FL needs to know what root notes you want to use for your progression. Chords are built from the root note and are then using the notes in the scale to form a coherent and reasonable harmony. If this post is popular enough, and with high enough demand I don't even mind making my own .fsc files, the exact files that riff machine uses, that have cool and interesting basslines to build chords from for you to download
Right-click on the instrument you intend using, open piano roll. To access the riff machine tool in Fl Studio, mouse over to the piano roll option, scroll down to tools, and then clicking on Riff Machine will load up the riff machine or just by using shortcut Alt+E will load up the riff machine tool instantly.
This Max For Live arpeggiator, riff generator and sequencer turns even the simplest of MIDI patterns into an evolving musical continuum, and comes with a collection of hybrid electronic/acoustic Instrument and Effects Racks capturing the classic and contemporary sounds of electronic music legends Coldcut.
So I'm not sure what's up with this part - perhaps it was an alternate track where they thought they might want to drop in the sidestick in place of the snare? Or perhaps it was a previous iteration of a guide drum track, that they used before they found a snare drum sound they liked and continued to refine the drum programming.
And rather than this track being delayed, perhaps the main drum track is advanced slightly before the beat? Sure enough, when we compare this "late" sidestick drum part with the rigidly quantised bass part, it's absolutely in time with the bass - so we now know the main drum track has been brought forward by 7ms. Everything else, including the other drum parts, runs "late" to this. Suspecting this might be an error, I checked it alongside the final mix and indeed, this drum track is too early - so it's likely an error somewhere along the line in preparing the files and about 7ms of the beginning of that track has been truncated.
The third drum track, called "Percussion" is interesting. It sounds like a vintage preset drum machine from the 70s - you know the ones with a few presets like "cha-cha-cha" to accompany the organ player. This is the part that's doing that in-between 8th note "chugging" that drives the main drums along so nicely, and it's not a part that was in any of the previous recorded versions of the song. On it's own, it sounds terribly cheesy, but in the mix it's fabulous and really adds some magic. This diverted my attention to a trawl through my various sample libraries of every obscure vintage drum machine known to man, both in terms of the character of the sounds, and any similar preset drum patterns - but although I could find things with a similar-ish sound character, I couldn't find a match, and certainly not a pattern preset match. To hear it clearly in the final mix, it's most exposed in the break down in the middle where the rest of the drums drop out. A second mystery!
Ok, moving on - next we have two tracks making a stereo snare track. This sounds great, and has a lovely short stereo reverb baked in to the recording (hence the need for two tracks - the right side is also delayed slightly for a wider sound):-
Ok, so lets think about the compression. Looking at the RG Jones studio pic from around that time (looking at some of the gear, it's a later pic from the 90s, but still retains much of the signature gear referenced in the SOS article), we get some extra clues to the gear that was in the studio. We can see the Lexicon 224X reverb up in the top of the rack (and it's LARC remote-control unit on the desk) and a couple of AMS units - one underneath the 224 (which looks like a DM2-20 Tape Phase Simulator), and the DMX 15-80S digital delay in the lower rack at the front, which would have been used for sampling the snare (and all the delay and thickening effects). We can also see a couple of 1176 compressors, another studio staple, and so the likely source of compression on those drums parts was probably either those 1176's, or the channel compressors in the SSL desk itself.
I tried both to see how close I could get, and the SSL's E-series channel compressors seemed again to be the obvious solution - a nice aggressive smack, coupled with some overall SSL bus compression (also pretty much a standard on SSL-mixed records) seemed to work great - here's a dry pass, and then a compressed pass of the example drum patterns:-
Ok, at this point, the only element we haven't identified is this pesky vintage drum machine loop "Percussion" track. This was bugging me, because it didn't make sense on a number of levels. Alan Tarney took his modern production gear into the studio to make the tracks - why would he also bring an old, cheesy vintage drum machine and set it up as part of that rig? Would it really be something the studio had lying around in a gear closet somewhere? How would you even sync it up to the BBC/UMI sequencer? Those old preset drum machines often didn't have sync inputs or anything like that. It probably wasn't something sampled into the AMS because the loop changes with an extra accent every fourth bar, which is probably too long to practically sample. More Linn chip sounds? Hmm, it wasn't exactly common for people to spend a couple of thousand pounds on a drum machine only to put terribly cheap-sounding replacement drum chips in it, but I guess it's possible. But the pieces didn't fit, at this point.
And that percussion track really makes a big difference in the drum track, for all it's simplicity.
Let's leave this a mystery for the moment, and turn to the instrumental parts.
While we've been looking at this sound, let's go back to the original demo (which inspired Tarney to "recreate" for this version) and see what they did there. Interestingly, the main riff is intact, but it's just the PPG bell sound - none of the pluckiness that the Juno contributes in the final version. So it's likely that when recording the riff, the band probably pulled up the sound they were using previously for the part, or something similar, and layered it up to make it sound thicker and more defined. This gives us the opportunity to only hear that initial PPG part and thus help get closer to it in the recreations - and the demo does indeed seem to be a PPG Preset layer - a layer of the last two patches in the preset bank - "PPG Wave 063 A+B".
The bottom line is it looks very much to me like the main signature riff is the combination of the PPG & Juno 60. While the Juno 60 may have started the riff off in the studio, and it contributes some weight, bounce, attack and meat in the midrange, the PPG bell layer is an important element in the overall texture and makes it a more glossy, shimmering texture which catches the ear - it's effectively a very similar sound to the initial demo. Without that high PPG part, the riff won't sound like the record.
This bell sound is also combined with a string part in the middle eight section - the string part is the same Juno 60 string sound as used in the choruses. I used the modified factory "Strings 1" patch I made earlier and it's almost indistinguishable from the recorded part.
There's an echoey arpeggiated effect or motif that starts in the middle eight. It's a plucky, sequenced/arpeggiated sound which fades in and out of a reverb wash and has a large delay/echo on it, likely from the AMS - the effects are printed onto the two hi bell tracks in stereo. The sound could probably come from any of the synths, but from the sound I'd guess this is the Juno 60, with a plucky sound similar to the main riff pluck the Juno is already doing - and it must be either sequenced from the BBC, or it might possibly be using it's internal arpeggiator.
I edited the 27 "Guitar" preset in the Juno 60 and got a plucky tone pretty close to the part, and programmed in a looping four note descending sequence, with lots of delay. To recreate the middle section, it's a bit of a science experiment to automate it into a lush reverb, panning around the stereo field, and automating the delay and reverb level etc.
After the middle eight and intro refrain, this sequence repeats a few times in the close of the song but this time the sequence rises by a full tone every couple of repeats until a longer fall, like this:
This pitch change rise could have been programmed into the sequencer, played live with the arpeggiator, or had the pitch changed on the AMS delay on the fly perhaps. In any case, to me this isn't as effective as the earlier parts without the transposition coming out of the reverb, and in the final mix it's fairly low - in fact, I'd never consciously noticed it before, but once you know it's there, it's fairly easy to hear.
Lastly for these tracks there's a single mono white noise rising swoosh sound in the middle eight leading to the breakdown riff. It sounds too beefy to be the DX7 so I checked the Juno 60, and it has a preset called "Surf", a white noise sound, which with some filter tweakage seems to be a good tonal match - so my guess on this is that they quickly modified this patch to give a rising swoosh effect - or just dialled it up manually. It has a slightly flangey/chorusey quality, so they probably ran it through the AMS on the way to tape. Very few sounds have been tracked totally dry - they liked to commit a sound to tape, and this was more common when studios didn't have a huge amount of effects units to use in the final mix.
After plotting several different songs against their own tempo like this it seemed that in addition to electronic music a lot of pop and rock has this type of a pattern, too. The most striking and clear patterns can be seen in music that makes use of drum samples in a quantized time base (aka. a drum machine): the same kick drum sample, for example, repeats four times in each bar, perfectly timed by a computer so that they align in phase.
f448fe82f3