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: strange spectrogram (Read 7710 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

strange spectrogram

i received a VinylRip file from a friend claiming it was 24 bits 192 Khz, but when i analysed it on Audition and Audacity i found an odd spectrogram…more evident on audacity


i really would like to know if you guys can tell me just by looking at these spectrograms if my friend sent me a transcoded file, if this Vinyl was not ripped at 24 bits  192 kHz and later was transcoded to this parameters.
in certain cases where the higher frequencies are cut and the spectrogram reveals a flat top, these are easy but some  i really found  hard to separate the good from the bad, so, please, i need an opinion of an expert or someone with experience in analysing spectrograms to somehow tell me if this flac is a transcode
thanks

strange spectrogram

Reply #1 –
Looks fine to me.

You need to consider a few things:
  • It is physically impossible for a human to hear above 20 KHz.
  • It's unlikely any of the equipment used in recording, mixing, and mastering the source material was designed to reliably handle anything above 20 KHz, and may have even filtered out such frequencies, for various reasons.
  • Due to physical limitations of the cutting head, and vinyl's relatively slow speed, a lowpass filter was applied to prevent anything above 20 KHz from getting to the cutting head, and thus into the groove.
  • 0-20 KHz is fully and perfectly accommodated by a 44.1 KHz sample rate; your friend is wasting gobs of space ripping at 192, for zero benefit.
  • The only thing you see in the rip above 20 KHz is noise: the sides of the groove wall rushing past the needle; artifacts of clicks & pops from crud and scratches; the extra randomness from the way noisy transients (snare hits, cymbals) manifest in the groove; and the needle's inability to accurately follow the groove's contours during those transients.

strange spectrogram

Reply #2 –
@mjb2006: ?

@sisiphus1: Maybe you can upload (uploads sub forum) or link a 30s sample?

strange spectrogram

Reply #3 –
@Juha: ? What? I answered his question and then explained why he shouldn't read too much into the "flat top".

While there's no way to know with certainty if something is a transcode by looking or listening, we can see that his graphs show the musical content bandlimited to ~20 KHz (indicating source material going onto the vinyl was lowpassed), and there are occasional spikes of noise (probably vinyl surface noise), plus broadband background noise (purple/blue all over everything). If this vinyl rip were done at a lower bit depth and sample rate and then converted to 24/192, surely the pop/click noise wouldn't extend as high as it does. And unless dither was added (unnecessary for an upconvert), the background noise can be assumed to have been recorded, i.e. the rip was recorded at this bit depth. I did allow for the possibility of an elaborate fake when I said the spectrogram "looks" fine to me; i.e. it is what I would expect a 24/192 vinyl rip to look like. I don't know why he thought it looked odd.

strange spectrogram

Reply #4 –
Quote from: sisiphus1 on
i really would like to know if you guys can tell me just by looking at these spectrograms if my friend sent me a transcoded file


Looks fine to me, in terms of sample rate. Music apparently up to 20KHz, complete noise above that: it's a typical 192KHz spectrogram. You can't say much more about it just from looking at a tiny spectrogram.

Well, you can say that it's a complete waste of space, like mjp2006 said. 

strange spectrogram

Reply #5 –
Here is an example of 24/192 recording from vinyl I just did (random pics from my collection).



My chart (EPC205CMK3) frequency range is 5Hz-80kHz. I had a steep HP filter @ 20Hz.

Recording path was TT -> ProJect Phono Box MM -> E-MU 0404 USB -> Reaper (ASIO path).

strange spectrogram

Reply #6 –
What's all that "tall grass" between 20K and 50K? Is that all simply unhearable cymbals and crashes? I assume it's not pops & clicks

strange spectrogram

Reply #7 –
Quote from: dhromed on
What's all that "tall grass" between 20K and 50K? Is that all simply unhearable cymbals and crashes? I assume it's not pops & clicks


Those 'spikes' follows percussive instruments and other peaking instruments so probably overtones of them ... no noticeable pops & clicks in these samples.


Also, notice the "noiseless" area before noise @ hf area. OP had it filled with noise ... maybe it's was just because of lower quality equipment (chart, pre-amp, audio interface) they used for recording?
 

strange spectrogram

Reply #8 –
Assuming good faith on his friend's side, I'd say noisy equipment with a source that's already lowpassed at 20KHz.

strange spectrogram

Reply #9 –
Quote from: dhromed on
What's all that "tall grass" between 20K and 50K? Is that all simply unhearable cymbals and crashes? I assume it's not pops & clicks

It is the nonlinear distortion from the disk wear.

strange spectrogram

Reply #10 –
Cymbals/snares are like white noise, full of essentially gobs of random overtones across the audible spectrum, and beyond. If you filter out the ultrasonic frequencies, as I'm sure was done when cutting the master for the vinyl, if not sooner, the "tall grass" from these instruments won't be going onto the record. Yet, it will still appear in a rip, so it's either in the groove or added by the playback equipment. If in the groove, it's only a result of wear, imperfections in the cutting process (perhaps unavoidable), or (I'm guessing) worn-out pressing plates. In any case, it's not music, it's not coming from the master recording from which the vinyl master was derived. That's the point I wish to impress upon people who worry about whatever is above 20 KHz in a rip.

(I assumed pops and clicks for the highest spikes because they were so infrequent and stick out like a sore thumb from the music. But with the graphs being so low-res, they could just be musical transients in just the right spots to be visible.)

strange spectrogram

Reply #12 –
thanks for your input
now i begin to understand a bit more about this spectrograms