Recording standards and CD remastering

[Dave2002]

I do a lot of listening using headphones, which means that I can sometimes hear problems which may go unnoticed if heard via loudspeakers.

A few days ago I downloaded the track Summer from this CD of Frank Bridge's music - http://www.discogs.com/Frank-Bridge-...elease/3822914 from iTUnes.

Ignoring the issues about listening via a digital compressed format, this particular recording has a fairly continuous low frequency background noise. Other recordings also have evidence of low frequency noise - so this problem is not an isolated one.

Questions which come to mind are

"Was this low frequency noise present in the original recording?"
"Did nobody hear it, when the recording was made?"
"Has the LF noise been introduced later, due to some other process?"

It is possible to reduce LF noise considerably using modern digital filtering techniques, and this can be very effective if done carefully. The particular recording sounds rather good, apart from the LF noise, so it's a shame that this hasn't been removed, or that more care wasn't taken to avoid it in the first place.

However, I'm not convinced that recording engineers of the past were unaware of these problems - so the problems could have been introduced later, though how? If the CD remaster were made from an LP though, then it would be easy enough to see how LF noise could get in - though the best LP replay systems can largely avoid this.

Comments!

[amateur51]

I do a lot of listening using headphones, which means that I can sometimes hear problems which may go unnoticed if heard via loudspeakers.

A few days ago I downloaded the track Summer from this CD of Frank Bridge's music - http://www.discogs.com/Frank-Bridge-...elease/3822914 from iTUnes.

Ignoring the issues about listening via a digital compressed format, this particular recording has a fairly continuous low frequency background noise. Other recordings also have evidence of low frequency noise - so this problem is not an isolated one.

Questions which come to mind are

"Was this low frequency noise present in the original recording?"
"Did nobody hear it, when the recording was made?"
"Has the LF noise been introduced later, due to some other process?"

It is possible to reduce LF noise considerably using modern digital filtering techniques, and this can be very effective if done carefully. The particular recording sounds rather good, apart from the LF noise, so it's a shame that this hasn't been removed, or that more care wasn't taken to avoid it in the first place.

However, I'm not convinced that recording engineers of the past were unaware of these problems - so the problems could have been introduced later, though how? If the CD remaster were made from an LP though, then it would be easy enough to see how LF noise could get in - though the best LP replay systems can largely avoid this.

Comments!

Well, perhaps, if it bothers you so much, you might stop listening through headphones ...?

[johnb]

Out of curiosity I've just listened to the 1 minute samples of each of the "The Sea" tracks of the recording on Qobuz and wasn't able to detect any low frequency noise. This might, of course, be due to my headphones, my hearing or just the samples I listened to.

http://player.qobuz.com/#!/album/0724356685550?pt=1033708

The release on both iTunes and Qobuz seems to have the same remastering date (1989) and both seem to be later releases than the one that your link points to.

[richardfinegold]

Is it possible to digititally reedit the track yourself? There is a lot of software out there for this, intended primarily for digitizing lps or 78s. I am not sure if one can do this with downloads but it should be possible.

[cloughie]

Well, perhaps, if it bothers you so much, you might stop listening through headphones ...?

ams maybe you type what others think!

[richardfinegold]

Another thought Dave--perhaps you acquired some RFI during either the download or playback process. I had purchased a cheap headphone amp from Pro-Ject that seemed horribly affected by RFI to the point where it was unlistenable. I now use the headphone jack from my Mac if I am forced to use cans.
Again, if you analyze the track with digital software you could tell if the LF noise is in the download or an artifact of your playback system

[MrGongGong]

Ignoring the issues about listening via a digital compressed format, this particular recording has a fairly continuous low frequency background noise. Other recordings also have evidence of low frequency noise - so this problem is not an isolated one.

Comments!

I think you might be ignoring the thing that is the problem
I'm not an expert on compression artifacts but i'm sure there are folks who will be along in a minute who are !

OR
It's your body reminding you that you are ALIVE !
OR
It's the VLF humm that some folks hear all the time
OR
You have an earthing issue

and so on ............

[amateur51]

I think you might be ignoring the thing that is the problem
I'm not an expert on compression artifacts but i'm sure there are folks who will be along in a minute who are !

OR
It's your body reminding you that you are ALIVE !
OR
It's the VLF humm that some folks hear all the time
OR
You have an earthing issue

and so on ............

Those who have tinnitus have learned to filter out all manner of extraneous/intraneous auditory artefacts as a matter of course - did you hear that whistle then?!?

[Dave2002]

Is it possible to digititally reedit the track yourself? There is a lot of software out there for this, intended primarily for digitizing lps or 78s. I am not sure if one can do this with downloads but it should be possible.

This indeed may be possible - though I'm not sure about this particular piece. Years ago I made a number of recordings and "succeeded" (!!) in injecting hum through the sort of things that MrGG has mentioned - lack of screening, crossed wires, earth loops etc. I was able to remove the hum rather successfully using Audacity. I've tried since on other recordings, and it is possible to "over cook" the LF noise removal, but with care it is possible to reduce much noise to normally acceptable levels.

Really my point is whether recording engineers - of any period - were aware enough of LF problems (I'm sure many were), and took steps to avoid them, or whether problems tended to (and still tend to) get introduced after an acceptably good recording has been made, possibly during later remastering processes.

[johnb]

Surely the first step is to establish whether the low frequency background noise you hear is present on the CD/lossless downloads or if it is due you problems with the iTune encoding or whether it is being generated on your own system.

The short samples you can play on the Qobuz and iTunes websites might or might not be suitable to show the noise you hear, but these are the links:

http://player.qobuz.com/#!/album/0724356685550?pt=1033708

https://itunes.apple.com/gb/album/bridge-sea-summer-cherry-ripe/id688863909

[Dave2002]

Surely the first step is to establish whether the low frequency background noise you hear is present on the CD/lossless downloads or if it is due you problems with the iTune encoding or whether it is being generated on your own system.

The short samples you can play on the Qobuz and iTunes websites might or might not be suitable to show the noise you hear, but these are the links:

http://player.qobuz.com/#!/album/0724356685550?pt=1033708

https://itunes.apple.com/gb/album/br...pe/id688863909

If the m4a file is loaded into Audacity, then it's possible to look at various spectra. It's not quite as simple as one might hope - very much different from looking at spikes with known inputs, which one sometimes sees in magazines and reviews of equipment. However it's just about possible to detect something. If the sample is taken right at the end of the track where most of the music has faded out then there are peaks around 50-55 Hz, 90-100 Hz and a fairly sharp one around 190-200 Hz. The peaks are at around -60dB, which seems low - but then one has to look at the peak levels when the music is playing. Typically these are at around -30dB as measured in Audacity, so this gives an effective SNR of about 30dB.

Using the same sample right at the end of the track to generate a noise profile it seems possible to remove some of the pitched noise, and to improve the SNR, at least to the point where to me at least most of the LF effects seem to have disappeared.

If the effects are due to the frequencies suggested, they will have a pitch effect around G to G sharp, and if this persists then this can affect
harmonic relationships within the music.

[Flosshilde]

You have an earthing issue

I read that as 'earthling issue' - sometimes when I read some of Dave's more technical posts I wonder what planet he's on

I try & avoid using headphones - it always sounds as if the music is in my head, rather than inhabiting the space around me. But when I do listen through headphones, it's amazing how much detail I can hear - I never realised until recently that Sammy was so vocal, for example.

[Eine Alpensinfonie]

A more general comment on recording standards - the Decca sound remains for me the best of all, with a depth of sound not found with such consistency elsewhere. They were particularly impressive in opera recordings.

[pastoralguy]

Ignoring the issues about listening via a digital compressed format, this particular recording has a fairly continuous low frequency background noise.

Comments!

You sure it's not the viola section...?

[richardfinegold]

I read that as 'earthling issue' - sometimes when I read some of Dave's more technical posts I wonder what planet he's on

I try & avoid using headphones - it always sounds as if the music is in my head, rather than inhabiting the space around me. But when I do listen through headphones, it's amazing how much detail I can hear - I never realised until recently that Sammy was so vocal, for example.

Headphone listening could be the subject of a thread. I'll start one.