though for some reason she performed omanko instead of love in stereo which was in the folder

Share this post


Link to post
Share on other sites

okayyy so i’m using this tracklist

 

1. Don’t Forget

2. Descending

3. Innocent Kind

4. I Pray For Rain (tagged as “I Pray 4 Rain”)

5. Voices Carry (tagged as “Voices Carry (Sky’s Version)”)

6. Legends

7. The Freezing (tagged as “The Freezing (Sky Ferreira & Jorge Elbrecht)”)

8. Guardian

9. All My Lovers Die

10. Easy (tagged as Easy (Sky’s Version))

11. Downhill Lullaby

 

i tagged The Woods and Finders Keepers as a A-Side/B-Side single because i think those don’t really fit the sound of the album

Share this post


Link to post
Share on other sites
31 minutes ago, a wild one said:

 

originally it was a dropbox folder that had all the songs that leaked today + ntmt songs (its now replaced with a virus:flop:)

 

its basically the set of songs she performed this year, though for some reason she performed omanko instead of love in stereo which was in the folder

oh… nothing important left then i guess

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


  • Recently Browsing   1 member

×
×
  • Create New...