In its broadest scientific sense, quantisation means taking a continuous value and making it fit against a scale of pre-determined values. In MIDI sequencers like Cubase, the timing of notes is quantised against the internal resolution of the sequencer, which is measured in 'Parts Per Quarter Note' (or PPQN), as you perform them. The higher the PPQN, the more accurately the timing of your performance is rendered. However, while the issue of PPQN was really important maybe 15 years ago, modern sequencers offer such a high internal resolution that nobody really talks about the resolution of their sequencer any more.
Although modern sequencers do a pretty good job of accurately representing the timing of your playing, even the best player will deviate from the metronome click when recording. So it can be useful to further quantise your playing against a much larger or smaller resolution (depending how you look at it), such as a quaver (or eighth note), enabling you to effectively take an out-of-time recording and make it completely accurate against the sequencer's precise musical time grid. There are both functional and creative uses to Cubase's Quantise features, which is what we'll be looking at in this month's technique article.
Let's start with the basics. Say you've played a MIDI part into Cubase (badly, as shown in Figure 1) and you want to simply tighten up the timing by quantising the start positions of the notes. In the Project window, you would choose the quantise resolution from the Quantise pop-up menu on the toolbar, select the MIDI part (or parts) you want to quantise, and select 'MIDI / Over Quantise', or press 'Q'.
You can also do this from the Key Editor, which features a Quantise pop-up menu on the toolbar that's linked to the one on the Project window (choosing a resolution in one window automatically updates the other with the same selection, for example). However, if there aren't any notes selected in the Key Editor, all the notes displayed in the editor will be quantised, unless the 'e!' Edit Active Part Only button is active, in which case only the notes in the currently active part will be quantised. You'll notice the resolution specified in the Quantise pop-up menu also determines how the grid is drawn in the Key Editor, which can be quite useful when you get into more complex rhythmic quantise patterns.
Once you've quantised your part, all the notes will move to the nearest beat specified by the resolution you chose from the Quantise pop-up menu. If you look at Figure 2, you'll notice that I quantised my original recording with a resolution of 1/16 (a semiquaver) — in order to preserve the correct timing of the music, it's important to make sure that you choose the shortest note used for the rhythms in your performance as the quantise resolution. The shortest notes used rhythmically in my performance were semiquavers, which is why I chose 1/16 as my resolution — if I had chosen 1/8 (as in Figure 3), you will notice that the basic rhythm of my original performance would have been destroyed.
Selecting the 'wrong' resolution isn't always a bad thing, though, as you could end up with a creative variation on your performance by over-quantising. Alternatively, if the shortest note is a semiquaver, you might want to tighten the timing a little by quantising with a shorter resolution, such as a 64th note, although there are better ways of achieving a more human feel with quantisation, as we shall see shortly. As a footnote, it isn't just straight note values that are available: if your rhythms are based on dotted notes (such as two dotted crotchets, or quarter notes, per bar of 6/8, for example) or triplets, there are suitable options also available in the Quantise pop-up menu.
On the subject of changing resolutions, remember there's nothing to prevent you from making selections of different notes in the Key editor and quantising them with different resolutions. This can be handy if you play a piano part with a slow-moving bass line underneath a fast arpeggiation for example.
Once you've quantised notes, you might want to change your mind and return to an unquantised version and try something else. Cubase's multiple undo feature can come in handy here, but what if quantising wasn't the last thing you did, or what if you quantised the notes in question last week and many different versions of the Project ago? Fortunately, Cubase offers a dedicated Undo Quantise command (which you can find in the MIDI / Advanced Quantise menu, although I would usually assign it to the 'U' keyboard shortcut, as it was in 'classic' Cubase) to return notes to their original positions, which are always stored in the Project.
This is pretty handy, but it's important to note that Undo Quantise always returns notes to the positions they occupied when they were first input into Cubase. This matters when you use Quantise multiple times, because if you play in some notes, quantise them, and then further quantise them again, Undo Quantise will undo to the original positions, not the intermediary quantised positions. If you want to reset the currently quantised notes as the original positions for a set of notes, you can do this with the Freeze Quantise command.
It's Not Just About The Start
Quantise Ends does exactly what you would expect, and moves the end of the note (leaving the start position where it was) to the nearest grid position specified by the current Quantise Preset. Quantise Lengths quantises the length of the note to the nearest value specified by a separate Length Quantise pop-up menu on the Key Editor's toolbar. By default, Length Quantise is set to Quantise Link so that the value is taken from whatever the Quantise menu is set to, although you can set this to an independent value if you wish.
It's important to remember that Quantise Lengths doesn't make all the notes the length set by the Length Quantise menu. This is possible with a command called Fixed Lengths, which uses the resolution specified in the Quantise pop-up menu and makes the specified notes the length set by that resolution. For example, if you want to make a set of notes precisely demisemiquavers in length, select 1/32 in the Quantise pop-up menu, select the appropriate notes if you're in the Key Editor (or none at all to process all the notes in the Key Editor) or the appropriate part in the Project window and choose MIDI / Functions / Fixed Lengths.
So far we've looked at a fairly functional use of quantisation — moving notes to rigid metronomic positions — but it's also possible to use other rhythmic templates and a variety of other options in Cubase's Quantise Setup window. To open this window, select either 'MIDI / Quantise Setup' or choose Setup from the bottom of the Quantise pop-up menu. The Quantise Setup window is where you set up the quantise templates we've already been using from the Quantise pop-up menu, and once this window is open, any settings you make automatically become the currently selected quantise template.
The Grid and Type menus at the top of the window allow you to define the type of quantise templates we've looked at already: Grid allows you to pick a note length (for example: 1/4, 1/8, and so on) to use for the resolution, while Type sets a modifier for the note length: Straight, Triplet or Dotted. Speaking of triplets, there are times when you might need to quantise against other divisions of a beat other than three (ie. triplets), and Cubase allows you to set up such patterns with the Tuplet setting, which is set to 'Off' by default.
Take the third Opus-90 Impromptu by Schubert, for example, where you have crotchet sextuplets (six notes divided into the space of a quarter note). To set up a suitable quantise template in Cubase you would set Grid to 1/4, Type to Straight and Tuplet to six.
There's also a Swing slider where you can specify how much of a swing feel (as a percentage) to apply to the quantise pattern. For example, say you have a straight-16th performance (much as in the example figures) and you want to inject a swing feel: set Grid to 1/16, Type to Straight, and add as much swing as you can handle. To carry out the quantise operation you can either use the normal commands or click 'Apply Quantise' in the Quantise Setup window — see Figure 4 on the next page.
As you were dragging the Swing slider, you might have noticed that the grid in the Key editor is automatically redrawn to show where the beats fall in the new quantise template. You can also make Cubase automatically apply the actual quantising to the notes while you experiment with settings in the Quantise Setup window by enabling the Auto checkbox just below the Apply Quantise button.
In addition to the Key Editor showing the quantise pattern in its grid, the Quantise Setup window also offers a graphical display of where the beats will be in the current quantise pattern, represented by dark blue lines. This is particularly useful for another Quantise Setup feature called Magnetic Area. Normally when you use Quantise, notes are moved to the nearest beats as specified by the quantise template; however, the Magnetic Area allows you to specify a distance from the beats in the quantise template (as a percentage) and only if a note is within the specified distance from a target beat will it be moved. The Magnetic Area is indicated on the Quantise Setup window's grid by a light blue region drawn around the darker blue 'quantise beat' lines.
The opposite to Magnetic Area is the Non-Quantise setting, which enables you to specify a distance in ticks (where there are 120 ticks in a semiquaver) from the target beat where notes within this distance will not be quantised. This is pretty useful for rendering a more natural-sounding performance, as you quantise only the notes that are badly out of time, leaving the ones that are almost on the beat alone. The Non Quantise distance is indicated on the Quantise Setup window's grid by light red regions that are drawn around the beats.
Another feature that can be useful to make your performances sound more in time without being completely mechanical is Random Quantise. Here you specify a value in ticks so that when a note is quantised to the nearest beat specified by the other parameters in the quantise template, it is offset by a random amount from zero to the value specified by the Random Quantise setting. This is similar to the final Quantise parameter, Iterative Strength, which is used in conjunction with a separate Iterative Quantise command in the MIDI menu. Iterative Quantise works similarly to normal Quantise except notes are moved closer to the quantise grid by the percentage specified by Iterative Strength. Repeated use of Iterative Quantise will keep moving the notes closer to the grid, so you can experiment with how relaxed you want to the timing to sound, and eventually Iterative Quantise will end up completely quantising the notes as if you'd used the standard Quantise command.
There are two other features in the Quantise Setup window that can come in useful. Firstly, the Move Controller option specifies whether controller data within a note being quantised should be moved along with the quantised note, and secondly, the 5.x Import button allows you to import a quantise pattern from an earlier version of Cubase that was saved as an old Cubase VST 5.x Part (.prt) file.
Once you have a quantise template you want to use again, you can store it as a Quantise Preset so that it appears in the Quantise pop-up menu by clicking the Store button in the Quantise Setup window. The Preset is automatically created (and selected) and is given a name by Cubase based on the current settings in the Quantise Setup window. If you want to change the name, with the Preset selected, double click the Presets pop-up menu in the Quantise Setup window, enter a new name in the aptly-titled 'Type In A Preset Name' window and press Return or click 'OK'. The currently selected Preset in the Quantise Setup menu can be deleted by clicking the Remove button.