Archives

To book Steve Smart for your mastering project, click here or contact Lynley on 02 9698 5888 or mastering@studios301.com

This is the third of a three part series:

• From the Desk of Steve Smart: Mastering for Vinyl – Part 1
• From the Desk of Steve Smart: Mastering for Vinyl – Part 2
• From the Desk of Steve Smart: Mastering for Vinyl – Part 3

Going back to your time as a vinyl cutting engineer, what was the day-to-day process?

First of all, when the engineer finished mixing, either they or the assistant would put together the many different takes of the mixes onto two reels of tape – for side 1 and side 2 of the LP. They put the gaps the way they wanted them, but there would be discrepancies between the individual tracks – the volume level and the individual sound/tone – which affected the continuity of the album. And that’s where the first mastering engineer truly was a master engineer. He would get the master tape, play the reel through, make some notes, set-up some equalisers and compressors for each individual track and then, with those settings in mind, and in real time, transfer or cut that tape on to a disc by means of a cutting lathe.

Would you change the settings on the mastering gear between songs or would you do a quick patch to other units?

Well, we had the luxury of four equalizers, where two equalizers were used for the operation of each song. One equalizer being for the music that went to the cutter head, the other equaliser being for the music that went to the brains of the lathe. This “brain” was called the preview, because the lathe had to mechanically know what was coming up at the same time as what was going into the groove itself.

So there was one bank of EQ (two units) for one song, and while that song was actually cutting, you’d set up the second bank of EQs. During the crossfade or the gap between songs, you’d simply crossfade between one batch of EQs to the other. Then, while that next song was travelling along, you’d set up the next bank of EQs for the song following.

It was very attention grabbing.  You had to set up your EQs and your compressors and you had to also drive the lathe at the same time. Fortunately, everything was on remote control to the central desks so you could keep it all within hand-grabbing distance.  But it was quite fast and furious and if anything went horribly wrong, that whole cut just simply went in the bin – it was ruined.

What would be the success rate for cuts generally?

Well, a good cutting engineer knew exactly what to do – he could play a tape, knew what potential problems may be laying in the music itself, and note the corrections that had to be made.  Let’s just say if there was a ballad followed by a big rock track, a cutting engineer would guesstimate how much real estate would be consumed by the various amounts of “energy” in each song. The ballad would take up less room, the rock track requiring a lot more, so you’d plan your consumption of the side of the record to accommodate. If you had, say, 25 minutes per side, you could get away with cutting that at -2dB, or -1dB if you are lucky, depending on what sort of music it was.

But you also did little test cuts of each track just to make sure that it translated to the actual lacquer and played back as you intended, because it was also the slight change of the physical playback of the groove itself that influenced it greatly.

How does the audio content affect the grooves on a record?

If it’s a mono signal, the groove remains at the same depth, but moves from side to side, known as a lateral excursion. If it’s in stereo, and the content of that groove exceeds a certain degree of phase, the groove depth (known as vertical excursion) can change anywhere between 10 to 50 microns. It’s always noted that in record cutting, a groove is half as deep as it appears wide (by physical design). So when you’ve got stereo information in low frequency content, you don’t really have a lot of depth to play with before you actually end up with a groove that’s so small, that it’s smaller than the point of the playback stylus on the turntable. So, that stylus will want to force itself towards the centre of the groove.

Conversely, if you’ve got a groove that’s too deep, then when in the cutting process, it can actually cut so deep that it can go into the aluminium substrate of the lacquer itself. That will completely sever the tip of the ruby stylus, which is very costly.

Essentially, the thing that dictates the depth/width of the groove is the amount of time that the music runs for, versus the song volume levels that you wanted to cut the record at.

Is there anything to bear in mind with the high frequencies when cutting?

It’s interesting because I think it begins when you are recording an actual track itself.  If you recorded a kick drum onto a tape machine at excessive level, it would start to overload, distort and bleed into the other tracks on the multi track tape. Also, there was a very strict level of what you could record a hi-hat at. The medium of tape itself dictated at what level you could record. Interestingly, by the time you push the mix up, you generally find that those things were all in context. Things were, I guess, forced to be where they were.

If the mixing engineer or producer decided that he just wanted the hi-hat a little bit brighter or louder on the day, it was then up to the cutting engineer to decide whether he could get away with that or not. If the high frequencies were too excessive when cut to the lacquer, it would saturate and distort when played back. Also, high frequency grooves on lacquers are very jagged and would physically tear when you separated the metal from the lacquer (during the subsequent plating process).

So you had to, through experience, make an educated assumption of how much you can get away with.  On some lathes, particularly the one that I’ve used and (I’ve still got a piece of the equipment from it), there were high frequency limiters installed for the sole purpose of protecting the cutting amps and the cutter head from overloading on high frequency. This is because the high frequencies demand the most amount of electric current from a cutter head.  It has to generate – it has to resonate – that stylus at an incredibly high rate for it to carve that high frequency signal into the lacquer.

But, you know, the art of a good cutting engineer was just how much you could get away with.

What other roles were involved in the process?

There was a whole team of people and excellent manufacturing facilities required.  I mean you had to have great quality lacquers, a great cutting lathe, a plating department…. And top quality silver, top quality nickel, and great quality vinyl to press records from…. Plus all of the fitters and turners that looked after the pressers and fed the steam lines. If everybody got it right, that whole company could produce a really high quality record.  And there were companies, for example, like Deutsche Grammophon who used to be just be renowned for cutting fantastic classical records. They used to have it down pat, you know, they just really knew their stuff. It became an art form and engineers / producers who mixed records sought out specific cutting engineers, because they knew that they could get a really, really good sounding result on to the record at a really good level.

Part of the level thing was in order to cheat the surface noise, which was the biggest drawback of vinyl.  It was that increasing surface noise each time you plated it, it just got louder and louder. And each time the dog sat near the record collection, you know, it just got dirtier and dirtier.

What about the vinyl wearing out?

The vinyl wearing over continual playing is an issue,  as is the static dust that accumulates on a piece of black plastic like vinyl. They just got worse and worse over time and so, you know, major record companies used to make a killing out of just repressing.  I think while I was working at CBS, I had to cut Bat Out of Hell  by Meat Loaf more times than I’d care to remember!

People kept buying it because it kept wearing out?

Well, it kept selling and it also kept wearing out and, you know, you can only get so many records pressed off the metal stampers (that pressed the vinyl) before they physically wear out.  They just get blunt, you might say, and the cut sounds distorted and noisy.

You know, with all that said that vinyl records are great, they’re possibly one of the most environmentally unfriendly music carriages of all time. They are a petroleum based extract.  The waste products that come from manufacturing a vinyl record are insurmountable.  Getting rid of a bad bath of nickel and silver isn’t cheap these days. Apart from that, they’re hardly portable and they’re noisy and obtaining a vinyl cutting lathe, obtaining the nickel and the silver and the processing that goes with it, the presses, it’s no easy avenue to get into. It’s quite expensive accumulating all of that sort of equipment and putting it all back together and having enough work to – supplied to you – to meet your demands.

Once you’ve mastered something for vinyl, what is the process to actually get it onto vinyl?

In our vinyl mastering process, we aim to have the clients totally happy with the sound of that song, and possibly with some information garnished from us, go to a reputable vinyl cutting house.

Once the master leaves 301, the complete process of vinyl pressing is a series of stages.  When something is mastered at 301, we supply the file, and the vinyl cutting engineer then uses a digital workstation to play that file back. He can simply lower the cutter head on to the lacquer and cut one side. This is the lacquer – well the record is actually two lacquers – Side 1 and Side 2. Those lacquers are then dipped in to a large chemical bath, which is charged with electricity, called an electrolysis galvanic plating process. The lacquer is dipped into the bath and coated with silver, then with nickel, growing the “plate” on its surface. This plate is then separated from the lacquer and is called a “father” (The lacquer is rendered useless and destroyed). The father is plated again, producing a “mother”, which can be played and checked. The mother is then plated again to produce a “stamper”. This whole process is repeated for the second side, then the two stampers are finally loaded into the pressing machine for bulk manufacturing of the vinyl.

People sometimes shortcut the three step plating process simply by generating a plate straight from the lacquer and putting that into the press. This is most commonly known as DMM cutting (Direct Metal Mastering).

The father and mother plating process is deemed necessary to mass-produce a record into thousands.  If you are only doing a 500 run, you can go for the single plating process where you simply grow the metal stamper straight on to lacquer and then bolt that metal stamper into the press and press from that.  That saves a lot of money, as it saves time and cuts down on waste.

I’ve noticed some records are heavy and some are light, why is this?

It depends on what sort of press is used. There were three pressing plants in Australia – EMI, Festival and CBS, and they used various methods for pressing (the CBS press where I worked was steam-rammed). You can hold the pressings from different plants up and you can definitely tell the difference between where they were pressed, just by physically looking at them.

Was one better quality than the other?

No, just different.  Absolutely just different.  Different ways of pressing the same thing. the CBS records were thinner and more flexible. The EMI and Festival records were much heavier.

What about coloured vinyl?

Standard vinyl is pressed using black plastic and the coloured vinyl can be whatever the client orders. When CBS was pressing coloured vinyl, they would have to change the bag of vinyl (that was feeding to the presses) to a particular colour. It would take a few attempts to get the colour to run clean (eliminating any black residue).

I used to like sneaking into the pressing factory when they were doing this, because the first records that were pressed were useless due to the remnants of the black vinyl in there.  So, the first twenty or so had streaks of black and you always ended with really cool designs in them. I souvenired a few of those in my time!

To book Steve Smart for your mastering project, click here or contact Lynley on 02 9698 5888 or mastering@studios301.com

Steve Smart is one of the few mastering engineers in Australia who started his career as a Vinyl Cutting Engineer (back at CBS records in the 1980’s). Based on his experience of seeing Vinyl come, go, and come back again, we asked him a few questions about Mastering for the format in the 21^st Century.

This is the first of a three part series:

• From the Desk of Steve Smart: Mastering for Vinyl – Part 1
• From the Desk of Steve Smart: Mastering for Vinyl – Part 2
• From the Desk of Steve Smart: Mastering for Vinyl – Part 3

To book Steve Smart for your mastering project, click here or contact Lynley on 02 9698 5888 or mastering@studios301.com

Vinyl vs CD vs Hi-Res files

The difference between vinyl and CD in the very beginning was that the CD was deemed to be indestructible and you didn’t have the surface noise of vinyl, you didn’t have that crackle. They were immune to dust because they were read by a laser and so all of the  – you know the comparisons – that’s why I think they were welcomed so readily ‘cause you could (sort of) be rough with them and abuse them a little bit. Also in the early stages, the dynamic range was far superior to vinyl and the frequency response was restricted, but also far superior to vinyl. It had a frequency of 20 Hz to 20 kHz. With Vinyl, you couldn’t really get away with too much low frequency and you couldn’t get away with too much mid frequency and so a lot of vinyl records don’t have a lot of bottom end.

Is there a general frequency range that vinyl can reproduce?

Well you can get away with 20Hz to 20kHz on vinyl, but the mechanical and physical constraints of cutting a 20 Hz frequency, is such a violent excursion of the groove on the record. 20 Hz on a groove produces all manner of distortion factors, just by the tracing of the groove and the tracing of the needle itself. Same for that fact with high frequencies. So if you brought the frequency range in a little bit, you could get away with a really nice sounding cut.  It’s said that in vinyl, all you really had to do is get the bottom end right and everything else would fall into place.

But in CDs, all you really got to do is get the top end right and everything else will fall into place.

Is that because there is so much more high frequency content available on the CD medium?

Yeah, a lot of people say that vinyl sounds warmer. You know, what is warm? I guess it’s a colourful explanation of the way something sounds. So in warmth, I’d say that there aren’t exceptionally high frequencies just due to the limitations of the vinyl medium itself. There’s more presence, and I think the mid frequencies are catered to a little bit better. Because most of the information is in the mid frequency area anyway, a lot of cutting engineers paid particular attention to the mid frequencies just to get them right so that it sounded like a really good, level cut.

The Loudness War

I think what’s happened is that CDs got out of control. Maybe it is due to the fact that the artist/label demands it to be louder than everybody else’s master. The whole loudness war started with the 7” single and in cafés and jukeboxes because every artist wanted their 7” single to be played back in the café louder than the other records. That’s the first loudness war. But the loudness of a record or a loudness of a CD doesn’t really matter on radio or television because they have multiband limiters to take it to a certain level anyway during broadcast. And so for that fact is iTunes, it stops at a certain level. If it’s too loud, it gets turned it down. If it’s too quiet, it get turned it up.

You mean the iTunes player?

Yes, iTunes playback. The whole loudness war got really out of hand, especially when there’s only a certain amount of level you can put on to a digital device, after that it starts clipping and goes into red distortion. Digital distortion isn’t as nice as analogue distortion. It tends to end up just being a crackly annoying sound. Analogue distortion can take on some sort of interesting facets where it becomes quite pleasing. So with everyone wanting their CD louder than the next guy’s, someone ended up writing software and implementing that into devices at the hardware level, which is called “Look-Forward Limiting“. This is where the limiter basically looks at the transient’s absolute peak, or the most significant bit of the program itself, and chops it off. It just squares it into a square wave, instead of a nice sine wave.

Hence the distortion?

Hence the clipping distortion. CDs just got louder and louder and louder and of course the one thing that really started to suffer was the music itself.  The dynamic range started to decrease significantly. That nice transient deluxe kind of quality of the whole high frequencies where the most energy is, started to just simply get squared off.

So CDs, I guess, now started to sound more and more and more horrible and horrible…. and horrible.

You mean harsh?

Harsh, hard, abrasive and very confronting. The basic dynamic in digital is on and off. It’s a word. It’s a 0 or a 1. The basic dynamic in analogue is loud or soft.  So there’s a big difference there between the two in a way that you’re listening to them. Either your ear is switching on or off or your ear is detecting loud and soft. It goes against the grain for the brain to switch on and off. It can slowly switch on and slowly switch off.

But I think that’s what’s encouraged this resurgence in vinyl music, apart from just the pure fascination of it, is where basically you do have limitations which you’re stuck to – if you make a record too loud, it physically won’t playback.  The grooves will make contact with each other and cause skipping and jumping so that record is basically a failure. The way to make a record sound loud is in sound, not just its pure level.

You mean generally, or on vinyl?

Particularly on vinyl and, I guess in generally, you can apply that as well.  You can make a record sound loud, without it actually being loud, just by the choice of frequencies that you apply to it.  Commonly, if you apply upper mids to a piece of music, it’ll sound louder.  It’s just the way the human ear works.  So you can only go to a certain level with vinyl records and then you physically can’t go any louder. Vinyl records are pretty much generally within a couple of dB, all very much the same level.  And the beauty of a good cut and a bad cut is in the way it sounds, not actually it’s levels.

But unfortunately what’s going on with CDs of late has now been misconstrued. What is a good CD, is it loud or soft? Because it’s irrelevant if it sounds good or not.  It’s whether it’s loud or not.  I’ve been asked as a mastering engineer to just smash it. And, as a mastering engineer, that’s almost like destroying your work.

Limiting

You do what you’re instructed to do by the client, within your recommendations. But then if they bring out a CD that’s had look-forward limiting applied in the recording process and look-forward limiting applied in the online mixing process and look-forward limiting applied to the mastering process, versus something that hasn’t had any compression or limiting at all, it’s almost impossible to match that without destroying the pure sonic integrity of the music itself.

You know, there’s a lot of CDs out there that I’ve bought because I like the music, but I don’t play very often.

Because they’re too loud?

Well they’ve been, I guess, destroyed by this ever increasing need to have a loud CD. I mean, I can to listen to it but I can’t exactly be too close to the speakers. That’s probably being a little too fussy, but I’ve got to pretty much ”put the stereo on and go in the backyard” because it’s just too annoying to be too close to the speakers.

But you know, when you get a vinyl record, you are almost drawn to the speakers. You’re invited in to what’s being played for you.

So all things being equal, does vinyl sound better than CD?

I think vinyl does sound in some ways better than CD because of the limitations that the medium applies on the music. But CDs can – and do – sound better than vinyl if they are treated technically properly.

Admittedly, I’m not a big fan of the 44 kHz / 16-bit sound. I don’t really think, that in its vocabulary, it truly understands harmonics. I’ve recorded pianos on analogue and I’ve recorded pianos on digital.  And the analogue recording sounds like a piano.  The digital recording sounds like a keyboard.

It’s the same microphone, at the same level – I remember reading somewhere, some bizarre misconception that they didn’t really want the public to have any more capability of digital recording because it was getting far too close to the realms of military spec!

Really?

So they restricted it – Well, apart from the obvious case that you can’t really fit that much data on a one-sided 700MB-ish disc, it can only be done either at the sacrifice of the sampling frequency or at the sacrifice of the bit depth.

So there was an equal trade-off between 44 kHz, which was deemed acceptable and 16-bit, which was deemed acceptable for digital.

So what about 24/96 or any variation thereof, as opposed to vinyl?

I much prefer the sound of 88 kHz / 24-bit. I think that just seems to sound a little bit better than 96 kHz / 24-bit.  To me anyway, sitting in here, I just find that there’s a little bit more subtlety about the recording itself at the 88 kHz frequency.

Why is that?

I’m not quite sure whether what I’m hearing, but there does tend to be a coldness about the 192 and 96 kHz sampling frequencies.  Whether it’s paying too much attention to the real sound and not masking some details of it, and you’re actually hearing too much information, or – I’m just not sure, but 88 kHz / 24-bit is capable of recording and playing back natural sounding instruments like trumpets, and particularly wind instruments, very well. You end up hearing a lot of that natural distortion and intermodulation that you get from trumpets and trombones.

In the harmonics?

Absolutely.

So analogue doesn’t necessarily sound better than digital?

I think that, you know, it’s a personal thing.  When you’re listening to vinyl as versus CD, vinyl has more inherent side effects to the music than the CD.  If the digital information is properly recorded and properly mastered and played back with a much more musical and sympathetic approach to doing the job, rather than actually just flippantly slamming it down, CDs can sound better than vinyl. In fact, they do.  But, you know, it’s just simply the constraints of the vinyl medium that prevent you from overdoing it, in which case, you know, there’s quite a lot of CDs out there that are overdone.

As in there’s too much room for abuse with what you can do with it?

Absolutely.  There’s just far too much room for error. There is no room for error in vinyl, either it’s a good sounding record or it’s shit. And it’s quite clear and obvious.  If there is a problem, physically it will not playback correctly and therefore it probably won’t get through quality control. But who’s to say what’s deemed acceptable in the days of CD mastering.  There’s very little quality control.  Well, [they’re] certainly not played and checked by someone when the CDs are being pressed.

Is that why we listen to our masters in their entirety before we send them out?

That’s correct.

However….

In saying that, I think that CDs can sound very much like we want them to sound.  Warmer, much more in characteristic with, the sound that you get from a vinyl record.  A mastering engineer who has actually cut records for a long period of time and who understands the process and has the right equipment, whether that equipment be analogue or digital, can master a record sensibly and not make it harsh and un-wanting.

Because of their experience more so than anything else?

Yes, experience.

When you’re mastering nowadays for vinyl as opposed to CD, are you taking a different approach to each master?

Yes, definitely. There’s a requirement of the way people want their stuff to sound as far a CD is concerned.  And then there’s the complete no-go zone for mastering it for vinyl, where you just can’t get away with that sort of stuff.  So, the difference between mastering specifically for vinyl and CD is that you have to pay very strict attention to the low frequencies. You have to whittle out all of that dangerous overly wide stereo low frequencies and you have to pay serious attention to the high frequency transients and make sure that they’re not too fast or not too abrasive or don’t contain too many extraneous unwanted high frequencies. So you, musically I suppose, smooth it over a bit and make it a little bit softer.  There’s also no need whatsoever to apply any of the digital gain structuring, such as the look-forward limiting to turn it up because it’s completely useless, because there’s only a certain level you can cut that vinyl record out at, which is dictated by its duration.  So then comes the artistic and loving application of compression. Whether that be analogue or digital, whichever you choose.  But in good old school compression and good old fashioned limiting, it makes up this lovely musical piece of mastering.  But I tend to utilise the approach of mastering for vinyl, and mastering for CD, in everything I do.

Either it sounds right or it sounds harsh and brittle.  And depending on the original source recording, unfortunately some of the recordings you get aren’t really done to their best possible advantage, but you get what you’re given to work with and you try and bring the best out of that as you possibly can, with the equipment you’ve got.

When you’re mastering for CD, you’re pretty much hearing in the studio how the CD is gonna end up, right?

Very much. I mean there’s not gonna be much difference between what you’re hearing in this studio and what you’re hearing at home. CD replication and duplication is so good these days that what you give is what you get.

How about vinyl?

With vinyl, we try and make the assumption with garnered knowledge as to what’s necessary to get that song onto a piece of plastic and playing back the best it possibly can.

Is there a bit of imagination as to how ‘what you do now’ affects the final Vinyl pressing?

Read Part 2 of this blog series here

[Steve Smart is one of the senior mastering engineers at Studios 301]

To book Steve Smart for your mastering project, click here or contact Lynley on 02 9698 5888 or mastering@studios301.com