The making of: Heaven Is Oblivion S2

I’ve recently been doing a spot of educational work and last February performed a guest lecture to music technology students at Confetti Studios in Nottingham. As a demonstration of shifts in the music making process I made a new version of 1996’s Heaven Is Oblivion single – here follows an explanation of how the chorus section was made with slides from the presentation. Should you have no interest in music technology, but would like to listen to the finished track click here.

Heaven Is Oblivion first appeared as the final track of the episodic Ultraviolence album Psycho Drama. Although, like most of the album, the subject matter was dark the track was upbeat and released as a single in the following year as it was believed by record label Earache, myself and several names in the music biz to have some mainstream crossover potential. (audio example)

All of the Psycho Drama album  was recorded at my home which I believe in retrospect to have resulted in some substandard vocal recording. However, around £9000 had been spent upgrading my home studio to a standard capable of producing the electronic sound I wanted and so there was no further money left for microphones and acoustic treatment or for hiring a commercial studio. There was however an additional budget for the single version and a recording was made during a day at Protocol Studios in London, with original vocalists Didi Goldhawk and D Quest. The Protocol mix was hurried on unfamiliar equipment and was rejected but the 24 track vocal recordings from the session were used for a new, shorter home version that would become the single, mixed on my Tascam M1600 desk. The 2011 version was recorded within a PC computer using Cubase 5. It is my opinion that mainstream DAWs are very similar to one another in nature and quality but differ in respect to interface, quirks and minor functions – in short I prefer Cubase 5 because I am used to it. I like to mix digitally although it took time to get used to after spending years with an analogue set up, where you would constantly ironing out imperfections and seeking complete separation and purity of sounds, where you might need to add “faults” or ” humanity” to a digital mix. Both domains have pluses and minuses and it would be folly to claim one to be “better” than the other.

The chorus section of Heaven Is Oblivion is loosely based on a triumphant four bar melody section from Beethoven’s 9th Symphony, with the more whimsical 4 bars, both appropriate to the subject matter. The simple backbone of the melody was originally created using an Emu Proteus 2 Orchestral module which was supplied with a bank of 128 sounds which were a mixture of synthesised and sample sounds suited to replicating classical music and cost around £800. Here it is replaced with the modern equivalent East West’s Symphonic Orchestra, one of many minorly different VST plug-ins on the marketplace  containing many gigabytes of orchestral samples.  It sells in various iterations, costing from around £100 for the budget “silver” version heard here – the multiple microphone placements and expressions are surplus to this kind of work. It is worth noting that although Heaven Is Oblivion may be heavily influenced by classical music it was not my intention to produce classical music cheaply by using electronics, but more to use electronics to create a new sound that would create the deep textures and resulting emotions possible with orchestral music. (audio example)

The biggest change  in creating music at home over the last 16 years has been, for me,  the physical size and monetary expense of the equipment, and no more extreme example of the latter can be found in this choir section. 16 years ago, in order to have usable choir samples, I had to buy an sampler with at least 8mb of RAM, in this case an Akai S2800 plus a 270MB hard drive all piling in at a whopping £3000. The actual choir sample CD, the decent Peter Sidilocheck’s Classical Choirs was a more reasonable £50. As I didn’t own a CD-ROM (noisy and unreliable) I had to play each sound into sampler from an external CD player.  For the remake I have used East West’s Symphonic Chiors, which weighs in at £300 and 40GB and will run on a modern domestic PC. Symphonic Choirs contains a “Wordbuilder” function whereby the virtual choir “sings” words entered into a text box by the user. Although the technology is still embryonic, sounding nothing like a real choir performing from a lyric sheet, I have used the function can garner some interesting hooks that hint at the message of the track, and the result is more interesting than the generic sampled choir of the 1996 mix. (audio example)

The original bland keyboard sound of the Roland JX1 is replaced by Camel Audio’s Alchemy VST synth, its morphing additive sounds are rather like saples that can be fully manipulated as synth sounds, lending itself well to distinct electronic orchestral sounds, as hear in this synth- wind preset. The arpeggiator sound of the finished track is also courtesy of Alchemy, with some automation of the morph control providing additional interest as well as a dramatic whoosh for the bridge sections. (audio example)

Some straight old skool synth to keep the sound pallet from straying into faux classical territory courtesy of Native Instrument’s Prophet 5 emulator, the Pro-53, replacing Oberheim’s ultra-phat but non-editable Matrix 1000. (audio example)

Plush layers and spot FX were provided by the S2800 which I have spiritually replicated using ReFx’s Nexus sample player. Nexus contains many expertly programmed, lush presets and I have been careful not to over-use it, as productions can quickly sound like a Nexus fest. Such temptations are not beyond the resistance of mainstream producers and you will hear Nexus recognisably in many chart productions. I have used Native Instruments Battery as a sample player, as I have no need for the extra functions (such as complex keyboard mapping) of their more extensive Kontact counterpart. Here it plays a short classical music sample with a little manipulation within Battery for an electronic feel. (audio example) Now we can hear all the chorus layers together, first raw (audio example) and then panned out with a little EQ and reverb. (audioexample)

I do not have the original vocal samples from Heaven Is Oblivion, so I decided to create something completely different using Image Line’s budget DAW Fruity Loops (now rebadged FL Studio) speech generator.  The sounds were manipulated within Battery and through Prosoniq’s classic (and discontinued) Orange Vocoder. I have lifted a single vocal sample from the original CD which adds a euphoric touch, used sparingly save for in the climax of the piece in the second bridge. I believe these vocals to be at least “as good” as those in the original track, so I have turned a limitation into an advantage – something essential when working within a limited budget.

Solid State Logics’s X-EQ is the only equaliser used in Heaven Is Oblivion S2. Firstly because using one type of EQ can gel the sound in the same way as an analogue mixing desk might, and also because X-EQ has a beautifully transparent sound that can also be automated and used for filter sweeps (as heard on the kick drum early in the finished track) – the filter sweep can also act as an EQ duck so you get a “two for the price of one” dramatic effect together with non-intrusive sound separation. The brilliant graphic interface thankfully makes no attempt to replicate traditional controls and also makes use of a spectrum analyser which can be seen in green behind the EQ curve. Careful to turn it off after use, though, as instances quickly gobble CPU power.

When mixing a traditional band  an aim would normally be to place the instruments into a convincing space. Reverb in electronic music is very different in that it is instead used to embellish and fatten sounds and to place them into a deliberately unworldly sound space. The artefacts associated with actual spaces are generally undesirable. The versatile sound of SSL X-Verb wins over convolution or plate emulations – again the modern GUI makes the usable presets easy to edit and wins out over old fashioned dials and pretend LED readouts. Most of the delays in the finished mix are also X-Verb.

The initial mix of Heaven Is Oblivion S2 runs at 2’40” and took around ten hours of programming to complete. It can be heard here. It is adequate in demonstrating the above points and was deliberately made in such a way as to allow for easy remixing – for instance a dubstep or hard trance version could be made by adding rhythms and only minimal changes to the core elements. However (as is often the case) I grew emotionally attached to the track as well as becoming curious as to what would be possible if it were extended to a full four minutes or so, so spent a further fifteen hours or so making what I consider to be a finished version – click here to listen to it. I feel it to be superior to the original from a technical point of view and possibly from an artistic one, as the increase in dynamics and sonic textures make for a more fulfilling listening experience. Of course, the new technology makes this much easier so on a level field I would say the new version is about “as good” as the ’96 version was fifteen years ago. It is unfortunate that the this music will not see a commercial release, but it was enjoyable to accomplish such a piece of music without feeling under pressure and I hope that Ultraviolence fans will appreciate hearing it free over the Internet. There are many more elements of the track that I haven’t had time to go into here – perhaps I will expand on this article at some point.

Links: Confetti Studios, Steinberg, East West, Native Instruments, Camel Audio, ReFx, Image Line, SSL, Earache Records