Guides›Making beats from packs

Making two different packs sound like one record

Combining samples from different sessions without the seams showing. The problem is rarely the notes, it is the rooms.

Two samples in the same key at the same tempo ought to fit together. Often they refuse to, and producers assume the problem is musical when it is almost always spatial.

1.Why it does not glue

A recording carries its room with it: microphone distance, wall reflections, how long the space takes to decay. Put two samples recorded in different rooms side by side and the listener hears two places at once. They cannot usually name what is wrong. They just hear that it does not sit.

Four differences are doing it.

• Room. Different spaces, different tails. The biggest offender by far.
• Tone. Different microphones and preamps produce different frequency balances. One sample is bright, the other is dull.
• Dynamics. One has been compressed hard, one has not, so they breathe at different rates.
• Noise floor. One has hiss, one is silent. When the noisy one drops out, the silence is audible.

2.Fixing the room

You cannot remove a room. You can put both samples in a third one.

• Send both to the same reverb. One reverb, a send from each. A small amount is plenty. A shared space over the top is what convinces the ear that two things belong together.
• Shorten the wet one rather than wetting the dry one. If one sample is drowning in room and the other is bone dry, gate or trim the wet one. Matching downward is easier than matching upward.
• Or commit to the mismatch. Two obviously different spaces can sound deliberate if the arrangement treats them as separate elements rather than as one part. The failure mode is the half-measure.

3.Fixing the tone

• Match the dull one to the bright one, not the reverse. You can take brightness away convincingly. Adding it brings up noise and artefacts.
• Use one filter across both. Bus them together and roll off the top and bottom of the pair as a unit. Anything that happens to both of them reads as a shared origin.
• Saturate the pair together. One saturator on the bus, driven gently. Identical harmonic distortion on two sources is one of the strongest glue effects there is.

4.Fixing the dynamics and the noise

• One compressor across both. Not one each. A single compressor on the bus, moving two or three decibels, makes them rise and fall together.
• Make the slower one follow the faster one. If one sample has obvious pumping, compress the other into roughly the same behaviour rather than trying to un-pump the first.
• Add noise to the clean sample. Vinyl crackle or tape hiss, very low, across the whole thing. Noise reduction on the dirty sample will take the life out of it. A shared noise floor is the oldest glue trick in sampling and it still works.

Noise is the shallowest of these tools though, and it does less on its own than most people expect. What makes a soul sample sound vintage sets out what actually carries that impression, and why stereo width and transients do more than crackle.

5.The order that works

1. Get both in tune and in time first. No amount of processing fixes a pitch mismatch.
2. Bus them together.
3. One filter across the bus, top and bottom.
4. One saturator, gentle.
5. One compressor, two or three decibels.
6. One reverb send, shared.
7. A low bed of noise across everything.
8. Mute and unmute each in turn. If either sounds wrong alone now, you went too far.

The shortcut

If both samples came from the same pack, most of this is unnecessary, because they were recorded in the same room on the same day through the same chain. That is the practical argument for building a track from one pack before reaching across three.

When you do reach across, stems help enormously, because you can drop the drums from one sample and use the other, which removes the most obvious source of two rooms at once.