Final Thoughts

I am really pleased with how my individual track has turned out. I think that the final mastered mix has a big 80's style sound which is both strong and clear. The mastering process has given my song a professional sounding polish in my opinion. The final mastered track is shown below:

Final Mastering

I mastered 'Poker Life' by inserting the following plug-ins into the stereo output in the following sequence:

1. Multiband compressor.
2. Stereo Enhancer
3. Maximiser
4. Limiter
5. UV22HR

Multiband Compressor

I solo'ed each band in turn and moved the edge of each band left and right while listening to the track being played. The top band was set to mainly filter the very high hissy sounds (e.g. top end of cymbals). It is these sounds that add sparkle to the mix. The bottom band was set to mainly filter the kick and low bass synthesiser sounds. The low-mid band was set mainly to capture the snare as I wanted to be able the emphasise the snare hits to ensure that the strong rhythmic feel of the track came through in the mastered mix. The bottom end of the vocals also appeared in this band. The remaining high-mid band filtered the synthesiser sounds and the higher vocal frequencies. I used a compression ratio of 2:1 with a medium threshold for the 1st and 4th bands. I also increased the gain on the 1st and 4th bands to compensate for the medium compressor threshold. This had the overall effect of producing a strong bass without it becoming muddy. I used a ratio of 3: 1 and a higher threshold value for bands 2 and 3. This allowed the vocals and synthesiser sounds to sit well together in the mix without seeming to be over-powered by the strong bass synthesiser and kick. I slightly increased the gain on band 2 to further emphasise the snare hits.

Stereo Enhancer

The stereo enhancer adds width and delay effects to the mastered track. The overall width control allowed me to give the track a larger sound stage. The small delay also gives the track a big 'larger than life' 80's vibe which I wanted. 

Maximiser

For commercial reasons, it is important that the track be as loud as possible when mastered. I used the maximiser to increase the overall level of the stereo output. The maximiser can also be used to add some harmonic distortion via the Optimise control. I kept the optimise control turned down quite low to ensure that the clarity of the track was not lost.

Limiter

The limiter was used to prevent the stereo output from exceeding 0dB (which would introduce undesirable clipping distortion). I used the input control to drive it reasonably hard so that it was squashing the output signal by 2 dB. I found that this helped the guitar solo stand out in the mix.

UV22HR

This is used as the last plug-in in the mastering chain. It dithers (i.e. randomises) the least significant bit of the digitised waveform to reduce harmonic distortion due to rounding errors when the analogue waveform is re-created from the digital output data.

Multiband Compression

Multiband compression is often used in mastering to give a final polish to the overall mix. It turns out that using a full-band compressor to compress a complete mix is not ideal because the low-frequency sounds in the mix will determine the compression applied to everything else. The kick drum and bass line would probably dictate how the mix will be compressed if you used a full-band compressor.

A multi-band compressor comprises a set of filters that splits the audio signal into several frequency bands. After passing through the filters, each frequency band is fed into its own compressor, after which the signals are recombined. The main advantage of multi-band compression is that a loud event in one frequency band won't trigger gain reduction in the other bands. For example, a loud kick drum will not pull the whole mix down with it, only the low-frequency sounds (kick and bass instrument) will be compressed, leaving the mid-range and high frequencies untroubled. Similarly, a loud event in the mid-range won't affect the extreme high or low frequencies for the same reason.

The 4-band multi-band compressor in Cubase allows you to control the range of frequencies in each band to be compressed, the type of compression to be applied to each band (i.e. the threshold, ratio, attack and release) as well as the overall level of gain to be applied to each frequency band. A multi-band compressor is therefore a very powerful tool that can provide an improved finish to the final mastered track. Using a multi-band compressor's controls appropriately is absolutely critical. A multi-band compressor can ruin a mastered track if it is not used correctly.

Mixing and Tidying Up.

I wanted more control over the drum sound from Cubase's Groove Agent One plug-in. I discovered that if the Groove Agent One plug-in is created via VST Instruments option on the Devices menu, then up to 16 stereo output channels can be assigned to it. I assigned and configured 5 outputs as follows:

1. Kick
2. Snare
3. Toms
4. Other Percussion (e.g. hand claps)
5. Cymbals

 I could then apply techniques such as using parallel compression and eq shaping to each individual drum track to get a big drum sound. The 5 channels also meant I could adjust and get a good balance between the kick, snare and cymbals. I routed these 5 stereo outputs from Groove Agent to a 'Drum' group channel. This meant I could easily adjust the overall drum level within my mix using a single group channel slider in the mixer.

I applied compression to the vocals and bass synthesiser. I eq'ed the string/pad synthesiser sounds to boost the high frequencies to help make these synthesiser sounds stand out in the mix.

I mixed the track by moving all of the faders to the bottom of the screen. I set the drum group faders to a reference level (approx -6dB). I then gradually brought up all of the other faders in sequence (bass, synthesisrs, rhythm guitars, vocals and solo guitar) while ensuring I maintained a balanced sound.

Finally I mixed the whole song down to a 24bit 44.1Khz wav file ready for mastering.

This un-mastered mixdown is shown below:

Adding Vocals and More Rhythm Guitar

I decided to add vocals based around the concept of a game of poker imitating life. I am no singer so I used the Cubase pitch correction plugin to help ensure that I did not sound too off tune as shown below:

I also decided to duplicate the audio track and use the pitch correction plug in to pich my voice up by an octave. This allowed a simple harmony to be added when both vocal tracks were played together. The transpose value was therefore set to 12 to raise the pitch of my voice by one octave as shown below:

I added a 'wail' type vocal sound and saved it as a sample. The start of this sample was actually sightly off-key and so I used the audio editor to top and tail it with a fade in to just the 'in-tune part. I then time-stretched the vocal wail to be slightly different lengths on the normal and pitch shifted tracks. This gave me an interesting overlapping sound as though two people were singing. Finally I used automation of the panning control for the vocal wail sample to move it left and right.

I added delay and reverberation effects to my main vocal to get a better and fuller vocal sound.

On listening back I realised that some of my vocal phrases sounded a bit slow and ponderous in the mix. I did not want to re-record them as it had taken along time to get them right. I therefore the snip tool to cut up each vocal phrase and then used the time-stretch mode to slightly shorten the length of each phrase.

I felt that the bridge section needed a clean rhythm guitar sound to emphasise the descent into chaos. I therefore recorded a set of chords using the VST Amp Rack plug-in. I then added a virtual wah-wah peddle and used automation to move the wah-wah pedal to create the characteristic wah-wah sound.

  

 

Adding a Guitar Solo

I wanted an electric guitar solo with a big sound to complement the appeggiated  chords I  had created. I first tried to record the  guitar solo myself but it sounded very amateurish due to my lack  of skill in playing the guitar. I  therefore asked Mike Ohman, a fellow student and an accomplished guitarist, if he would mind recording a guitar solo for me.

Mike plugged his guitar directly into my Focusrite IO interface and spent the best part of 90 minutes working out and playing a guitar solo to fit in with my backing chords, bass and drum beat. We used the new VST Amp Rack plugin introduced into Cubase 5 to create a big epic guitar sound.

We also added a ping pong delay to further create a big 80's style guitar solo sound. I was really pleased with Mike's contribution to my project.

Adding Ambient Strings

I needed the track to move forward and develop. I decided to add some floating ambient string sounds. To create a really full sound I decided to use the Cubase Arpache SX arpeggiator MIDI plug-in to automatically create string arpeggios. I experimented with different virtual synthesiser sounds and arpeggio patterns until I found what I wanted.

I found I could play a sequence of chords and broken chords on the keyboard and get a very full sound texture which I was very pleased with.

The following two sound files give examples of two string sections I created using Cubase's Arpache MIDI insert plug-in: