Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Proper way to compare spectrograms from CD vs. 24bit Digital Download (Read 2254 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Proper way to compare spectrograms from CD vs. 24bit Digital Download

I want to compare a single track from an album digital vs. cd.

Digital album is from bandcamp and come as 24bit 48kHz strict.
CD obvious 16bit 44.1kHz.
CD backup with EAC secure mode and gap detection.

I create spectrograms with following command:

Code: [Select]
sox INPUT.flac -n spectrogram -X 50 -Y 2000 -o OUTPUT.png

To downsample the WEB i use the following command:
Code: [Select]
ffmpeg -i INPUT.flac -sample_fmt s16 -ar 44100 OUTPUT.flac

Image with noticeable names are attached.

What i''m doing wrong? The kHz on WEB dither and original are very low compared to CD.

Re: Proper way to compare spectrograms from CD vs. 24bit Digital Download

Reply #1 –
Quote
What i''m doing wrong?
You MIGHT be wrong in ASSUMING the Bandcamp download has more high frequency content.     ;)

Re: Proper way to compare spectrograms from CD vs. 24bit Digital Download

Reply #2 –
In that case I could have saved myself the purchase on Bandcamp. One can probably only rely on CD.

I know that bandcamp don't even say what you will get but I expected at least CD quality.

Re: Proper way to compare spectrograms from CD vs. 24bit Digital Download

Reply #3 –
Quote from: critkitten on
What i''m doing wrong?

Everything. You don't use spectrum analyzer to determine audio quality...
gold plated toslink fan

Re: Proper way to compare spectrograms from CD vs. 24bit Digital Download

Reply #5 –
Quote from: critkitten on
Quote from: Markuza97 on
Quote from: critkitten on
What i''m doing wrong?
Everything. You don't use spectrum analyzer to determine audio quality...
Okay you say that the cd can the bad one in this case? I just want to compare both versions.

The only way to compare them is by listening.
Just because something looks nice on the spectrum doesn't mean it actually sounds good.
gold plated toslink fan

Re: Proper way to compare spectrograms from CD vs. 24bit Digital Download

Reply #6 –
The original web rip has a cutoff at 19 khz and so has your resampled version. You arent doing anything wrong. Its likely a lossy transcode.

Re: Proper way to compare spectrograms from CD vs. 24bit Digital Download

Reply #7 –
Technically you're not doing anything wrong.
Seems like web versions got more processing than CD. By spectrograms impossible to tell if for the better or worse (going full Nyquist isn't really desirable). You might try to search here for a way to see if there's anything significant on more than 16 bits, but again some of the processing might make it look like there is while there's not.
Such lowpass always raises doubts if there wasn't lossy encoding at some point...

Re: Proper way to compare spectrograms from CD vs. 24bit Digital Download

Reply #8 –
Quote from: rutra80 on
Such lowpass always raises doubts if there wasn't lossy encoding at some point...
For bandcamp that is quite shady, i expected this sort of behavior from streamers and not digital download sites. Or at least i havnt had many transcodes on bandcamp when they really dont need to be transcodes.

Re: Proper way to compare spectrograms from CD vs. 24bit Digital Download

Reply #9 –
First I thought it was because I was creating the wrong spectrogram because there were different bit depths.

I just mention bandcamp to make clear where i buy it.
i don't know if it's bandcamp fault if even.

I will write to the label and describe the problem. Let's see.

Re: Proper way to compare spectrograms from CD vs. 24bit Digital Download

Reply #10 –
Spectrograms are not worthless. They are not good to decide whether this is going to sound better than that, but surely they can get you an idea whether a 96000 file is just an upconverted CD. Easy to fool by a "malicious salesperson", but not useless against WTFs.


But different masterings might be quite different both on the eye and on the ear. Nothing says that if you get a 2022 version of a 1987 CD, it will sound the same. Bandcamp or not. Bandcamp might be more prone to it because well if some musician think they improved the mix, it is easier to upload than to go to the CD pressing plant. So if this song on CD and same song on download are different ... nothing special about that.

As for Bandcamp (who cancels my purchases after I've paid, so ... not putting much more money there):
Bandcamp requires the upload to be as e.g. WAVE rather than MP3.  That doesn't prevent anyone from decoding an MP3 to WAVE, it works more as a polite "hey, this isn't what we like" and might at best educate someone on how not to export music from their home studio.
* Different masterings are different. Just because a spectrogram