Although I was totally immersed in technically complex patches, noodles and synthesis stuff, I suddenly had the need for a versatile keyboard sound with a strong analog flavour, standing out between other instruments and just a minimum of knobs to control the sound while playing. It had to be able to do a bright sharp leadsound and with the tweaking of a single knob also a phat bass at the lower end of the keyboard. The filter is under the modwheel for expressive play and the patch is slightly velo sensitive.

Adding a very bright gated reverb with very short initial delay can further enhance the sound.

Later I found that when the second patch (just another patch I happened to be working on) is loaded in another slot the sound becomes more sentimental, in the middle of the keyboard there is some resonant tail that vaguely reminds of a sitar-like resonance.

I almost sold the Nord Modular already, since most synth patches somehow don't suit me, which may also be because of the lack of reverb and long delay FX inside the Nord, and my A3k Sympler is currently down because of an OS-upgrade, so I don't really have any (realtime) FX box at the moment. So what I did is go though most of the existing patches and edited the ones that i liked to get a warmer sound (which I'm after, I don't like most FM stuff).

So here's a patch with a nice old synth character.

Your patch uses the Osc´s/LFO´s in a quite tricki way... This slave connection from the main Osc to the Sub Osc is cool... Maybe an old invention or maybe yours? Another cool thing is the feedback loop with a Partial Generator... subtile... but realy usefull in the mix.

I decided to compare the LFO settings of this patch to the high resolution Osc´s and made this test patch. Knob 4 changes between LFO and "real" Osc´s. I am posting this test patch because I was astonished about the result, because of this Sub Oscillator arrangement the LFO´s as Osc win with big distance. At least for fatness and charakter... The highend could be more precise... Hmm maybe a tackpatch... Both Osc configurations running simultanious.. Maybe sombody else finds the direkt comparison informativ... It's also remarkable that the tuning behaviour between Master and Slave Osc´s ist different for the LFO´s... Tempered scale against math divisions?

What I found with the modular is that it's sounds lack "colour" in some way, which has the benefit that you can start with clean uncoloured sounds which wouldn't be possible if the sound character would be more obvious(=coloured).

However, I tried to colour my synth patches to have a definite character. A really good way is to use EQ's. Shure, you can do it on your mixing desk, too, but that will unlikely save you setup-time.

I did a bass patch which uses 3 EQ's. Without them it sounds "OK", but when you switch them on, the sound gets more "defined", more recognizable, louder and more energetic, which is what I'm after.

What I found with effect processing is that it often sounds WAY better when you take only certain frequencies out of the sound to stuff it through an effect and colour the sound of the effect afterwards by using an EQ.(and then add it back to the dry signal at a certain ratio)

I used this technique is some previously posted patches (I added a chorus to a highpassed signal) and did it again in this bass patch, but a different effect:).

Even though there are people who claim that modular sucks when it comes to its D/A conversion and all the people who complain about its supposedly [?] 'thin' and 'digital' sound: you can do many things to make it sound a bit warmer... Feed its signal through a tube compressor or a tube eq. that's why I still have my tl audio eq 1: there is no need to make genuine analog sounds 'more analog' and warmer, but sometimes a bit of tube eq helps the modular to sit better in your mix and sound better in every respect.

One of the most 'analogue' sounding solutions I heard was Rob Hordijk's 'one sample delay' using a Crossfader Module in 'feedback'.

The trick is very simple. Problem with most digital synths is there is quite a lot of energy in the area above 8kHz. On an analog this is mostly filtered away by the circuitry after the VCO's. Imagine that if the frequency is 200 Hz there are 60 harmonics present between 8kHz and 20kHz. And although their amplitudes are low their sonic energy is high. It can eg create a buzzing that can completely drown your hihats.

Getting rid of this frequencyband can make the sound more analoguish. Even if there is a filter with resonance after the VCO's, as if the resonant peak is in this above 8kH area it can spoil all your other subtle effects in this high. No exciter can cure it.

Its all about the cheapest way to get rid of it. A 6dB LP can help and is reasonably cheap. But one can aslo make use of the fact that the output on a module lags one sample on its input. If a mixer is fed back this creates a slight filtering effect, in fact a 6dB LP filter. But the input signal has to be reduced to avoid overflow. The effect is simply best explained with the crossfade mixer. If the output is fed back to input2 the crossfade knob shows the filtering effect. Regrettably it is 1.7% dsp, but it can never cause overflow. Still a 6dB LP filter would do the same job for less DSP. But there are times when two signals are mixed and the third input of the mixer is unconnected. Then this input can be used to do the trick. But care must be taken the other inputs are turned back to avoid too much signal boost. Many times you get this opportunity to do something about the sound for free using a mixer module that way.

My personal favourite is to use a static 12dB filter with some feedback of the LP and a mix of the LP and HP that put the filter in what is called an elliptic filter mode. This makes it possible to get a notch at around 10kHz to 12kHz and boost the low end by the feedback from the LP output. It is the most expensive way, 3.6% DSP, but also does the most rigid job. It also emphasized the spatiousness of unisono sounds. It is a good trick for padsounds layered under vocals.

Another trick when using a chorus is to filter the input of the chorus with a 6dB LP somewhere between 2.5kHz and 5kHz. This also prevents too much buzzing in the very high.