rayzoon technologies

Getting The Most Out Of Jamstix Kits
A Guide To The Internal Audio Engine And Advanced Usage Of The Mapping Functions

In this article I will talk a bit about the internal audio engine of Jamstix and how to get the most out of it using the editing functions that the mapping tab provides. You will see how you can adjust Jamstix kits to the needs of your specific projects and how to get different sounding kits out of the same sample sets.

How the engine works

The Jamstix audio engine uses CMAS technology, which stands for Close-Mono Ambient-Stereo. This means that every sound played consist of three samples: one is a mono sample recorded close to the instrument and the other two are a stereo image recorded with some distance. This approach allows you to control the ratio of close mic to ambient mic signal for each sound.

Starting with Jamstix 1.4 and DrumPak #2, the audio engine also supports a second mode, called CS (Close-Stereo), which uses a stereo sample set recorded close to the instrument. This mode requires 33% less memory and CPU usage than the CMAS sounds and is great for situations where you do not need any ambience and would like to use a room simulator or other DSPs on the host side to develop your ambient sound of choice.

Sustain

The sustain slider is your friend! It's essentially a dampening function. By sliding to the left, you are dampening the sound by shortening its sustain. Use this slider to do any of the following:

You want to reduce sound length, such as to reduce the boominess of the kick or to increase the 'crack' of the snare
You want to reduce ambience without losing the attack and color of the ambient signal

Pitch tuning

As far as the pitch slider is concerned: a little goes a long way! Some sounds are very sensitive to pitch changes and may quickly sound unrealistic (which could be what you're after!).

Some sounds can benefit greatly from pitch changes. The best example is the kick. You can dramatically alter the sound of the kick without it sounding unrealistic. Experiment a bit and see what pitch fits your specific song best!

You can also turn a standard snare into a deep snare by slightly lowering the pitch

X-Samples + Low Mem

You may have wondered what those boxes in the internal sound panel really do. Let's discuss them in detail:

The X-Samples are multiple samples of each sound at the same velocity. Without them, the sound of a drum hit repeatedly at the same velocity will always sound the same, which is hardly realistic. Naturally, having all these extra samples takes up memory space so the X-Samples checkbox allows you to turn this feature off. The result will be less CPU usage at the expense of realism. If you are really hurting for RAM and your song keeps the drums pretty much at the same loudness level then you're better off using Low Mem rather than disabling the X-samples since your song doesn't use many velocity layers to begin with.

The Low Mem box does a similar thing but affects the number of velocity layers loaded for each sound. Turning Low Mem on will save memory at the expense of realism. If you must save RAM and your song has a lot of velocity changes in the drums then disable X-Samples rather than using Low Mem.

I generally recommend to use X-Samples and not touch Low mem, if at all possible, to get the best drum sound out of Jamstix. Most modern hosts have VSTi freeze functionality so you can 'freeze' your Jamstix performance at any time and recoup CPU and RAM without losing the ability to rework the arrangement at a later time as you would if you mixdown Jamstix.

Voice Reduction

Chances are you never really worried about the voice reduction slider but let's talk about it a bit. The Jamstix audio engine plays a sound for every note triggered and the only limit to the maximum number of sounds at one point in time is set by the CPU usage. Some cymbals have very long sustain so repeatedly hitting them will produce more and more sounds playing concurrently, using up a lot of CPU power. A good example is the BrushPak with its 22" sizzle cymbal. It is an absolutely beautiful instrument and Jamstix will happily play 16 or more ride sounds at the same time for a fast rhythm. This produces a very realistic wash of sounds just like the real cymbal.

If we all had 5GHz CPUs then we'd slide voice reduction all the way left and stop reading. In the real-world you may run into a siuation where CPU usage is too high and you can reduce it by moving the voice reduction slider to the right. This raises the volume threshold where Jamstix decides that a sound has become quiet enough to kill it. If you go too far, you will notice sounds being choked. If that happens, move the slider back left.

Outputs

You may find yourself in a situation where you would like to alter the sound of, for example, the hihats on the host side but not affect the other sounds. You can do this easily by taking advantage of the output routing. Jamstix supports up to 17 stereo outputs and each individual sound can be routed to anyone of those. You can send the hihat sounds to a separate output and then use EQs and other DSP plugins on the host side to alter the hihat sound.

The concept of audio outputs is simple: it splits the audio return to the host into separate units that can be manipulated separately in the host. You can achieve dramatic results by splitting the Jamstix output across multiple outputs and then trying various effect plugins on those outputs. You can use EQing to perfect the frequency spectrum for a specific song or you can use flangers and wah-wahs to create a wild electronica kit. Experimentation is key!

Some Examples:

If you'd like to sharpen the hihats and get them 'closer' to the listener, use an EQ to reduce lows and mids and boost highs.

If the snare is too hard, send it to its own output and reduce the mids and/or lows and boost the highs.

If the kick is too noisy, send it to its own output, reduce the highs and boost the low end.

Keep in mind that more outputs mean more CPU usage as more audio buffers are handled by the host. This is especially crucial for hosts that use DXi-wrapped VST plugins, such as Sonar and Project 5.