BrettM

My two cents: Consistent stages across all envelopes. More stages, maybe like those on the PX-5s. I'd also like an additional parameter called something like “Init Time” which allows you to control the timing of the onset of a wave so you can slightly delay its playback if necessary. A bit like the current delay parameter on some of the LFOs. Preferably out to about half a second. LFOs that can go into audio rates Real time control over sample start, loop and end points Cross fading option of sample loop points Better PWM Wavetables Multifunction keys across the entire keyboard An audition key which allows you to quickly compare changes to a tone with its currently stored sound

That's right Alen, the G1 does not behave this way. It's got a mixer you can't mix with. It works the way Ted said: It does work properly in the step sequencer, but not in performance mode. This was the topic of one of my first posts: http://www.casiomusicforums.com/index.php?/topic/5352-mixer-in-performance-mode/#comment-12826 Like Ted and XW-Addict, and probably every other G1 owner, I've learned to live with it and use the knobs as faders in performance mode if necessary.

Thanks for this information guys. One frustration I have is that I when I use the sliders in the mixer to alter individual channel volumes, I only hear that channel, not the full mix. This is how it always works, not just for this set up, and seems odd behaviour for a "mixer". (Not sure if this is a peculiarity of the G1 because of the way its sliders are set up to control the solo synth.) If I could alter the volumes while hearing the full mix then I would be able to truly create a real-time adjustable drawbar organ on the G1.

All channels/zones whose DSP Lines are enabled do seem to be linked. They share volume, pan position and many other parameters unfortunately. I think it's either the zone 1 or the lowest channel number settings that takes precedence. This is especially frustrating with respect to volume because it means you can't blend/mix the tones involved. But there is a work around, at least for volume: use expression (CC#11) instead - http://www.casiomusicforums.com/index.php?/topic/11206-expression-cc11-vs-volume-cc07-and-performance-dsp/ You can assign a knob to change expression for specific zones and thus mix the tones. Unfortunately expression is not saved as part of a Performance, so you have to reset it each time. The multifunction keys send CC#11 on the G1, so you can use a few to simply set the appropriate values. Then all you need to do is press them once to get your desired mix. One gotcha that occurs due to expression not being saved is that it carries over between Performances so you can end up with your previous expression settings affecting the mix in a new Performance. For this reason I always create a few multifunction keys to restore expression values back to 127 in any Performance where I alter them. I just have to remember to hit the multifunction keys before moving on to a new Performance.

Since purchasing my G1, I've long been interested in layering tones as a means of creating new waveforms. Sure I can use the sampler with overdubbing, but I've also been trying to create something similar to the P1's hex layer mode http://www.casiomusicforums.com/index.php?/topic/5483-the-nona-layer-9-layers/#comment-13564 Some time ago I wondered whether the keyboard could be used to control the other MIDI channels in addition to the usual channel/part/zones 1 to 4, given that an XW can function as both a MIDI controller and a MIDI sound source. And it can, for a maximum of 16 possible layers! I doubt that I will ever go this high since the polyphony would take a huge hit and the sound would probably get pretty muddy, but it's nice to know that I can now layer more than the usual four performance zones. No doubt the P1 can do something similar and include a hex layer in zone 1 for a maximum of 21 layered tones! The trick is to route the XW's MIDI back into itself. In researching this possibility on the forum, I noticed that Korakios had a similar idea, http://www.casiomusicforums.com/index.php?/topic/6901-connecting-the-midi-out-to-midi-in-port/#comment-21195 and that some very experienced forum members strongly cautioned against it. They were worried that you might create a set up in which MIDI messages generated more MIDI messages in an infinite loop that overwhelmed and possibly damaged the system, but neither Korakios nor myself have experienced this problem. That's not to say it couldn't happen, so be careful and do this at your own risk. Here is the set up I use: Connect the MIDI In to the MIDI Out/Thru ports using a standard MIDI cable, i.e., plug the XW back into itself. Performance→ MIDI Edit→ Gen Out is usually on for all 16 channels, by default; but turn if off for channel 1, just to be on the safe side. This controls the internal sends of MIDI information and thus disables all internal MIDI sends on channel 1. Set Performance→ MIDI Edit→ MIDI Out to off for all channels, except channel 1. This ensures that only channel 1 is outputting MIDI information via the MIDI cable. Steps (ii) and (iii) mean that channel/part/zone 1 will receive it's information via the cable rather than internally. Set the MIDI in channel to 1 for each part 5 to 16 so that their sound source is controlled by channel 1, the keyboard: Performance→MIDI Edit →Ext In →01 for each part. (Use the Part +/- keys to move between the different parts.) . The standard internal routings are still set up because they were not disabled in step (ii). If you don't want both the step sequencer and the keyboard to control parts 8 to 16, set the Gen Out to off for these channels in step (ii) above, or just don't use the step sequencer. I don't really know why you'd want the step sequencer to be intruding into you're playing, but it might make for some weird/interesting performance effects, especially via the control parts. In the Mixer, assign the Tone you want for each part, set it's relevant volume, pan position, etc. (As in Step iv, use the Part +/- keys to move between the different parts.) Finally, ensure that the XW is receiving MIDI: Setting→MIDI In → On. This is usually off by default and will require a restart when changed. A message comes up on the screen to this effect. Thus the keyboard is sending MIDI over channel 1 to the sound source on channels 1 and 5 through 16. Channels/zones 2, 3 and 4 are unchanged and behave normally. To save this configuration you need to save a Performance and a Step Sequence. The Step Sequence remembers the settings for channels 5 to 16 and the Performance remembers everything else, including the relevant Step Sequence. I've found it useful to be in the Mixer while blending all 16 tones. Here I can set individual part volumes, etc. and, more importantly, mute different parts using the 16 step buttons to listen to them in isolation or smaller groupings. Unlike a hex layer, this technique layers complete PCM tones rather than individual PCM waves but does not allow for control over velocity switching. For G1 owners, it gives access to layering beyond the usual four layers available in a Performance. For P1 owners it provides an alternative way of layering that opens up some possibilities not available in a hex layer, in particular the possibility of having different vibrato rates for each tone, provided they're saved as separate user tones. In a hex layer the pitch LFO is only available globally to all six tones as a group. Different vibrato rates on different layers can be a good way to create lush pads and string tones. === edit === For completeness I should mention how to back out of this. It's ok to leave the Performance and Step Sequence saved, but it's probably sensible to disable the MIDI in as per step (vi) when not needing this set up: Setting→MIDI In → Off. Then switch off the synth and remove the MIDI cable.

Excellent!

Hey Alen, I just noticed a subtle difference between Revisions 1 and 2 of your XW Companion: the cover page of Revision 2 refers to "Volume 1" whereas there is no such reference in Revision 1. Does that mean you have other volumes in the works?

Thanks again Alen, this is truly an invaluable resource. I've read both revisions several times now and I'm still learning new things.

Have you managed to sort out this problem Imanji? The only thing I can think of is that you may have a very light touch which results in key velocity giving a low volume. Does the volume increase if you play harder? If so, you can alter the touch sensitivity via Performance->Controller->Touch.

I just tried the above and it does create audible stepping as LFO1 moves between discrete values of Pitch. So a better method might be to combine LFO1 and LFO2 to drive the oscillator directly. Leave Synth1+Oscillator+LFO1 Depth at its default of +63 in step (i) above, and also set LFO2 Depth to +63. Then set LFO2 Sync to LFO1. Now the two LFO's are applying their combined depths to synth1's oscillator directly. The rate is determined by LFO1 and adjusting the depth of either or both LFOs alters the range of the pitch sweep. It doesn't have the full range of the above Virtual Controller technique, but there is no stepping.

I agree with AlenK, it sounds like a sweeping self-oscillating filter. Self oscillation invariably produces a sine wave. Infact, that's how many of the old synths (especially Arps) produced sine waves. The XWs contain sine waves aplenty, so resorting to a new keyboard or sampling seems a bit drastic. The only hiccup that I can see is the range of the oscillation. It's probably wider than you can achieve using the pitch bend wheel, even set to +/- 2 octaves, and wider than the LFOs can produce using the standard routing. So set up a non-standard routing: i) In a new Solo Synth patch using the default synth1 oscilator sine wave, zero out the LFO1 Depth parameter. ii) In one of the Virtual Controller slots, set Source = LFO1, Depth = +127 and Destination = Synth1+Oscilator+Pitch iii) In LFO1, set your preferred Depth and Rate, probably around 2 or tempo synced to the song. A Depth of 127 will sweep the entire keyboard, which is way more than you need, so find a spot somewhat less than this. Also crank up the reverb in this case. I hope this helps.

You can route LFO2 to modulate the rate of LFO1 using one of the Virtual Controller slots. LFO1 can be a descending sawtooth modulating the filter cutoff at 6 Hz or whatever rate sounds best. Use the LFO1 Depth parameter in the Total Filter block for this. Perhaps the best way to steadily slow the rate of LFO1 would be to set the rate of LFO2 to zero, thus turning it into an envelope. LFO2's Rise parameter then effectively acts as an attack parameter, controlling the rate at which LFO2 is applied to LFO1. Set LFO2's Waveform to Pulse, to give it a positive offset, and set the Virtual Controller Depth to -128 so that a negative value is sent to LFO1's rate, thus slowing it down. Section 6.2 of AlenK's XW Companion describes how to turn an LFO into an envelope in this way, although the discussion is about pitch envelopes rather than filter envelopes. Which reminds me, listening to the song, I would also suggest trying to route LFO1 to oscillator pitch instead of filter cutoff. What sounds like a filter sweep may be a pitch sweep, or both.

Tried this, but it never really worked to my satisfaction, so went back to simply layering a bass tone with vibrato. The tones that seemed to work best were Square Lead 1, shifted down an octave and with plenty of cutoff applied, or Synth Bass 2. Elgin Baylor, thanks for the suggestion about using chorus. This led me to try chorus DSPs instead of the rotary DSP. I especially found the dual chorus+tremolo and chorus+auto-pan DSPs effective, for mono and stereo rotary respectively. They both give precise control over the LFOs involved in the chorus and tremolo/auto-pan. Knobs can be set up to dial in the desired rates for fast and slow rotary effects. The chorus+auto-pan was especially good for creating a real close miked Leslie sound. I'm not suggesting that the chorus DSPs are better than the rotary DSP, but they may offer some viable alternatives.

The fact that expression is not saved within a Performance creates a pretty serious gotcha - it's not reset between Performances either. So if you lower, or zero, expression for a zone in one Performance and then change to another one, you're left with the previous expression value in the new Performance. Worst case scenario is silencing a zone by zeroing its expression and then changing to a new Performance only to find that one of its zones is off. If the new Performance doesn't have a quick way to reset expression for the zone back to 127, via a knob or multifunction key, say, then you would have to go back to the old Performance with the expression control. This could get very confusing in a live situation. So the rule is: if you change expression in a Performance, reset it before leaving that Performance. Expression is the only way to mix zones that are tied by the common DSP line, but may not be the best way to change the volume of a zone otherwise.

I had a few more thoughts whilst dozing off last night: i) Set the Oct. Shift parameter of Zone 2 to -1 for a much beefier bass; although this would effectively create a 32' pipe making it more like a pipe organ than a Hammond. ii) Bass vibrato is easy to create because it's built in to every tone, but I was thinking about how to produce the additional tremolo. One way would be to record a knob twiddle of expression (CC#11) in the phrase sequencer, as described in Section 6.5.2 Tremolo of AlenK's XW Companion Rev 1. Playing back the looped phrase produces tremolo. Master tempo is used to change the speed of the tremolo. iii) The above technique can be expanded to recording two knob twiddles: one for the tremolo and the other for vibrato. The knob controlling the vibrato would be set to one of Performance Zone 2's pitch shifting parameters such as Transpose or FineTune. This technique would allow you to get closer the 90 degree phase shift between the vibrato and tremolo, though it may be difficult to record. iv) For even finer control, with exact 90 degree phase shifting, use two control tracks in the step sequencer sending K1 and K2 values to the knobs in the Performance which are in turn directed to pitch shift and expression of Zone 2, see Section 5 of the XW Companion. You could even create two step sequencer patterns that can be instantly switched from the front panel – one for fast rotary and the other for slow rotary, simply by changing the Step Size parameter of each pattern. v) G1 owners, especially those using my G1 Drawbar http://www.casiomusicforums.com/index.php?/files/file/522-g1-drawbar-pack/ which creates organ registrations via user waves, should note that each user wave has a Pitch LFO for vibrato plus a AMP LFO for tremolo ready and waiting to be applied! You just need to copy DBL88888.ZWT to another user wave slot, set the volumes of splits 2 to 5 to zero, and apply the Pitch LFO and AMP LFO to the wave, which now only sounds split 1, i.e., the bass 16' drawbar.

I've just stumbled on a nice interaction between the knobs and virtual controllers on my G1. If the solo synth engines in each model are genuinely the same, this probably applies to the P1 as well. I've long known that whatever destination is set for one of the first four virtual controllers is also affected by its correspondingly numbered knob. The effect of the knob occurs irrespective of the source, even when it is off, or if its depth is zero. What I didn't know is that the knob can be used to alter the depth of the source in real time. So if you have the source set to modulation, say, i.e., the mod wheel, then the knob will determine the depth of the mod wheel. You can even turn the mod wheel off simply by using the knob. Depending on the destination, the off/zero position will either be fully anti-clockwise or at 12 o'clock. This even allows you to have multiple destinations set up for the mod wheel and choose between them on the fly by simply adjusting their corresponding knobs. Presumably the knobs work the same way for other virtual controller sources as well. Virtual controllers 5-8 don't have corresponding knobs so are free from this interaction.

I wonder if the XW rotary effect can be improved using the synthesis resources on hand, rather than reaching for an outboard effects unit. Two things got me thinking about this: i) the main problem is that the XWs only simulate one rotating speaker instead of two (AlenK's XW Companion Rev 1, p. 23) and ii) the effect is primarily composed of vibrato and tremolo, 90 degrees out of phase with each other http://www.soundonsound.com/sos/feb04/articles/synthsecrets.htm. Let's assume that the XW effect handles the rotating treble horn satisfactorily, so the question is: can we simulate the second rotating speaker by layering a suitable bass sine-wave-like tone under the main organ tone? According to (ii), the bass tone should have vibrato and tremolo, but let's say that vibrato is the more predominant effect, which means that any PCM bass tone is a potential candidate because they all have in-built vibrato. The only requirement is to choose a tone that is suitable dull and sine-like. To this end, I created a Performance with an organ with the rotary DSP enabled in Zone 1 and a dull, sine-like bass tone in Zone 2. I also set up three knobs to control the volume, vibrato rate and vibrato depth of the bass tone in Zone 2. By blending these three knobs, I certainly created a richer sound than was available in Zone 1 alone. G1 owners can use one of the PCM organ tones in Zone 1; P1 owners, of course, have access to the full drawbar organ sound source in Zone 1. Maybe an XW owner with a Ventilator or Burn pedal, or a real Leslie!, could make the comparison and let us know how much of an improvement can be made by simply layering a bass tone with the right amount of vibrato.

Cairnsfella, You can't set the timing of an envelope Sustain parameter, because Sustain is a level parameter, not a timing parameter, whose period is determined by key press duration, by definition. But I think the “auto-sustain” effect you're after can be achieved using one of the genuine timing parameters, viz. Attack Time, Release Time1 or Release Time2. Here are the results of a few envelopes I investigated: Envelope .........................1......2......3......4 Init Level......................126......0......0......0 Attack Time.....................127......0......0......0 Attack Level....................127......0......0......0 Decay Time........................0......0......0......0 Sustain Level.....................0....127......0....127 Release Time1.....................0....127......0....127 Release Level1....................0....126....127....126 Release Time2.....................0......0....127....127 Release Level2....................0......0....126....125 Total Time of Sound in seconds...24....8.5....8.5.....17 So you can get up to 24 seconds of “auto-sustain” from Envelope 1, so long as you hold the key down. Adjust Attack Time for a duration less than this. You can also set Decay Time to create an “auto-release” effect. If you just want to tap a key and let the release stages of the envelope handle the rest, try Envelopes 2-4. Note that if the level parameter between two times is the same, e.g., 127 to 127 instead of 127 to 126, the XW effectively cancels the first stage, i.e., it acts as if the time parameter is zero. This obviously only applies to the solo synth or user waves on the G1. PCM tones don't have such advanced envelopes. On the G1, you can sample a PCM tone and play the sample to give it one of these envelopes.

A number of discussions have advocated the use of CC#11 to control volume, rather than the more obvious volume parameter, CC#07; e.g., http://www.casiomusicforums.com/index.php?/topic/4595-how-to-set-sliders-1-4-to-control-volume-of-zones-1-4/?hl=express#entry7988 I've just stumbled on an even better reason, at least for an XW-G1, but I suspect it works the same for a P1. You've probably noticed that, within Performance mode, all zones whose DSP Line is enabled are tied together. They obviously share the same DSP, but they are also forced to share the same volume, pan position and reverb send. I've always thought that this meant you lose the ability to mix or blend these zones independently, but that's only the case if you try to do it via volume (CC#07). If you use expression instead (CC#11), you can mix the relative volumes of all zones whose DSP Line is enabled. The only shortcoming is that, unlike volume, expression is not saved within a Performance so you have to set it each time. Assigning a knob to send CC#11 to the desired zone is a good way to do the mixing. It's also a good way to "turn off" a DSP enabled zone on the fly while leaving another DSP enabled zone still sounding: just turn it's expression down to zero. I haven't tested this, but I suspect that the same thing happens for channel/parts accessed from within the Step Sequencer. You can probably use one of the control parts to adjust the expression of channel/parts whose DSP Line is enabled to get the desired mix between all parts sharing the DSP.

CairnsFella, I think I have experienced what you refer to in issue 1: growly, distorion-like artefacts in the sampler on some keys. Like you I thought it may be due to some kind of overload and tried reducing the gain setting in the editor during saving. That didn't work. Two things do work for me. First, don't make the sound too loud in the underlying wav file. I do my preparation in Audacity and if I keep the maximum amplitude in the waveform BELOW -5dB or -6dB, the problem doesn't arise. This suggests that it does have something to do with overloading the sampler - amp, pitch shifter, internal digital waveform representation or whatever. Second, the problem only seems to occur after editing Points directly on the G1. Set your loops in the computer editor and don't go into Points and that seems to work, even though it's a lot easier to edit loop points on the G1 itself than in the editor. If you really want to edit points on the G1, keep the volume down, as per previous paragraph.

Now that you've mentioned the song, the way the sound brightens up during the swell leads me to think it is probably a filter swell, as per my first suggestion.

Hi Geoff, As Brad said, the standard way to create a swell would be with an amp envelope, unfortunately the envelopes for the PCM melody tones are so minimal (only attack and release) you can't use that technique in this case. But there are many other approaches. You could route the PCM tone through the solo synth, section 3.4.3.3 in AlenK's XW Companion Rev 1, and do a total filter swell instead. This will probably swell the timbre more than the volume, so you may not want that. So you could set up one of the knobs to control volume and record yourself playing one cycle of a swell in the phrase sequencer, see section 5.1 in XW Companion Rev 1. Or, given that this is the G1 specific forum, you could sample the PCM tone in the vicinity of the keys you want to play, and play the sample in stead. The sampler has the full suite of amp envelope parameters available to each split and is a great way of expanding the limited envelopes available on PCM melody tones.

It seems pretty simple, right: select bend up or bend down as the source of one of the virtual controllers and set the desired depth and destination. The catch is that the pitch bend wheel is hard wired to ... pitch bend. I was recently working on a patch that used bend up, but not bend down, and I thought I could make it more playable if I routed bend down to control Total Filter cutoff. The wheel spring back is very useful for some effects. Problem was I got filter close plus pitch bend down. So the question was, how to cancel the pitch bend. Unlike the mod wheel, whose default vibrato can be cancelled by zeroing Mod Depth in the LFOs, there doesn't seem to be a comparable parameter for the pitch bend. Zeroing the BendRngDwn parameter in the overarching Performance doesn't work because that just stops all bend data being sent and the filter doesn't close. My solution was to cancel the pitch bend with a second virtual controller: source Bend Down, Depth +8, destination Synth1 pitch. To cancel a pitch bend up, you'd use -8 depth. The value of 8 is linked to the bend range set to Zone 1 in the overarching Performance, which defaults to 2. This parameter ranges from 0 to 24 semitones, so it seems that each increment in the Performance corresponds to 4 in virtual controller depth; so you't need a virtual controller depth of 96 to cancel a full 24 semitone pitch bend. Interestingly I didn't need to go beyond the default of 2 to get a full filter close. Unfortunately you have to use a separate virtual controller for every enabled oscillator, so it uses up your virtual controller slots pretty fast. If there's a better way, especially one that doesn't burn through virtual controllers, please let me know.