For those of you who want info quickly, here are the main points we’ll cover in more detail as the video progresses.
1. The master or masters made for streaming can be used for vinyl production, but it’s not ideal. Many of the choices made with streaming in mind will negatively impact the print quality.
2. The loudest the master should be is -12 LUFS - any higher and the needle will likely distort during playback, or need to be adjusted before printing, resulting in a truncated dynamic range.
3. Aside from classical music, the lows up to about 100Hz should be made mono with a gradual filter - otherwise the needle will likely skip or jump out of the groove.
4. Sibilance isn’t tolerated as well on Vinyl, meaning de-essing during mastering might be necessary.
5. RMS and Peaks need to be controlled - great changes in dynamics can cause the needle to skip.
6. And lastly, softer songs with less high-frequency info are best if saved for the middle of the record.
When mastering for streaming, heavy limiting and clipping are often used, mainly to get the level to -10 LUFS or higher.
Both limiting and clipping cause distortion, with the latter causing significantly more. By its nature, increased distortion occurs during the playback of a vinyl record, meaning a heavily distorted master will only sound more distorted when printed.
If you were to use the master made for streaming for this process, the print engineer will make some needed changes, including reducing the gain of the track to suit the medium better.
Whereas a heavily limited and clipped master can work well in streaming, most of the perceived benefit will be lost when preparing the master for vinyl. So, you’re losing dynamics, adding distortion, and likely altering the frequency response, all for that engineer to turn it down out of necessity.
Furthermore, a good deal of the harmonic distortion created from limiting and clipping will occupy the high mids and highs - an area that needs to be carefully controlled to avoid unpleasant distortion during vinyl playback.
So, you might save some time or money, but simply put, your record won’t sound as good as it can for anyone who purchases your vinyl.
Controlling the stereo width of low frequencies is a good idea in just about any master if you want a driving and focused low end, but it becomes a necessity when working with vinyl.
The needle moves laterally for mono information and vertically for stereo information.
Lows, having the majority of the energy, will cause aggressive vertical cuts or dips if they occupy the side image or have differing information between the left and right channels.
As the playback needle moves across these large dips, it can jump out of the groove.
This is especially problematic for genres with heavy bass or an aggressive kick like pop, edm, hip-hop, etc.
There are multiple ways to do this, but I believe a gradual HP filter on the side image is the best method.
Depending on how aggressive the lows are, use either a 6dB per octave or a 12dB per octave filter on the side image, and at the highest, set it to 100 Hz. In the past, I believed a higher setting was necessary, but it appears that today’s turntables are capable of handling some variation in the low mids.
Since we’re using a sloped filter and not a hard cut-off, the more problematic frequencies will have more attenuated from the side, whereas the lows that pose less of a problem won’t be cut as aggressively.
Classical music typically doesn’t need this filter due to its lower overall amplitude and less energy in this region, but if the composition has aggressive lows in the side image, a subtle cut may be needed.
Controlling sibilance is a surprisingly important aspect of mastering for vinyl.
To explain why, let’s take a look at the RIAA EQ curve.
When cutting a track onto vinyl, something similar to a tilt filter is used.
The lows are attenuated greatly, while the highs are amplified - this helps the needle, which is less sensitive to higher frequencies, properly cut these hard to replicate frequencies.
However, if the highs are too aggressive before this tilt filter, and too aggressive of a level is fed into the needle, we get distortion.
During playback, the opposite curve is applied to ensure a flat response; however, the distortion is already embedded into the vinyl. Any duplications of this original cut will have the same distortion.
This is why controlling sibilance, cymbals, or any other high-amplitude high-frequency instrument is important, both during mixing and mastering.
On the mastering side of things, using a low-distortion frequency-specific compressor is the best option.
I enjoy the Weiss DS1 MK3 - in particular, the de-esser that’s bundled with it and uses its algorithm.
You’ll achieve smooth de-essing more often than not; however, a softer knee, moderate ratio, and slightly longer release will ensure that the sibilance is reduced while minimal distortion is introduced.
Alternatively, an MB dynamics processor is a good option. So long as you can ensure that the majority of the peak is attenuated without adding distortion, and while retaining the original timbre of the track, then you should be fine.
This is similar to our side-image low-frequency issue, but presents itself differently.
In short, drastic changes in dynamics cause drastic changes in the lateral movement of the needle, although vertical changes can occur if the side image is high enough in amplitude. During playback, a consumer-grade needle may skip if it's shifted too greatly.
You might think this only applies to individual tracks, but it’s usually the transition from one song to the next that causes this problem.
If a song fades out, and the next starts with a massive hit or large dynamic shift, this may be enough for the needle to skip. A solution may be to alter the sequencing of the tracks, but a better option is to use low-distortion RMS compression to help control dynamics.
Since I’m not relying on limiting for controlling dynamics when mastering for vinyl, an RMS compressor, mixed with subtle peak down compression, is a great combination for controlling dynamics.
1dB of RMS attenuation and 1 to 1.5dB of peak down attenuation will be enough to keep large lateral or vertical cuts from occurring.
Additionally, this amount of compression shouldn’t be too audible and won’t greatly impact the perceived dynamics of the master.
I find this to be the most interesting part of mastering for vinyl.
As the needle moves around the record during playback, its speed remains the same. However, the circumference of the disc is reduced the closer the needle gets to the center.
This results in a lower velocity near the record’s center, and a lowered ability for the needle to reproduce high frequencies. On the outside of a 12” record, the needle is moving about 20 inches per second. By the time it’s at the center, it’s about 8 and 1/2 inches per second.
Subsequently, from the outside of the disc to the inside, there’s a drop of 3dB for 15 kHz and above.
If the track has cymbals, high-order harmonics and overtones, airy reverb, or any defining factor that occupies the highs, there will be a noticeable drop in amplitude.
With that in mind, many artists choose to put softer songs toward the record’s center. Maybe an acoustic track that mainly occupies the lows and mids.
Alternatively, you could boost the highs of the tracks you know will be sequenced toward the center, but as we covered earlier, this may cause unwanted distortion.
The best solution, but perhaps the most costly, is to create a shorter run time for each side of the record.
The highs will be retained up to about 18 minutes on a 12” 33 1/3 RPM record.
Up to 22 minutes will still sound good, but anything longer than that and you’ll run into issues.
So, if your album is longer than 44 minutes, it might be best to either cut a song, create a shorter version of a song, or create a double-disc LP.