Interview with Nick Dellow – Audio Engineer

In 2017, AVID Easy released a 2-CD collection of four Judy Garland LPs: A Star Is Born (Columbia, 1954), Miss Show Business (Capitol, 1955), Judy (Capitol, 1956), and Alone (Capitol, 1957) for which Nick Dellow was in charge of restoration and remastering.  The bonus tracks on this set included Garland’s four sides recorded at Columbia in 1953. The release got glowing reviews.  Now in 2019, Dellow is again at the helm of another 2-CD Garland set for AVID devoted to four other Garland LPs: Judy in Love (Capitol, 1958), Judy Garland at the Grove (Capitol, 1959), That’s Entertainment! (Capitol, 1960), and The Garland Touch (Capitol, 1962).  The bonus tracks on this one include “It’s Lovely To Be Back in London” (Capitol, 1957) based on the ultra-rare Capitol 78 rpm released only in England, the mono “Zing! Went The Strings Of My Heart” (Capitol, 1958), and sides from the animated feature, Gay Purr-ee (Warner Bros. Records, 1962).  Dellow has taken great pains in restoring these more than a half-century old tracks, including giving most of them the right pitch for the first time. Dellow lives outside of London, near Watford.

The Judy Room: Can you give us a brief biographical sketch?

Nick Dellow: I trained as an industrial journalist and landed up editing magazines on such esoteric technologies as engineering ceramics, carbon fiber composites, and superconductors.  I eventually ran my own company publishing market reports on advanced materials.  I closed this in 2013, after 30 years, and now concentrate on sound restoration, though I had been working as a professional sound restorer since 2008.

I began collecting vintage recordings when I was 18.  While my friends were playing in local rock bands, I was listening to 78s of British dance bands from the 1920s and 1930s. No wonder they thought I was eccentric!  I was lucky in that a well-known collector and dealer of 78s lived locally.  He organized record fairs (record bashes) and took me along, so I hit a rich source of 78s right from the start.  Pretty soon, I met other collectors and eventually I started interviewing the musicians that had actually been around in the 1920s making these records.  We are talking about the early 1980s.  I still have all the taped interviews and have slowly been digitizing them and uploading the results to my YouTube channel.

Ever since I began collecting 78s (and later on, as my interests broadened, LPs and 45s) I have been interested in the how these vintage recordings could be restored.  Restoration work appeals to me for two reasons – firstly, my love for the music contained within the grooves, and, secondly, I have always had an interest in technology (hence my former career in advanced materials) and I am fascinated by the technical developments that are continuously being made in sound restoration, especially with regard to software.

JR: Can you tell us a little about the software you use?

ND: I use a wide range of software, which basically falls into two main categories: restoration software and re-mastering software.  Restoration covers such processes as declicking, decrackling, and denoising, and re-mastering covers such processes as re-equalization. The DAW (Digital Audio Workstation) I tend to use is Sony SoundForge. It’s simple, straightforward and flexible.  I use quite a lot of VST (Virtual Studio Technology) plug-ins and they work very well with Sony SoundForge.  I have evaluated 100s of VST plug-ins and probably use around 20 or so regularly.  What I like about these is that you can evaluate trial versions for free and if you like them, then you can upgrade to the licensed version. Most VST plug-ins are reasonably priced.  I am also a fan of the iZotope RX range of noise reduction and audio repair software.  The RX spectral editor is an essential bit of kit for removing glitches. 

   

JR: You have now restored and remastered two 2xCD sets from AVID devoted to Judy Garland.  Can you tell us about each of these projects?  What were the main challenges you had in your work?

ND: Firstly, and most importantly, one needs access to the cleanest possible sources.  Secondly, the best possible transfers are taken by making sure that the cartridge and stylus tip match the characteristics of the record groove.  I have a choice of styli, cartridges, tone arms and turntables, though I tend to use a very well built but extremely simple turntable built by a British company called Nottingham Analogue, together with their 12-inch carbon fiber tone arm.  I also have a very nice ex-BBC Technics turntable that works on a completely different principle!  Having a choice of equipment is a necessity with vintage material, though I maintain that simple, well built units are better than all-singing, all-dancing gear.

Once the source material has been successfully transferred to the digital domain, the restoration process can begin.  Clicks and crackle are reasonably easy to remove from both shellac and vinyl sources as their characteristics are, for the most part, clearly differentiated when compared to the wanted music signal.  Hiss and other types of surface noise that one finds on worn 78s, LPs and other source material is more problematic as it resides within the music signal and it is often difficult for software filters to distinguish it from the music.  However, it is possible to separate out a fair proportion of this chaff from the wheat without dulling the music, though it can be rather time-consuming and one always has to be very careful not to go too far!

JR: How do such projects as this Judy Garland CD set for AVID get off the ground? Did you propose them, or did AVID?

ND: I proposed these projects, though usually, AVID’s own team decide which artist or artists is/are going to be covered by a particular CD set.  You might say that Judy Garland was a “special”, though of course nothing is carried through to production without it being commercially viable.  I could see that there was a need for reissuing this valuable material.  The big majors, by and large, aren’t that interested in reissuing vintage popular music, beyond the perennial favorites such as Elvis and The Beatles.  They are sitting on vaults of material I would love to be able to work on, but in the meantime, I will continue to work with the best source material I can find and restore it to the best of my abilities in the hope that it will provide enjoyment for others.  This is as much about dissemination as it is about taking pride in one’s work.

Needless to say, working on these Judy Garland CDs has been a labor of love.  What has made this a particularly rewarding experience is the fact that Lawrence Schulman has acted as a consultant on the project.  His advice has been absolutely invaluable. For instance, I could not have carried out the repitching without Lawrence’s help.

JR: After listening to so much of Judy Garland’s voice, how would you describe it?

ND: I have been listening to Judy Garland’s voice since the 1980s when a friend played one of her Capitol LPs to me.  I’m ashamed to say that all I knew about her before this was her performance in The Wizard Of Oz!  I collected jazz and I wasn’t that interested in listening to an “easy listening” singer, as I was led to believe Garland was.  They say that ignorance is bliss, but in this case, it was just, well, ignorance.

When I listened to the LP that my friend played for me – which happened to be “Judy In Love”, one of her best Capitol LPs – it was something of a revelation.  With music being such an important part of my life, such occasions carry a special significance.  I remember the same feeling – a shiver down the spine – when hearing, for the first time, soprano saxophonist Sidney Bechet play Petite Fleur or Bix Beiderbecke’s cornet solo on his 1927 recording of “Singin’ The Blues”.  Listening to music is first and foremost an emotional experience, and all these performers play with genuine passion and intensity.  It’s as if they have to let this passion out of their very heart and soul.  I’m more than happy to be the recipient, and recordings allow me to get my fix whenever I want!  For some reason, I have never been as keen on hearing performers play live, though obviously if I had a time machine the first stop off would be at Carnegie Hall on Sunday, April 23, 1961.  That is one concert I would have dearly loved to have been able to attend.

After hearing “Judy In Love” I started buying as many of her recordings as I could find.  I quickly came to appreciate the unique aspects of Judy Garland’s voice.  First, there is her beautiful tone, which developed over the years.  The later Capitol LPs, including those I have restored for the two AVID CD sets, capture a voice that had matured like a fine wine, with a somewhat richer-sounding timbre than the younger Garland.  It isn’t a “pure” tone, but it is all the more human for that.  The CBS TV Shows also contain wonderful performances, but I think they were tough going and took their toll, both physically and emotionally.  Despite this, even later on in her career, she can still knock me out with performances such as “How Insensitive”.

It seems to me, and I know I speak for many other fans, that Judy Garland projects her voice towards you and you alone.  She has a very empathetic nature, and the fact that she allowed that private, vulnerable side of her personality to be exposed is something that created such a heartfelt response from her audiences, even though, sadly, it didn’t do her any favors in her personal life.

Garland always strikes me as a powerful singer, and even on tender ballads, one has the feeling of a great deal of power held in reserve.  She sings with her whole body, starting – as with opera singers – from deep down at the diaphragm level.  On her CBS shows, it is fascinating to see how she opens up her throat by tilting her head back, and then to observe how she changes the shape of her mouth, the position of her tongue and even the proximity of her hand close to her mouth, especially on long notes, as if to funnel the voice.  In this respect, I often think of her as using her body as a musical instrument.  Some blues singers use their bodies in this way when they sing.  Ethel Merman did it to a certain extent too, though she has about as much subtlety as a sledgehammer on steroids.

Judy Garland is also a very intelligent singer and yet her approach is always warm and emotive.  She doesn’t “act” when singing the lyrics, she lives them and that gives added power to her voice.  And through her innate understanding of what is being conveyed by a song, she can emphasize a line or even a single word in a lyric and so give it a depth of meaning that perhaps even the songwriter didn’t envisage!  Oh yes, and I mustn’t forget to mention her delightful wry sense of humor, which often informs the songs she performs in a very positive way.

She is also often quite playful with the rhythm of a song – anticipating a beat and then going the other way and holding on until you almost feel she’s going to go too far, but then she lands right where she knew she wanted to.  In this respect, her approach is like that of a jazz singer. She could have certainly been a jazz singer if she had decided to go down that route.  Her ability to sing comfortably and with confidence across different genres is both a strength and a weakness.  It is a weakness because people love to categorize singers – jazz singer, pop singer, blues singer, or whatever – and tend to downplay the abilities of those who can cover more than one style.  So, therefore, few would regard Judy Garland as having specific abilities as a jazz singer….but why not?!  One thing I must add on this point is that she is definitely NOT an “easy listening” singer.  There is nothing “easy” – i.e. simple – about Judy Garland’s voice.

JR: Do you have a philosophy insofar as restoration?

ND: It’s a simple one. Respect the source material.  And don’t over-egg the pudding!

JR: What was the most difficult track to restore and remaster on the 2019 set?

ND: That’s a difficult one to answer as they all have their plus and minus points when it comes to restoration.  A couple of the tracks on “Judy Garland At The Grove” were tough, simply because this was a live performance album and sometimes there are drops in the signal, as well as other specific problems that need to be dealt with.  The tracks from “Gay Purr-ee” also required a bit of extra effort, but everything was “do-able.”  It just takes time. When you’ve had to try and restore early 1920s acoustic recordings from well-worn 78s, it’s a pleasure to work on transfers taken from early 1960s LPs, believe me!  Having said that, I very much enjoy restoring 78s and I am always learning. Audio restoration engineers, like the records, should never keep still!

JR: The sound on both AVID sets is amazing.  How is it possible to get such good sound from a “needle drop”, especially given the fact that there are some people who are highly critical of public domain releases because they are not based on original master tapes?

ND: Thank you very much for your kind words.  When you say “needle drop”, some people immediately throw their hands up and say “But it can’t sound as good as a master tape.”  It isn’t quite as straightforward as it seems though.

The master tapes should theoretically sound better, but I can point to many CD issues that use master tapes as the source material that sound duller than the LPs that were issued from the same master tapes!  This is often due to the tapes being incorrectly remastered to digital.  However, the sound quality also depends on the physical condition of the master tape used.  In fact, the tape source used when remastering is often not a “master” as such, but a second generation tape that was made during the mixing process, when the mixing engineer would adjust the levels of the original master tape or tapes (if multi-tracked) to achieve the desired sound, prior to further processing to produce the acetate master for LP production.

Moreover, the tape itself may have decayed over the years.  All magnetic tapes are prone to decay – due to changes in the magnetic properties of the tape over time and sometimes because the oxide (recording) layer physically spalls due to problems such as the so-called “Sticky-shed syndrome.”

The standard preservation protocol for magnetic tapes is to make at least two copies of the original tape, a master copy, and a “use copy.”  But you might be surprised by how many companies didn’t do this as a matter of routine.

Original LPs pressed using early ‘stampers’ and good quality vinyl generate a quality of sound that is almost invariably of a high enough standard to use as the source material when it comes to a CD reissue.  The audio information is there in the LP grooves – it is just a case of knowing how to extract it when transferring and remastering.  Selecting the right stylus type and size, the best cartridge (either moving coil or moving magnet), the correct equalisation curve and so on are just some of the keys that unlock the sound and ensure that the best possible transfers can be made using LPs as the source material (the same applies to 78s and 45s).  And then there are additional techniques that the remastering engineer can apply, such as re-equalization, but, as is always the case, these techniques need to be applied properly and with the greatest of respect for the original sound and the original artist.  Once in the digital domain, remastering becomes a much easier exercise, but one has to be very careful not to abuse the power one has at one’s fingertips!

JR: Given the richness of analog audio data that can be gathered on an LP vs. the 44.1 kHz allowed by a red-book standard CD, wouldn’t you have liked to be able to issue your restoration on a medium allowing high-resolution sound, such as SACD, DVD Audio or Blu-ray Audio?

ND: This becomes a very subjective matter and in this respect, there are no verifiable demarcation lines along which one can offer a clear answer concerning the vexed question of analog versus digital, or 44.1 kHz versus higher sampling frequencies.  However, empirically speaking, double blindfold tests prove time and time again that human beings cannot tell the difference between analogue recordings and digital recordings set at 44.1 kHz “Red Book” standard.  Ergo, CD quality is of a high enough level to fully capture the frequency response and dynamic range detectable by humans, even amongst youngsters whose hearing range is usually considerably better than those older generations who insist that analog is superior.

What does worry me is the proliferation of mp3s.  Here you are dealing with a whole different kettle of fish.  mp3 technology relies on compression to allow what would otherwise be large size audio files to be stored and processed with relative ease.  With compression, there is obviously going to be a trade-off between the amount of data generated and the sound quality of the results, and that starts the alarm bells ringing as far as I am concerned.  The loss in quality is generally imperceptible when the compression rate is low (i.e. 320 kbps bit rate), but the person or organization generating the mp3 can select the bit rate and there is no “red book” standard employed here!  Artifacts not present in the original recording start to become audible as the compression rate is raised, and there is a general loss of sound quality.  In many ways, it could be suggested that mp3 is the digital equivalent of the cassette tape – with convenience overriding sound quality.  Of course, there are those out there who think cassette tapes are the best thing since sliced bread.  Each to their own. The fact of the matter is that many people are nowadays regularly listening to compressed music across streaming services on “smartphones” that have a relatively poor quality sound.  OK, such audiences are potentially being exposed to a far wider range of music than was the case with previous generations of listeners, but I’m not so sure it is a good idea to spread things so thinly at the expense of sound quality.

JR: When do you know that you are done working on a track?

ND: That is a very perceptive question.  As a bit of a pedantic perfectionist, I am never really entirely happy with my restoration work, so it is tempting to keep on tweaking.  However, experience generally leads one to know when enough is enough and that to go any further would start to reduce the sound quality, not improve it.

There is only so much one can do with sound restoration anyway, especially if one always respects the original sound of the source material.  Of course, not respecting the source material allows the restoration engineer to over-process to his or her heart’s content and that can produce some truly awful results, with the “restored” tracks usually swimming in reverb and echo.  An analogy would be to restore an ‘old master’ oil painting by covering it with modern acrylic varnish.  We don’t want vintage material to look or sound shiny for the sake of it.  When an old master is properly restored, the layers of old varnish are carefully peeled away to reveal the original, and any damage is repaired where possible.  The process is exactly the same with sound restoration.

JR: On the new set, you have lowered the pitch on many tracks.  How does that process work?  And why are these tracks sharp in the first place?

ND: Lawrence Schulman and I established that most of Garland’s Capitol recordings were mastered sharp…even if they weren’t actually recorded sharp.  I don’t think there is anything particularly unusual about this.  Many times I have come across tracks on 1950s and early 1960s LPs that are out of pitch. Sometimes the problem is due to a technical issue with the tape machines or acetate master cutting machine at the mastering stage. But the fact that some tracks were mastered sharp was often the result of a deliberate policy carried out by record companies in order to “brighten” the sound, and that is especially true when it came recordings of popular music singers.  Judy Garland’s voice was very natural and unforced, so restoring it to the correct pitch has been an important aspect and has certainly helped to reveal the true beauty of her singing, whatever the track might be.

JR: Insofar as pitch is concerned, which is more challenging: LPs or 78 rpm?

ND: I would say 78s because the pitch can vary enormously and almost invariably requires some adjustment.  The pitch can even vary between individual tracks recorded during the same session!  There is a myriad of reasons why this should be, even down to the grease on the bearings on the cutting table being a bit stiff, or the wax master not being at the right temperature.  Each recording engineer probably had his favorite speed too!

To establish the correct pitch, one needs to first work out the musical keys that are being used by the band or singer during the recording.  Often there are clues.  For instance, if one plays a 78 of a 1930s dance band at 78 rpm and the main (home) key at that speed is B, you can be sure that the correct key is either Bb or C. Dance bands would rarely if ever record in the key of B, but because of the transposing instruments they used they often recorded in either Bb or C.

JR: Which is a better source: shellac or vinyl?

ND: Technically speaking, it is shellac.  A 78 has a large groove size and spins at a much higher speed than an LP, and one can obtain more information from a larger grooved record spinning at higher speed.  The downside is the increased surface noise of shellac-based 78s.  I say “shellac-based” rather than “shellac” because in reality 78s are made of a composite material that, as well as shellac, also includes mineral fillers that give additional properties such as increased hardness.  However, these mineral fillers are hygroscopic and so over time can expand and break through the groove walls on a micro-scale, leading to increased crackle and hiss.  This is particularly the case if the records are stored incorrectly, such as in humid atmospheres.  78s are also prone to damage by mold.

The physical manufacturing process for producing a 78 is very similar to producing an LP. A master (wax for 78s, acetate for LPs) is electroplated and this produces a metal master (with the grooves protruding).  This is then further electroplated to produce a “mother” (standard grooves), which is then electroplated again to produce a “stamper” (grooves protruding).  The stamper is then used to press the record.  I have owned metal mother 78s and the sound quality is utterly astonishing.

JR: In the age of MP3, do you think there are still people who care about the quality of sound?

ND: Every generation has people who care passionately about sound quality, though these will only ever add up to a tiny percentage of the general listening public.

Of course, there is sound quality from a technical aspect and sound quality concerning performance and repertoire, though the validity of both is partly determined subjectively.  In other words, it’s a matter of opinion!  I’m happy for people to listen to whatever floats their boat.  I am approaching 60, but I have no plans to join the ranks of those older generations who bemoan and denigrate the popular music that younger generations listen to.  Popular music, like popular culture, is never frozen in time.

Every generation has its own tastes, and the popular music they like will always range in quality from the humdrum to something superlative.  The best popular music from each generation transcends the ephemeral and becomes timeless.  That’s true for The Beatles and it’s certainly true for Judy Garland.  There is something absolutely timeless about her voice.   When you play a track from, say “Judy In Love”, to a young person today they don’t say “That sounds old fashioned”, they say “Wow, who’s that?”  I speak from personal experience on this point.  And it happened to me back in the 1980s when I first heard her voice on “Judy In Love.”  Then I was the one saying “Wow, who’s that?”.

JR: Are there restorers, living or dead, whom you admire?

ND: There are a couple of people who have been an influence.  The man who led the way in restoring 78s and whose work has withstood the test of time is the late John R.T. Davies.  He was a pioneer and an experimenter in the days before digital domains, overcoming seemingly intractable problems in order to extract the best quality transfers from 78s.  His work taught me to respect the source material.

I interviewed John in 2004 and asked him about his philosophy when it came to transferring and restoring 78s.  He said: “When you are re-mastering a nugget of sound, be it a record from 1930 or a cylinder from 1900, that nugget of sound is the important thing and should be disturbed as little as possible.  The performance is the most important consideration.  I want to retain as much as possible of the original recording, even if it happens to be in a rather low proportion to the overall sound on the recording.  I believe that if you, as it were, palm off to the public a sort of dumbed-down version of the original, beautifully cleaned but lackluster, you are not doing the music or its audience any favors at all.  In fact, you will drive potential listeners away from the music.”  I also asked John what was the most important piece of equipment he used when carrying out restoration work and he pointed to his ears.  I think that says it all really!

JR: Your source material – the 78 rpm of “It’s Lovely To Be Back In London” to name but one – is often very rare. What does your collection look like?

ND: I have a collection of about 9,000 78s and about 100 phonograph cylinders, ranging from early music hall artists to ragtime, jazz, dance music, and post-WWII jazz and singers.  I also have about 500 LPs.  To be honest, a lot of the recordings I own aren’t particularly valuable as far as their financial value goes.  I just love the music and I have always been driven to having the best quality sound, and that means, in general, trying to get hold of the original source material.

© 2019 The Judy Room & Judy Garland News & Events