Vinyl Master Requirements: What Every Plant Wants, and What Changes Between Them
Updated September 2026 · 6 min readThe most expensive mistake on a first vinyl project usually isn't the plant or the colour. It's sending a master nobody looked at with cutting in mind, hearing nothing for six weeks, and then finding out at the test pressing (TP) stage that the record is quiet, or that the sibilance on the vocal has turned into a rasp. By then the lacquer is cut, the plating is done, and somebody is paying for a re-master and a second cut.
A master made for streaming or CD is not automatically unusable. A clean one that hasn't been squeezed flat is often cut without a separate pass, and a good cutting engineer will make small adjustments as they go. The risk is in the ones that leave no room to work with: heavily limited, bass-heavy in the sides, or bright and sibilant, especially on a long side.
Most of that is avoidable, and it comes down to understanding one thing: there is no single document called "the vinyl spec". There's a set of requirements nearly every plant shares, and a set that changes from plant to plant. Knowing which is which tells you when to finalise your master.
The two kinds of requirement
Almost everything a cutting room asks for falls into one of two groups. The left-hand column below holds true wherever you press. The right-hand column is why you can't finalise a master against a generic spec and expect it to fit whoever quotes best.
| Parameter | Same at every plant | Changes from plant to plant |
|---|---|---|
| Master type | Vinyl-ready, not a streaming or CD master | Whether the engineer will adjust what you send, or cut it exactly as delivered |
| Peak level | Headroom below full scale, nothing clipping | The exact ceiling they ask you to leave |
| Low end | Centred and in phase. Out-of-phase bass moves the stylus vertically and can lift it out of the groove | The frequency below which they want it mono, often somewhere in the 150 to 300 Hz region |
| High end | Sibilance under control, and harder to track cleanly closer to the label | Whether the plant applies its own filtering as standard, and at what settings |
| Dynamics | A heavily limited master cuts quieter, not louder | How much the engineer will compensate before handing the problem back to you |
| Side length | More minutes per side means lower level and more surface noise | Suggested and maximum lengths, and how hard they push back when you're over |
| Inner grooves | Groove velocity drops toward the label, so bright or sibilant material is harder to track cleanly at the end of a side than at the start | Whether they'll flag a risky running order, or treat sequencing as entirely your call |
| Cutting medium | Either route produces the master disc your record is pressed from | Lacquer or Direct Metal Mastering (DMM). Not every plant offers both |
| Deliverables | Uncompressed WAV or AIFF, never MP3, plus a tracklist with sides, running order, durations and gaps | Preferred sample rate, and whether they want one file per side or separate files |
Ranges shown are what we see across EU plants, not a specification. Your plant's own figures are the ones that count.
Three rows in that table catch people out more than the rest.
Dynamics works backwards from digital instinct. Squashing a master flat before it reaches the lathe doesn't make the record louder, it takes away the room the cutting engineer needed to make it loud. The lathe has physical limits of its own, and it handles transients better than it handles a solid wall.
The inner grooves are physics, not folklore. The disc spins at a constant speed, but the circumference shrinks as the stylus moves inward, so the groove passes under the needle roughly two and a half times slower at the label than at the outer edge. The same audio is squeezed into less physical distance, and high frequencies are what suffer first: sibilance, cymbals, bright synths. Level makes it worse, but the driver is high-frequency content rather than loudness on its own. If one track on the record is going to fight the format, it does better early on a side than last.
Worth keeping that one in proportion, though. Sequencing softens the problem; side length is what actually determines how far into the difficult radius your music goes. A short side leaves the cutting engineer room to keep the programme away from the innermost grooves; a long one takes that room away and the last track ends up in the difficult radius whatever order you chose. Where exactly the line falls depends on the format, the speed and the material. Side length, speed, disc count and format are the four things that move this, and all four get decided in the format spec before anyone quotes.
The bass mono point is the clearest case of why plant-specific matters. Everyone wants the low end centred. Collapsing to mono at 300 Hz when your plant only needed 150 costs you stereo width you never had to give up, and you won't get it back once the lacquer is cut.
Why the final master waits for the cutting room
Put those two lists side by side and the sequence writes itself. The dependency is on whoever cuts the record, which is usually the plant's own cutting room but can equally be an independent mastering and cutting house you engage separately. Either way, if you finalise against one set of published requirements and then change route for price, lead time or capacity, some of that work may need revisiting. Often it is only the delivery format or a conversation between the two engineers. Sometimes it is a second mastering pass, and that isn't free.
Almost all of the work can start before you've picked anyone. What has to wait is the short list on the right, and it's short on purpose.
One requirement that isn't about the audio
Alongside the master itself, a plant needs to know it is allowed to manufacture the record. You have to own or have cleared the rights to everything on it, samples included, and that sits on your side of the line rather than with the plant or with whoever manages the project. Some EU plants go a step further and ask for mechanical rights documentation before production starts.
Handle it at the same time as the master, for the same reason. Sorting it out early costs nothing. A run held at the plant because a licence is missing costs the same slot in the calendar as a run held for a master that had to be cut again.
What we do about it
We don't publish our own audio requirements, and we're not going to. We don't operate a lathe, so we have no tolerances of our own to give you, and inventing a generic set would be worse than useless. It would produce masters that satisfy nobody's actual cutting room.
What we do instead is get the chosen plant's requirements in writing and pass them to you as part of your project spec, alongside their artwork templates, before you finalise anything. If your mastering engineer has questions for the cutting engineer, we'd rather the two of them talk directly than route it through us and lose something on the way.
If you don't have a mastering engineer yet, that's worth sorting before you have a release date to hit. We raise the rights question at the same stage, and for the same reason, but clearing rights stays yours to do.
A master is right for vinyl or it isn't, and the test pressing is where you find out. Getting the plant's actual requirements in front of your engineer before the lacquer is cut costs nothing. Finding out afterwards costs a re-master and a second cut. Microcut Studio handles that handover as a standard step on every project, along with the artwork templates for the format you're pressing.