Korg Forums

[PoisonJam7]

Hello. I am admittedly a synth newb; I don't my LFO's from my Oscillators. My experience is mainly as a guitar and bass player, but I'm doing a project which requires quite a bit of keyboard sounds. So, I purchased a MicroKorg since it has gotten such great reviews and fit my budget.

Anyways, I'm trying to re-create the iconic sounds of the OB-X and the Oberheim polyphonic as heard throughout "The Camera Eye" as recorded on Rush's "Moving Pictures" album. I'm sure many of you are familiar with this sound...

After fooling around with the MK for a while, I was able to create a pretty close approximation of the "Big LFO sweep" sound that you hear after the initial 7 notes. (It's hard to describe...but it's the part that starts at about 28 seconds into the song) But, I'm having no luck in creating the main sound. The one that goes "weeow, weeow, weeow, weoow" etc.. As I said, I'm a synth newb, but from what I've read, this was originally created using an OB-X. I saw them twice on the Time Machine Tour and it looked like Geddy was playing sampled sounds on his Moog Little Phatty. (Probably sampled from the album)

Regardless, I'm still at a loss as to how to create this sound. I'm not really looking for a patch file, but it certainly wouldn't be denied, just some guidance about how this sound was constructed and how I can go about re-creating it on my MicroKorg.

Thanks for any input.

(While you're at it, don't be hesitant to include some info or patch links related to the iconic Moog Taurus intro to "Tom Sawyer" or the Polysynth of "Subdivisions" or ANY Rush song for that matter! )

P.S - Most of the info I have about this song came from the Rush Tab Project website, and the subsequent Camera Eye Keyboard Tab. As seen here.


*Edit* Here is a recording of the patch I made, as described earlier. The first example is with the sweeping LFO, and the second is without it. I would upload my patch, but I don't have the Korg software installed on this computer. I think i've gotten pretty close, but I'm still missing the "weeow, weeow" sound.

Example 1

Example 2

[xmlguy]

You need to play with the filter envelope so that a low pass filter center point increases and decreases using the same timing that you are trying to synthesize. You also need to play with the filter resonance.

[PoisonJam7]

Thank you for that information. i will try fooling around with those settings tonight and see if I can come up with anything.

[PoisonJam7]

After a bit of fooling around, I was able to come up with a pretty close approximation of the sound. There is one problem I'm still having though:

The patch makes the "weeow" sound like I was looking for, but it seems to be less pronounced and noticeable the higher up I play. For example, when I play the opening chord, it sounds good and pretty close to the record. But when I start to play the second theme (the one that the guitar plays after the drums enter) the effect is less pronounced and doesn't sound as close to the recording. Is this maybe the limit of the microKorg's synth engine in being able to re-create Oberheim sounds? Or is it (most likely) my inexperience?

I've uploaded my patch and was hoping someone could take a look at it and maybe help me tweak it to get it sounding just right? Thanks for any help.

Patch Download

[xmlguy]

It would be easier to listen to an mp3 of your example, if you could record and post it. I am a sound designer but I do not have the Microkorg, just a bunch of other synths like the R3 that are very similar in design.

Another tip is to use a virtual patch to use the keyboard velocity to affect the intensity of the filter sweep. This will let you control how much of a wah that you get based on how fast you hit the chord. The sound of the wah is affected most by the filter resonance, so be sure to try out different amounts. You also need to carefully control the starting and ending points of the filter to be in the correct frequency range to match what you hear on the Rush sample.

The problem with the wah sound for higher notes might be that the ending point of the filter sweep is not going high enough, or you may not have enough frequency content in the higher range for the filter to have as much effect. You can increase the higher harmonics by using a sawtooth and/or square wave, which are rich in harmonics compared to sine waves or triangles. The cool thing about saw and square waves is that the filter can turn them into a sine wave as the harmonics are filtered out, leaving just the root note. The reverse is true as well, the saw/square can start as sine wave when the filter center point is low and then the harmonics reappear when the center point is raised to let them through. You can also try using a band pass filter instead of a low pass.

I also forgot to mention that the robotic sounding filter pitch shifts are done by using an LFO using a sample and hold on the filter centerpoint, which gives that random variation that occurs at the beginning of the song for the held chord. The wah produced by the filter envelope is in addition to the S&H LFO modulation.

But most importantly, do not give up trying. It always takes a lot of time and experience to dial in the sound you want while you are still just learning synthesis.

[X-Trade]

I suggest you play around with the filter key tracking a bit.
After that you may have to revisit the cutoff setting to get it to work across the whole keyboard.

Afaik on the microKorg you can't control filter EG intensity with velocity.
_________________
Current Gear: Kronos 61, Karma, RADIAS-R, ESX-1, microKorg, MS2000B, MiniKP, AX3000B, nanoKontrol, nanoPad MK II, Novation ReMote37SL, Akai APC20, MOTU MIDI TimePiece 2.
Past Gear: Korg TR61, Poly800, EA-1, Kawai K1
Software: Cubase Studio 4, NI Reaktor 5, FM8, Ableton Live 9. Apple OSX(10.8.3 Mountain Lion) on 15" MacBook Pro

[xmlguy]

Hi xtrade, I think a virtual patch from velocity to filter center should work, with the virtual patch mod intensity serving the same purpose rather than having the EG intensity as a destination. At least that would be what I would try. On the R3, the 2nd filter would also serve as the destination, but alas, the MK has only the one filter per timbre.

[PoisonJam7]

Thanks for the advice guys. I think I'm getting closer!

After tweaking the resonance, I was able to get the wah effect much more pronounced on both the lower and higher keys. However, this made the sound much softer, so I tried to counter the effect by tweaking the cutoff a bit. There is a fine line you have to straddle between cutoff and resonance if you want the wah effect to sound right, I have learned.

I messed around with lots of settings based on the advice posted here and created something that I think was closer then my last patch. I also made a virtual patch to control the S&H LFO that I can dial in after playing the intro chord, so that part of the sound is now covered. Before I was using a completely separate sound patch to do this, but now that I have learned how it is being created I was able to include it in this one.

I have uploaded an MP3 example of this new patch:

Example 3

A few new issues though:

The notes aren't sustaining as long as before, and the more notes I play, the quieter it gets. I thought the keyboard tracking was causing this at first, but I tried lots of different settings with it and it never seemed to make a difference. You'll also hear that at the same time that it's getting quieter the more notes I add, it also starts to distort...paradoxically.

If I can solve the issue of the notes sustaining longer and the volume problems, I think I'll have a pretty close patch on my hands. Based on the above MP3, are there any more suggestions you guys can offer to fix these issues and further tweak the sound? Also, I have one more question (forgive me if this is a totally newb question) what is the filter centerpoint?

Thanks again for all the help! All this work makes me appreciate the simplicity of just plugging in my guitar or bass and playing right away...

[xmlguy]

The distortion is easy to solve. You just need to turn down the mix level for the patch. Remember that resonance boosts the center point significantly, so a bunch of notes with high resonance can cause internal clippng in the synth engine. A filter with resonance boosts and cuts at the same time, so it takes time to find a ballance between the filter wah and the overall audio level of the patch over time. When the filter is doing lots of filtering, you can use the Amp EG to boost the overall level coordinated with the Filter EG.

For a wah effect, the filter EG starts with a low center point, then raises it, then lowers it. So the audio level will start lower, go higher, then go lower. You can offset it somewhat by adjusting the EQ a bit. On other synths like the R3 that have a free envelope, you could directly counter act the filter EG to even out the audio level, but not on the MK. You may need to only adjust the amp EG to increase the sustain by decreasing the decay.

The issue with the level dropping with multiple notes is probably because you are running out of polyphony and some of the bass notes are getting cut off. You can only do a max of 4 note chords on the Microkorg, or less if you use laying or unison. That is the biggest limitation of the Microkorg: very low polyphony.

You can significantly fatten up the patch by using OSC2 as a sub oscillator tuned an octave or two below OSC1, and then mixing in just enough to get the bass speaker booming, but not enough to be noticeable as the root note on OSC1. There is a certain range where we cannot easily identify certain notes when they are unison harmonics below or above a dominant root. Gradually boost the mix level of the OSC subosc wave until you can clearly identify the note, and then back it down just a bit below that point. The root note will sound much fatter, but most listeners will not know why unless they are synthesists who know what to listen for. Whenever you hear a powerful bass patch, usually there are several of these subtle tricks involved. Another trick is to clear out a hole in the track in the frequency and time domains so that the bass does not compete with anything. It is very easy to muddy a track, but much harder to clean it up. You do not need a bunch of elements to have a powerful song, as 3 piece bands like Rush proves. Geddy Lee is a great example when he used synth bass pedals to complement his bass lines. The bass pedal can act as a sub oscillator to bass guitar, or for bass pads to fill inbetween the bass lines.

[PoisonJam7]

Thanks for the tips, xmlguy!

I'll work on my patch this weekend and try implementing what you have suggested.

[PoisonJam7]

Alright! I was able to create a patch I'm almost 100% satisfied with this weekend. There is just one last issue I'm having and I'm unsure as to what is causing it or how to fix it. It's also kind of hard to explain, so I've included a link to both an MP3 example and a picture of the waveform to aid in my description of the issue.

So, when I play a note and keep it held for any time longer then a few seconds, there is a drop in the waveform and what sounds and looks like the sample being re-triggered. This is most noticeable on the lower notes, and the higher up you play the smaller the drop; presumably because the waveform is getting shorter the higher up I play. What is causing this and how can I fix it?

Like I said, it is kind of hard to understand what I'm talking about with just words, so please take a look at this picture and listen to this MP3 to help.

Thanks in advance for any help you can offer.

[xmlguy]

There could be several causes of those breaks between notes, but the most likely cause is signal cancellation due to phasing of two of the same kind of waveforms. When you mix two sine waves together that are 180 degrees out of phase with each other, you get nearly complete cancellation.

Ever tune a guitar? Notice that when two strings are playing perfectly in tune with each other that there's just one single, pure note. But when you slowly increase or decrease the tension of one string, you start hearing a pulse that beats faster, the more out of tune it is. So you probably are mixing two waves from OSC1 and OSC2 that are slightly out of tune with each other. The amount of detuning is a set value, which results in low notes being more closely in tune than high notes, thus low notes have a slower cancellation pulse than high notes. But, this can be done other ways, because keytracking allows you to use the position on the keyboard to alter other things that can have the same effect, like adjusting the speed of an LFO.

It would help if you post the patch in text form here, like so:

From a tube posted in another thread here, at http://www.youtube.com/watch?v=ifOVWXKvLrQ

EDIT SELECT1
VOICE:-----SYT---SGL---PLY------------
PITCH:-----0--------0--------0-----2------1
OSC1:-----dig-------------- 54--------------
OSC2:------tri------off-------0-----0---------
MIXER:----127-----0--------0---------------
FILTER:---12.L--100------1-----42------20
FILTER EG:-0----40-------0-----56------off
AMP:---------120--cnt-----off-----45--------
AMP EG:-----0-----38-----90-----56------off
LFO 1:-------tri-----off-----off-----70---------
LFO 2:------Sin----off-----off-----28---------
PATCH1:-------LF.2-----Cut-----8----------
PATCH2:-------LF.1-----Ptc-----5-----------
PATCH3:-------nod-----LF.2---45----------
PATCH4:-------FEG-----AnP---32------------
DELAY:----------L-r-------off----56----115----
EQ:--------------320--------3---1.00----6-------

[PoisonJam7]

Yes, I'm well acquainted with tuning guitars...I've been doing it in some fashion for the past 12 years..lol So, yes, that sound phenomenon is familiar to me.

Here is my patch in text form:

EDIT SELECT1
VOICE:-----SYT---SGL---PLY------------
PITCH:-----0--------0--------1-----2------0
OSC1:-----squ-------------- 33-------0-------
OSC2:------squ------rng-------0------(-)7---------
MIXER:----127-----127--------0---------------
FILTER:---12.L--45------77-----11------3
FILTER EG:-26----53-------0-----127------off
AMP:-------37--cnt-----off-----0--------
AMP EG:-----0-----57-----127-----10------0n
LFO 1:-------saw-----tim-----off-----0---------
LFO 2:------S-H----tim-----on-----3.32---------
PATCH1:-------F.EG-----Cut-----7----------
PATCH2:-------LF.2-----Cut-----0----------- (This is for the robot sounds, so at the appropriate time I adjust the cutoff up from 0)
PATCH3:-------LF.1-----Cut---0----------
PATCH4:-------LF.2-----Cut---0------------
DELAY:----------str-------off----40----0----
EQ:--------------320--------4---6.00----3-------


After reading what you posted, and seeing the patch in text form, I surmised that the problem was probably being caused by the two square waves being used by Osc. 1 & 2. This was indeed the case, and changing the waveform of either oscillator seemed to fix the problem. Not without altering the sound though.

I guess the only thing to do now would be to change the waveforms on either oscillator and re-tweak the sound to get it back to where it was with two square waves?

[xmlguy]

Yes, that patch shows two square waves, with OSC2 detuned a bit and using ring modulation. That explains the heavy cancellation effect. You should research ring modulation if you don't know how it works. In this case, OSC2 will cancel out OSC1 due to the ring mod whenever in phase together. Take a look at the following video:

http://www.youtube.com/watch?v=mATt_HyfhOc&feature=related

Ring mod causes some very strange results if you don't know what it's doing.

[PoisonJam7]

Thanks for the link. I think I can handle it from here with the patch.

Thanks so much for all of your help, xmlguy! I have learned a lot about synthesis from your tips, and I couldn't have made this patch without your insight. So, thanks again.


One more thing, though. Sorry, but I have to ask....Tom Sawyer? The opening Taurus Bass? What's going on here?