In 1998, two years
after original JP-8000, Roland
released rack version 8080. Many thought of it as ''just a rack
version'', however, the 8080 is much more than that. With external
inputs it opens endless sound manipulation and efx possibilities. First
we will list 8080's main differences vs JP-8000.
10 voice polyphonic
compared to the 8000 which is 8
voice.
The JP-8000 does not
have additional noise generator
on OSC2. And that is not just ''noise'' but additional sine oscillator
(see oscillators section).
A 12 band filter bank
for processing internal or
external sounds.
JP-8080 has two
external inputs that you can use just
like an oscillator. There you can apply FM, Ring Modulator, Filter,
envelopes, effects to incoming signal.
True vocoder. Compared
to most other synths where you
plug the mic and sing while playing internal tone generator. On
JP-8080's vocoder you got another input for the carrier. If you want
you can use your voice as modulator and a jet plane engine as a
carrier. That's how real vocoders perform.
Voice Morph. Maybe not
the best term, but this
feature lets you use sound input as a controller. Just like you are
tweaking a mod wheel, but instead of mod wheel it is your voice (or
external sound of anything you imagine) changing any parameters you
desire - filter cutoff, oscillator pitch, amplitude etc. Powerful and
unique function.
Made of metal vs 8000
plastic.
Oscillators
Just like on
JP-8000 there are two oscillators,
with additional noise oscillator on OSC2 and external audio input that
can be set on OSC2. This way you can apply amplitude modulation to all
incoming signals, or use frequency cross modulation (FXM) where
incoming audio signal will modulate the frequency of OSC1. You can also
process the incoming signal through the filter, apply envelopes, LFO
and effects.
It is imporatnt to note that it's not just pulse osc that offers its
modulation
(classic pulse width modulation - PWM), but each oscillator
has its own unique modulator set by two control knobs (control 1 and
2). Lets start with the famous Supersaw. With control 1 you set up
detune amount, while
with control 2 you set up amount as seen on image below.
Second oscillator offers Triangle Mod wave. It is a type of triangle
wave in which you can add new overtones with control 1 while with
control 2 you set up LFO depth. Image below shows an example.
Third oscillator is Noise. On first sight one may say it's just a white
noise. But it is more than that. It is a noise generating system where
you can build brown or white noise (any everything in between), or
crank up the resonator and build a pure sine wave - therefore getting a
sine wave oscillator. With control 1 you setup cutoff point, and with
control 2 you set up resonance. Note: this has nothing to do with
filter section. This is all happening inside oscillator 1.
As fourth oscillator we have Feedback Osc. With control 1 it is
possible to vary the sound of harmonics, while with control 2 we adjust
the amount of feedback to adjust the level of harmonics. Very
interesting oscillator.
Fifth oscillator is standard Pulse (square) with pulse width setting on
control 1 and LFO amount on control 2.
Sixth oscillator is a Sawtooth wave. It offers shape function, which
lets you shape various types of saw waves so that you can emulate
various synthesizers, as almost every analog synthesizer has a
different saw wave. Image below shows an example.
Seventh oscillator is Triangle wave. With control 1 you can deform it
as shown on image below.
Sound
demos
Below
are sounds that i programmed on
JP-8080. Most of them use a touch of chorus or delay, because
that is expected in most situations. Those sounds that are 100% dry
have a note. There is no any additional processing except
normalization.
plutonium.mp3
- Rich spectral ambient sound. Music
composed by Sealed resonances-oberheim_12dB.mp3 -
Exploring the
resonant 12dB filter. sweep.mp3
- Classic saw sweep sound. feed+xmod(guit-tap).mp3-
Emulation of guitar tapping technique using feedback oscillator and
xmod. bpf-organAM.mp3
- Organ type of sound with band pass filter and amplitude modulation
applied. As the sound is evolving i started to change the frequency of
the OSC1, to change AM ratio. Music
composed by Sealed ambient_atmosphere.mp3
- Ambient type of sound. On the second part of recording starting at
0:26, Roland JV/XV
synth owners might notice the sound reminds a little bit of the
legendary ''feedwave sample''. Click here to hear
the original feedwave recorded from the JV series. As you can hear they
are close, specially if you filter JV feedwave, you will get almost the
same sound. Note: i
purposely changed the name to feedwave - original name is
''Feedbackwave'', but i don't want anyone to mix it with feedback
oscillator on
JP-8080. It is just a coincidence they have same names. The sound you
hear was not made on JP's feedback oscillator because that is a
completey different thing. xmod.mp3
- With xmod, some crazy sounds can be built. 100% dry sound. jup2.mp3
- Classic Jupiter filter, not as JP-8, but not that bad either. xmod+cur-rec.mp3
- Square wave with some xmod. gate1.mp3 -
EXTERNAL INPUT processing. Classic ''gate'' example of external sound +
filter + effects. And some envelope to remove clicks at start / end. raw-supersaw.mp3
- This is how raw supersaw sounds. 100% dry sound. While playing, the
detune function was put to higher values. Music
composed by Sealed feedback.mp3
- This is raw feedback oscillator in action. Both control 1 and 2 were
changed during recording. 100% dry sound. saw-12db-40res-wet-dry.mp3 -
12 dB filter test with
40% resonance. First part of recording contains delay, second is 100%
dry. Music
composed by Sealed saw-24db-40res-wet-dry.mp3 -
24 dB filter test with
40% resonance. First part of recording contains delay, second is 100%
dry. Music
composed by Sealed blur_pad.mp3 - Blur pad sound,
with
very small frequency width. roland-filtr.mp3
- Jupiter type of bass with cutoff sweep. old-saw-2.mp3
- Italo dance saw type of sound. Created from individual saw waves (no
supersaw here). gate2.mp3
- Same as gate1 example but with gentle FXM applied. AM_Noise.mp3
- Having fun with amplitude modulation and noise. 100% dry
sound. OSC.mp3
- Tweaking single oscillator. 100% dry sound.
Distortion
issue
Unfortunately with
JP-8000, Roland engineers overlook
one very important feature, the mixer level prior to filter input.
Technically, if you bring too much input level to digital filter, once
you engage the resonance, due to non linear circuits, result will be
distorted sound. Only solution is to reduce the level of oscillator
itself - that will prevent the distortion. Same issue had legendary JD
series. Roland later solved this problem on Super JV series with Tone
Gain setting where you can reduce the level of oscillator to -6 dB.
Later, with JP-80x0 series they unfortunately omitted this function.
What really puzzles me is that no one reported this issue on JP-8000
(do people program synths at all, these days?). If reported on time,
I'm sure Roland would solve this issue before 8080 was released - they
had 2 years of time. Here is a brief explanation of the problem:
When you switch to 12 dB, signal from the oscillator section overloads
the filter, making it completely useless. Output is distorted / clipped
sound that can be endlessly frustrating if you want to create some
gentle ethereal sweep pad and then distortion suddenly cracks in,
spoiling it totally.
Image above shows a recording at about -6dB, so that clipping can be
more apparent in time /amplitude domain analysis. The patch is typical
single mode, two osc sound, one octave apart, both saw waves. Filter is
modified by env and LFO. Resonance is at about 110 (of 127 max). AMP
level is at standard 64. All EFX=OFF
You can hear the audio example of above image, right here
Chord version (where clipping is more apparent) sounds terrible.
AMP level has no effects on this issue, as it is located after the
filter. Reducing it, just silent the whole sound that already features
clipping. I think the only solution would be to implement new function
that would reduce signal level prior to filter inputs - if Roland will
ever do software update. It is sad, but at the moment, this synth it is
not capable of creating basic resonant sweep sound without clipping.
The above
sound is available to download in both SYSex
and MIDI
file, so you can test it in your
own JP-80x0.
Don't worry. In most occasions you won't notice this distortion. You
can hear in sound demos that distortion is not
present, as long as critical values described above are not reached.
Solution
(for Roland engineers)
The simplest way would be to implement oscillator gain setting (for
example -6dB) that user could manually engage via patch setup (on image
below marked with red as GAIN) - so no need for external faders. You
just set oscillator gain like on Super JV/XP/XV series. If that's not
possible, then perhaps just signal attenuation prior to filter input
(marked with red as ATT on image below).
I also tested if this clipping problem can be solved by reducing
oscillator volume, and results were positive. If we look into structure
of JP80x0, it is possible to reduce the volume of an oscillator prior
to filter input, but as a result we loose one of the oscillators. For
example, we can set oscillator 2 to SAW wave, and oscillator 1 to
noise. Then we set oscillator 1 noise resonance to 0 and noise cutoff
freq to 0. Result of osc1 is pure silence. Using osc balance fader and
turning it toward osc1 position we can reduce the volume of osc 2 and
see how filter reacts to this change. After some volume reduction, the
clipping is gone. So yes, the filter distortion is happening due to its
input overload, not because of some "feature".
I would like clear out that AMP setting has not much with this issue.
The AMP level was at standard 64/127 value. The only thing we can do
with AMP is introduce more distortion due its own nonlinearity, but
that has nothing to do with the filter overload issue described here.
If we take a value of 64 as a standard 0dB gain, when we set it to 127
we are pushing Amplifier into its own nonlinear area which introduces
new distortions (due to additional positive gain). It is a normal thing
when we reduce AMP that distortion got lower. But this is true just for
values above 64 (say, reducing from 127 to 70). Below 64, we can't
reduce the distortion.
About
plutonium.mp3 - Rich spectral ambient sound. Music composed by Sealed
Deep Synthesis: Home