Need help patching an R3

[tpantano]

I'm trying to create some nice poppy leads for my R3, based off of owl city.

Anyways, I've been trying to make his lead sounds by turning osc 1 off, turning osc 2 to square, adding a bit of portamento, 100% low pass cutoff, a tiny bit of attack and full sustain, a tiny bit of release. however, none of the sounds are similar...

so, can you guide me through patching one of these 2 songs, to get me on the right direction?

http://www.youtube.com/watch?v=aI4JLa0hbUw (beginning here, dislike the song but that lead is common throughout his music)

http://www.youtube.com/watch?v=R6OQFnUrbW8 beginning lead there as well

thanks in advance, and not just from me but some other forum users!

[arzie]

Yes, i'm interested in this as well!
Huge thank you to anyone that can make it clear to us.

[xmlguy]

You have to listen rather carefully to a sound like that to be able to reproduce it. Here are some of my observations about it.

The basic tone is a very pure, single note without a lot of harmonics. That kind of note starts as a sine wave, which is a single frequency.

Most of the notes have a sort-of abrupt click to them, kind of like a string pluck or a sharp key hit on toy piano or music box. That means they have at least a fast attack amp envelope (EG2), but that's not enough to emphasis the note. You need a decay and sustain that cause a peak at the beginning of the sound. You can further emphasis this pluck sound by having a filter with a separate envelope (EG1), but with a negative EG1 intensity. This works by allowing the attack to be sharp but then quickly softens the note.

The notes also quickly stop (staccato) unless the note is held. That means EG2 needs a fast release.

In the middle you'll hear some notes that glide/slide between each other, but not when played staccato. The glide is done with portamento, but only when the notes are played legato. In other words, how the notes are played makes a difference to what you hear. This probably requires an external sequencer to do properly because the R3 arpeggiator and modulation sequencer do not let you mix staccato and legato notes. Just because you hear that an artist is using a synth like a microKorg to make a particular sound doesn't necessarily mean that you can reproduce it using only that piece of gear.

There is also an echo on each note, and the echo seems to bounce from left to right in stereo. That can be done in several ways, such as with a delay or reverb effect, then using a virtual patch to pan the main notes from left to right. Or you can leave the main note panned to center and use the L/C/R Delay, which causes an echo from left, center, and right. Getting the timing of the delay correct is very important so that the echo is synchronized to match the track. It's quite possible that this delay was done with external effects, but the R3 can do it too.

Now if you want me to tell you every value and knob that you need to adjust to reproduce this sound, sorry, I'm not going to do that. If the artists who worked on these Owl City tracks had to work to create this sound, then it seems natural for people to work to reproduce it too.

[tpantano]

Now if you want me to tell you every value and knob that you need to adjust to reproduce this sound, sorry, I'm not going to do that. If the artists who worked on these Owl City tracks had to work to create this sound, then it seems natural for people to work to reproduce it too.

Nope, that wasn't what I wanted. What you told me was very helpful. I simply wanted a very basic pathway towards what I'm aiming for, just to spark some ideas.

Thanks very much.

[xmlguy]

Now if you want me to tell you every value and knob that you need to adjust to reproduce this sound, sorry, I'm not going to do that. If the artists who worked on these Owl City tracks had to work to create this sound, then it seems natural for people to work to reproduce it too.

Nope, that wasn't what I wanted. What you told me was very helpful. I simply wanted a very basic pathway towards what I'm aiming for, just to spark some ideas.

Thanks very much.

Glad to help as best I can. On a patch like this with so many elements, it will take quite a bit work, so have patience with yourself. You might need to listen to the original sound dozens if not hundreds of times to pick up the difference between your work-in-progress compared to the original. A lot of synthesis involves very careful listening, and the more you practice, the easier it should be, but it still takes quite a bit of time and effort even if you've been doing it for many years.

There are different strategies you can use to duplicate a patch like this. The normal workflow to construct a patch is from "inside-out", from OSC -> Filters -> AMP -> Effects, with modulations at each stage. You can also use a reverse workflow to deconstruct a patch from the "outside-in".

First listen for the effects. Delays, Reverbs, Gates, Flangers, Chorus, Distortion, etc. all have distinctive sounds. Identify what you can, then imagine the original sound without effects.

Then listen for the "shape" of the sound - this is the Amp envelope. Does it cut flat on and off (gate)? Does it build slowly? Slow Attack. Does it sound plucked or strummed? Decay+Sustain. Does it continue to sound for awhile like a string vibrating - Release. Is it trigged with a pattern? Arp or Gate. Then imagine the sound without an envelope or arp/gate.

Then listen for the frequency characteristics typical of filtering? Can you hear certain frequencies being emphasized or cut when it sounds? Do these frequency effects occur quickly, slowly, or both? Then imagine the sound unfiltered.

Then listen for pitch characteristics? Is it a single tone? Sine wave. Is it brassy? Saw. Is it hollow sounding? Tri. Is it computer/digital sounding? Square. Does it sound like a specific real instrument? It may be a complex starting waveform like DWGS or a sample.

The goal of this strategy is to try to hear through the layers one-by-one to get all the way back to the original oscillator wave in your mind. Each time you are hearing the same source audio, but you are listening for different specific aspects. Then you try to ignore the aspects you've already identified in order to selectively hear something else.

Then you can build a patch based on what you've extracted from the inside-out, comparing just that one layer with the original to fine-tune it at each stage. Then start building up those layers. If something starts to sound wrong - then remove that layer and go back a step to try again.

[axxim]

nice exposée xmlguy

it reminds me of a story I heard about Salvador Dalí (don't know if its a true one) but once in an exposition of him, two young students saw one of his typical "few traces" art and as they saw the price for it, they began to joke saying that they could do those simple traces too and try to get such prices for it. Some minutes later he (Dalí) showed them a drawing he did while they were joking and asked them if they recognize it. They recognized themselves in that drawing and wondered. He replied: when you are able to do such a drawing in this short time like me, then you could demand such a price for those simple lines...

[xmlguy]

You're welcome, axxim. From the outside it's difficult to see the mental process and strategies along with some intuition that someone uses to craft any piece of art or other work. On synths it can look to a casual observer as if it's just a matter of fiddling with some knobs and faders and that the synth does most of the work. What they can't see is that I have the entire synth structure memorized and I intimately know how all of it works together.

I've built entire synth patches purely in my mind, and then reconstructed them in the R3 editor without the R3 attached. When I got around to physically hooking up the R3 to write the patch to memory, it sounded how I thought it should be, with a few tweaks here and there, naturally, to fine tune it. I have my own set of favorite patches in my head that I can reproduce on any synth in a matter of minutes. But I wasn't born with this knowledge, as I developed it by putting in enough time and effort to master the structure to my own satisfaction.

My suggestion is to not merely settle for learning just enough to do a particular patch, but instead work hard to achieve the goal of complete mastery of the tool. Go all the way. Learn everything about it. It pays off later when you need to get the most out of it.

I'll give you a little example. Research the "Endless" wave on the R3/Radias. It's worth finding the answer. It's even more important to understand how it's different from every other wave. It's yet more important to think about the implications of this wave with respect to polyphony, having two filters, EQ+effects, and a vocoder to shape it. This one wave can take many hours to explore its potential.

[tpantano]

So, back in the day when you were all first learning to patch, did you have resources to back up your learning? Or was it more a trial and error type of learning?

I'd like to learn some more on how waves themselves actually sound, like what even/odd harmonics do for a sound, what the filters actually do, etc.

For example, would something like http://search.barnesandnoble.com/Analog ... ynthesizer
this be worthwhile? or would it be best to simply learn on my own?

My keyboard instructor uses Roland Arrangers... so he's no help. Father hasn't used an analog since the late 80s, and has forgotten most of the patching technique.

[xmlguy]

When I first starting learning about synthesis, there were very few resources. The internet didn't exist. Here's the kind of resource I had to read and understand:

http://www.paia.com/manuals/docs/2720-b ... rticle.pdf

To read that article, I first had to ride my bicycle 3 miles each way to the main public library in my city. Then I had to talk to the librarian at the reference desk to get help about finding resources to learn about synthesis. I would have to open up drawers of index cards to see if some of the titles sounded promising. Then I had to write down the call numbers and search for the books on a shelf somewhere. I might find one or two 10+ year old books that covered the subject to some degree, if the books weren't already checked out or missing. Then I would head over to the periodical section and look through past issues of magazines that were about electronics, looking through issue after issue to see if they had any articles about synthesis. It might take a good part of a day to find an article like that one above. I would have to pay money to photocopy the pages in order to study it at home.

Then I had to telephone or write the manufacturer to get information sent to me, such as brochures or a catalog, which could take weeks or months to receive. Then I could order a synth module kit. After receving the kit I would have to solder it all together. But it doesn't include a power supply, which I had to separately build. This synth module had no keyboard or amplifier or speakers. It was a voltage controlled oscillator. It could make basic waves. The VC Filter and VC Amplifier were different modules. I had to make patch cords to hook them up together to make a sound. This was when writing patches required writing down patch cord positions on a piece of paper. You couldn't save a sound. You had to unhook the patch chords and repatch everything to make a different sound.

So lots of trial and error, and few people to help. Modular synths are probably the best way to learn analog subtractive synthesis, but they are very expensive and deliver much less synthesis per dollar than virtual analog and soft synths. So instead I usually recommend a reasonably priced VA synth with lots of knobs and faders, like the X-Station, Radias, or SH-201. The R3 software editor with its virtual knobs is a good option too.

[tpantano]

Wow, that sounds tough!

So, would a book like the one I just edited above be worth my while?

[xmlguy]

Wow, that sounds tough!

So, would a book like the one I just edited above be worth my while?

I would've sold my soul to Satan to have a book like that when I was learning synths. He didn't offer it to me at the crossroads.

I haven't read that book personally, but it sounds killer from the description. I view buying a book like an investment in my brain, so long as I connect with how the book is written. The right book at the right time can be life altering.

[tpantano]

Wow, that sounds tough!

So, would a book like the one I just edited above be worth my while?

I would've sold my soul to Satan to have a book like that when I was learning synths. He didn't offer it to me at the crossroads.

I haven't read that book personally, but it sounds killer from the description. I view buying a book like an investment in my brain, so long as I connect with how the book is written. The right book at the right time can be life altering.

Books can get you anywhere....
who would've thought that learning to build a PC was as simple as a $20 book?
or that said $20 book saved me $800 on my next PC?

I'll definetly check out this book, seems very promising.

[Yatmandu]

So, back in the day when you were all first learning to patch, did you have resources to back up your learning? Or was it more a trial and error type of learning?

I'd like to learn some more on how waves themselves actually sound, like what even/odd harmonics do for a sound, what the filters actually do, etc.

For example, would something like http://search.barnesandnoble.com/Analog ... ynthesizer
this be worthwhile? or would it be best to simply learn on my own?

My keyboard instructor uses Roland Arrangers... so he's no help. Father hasn't used an analog since the late 80s, and has forgotten most of the patching technique.

Although this is a very thorough book on the history of analogue synthesizers, covering who built them, how successful they were, who played them, etc, I don't think it's a good source on learning about synthesis itself. You're better off with the pdf above or, Sound-on-Sound's excellent multi-year tutorials on synthesis.

But the book you mentioned was a good read (IMO). It also comes with a CD of many of the popular synths, and the author is clearly a synth expert.

[tpantano]

So, back in the day when you were all first learning to patch, did you have resources to back up your learning? Or was it more a trial and error type of learning?

I'd like to learn some more on how waves themselves actually sound, like what even/odd harmonics do for a sound, what the filters actually do, etc.

For example, would something like http://search.barnesandnoble.com/Analog ... ynthesizer
this be worthwhile? or would it be best to simply learn on my own?

My keyboard instructor uses Roland Arrangers... so he's no help. Father hasn't used an analog since the late 80s, and has forgotten most of the patching technique.

Although this is a very thorough book on the history of analogue synthesizers, covering who built them, how successful they were, who played them, etc, I don't think it's a good source on learning about synthesis itself. You're better off with the pdf above or, Sound-on-Sound's excellent multi-year tutorials on synthesis.

But the book you mentioned was a good read (IMO). It also comes with a CD of many of the popular synths, and the author is clearly a synth expert.

Alright. I may just check into it for fun.

So, are there any resources you know of that I could check out that would help me put whats happening to the waveform to what I hear?

For example, its easy to relate an ADSR envelope to a real sound; when you see an acoustic instrument played the mind can discover how the sound slowly fades in over time, reaches a peak, drops to a softer level, and eventually fades or drops out.

However, I'm having a hard time relating wave form choices and filters to the sound I hear. I understand that wave forms have different harmonics added to the base frequency, and how filters cut these harmonics out (in subtractive synthesis), but what I'm not understanding is how a waves harmonics relates to the sound I hear.

So, in essence, I know how to program a synth, in that I know that envelopes vary a parameter over time, such as volume level; filters cut out select harmonics; wave forms are the base rich sound from an oscillator; LFOs vary a parameter over time based on a wave; etc; however, I lack the programming ability to relate what you would see on an oscilloscope (the visual pattern of a waveform) to what you hear, and because of that I can't envision a sound in my mind as a waveform.

[xmlguy]

One way to better understand the wave you're building is to view it with oscilloscope software or hardware.

http://www.zeitnitz.de/Christian/scope_en

http://www.picotech.com/picoscope-oscil ... tware.html

That way you don't have to guess what the signal looks like, you can see it.