In a bout of temporary insanity I took advantage of MH`s offer and bought Spectrafoo complete at a reduced price. I`ve been playing around with it, and among other things, I`ve made spectrograms of some of my LPs. Mostly out of curiosity about the HF content of vinyl vs. digital.
I got sidetracked right away by some of the strange spectrograms I found, because it seems to me that that digital recordings still keep their sonic signatures even after being transferred to vinyl. Maybe not a surprising find for you pros, but surprising for me at least.
But I am a complete amateur and newbie, so I was hoping that someone on this forum could explain to me what I am seeing:
1) Harry Belafonte at Carnegie Hall. RCA reissue, 180 g. vinyl. The strangest picture- some HF bands around 20 kHz on both channels. This is not an artifact, I kept the settings and made spectrograms of other recordings afterwards, and those bands weren`t there.
2) Jennifer Warnes, Famous Blue Raincoat, Sonet LP, bought by me in 1990. Bandlimited? If so, then why? 48kHz digital recording?
3) Jackson Browne, Running on empty All analog, I`m sure.
For the first one, it's not unheard for some recordings to have resonant peaks or tones just outside of the normal hearing range. Sometimes this is the fault of the recording gear, sometimes it's something in the chain itself.
For the second, I'm only seeing the thumbnail so I can't make out where the roll-off is. Could be a digital delay unit used in vinyl mastering, could've been sourced from an early CD (which usually had sharp rolloff at 20KHz).
For the third, could be a digital delay in the cutting process (probably not if this is an original pressing, but could be a later re-cut) or it could just be a limitation of the recording gear. Some of New Order's mid-80s stuff is bandlimited to 17KHz. Perhaps they were so visionary that they foresaw a future where MP3 encoders would lop off everything above that.
No idea why the second one only came out as thumbnail- I will try to reupload.
I have this Jennifer Warnes LP in another edition as well. The spectrogram of that LP seems to be band limited as well- but at another cutoff frequency! All very strange to me.
Looks to me like it's mastered from an old CD, although I can't really explain the difference between the two pressings.
I've seen this sort of bandlimiting before, like on this particular album: http://www.discogs.com/Tangerine-Dre...elease/2986910, which I am convinced was mastered from CD. If you zoom in on peaks, do the very top of the highs look to be slightly delayed as they near the cutoff point? I'm told that's usually evidence of the signal going through an analog brickwall filter, which were used in the early days of CD mastering to prevent aliasing.
On a tangent, I'm told the iTunes release of the album I linked to is itself a needle-drop.