Why Two Copies of the Same Record Sound Different

You have three copies of a record you already love in front of you. An original pressing at a price that makes you stop and think, a reissue from a decade ago, and a new 180 gram remaster with a sticker explaining how good it is. Ask which one to buy and you get a dozen confident answers citing a dozen different pieces of evidence.

They conflict because most of them are arguing about the wrong object. Nearly everything people cite is printed somewhere on the record: the year, the label, the catalog number, the weight on the sticker. Nearly everything that changes what you hear is not printed anywhere, because it happened in a room with a lathe in it.

Two copies of one title differ because of mastering first and manufacturing second, in that order and not close. Here is what happens between a master and your turntable, and how to tell what you are holding.

The Chain, and Where the Differences Get In

A record starts as a master: a tape, or since the mid nineties, a file. A cutting engineer plays that source into a lathe, whose cutting head carves a modulated spiral into a lacquer, a soft nitrocellulose coating on an aluminum disc. The lacquer is silvered and electroplated, and the plate comes off as a negative called a father. The father is plated to make mothers, the mothers to make stampers, and a stamper in each half of a press squeezes a puck of hot vinyl between them.

Four stages, and they do not contribute equally.

• The master and the cut. Almost all of the difference lives here. Two engineers given the same tape will set level, equalization, compression and bass differently, and hand you two audibly different records without either doing anything wrong.
• The lacquer. Neutral when it is fresh and plated quickly. It starts changing the moment it leaves the lathe, which is why plants plate within days rather than weeks.
• The plating. Each generation is a copy of a copy, and each one adds a little noise. Small amounts, not zero, which is the whole argument behind one step pressings.
• The press. Stamper condition, vinyl compound, cycle time, cooling, and whether anyone looked at the result. This stage cannot change the mastering, only how faithfully you receive it.

Most pressing arguments resolve themselves against that shape. An original that sounds better is a mastering difference credited to the year, and a noisy copy is a pressing problem blamed on the mastering.

What the Cutting Engineer Is Actually Up Against

Everyone calls cutting for vinyl a compromise, as though the engineer were grudgingly degrading something. It is better understood as a set of physical constraints that get solved, the way a mix gets solved.

Side length against level. A twelve inch side has a fixed amount of space between the lead in and the label. Loud music needs wide groove excursion, and wide excursion needs room between adjacent grooves, so an engineer can have a long side or a loud side and not both. Under about eighteen minutes there is room to work. Past about twenty two the level comes down, and the bass comes down with it, because bass demands the most excursion of anything on the record. This is why a beloved album sounds thin on one issue and full on another. Nobody changed their mind about the bass, they changed how much music went on the side. It is also why a double LP cut at 45 rpm exists: more groove per minute of music lets the engineer cut hotter, at the cost of two extra side changes and a higher price.

The inner groove. The platter turns at a constant rate, so by the last track the stylus reads less than half as much groove per second as it did at the outer edge. The same waveform goes into less space, the curvature tightens, and a stylus with a real tip radius eventually cannot trace it. That is why the quiet closer at the end of a side is a courtesy from whoever set the running order.

Bass has to behave. Out of phase low frequency content moves the cutter head vertically, which lifts it out of the lacquer and later makes your stylus jump, so engineers sum the bass to mono below somewhere around 150 to 300 Hz or limit its vertical component with an elliptical filter. A record with a wide, phasey low end on the digital version comes back more centered on vinyl.

The Original Pressing Question

The mistake is treating original pressing as a synonym for better. It is appealing because it is often true, it is cheap to verify, and it flatters whoever already paid for one. And it is often true for a real reason: an early issue was usually cut from a tape a generation or two from the master, by an engineer who had heard the record in the control room, when everyone still expected vinyl to be the version people owned.

Here is where it falls apart. Original is not one object. A record released in 1969 may exist as a US cut, a UK cut, a German cut and a Japanese cut, all from the first year and all different, because each territory cut its own lacquers from whatever tape copy it was sent. Within one territory there are often several, because lacquers get recut when stampers wear out. And your first year copy may have come off a stamper on its last legs.

Then there is the tape. The master you are romanticizing is over fifty years old, and some stock made from the mid seventies onward sheds its oxide onto the heads and has to be baked before it can be played at all, so a careful transfer done in 2015 may be the best that tape will ever give anyone. The year on the label does not tell you whether an original is the better copy, because it is not the year that decides. The mastering does, and it usually has a name attached if you know where to look. I do not pay the original premium by default, only where the cut has a reputation of its own.

Reissues Cut From Digital Files

Many reissues are cut from a digital file, and so is nearly every new record in the racks, because almost everything recorded since the mid nineties was mixed digitally and there is no analog master to go back to. The all analog chain is a specialist product, not the normal case.

What matters is which file. A 24 bit transfer of the original tape, cut by an engineer working from it, is a different proposition from a lacquer cut off the sixteen bit master a label made for a CD in 1987, with that decade's ideas about level and brightness baked in permanently. Both are digital and they are nothing like each other. The trouble with eighties reissues was that nobody remastered anything.

The trade has not always been straight about this. In 2022 Mobile Fidelity acknowledged that titles in its One Step series, marketed in a way most buyers read as fully analog, had a DSD step in the chain, and changed its labeling. Those records sounded the same the day after the disclosure as the day before.

What you are owed is the information, and the reissue programs worth your money print it on the jacket or the hype sticker: the source, the engineer, and where it was cut. Analogue Productions, Blue Note's Tone Poet and Classic series, Rhino High Fidelity, Speakers Corner and Craft's Small Batch have all worked that way. Buy the practice rather than the name, because a series can be retired and the disclosure is the part you are actually paying for.

The 180 Gram Question

A 180 gram disc is more rigid than a 140 gram one and resists warping better in storage. That is the complete list of what the weight buys you.

Weight does not touch resolution. The groove is cut by the stamper, and the stamper does not know how much vinyl is behind it. Depth, geometry and the modulation in the walls are identical at 140 grams or 200. Weight cannot improve a master, and it will not rescue a lacquer cut from a bad file.

There is a small argument on the other side that nobody puts on a sticker. A heavier puck needs a longer press cycle and more cooling, so a plant running behind on heavy vinyl produces non fill, a gritty crackling where the vinyl never quite reached the bottom of the groove, and dish warps from discs pulled before they cooled. Weight is on the sticker because it is one of the few things about a record that can be guaranteed in advance.

Half Speed, One Step, and the Audiophile Reissue

Half speed mastering. The source and the lathe both run at half speed, so a 20 kHz cymbal reaches the cutting head as 10 kHz, where the head has more headroom and better phase behavior and the whole system is doing something easier. That advantage is at the top end, which is where a real time cut asks the most of the cutting head. What it cannot do is improve the master, and it costs something: the engineer is not hearing the result in real time, and the low end is written an octave down where a cautious engineer tends to hold back. That is where half speed cuts get their reputation for coming out light in the bass. Abbey Road's half speed series is cut by Miles Showell and labeled as such, so you know who did it and how before you buy.

One step pressing. The normal route is father to mother to stamper, three generations between the lacquer and the disc. One step uses the father itself as the stamper, so the vinyl sits one generation out instead of three, and the gain is the two generations of plating it skips. The cost is that there is only one father, it wears like any other stamper, and when it is done the edition is done. That is the real reason those releases are limited and expensive, and it is an honest one.

Neither process adds anything that was not in the master. Both lose less on the way out, which is worth paying for only when the master deserves that much care.

Stampers Wear Out, So Copy Number Beats Pressing Year

A stamper is nickel being slammed into hot vinyl thousands of times. Every stamper has a working life, set by the vinyl compound, the press and the plant, and past it fine detail starts going and the noise floor starts rising, at which point a fresh stamper comes off a mother, and a plant chasing a hit against a deadline has been known to push past that. So the copy that matters is an early copy off a healthy stamper. A 1973 pressing made in the fifth month of a long run sits further down the stamper run than a 1978 repress cut from a fresh stamper off the same mother. The label looks older, the stamper was more tired, and nothing on the sleeve tells you so.

Late stamper copies get lumped in with dirty ones, and the trick is separating three things that all arrive as noise.

What it is	What it sounds like	Does it come off
Dirt	Broadband, moves around from play to play, and worse in some passages than others	Yes. A wet clean fixes it
A pressing fault	Repeats in the same place every rotation. A tick every 1.8 seconds at 33 rpm. Non fill arrives as a gritty, fizzy sound, usually in the first or last minute of a side. An off centre hole is pitch wobble once per revolution	No. It is permanent and cleaning will not touch it
Stamper wear	A lift in the noise floor with softened transients and a dulled top end, uniform across the whole side rather than in patches	No. It does not improve after cleaning

Dirt is broadband, moves around from play to play, is worse in some passages than others, and comes off. A wet clean fixes it, and what that actually takes is a short article of its own.

A pressing fault repeats in the same place every rotation. A tick every 1.8 seconds at 33 rpm is damage or a defect in the disc, it is permanent, and cleaning will not touch it. Non fill is the other pressing signature: vinyl that never fully filled the groove, leaving the stylus riding a wall that was never properly formed. It arrives as a gritty, fizzy sound rather than a single tick, usually in the first or last minute of a side where the vinyl had the furthest to flow. An off center spindle hole belongs here too, arriving as pitch wobble once per revolution and obvious on solo piano, and weight and pedigree are equally irrelevant to it.

Stamper wear is the one people misdiagnose as dirt. It is a lift in the noise floor with softened transients and a dulled top end, and unlike dirt it is uniform: the stamper wore over its whole face, so it costs you across the entire side rather than in patches, and it does not improve after cleaning. If a record is clean and still hazy in a way another copy of the same pressing is not, you are probably holding a late one. On a great many British records the dead wax will tell you how far down the stamper run you are.

How to Read a Dead Wax

The run out area between the last groove and the label is where a record tells you what it is. Reading it takes about a minute, it is free, and it is the only part of this you can do standing in a shop.

Hold the record under a light at a low angle and turn it slowly, looking for scratched and stamped characters in the blank vinyl. Telling those apart is most of the skill. Hand scratched characters went into the soft lacquer at the lathe, by the person who cut it, and initials, symbols and jokes all live there. Stamped characters were applied mechanically, usually catalog and matrix identification and the plant's own codes.

Read the matrix number first. It identifies the side and the specific cut, and a trailing digit or letter usually counts the recut, so a side carrying a 2 or a 3 is not the first lacquer. Then look for a cutting room or an engineer: RL for Robert Ludwig, BG for Bernie Grundman, a Sterling Sound or Masterdisk or TML stamp, a Porky or a Pecko Duck from George Peckham in London. A name lets you tell apart two issues the catalog number says are identical. Robert Ludwig's first cut of Led Zeppelin II was hot enough to make some period turntables mistrack and was recut quieter. Both say Led Zeppelin II on the label. They are not the same record.

Some plants encode the mother and stamper numbers in a letter cipher. EMI used the word GRAMOPHLTD, each letter standing for a digit from 1 through 0, so a string of those letters after the matrix says which mother and how far into the stamper run your copy sits. On a great many British records, the copy number question is answered right there in the vinyl.

The limits are worth stating. The dead wax identifies the lacquer, not your individual disc, so it cannot tell you how many copies came off your stamper or whether the source was tape or a file. It turns a guess into a lookup, so do the lookup: type the run out string into Discogs, where versions are separated by exactly these characters. That is how you match the object in your hands to a specific entry.

The Short Version

Buy the mastering, not the year and not the weight. If you can learn who cut a given issue and from what source, you know more about how it will sound than the pressing date, the label design and the gram count put together.

For a record you intend to keep, find the issue with a named cutting engineer and a stated source, and prefer a shorter side or a 45 rpm cut on anything loud or dense. Nothing about the year tells you whether that issue is a modern reissue or an original. Buy an original when it has a mastering reputation of its own, or when the reissue is silent about its source, but not as a default, because the premium often buys you a late stamper.

Ignore 180 gram as a quality signal. Treat digital in the chain as neither a defect nor a virtue, and ask which file rather than whether. Treat half speed and one step as ways of losing less on the way out, worth paying for only when the master deserves it.

Before you conclude a copy is bad, clean it properly and keep it stored so the next one does not go the same way. And before you blame any pressing, make sure the system in front of it is not the variable, because the phono stage moves more than most people expect and it is the first thing to touch a signal this small.

---

Or Bring Both Copies Here

The listening room at Free River Studios, 215 E. Lewis St #301 in Livingston, has a deck set up properly.

Bring both copies, and tell us what they are when you book so we can have the deck warm. It is free and it is by appointment, and you get to keep the answer.