Jump to content

Harsh trebles on 1930s jazz on CD - what is the cause?


Haydn

Recommended Posts

I'm a big fan of 1930s jazz, which I buy on CD and download. I've noticed a big problem with harsh treble sounds on some tracks. Recently I've discovered that the same music on original 78s doesn't seem to have this problem.

I've posted about this subject on the 'swing djs' forum -

http://www.swingdjs.com/phpbb2/viewtopic.php?t=2245

- and although we have discussed it, no-one there has explained simply clearly (in a way that I can understand) what actually causes the trebles to sound harsh.

I wondered if anyone here could explain what makes the trebles on some 1930s jazz on CD sound bad?

Here are some examples of songs which seem to suffer from harsh treble sounds:

A-Tisket, A-Tasket, Chick Webb and Ella Fitzgerald 1937

Doin' The Jive, Glenn Miller 1937

Gotta Pebble In My Shoe, Chick Webb and Ella Fitzgerald 1938

Free Wheeling, Artie Shaw 1937

Ya Got Me, Tommy Dorsey 1938

Edited by Haydn
Link to comment
Share on other sites

I also think bad transfers. It happened with vinyl also. One example" In the mid 80s I bought a new copy of the the "Swing Street" 3 lp set on Epic. I thought the set had long been unavailable and had not been reissued. I thought I had a real find. Apparently I was wrong. The vinyl was not very typical of mid 80s Columbia vinyl - clean but very thin. Not the vinyl of the 50s and 60s. Also no booklet - just the discography info on the inside front box cover. Obviously not the original issue. The sound was pretty miserable. WAY to trebly and although vinyl, VERY digitally scrubbed.

Link to comment
Share on other sites

Since many re-issue projects (especially LPs) came from the marketing side of record companies, perhaps the objective in some cases was to make old material somehow sound more "Hi-Fi", emphasis on "High", meaning brighter sound.

It was certainly the marketers who came up with the idea of reprocesssed stereo -- breaking down honest, clear mono with slight channel-to-channel delay and different EQs. Bad in "stereo", worse if you had the misfortune to hear it folded back to mono. I remember some particularly bad Decca things in the '60s, the Lunceford band, for example.

Link to comment
Share on other sites

popping in - as I'm in the middle of a major re-issue project right now -

one thing I've noticed is that old recordings regularly need a boost above 8K to "open" them up, get more needed highs - but in the process a certain shrillness results when frequencies in the 3-6 K range are boosted at the same time. What I do is 1) boost the old recording at about 8-11 K and 2) THEN go back and use a notch filter at the lower frequencies to get the shrillness out - it takes about 3 times longer per cut, but results in much better sound. My guess is that most engineers don't want to go to the trouble. But the results are gratifying; the other solution that lazy engineers use is to just roll off everything in the upper frequencies - also not conducive to good sound -

over and out -

Link to comment
Share on other sites

True, the bad habit to butcher old recordings in order to make sound them 'new' was pretty common. For instance the first good sounding Ellington came out with the big RCA box set, the previous cd reissues were definitely awful.

I have my sound issues with that box. Jack the Bear is unlistenable. The first few discs in the box have a whistling sound near the end of each number-- apparently caused by noise-reduction. I'm sort of reluctant to mention this latter problem because I didn't notice it until someone on line pointed it out. Now of course it jumps out at me. Hope I'm not doing it to you. ( I didn't need anyone to bring the Jack the Bear sound issues to my attention.)

Link to comment
Share on other sites

True, the bad habit to butcher old recordings in order to make sound them 'new' was pretty common. For instance the first good sounding Ellington came out with the big RCA box set, the previous cd reissues were definitely awful.

I have my sound issues with that box. Jack the Bear is unlistenable. The first few discs in the box have a whistling sound near the end of each number-- apparently caused by noise-reduction. I'm sort of reluctant to mention this latter problem because I didn't notice it until someone on line pointed it out. Now of course it jumps out at me. Hope I'm not doing it to you. ( I didn't need anyone to bring the Jack the Bear sound issues to my attention.)

I admit I just A/B compared the Blanton sessions' discs to the previous reissue, and I believed it was a big improvement at times.

Anyway I want to ask Jim to delete these posts, and I'll go directly to an hypnotist. Maybe it's not too late.

Link to comment
Share on other sites

True, the bad habit to butcher old recordings in order to make sound them 'new' was pretty common. For instance the first good sounding Ellington came out with the big RCA box set, the previous cd reissues were definitely awful.

I have my sound issues with that box. Jack the Bear is unlistenable. The first few discs in the box have a whistling sound near the end of each number-- apparently caused by noise-reduction. I'm sort of reluctant to mention this latter problem because I didn't notice it until someone on line pointed it out. Now of course it jumps out at me. Hope I'm not doing it to you. ( I didn't need anyone to bring the Jack the Bear sound issues to my attention.)

I don't think that's noise reduction, Joe (if you're talking about that big 20+ disc box that is). Iirc, there was a problem with the original lacquers (or mothers or whatever) were used for it. It was discussed before, though maybe elsewhere; I'll try to dig up the reference. Now that older 3-disc Duke set from 1989 or so was no-noised.

Link to comment
Share on other sites

The whistling sound is an artifact of the original masters, not noise reduction. Lots of discussion about this in sound engineering circles a decade back. I think the consensus was it is a result of the master cooling process. Jack the Bear was corrected for the cd/dvd set called The Centennial Collection. I burned a disc replacing this track and am a happy camper.

Link to comment
Share on other sites

there's also whistling on some old blues masters - Viola Lee Blues, by Gus Cannon, has this - I took a good transfer and notched the very end, which is noisiest - it doesn't disappear, but it can be cut by about half, I would say, and if you're careful (and you need a very good, probably digital, notch filter that can go down to a width of 5-10 hz) it is relatively unobtrusive. I found this on a few Victors cut from originals that are on some very good older Japanese reissues -

Link to comment
Share on other sites

Since many re-issue projects (especially LPs) came from the marketing side of record companies, perhaps the objective in some cases was to make old material somehow sound more "Hi-Fi", emphasis on "High", meaning brighter sound.

It was certainly the marketers who came up with the idea of reprocesssed stereo -- breaking down honest, clear mono with slight channel-to-channel delay and different EQs. Bad in "stereo", worse if you had the misfortune to hear it folded back to mono. I remember some particularly bad Decca things in the '60s, the Lunceford band, for example.

Oh yeah--the old Decca Jazz Heritage Series! Great music, unbelievably dreadful sound! One of my nightmares is that the big fire may have left some of these as the only versions from some more obscure bands like Jan Savitt or Lucky Millinder. Brrrr.

greg mo

Link to comment
Share on other sites

popping in - as I'm in the middle of a major re-issue project right now -

one thing I've noticed is that old recordings regularly need a boost above 8K to "open" them up, get more needed highs - but in the process a certain shrillness results when frequencies in the 3-6 K range are boosted at the same time. What I do is 1) boost the old recording at about 8-11 K and 2) THEN go back and use a notch filter at the lower frequencies to get the shrillness out - it takes about 3 times longer per cut, but results in much better sound. My guess is that most engineers don't want to go to the trouble. But the results are gratifying; the other solution that lazy engineers use is to just roll off everything in the upper frequencies - also not conducive to good sound -

Thanks for that Allen. I've found something similar. I edit the sound of recordings using a Mac program called Amadeus Pro which has a 31 band equaliser. I've noticed that most of the harshness (or 'shrillness' as you put it) seems to be around 2.5k and 5k, so when I lower the level of these frequencies a little, the harshness is noticeably reduced. I've also found that old recordings on CD often benefit from "opening up" the treble by boosting the higher frequencies - so that's where it gets complicated and time-consuming.

Some things I would like to understand properly:

1. What exactly is a 'transfer'?

2. Why does the problem of harsh treble sounds on transfers specifically affect old 1930s/40s jazz? What is it about recordings from this time that causes the problem?

Link to comment
Share on other sites

I seem to remember commenting on this issue in last year or so. There's a distinctive "whistle" that is audible on lots of Victor 78s. It becomes more and more prominent as you go deeper into the record. Many people who have lost the ability to hear high frequencies don't notice it, but it's definitely there. In fact, one way to tell if an engineer has lopped off too many of the highs on a transfer from a Victor 78 is the absence of the whistle. Chuck is right; the "whistle" is in the original master.

I just purchased a cache of mostly unreissued (or poorly reissued) jazz 78s and began to play them at home last night. Once again, I was blown away by how good they sound. Yes, there's some surface noise at times, but if you listen through the surface noise, there's a wonderful life-like warmth and presence to the sound, almost like you're right there at the performance. The shrillness on many CD transfers is due to excessive filtering of the highs.

there's also whistling on some old blues masters - Viola Lee Blues, by Gus Cannon, has this - I took a good transfer and notched the very end, which is noisiest - it doesn't disappear, but it can be cut by about half, I would say, and if you're careful (and you need a very good, probably digital, notch filter that can go down to a width of 5-10 hz) it is relatively unobtrusive. I found this on a few Victors cut from originals that are on some very good older Japanese reissues -
Link to comment
Share on other sites

The "original" issues of many of the Columbia LP box sets include rather nice booklets and generally have much better sound than the later reissues. The later reissues, as you note with the Swing Street box, omit the original booklets, have the discographical information printed on the box, and, in some cases, have inferior sound.

I also think bad transfers. It happened with vinyl also. One example" In the mid 80s I bought a new copy of the the "Swing Street" 3 lp set on Epic. I thought the set had long been unavailable and had not been reissued. I thought I had a real find. Apparently I was wrong. The vinyl was not very typical of mid 80s Columbia vinyl - clean but very thin. Not the vinyl of the 50s and 60s. Also no booklet - just the discography info on the inside front box cover. Obviously not the original issue. The sound was pretty miserable. WAY to trebly and although vinyl, VERY digitally scrubbed.
Link to comment
Share on other sites

"Why does the problem of harsh treble sounds on transfers specifically affect old 1930s/40s jazz? What is it about recordings from this time that causes the problem?"

I believe it is due the fact that they used so few microphones to record in those days - so everything is much more complicated in terms of balance. There is probably a more complicated tech reason for this, but it seems that even two-track recordings with multiple mikes are easier to eq, post-production.

Somehow I have the feeling that the separation of instruments at the initial recording stage, even if instantly eliminated in the "live" mix, gives them more of their own isolated sound and makes them easier to adjust in the mass - while recordings with one mike and multi instruments seems to make it harder to isololate a single one in terms of band width -

hope that makes sense, as I'm up to my ass in messy 1920s blues recordings right now -

Edited by AllenLowe
Link to comment
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   0 members

    • No registered users viewing this page.
×
×
  • Create New...