8 or 16 voices

[OneLittleFonzie]

Good day,

I have my rev2 for about 1 year now. I really love the sound and endless possibilities.
But, I only have 8 voices. When I had real analog synths before (juno 60, super jupiter, matrix 6), I never really had issues with running out of voices. But I do easily with the rev2. Is my memory wrong or am I doing something wrong? I have tried several priority settings.

I may go for the expansion but I would have to sell something else to fund that (probably my MPC1000).

Thanks for the advice

[RobustAmerican]

8 voices can be a little rough if your trying to use layered patches with longer envelope release times. Using both layers with long release times can bring you down to 2 usable notes depending on the sound in question. In layered patches, each key/note uses 2 voices, so a 2 note chord (interval) will use 4 voices. That's half of your available voices already with only 2 notes played!

Here's a few things that might help if you stick with the 8.

1. Try using more single layer patches. (Obvious I know but you can get amazing sounds out of 1 layer)
2. Try to use shorter envelope release times with layered patches.
3. Use the FX section to hide the short release times. (Reverb, Delay etc...)
4. If you use a DAW, edit the midi parts so all notes aren't sustained for the whole length of the chord.
You can usually get away with editing the lengths of the lower notes in the chord. The higher, and usually brighter notes in a chord tend to obscure the lower keys when they are edited to be slightly shorter. I'm talking like half a bar or less depending on your bpm.
4-B. You could also split the midi parts up. Break a 5 note chord up into 2 chords with 2 and 3 notes. Track them into your device using 2 passes.

Or just get the damn update   


Randy.

[MPM]

You’re not using all 4 audio outs are you?

[OneLittleFonzie]

8 voices can be a little rough if your trying to use layered patches with longer envelope release times. Using both layers with long release times can bring you down to 2 usable notes depending on the sound in question. In layered patches, each key/note uses 2 voices, so a 2 note chord (interval) will use 4 voices. That's half of your available voices already with only 2 notes played!

Here's a few things that might help if you stick with the 8.

1. Try using more single layer patches. (Obvious I know but you can get amazing sounds out of 1 layer)
2. Try to use shorter envelope release times with layered patches.
3. Use the FX section to hide the short release times. (Reverb, Delay etc...)
4. If you use a DAW, edit the midi parts so all notes aren't sustained for the whole length of the chord.
You can usually get away with editing the lengths of the lower notes in the chord. The higher, and usually brighter notes in a chord tend to obscure the lower keys when they are edited to be slightly shorter. I'm talking like half a bar or less depending on your bpm.
4-B. You could also split the midi parts up. Break a 5 note chord up into 2 chords with 2 and 3 notes. Track them into your device using 2 passes.

Or just get the damn update   


Randy.

Thank you

1. I actually prefer single layer sounds. Much nicer and cleaner than a damp forest of sound.
2. Doing that already (after some advice on facebook)
3. I only play everything live. I hardly record midi.

I'll get the update :-)

In general I know how to fix this (I do have quite some experience (30 years)). I was just wondering why I have this problem more prominent on the rev2 than every other analogue synth I've had. 

[RobustAmerican]

Thank you

1. I actually prefer single layer sounds. Much nicer and cleaner than a damp forest of sound.
2. Doing that already (after some advice on facebook)
3. I only play everything live. I hardly record midi.

I'll get the update :-)

In general I know how to fix this (I do have quite some experience (30 years)). I was just wondering why I have this problem more prominent on the rev2 than every other analogue synth I've had.

Sorry for the "basics 101" stuff OneLittleFonzie.   

Maybe it will help someone else. I got my first synth in 85 (still have it) and compared to my other instruments, the Rev2 seems normal as far as voice stealing goes. No better, no worse.

[OneLittleFonzie]

Sorry for the "basics 101" stuff OneLittleFonzie.   

Maybe it will help someone else. I got my first synth in 85 (still have it) and compared to my other instruments, the Rev2 seems normal as far as voice stealing goes. No better, no worse.

No worries :-)

I'm going to take all the advice and make some music this afternoon :-)

[AlainHubert]

Only in very specific circumstances will an 8 voice synth display voice stealing artifacts. When the patch has very long envelope release times, when a patch is layered, when a succession of chords is played through a sequencer, or if a particular synth is multi-timbral (which the REV2 isn't).

Otherwise, 8 note polyphony should suffice. That being said, I got the 16 voice REV2 because the price difference wasn't that much, and I knew that I like long pads patches... 

[LoboLives]

I’m nit really a fan of the sound of 16 voice analog synths. I’ve heard the REV2, The Prologue and the Moog One and I find the sound to be too busy sonically (messy isn’t the right word but it’s thr only one that comes to mind) and I have a bit of trouble getting it to sit properly in a mix, especially if you have other synths going as well. With long release times it doesn’t even sound musical I find, it just sounds like a wall of sound.

I look at it this way, Yamaha CS80, Oberhiem OBXa and Roland Jupiter 8. ALL some of the most praised and listed after synths of all time...and all of them 8 voices. Not saying you should base your decision on that but still it’s food for thought.

A lot of people disagree with me strongly but I think 10 voices on an analog synthesizer is the perfect number. Anything more than that isn’t musical imo.

[OceanMachine]

Odd perspective... 16÷2=8 or less when you're possibly dividing it by 3 in the case of the One. Also, one would think "too busy" has more to do with said layers and modulation possibilites more than voices. Furthermore, the PX you own is a 16 (albeit possibly cut in half in stereo mode or even quartered when bringing in the second layer) voice that can no doubt produce more complex (I.E. busy) sounds than anything you've mentioned.

[LoboLives]

Odd perspective... 16÷2=8 or less when you're possibly dividing it by 3 in the case of the One. Also, one would think "too busy" has more to do with said layers and modulation possibilites more than voices. Furthermore, the PX you own is a 16 (albeit possibly cut in half in stereo mode or even quartered when bringing in the second layer) voice that can no doubt produce more complex (I.E. busy) sounds than anything you've mentioned.

This is why I specified “analog” synth. PX and samplers are a different story.

[OceanMachine]

I fail to see your logic. It's almost literally an emulation of the REV2 with the addition of two sample oscillators. More importantly, you've avoided the main point.   

[Razmo]

Well... for ambient sounds like pads and ambiences, and also some sounds with a long ring-out, 16 voices is important in my opinion... otherwise you'd be set up with nothing but organ-type "on/off" like sounds to work with... maybe this is what some users need only, but in my case there sure as h*** is no way around long release times.

Simple emulations of bells and plucked sounds also have long release times, and every time I use a layered voice on my 16voice REV2, I hear the voice stealing... it's rather obvious.

There is a way to make voice stealing harder to hear on any polysynth, no matter how many voices, but Sequential do not support that voice allocation method in their synths... it's a simple mode, where every time you hit a note, that was hit previously, and is still sounding, then that exact voice is retriggered, instead of triggering a new voice that would steal another voice with a different pitch...

That allocation method makes note stealing harder to notice, and at the same time it mimics how a lot of real instruments work (if you pluck an already sounding guitar string again, you do not get two overlapping string sounds, that string is simply retriggered)...

I've talked about Sequential creating support for that playing style in here before... and with an 8-voice synth this method really has some benefits voice-wise.

The Novation - PEAK has exactly this feature, and I love it for it...

and some ideas on how to overcome voice stealing:

One that I often use is to simply make the release times shorter... it's basically the only way to fix it... and instead (if I want more "tail" in the sound), I simply use either reverb or delay to mimic the release...

[LoboLives]

I fail to see your logic. It's almost literally an emulation of the REV2 with the addition of two sample oscillators. More importantly, you've avoided the main point.

Being that it sounds “busy”. Not really because sonically having a piano or cymbal sample with long release times is completely different sonically than an analog synthesizer oscillator with a long release times. Samples on the PX can be short/medium/long/looped and offer a variety of tones where as more oscillators offer simply more of the same.

[OneLittleFonzie]

One that I often use is to simply make the release times shorter... it's basically the only way to fix it... and instead (if I want more "tail" in the sound), I simply use either reverb or delay to mimic the release...

That is pretty much what I do now. The RV-500 plays very well with the Rev2; not a fan of the internal reverb though.

[Razmo]

One that I often use is to simply make the release times shorter... it's basically the only way to fix it... and instead (if I want more "tail" in the sound), I simply use either reverb or delay to mimic the release...

That is pretty much what I do now. The RV-500 plays very well with the Rev2; not a fan of the internal reverb though.

Agree... it lacks proper modulation options... you can fake that though, by layering a copy of the sound in layer A into layer B, set this layer B's effect to reverb and 100% wet (control the mix with layer B volume instead)... then simply put some vibrato on layer B to simulate the modulation... and make the release time on the AMP envelope longer... this gives you a longer ambient reverb with modulation faked.... only drawback is that you again get half polyphony because of the layering... but I find this trick very useful for my ambient presets.

[OceanMachine]

Being that it sounds “busy”. Not really because sonically having a piano or cymbal sample with long release times is completely different sonically than an analog synthesizer oscillator with a long release times. Samples on the PX can be short/medium/long/looped and offer a variety of tones where as more oscillators offer simply more of the same.

It's not about your personal preference towards using the PX as a rompler, it's about you conflating busyness with number of voices. The problem being, this might confuse someone with little knowledge on the subject that's coming into threads like these looking for sound advice. The reason the classic lusted after synths have lower voice counts is due to the technology of the time (namely stability) than anything else. Eight voices being split between even two layers is a paltry sum and you let nostalgia guide your view it seems.

For the OP's use (a single layer), eight should be enough, except for when using a sustain pedal where that limit will become apparent rather quickly.

[LoboLives]

Being that it sounds “busy”. Not really because sonically having a piano or cymbal sample with long release times is completely different sonically than an analog synthesizer oscillator with a long release times. Samples on the PX can be short/medium/long/looped and offer a variety of tones where as more oscillators offer simply more of the same.

It's not about your personal preference towards using the PX as a rompler, it's about you conflating busyness with number of voices. The problem being, this might confuse someone with little knowledge on the subject that's coming into threads like these looking for sound advice. The reason the classic lusted after synths have lower voice counts is due to the technology of the time (namely stability) than anything else. Eight voices being split between even two layers is a paltry sum and you let nostalgia guide your view it seems.

For the OP's use (a single layer), eight should be enough, except for when using a sustain pedal where that limit will become apparent rather quickly.

I’ll take that as a compliment.

[Razmo]

Being that it sounds “busy”. Not really because sonically having a piano or cymbal sample with long release times is completely different sonically than an analog synthesizer oscillator with a long release times. Samples on the PX can be short/medium/long/looped and offer a variety of tones where as more oscillators offer simply more of the same.

It's not about your personal preference towards using the PX as a rompler, it's about you conflating busyness with number of voices. The problem being, this might confuse someone with little knowledge on the subject that's coming into threads like these looking for sound advice. The reason the classic lusted after synths have lower voice counts is due to the technology of the time (namely stability) than anything else. Eight voices being split between even two layers is a paltry sum and you let nostalgia guide your view it seems.

For the OP's use (a single layer), eight should be enough, except for when using a sustain pedal where that limit will become apparent rather quickly.

In my opinion, it depend... just because it's a single layer voice, does not means you will not run into voice stealing issues... I've made several ambient patches, with long release times, and I would definitely not want only 8 voices for those presets.... I don't like this "one size fits all" debate when it comes to synth... each to his own, each has different needs and requirements... some may be perfectly fine with 8 voices, other users (like me) would always go for at least 16 voices, if I can get them... I admit that Ambient is probably one genre that crave lots of voices compared to other genres, but I do not like these generalizations...

[OceanMachine]

just because it's a single layer voice, does not means you will not run into voice stealing issues...

... Never said otherwise. In fact, I specifically said if the sustain pedal is in play, then that would be the case. Personally I get by just fine with a six voice & would certainly do just fine with two more voices. Thus I said should be enough, which is my opinion, but if the OP is running into that limit on too many occasions, then he can always decide to upgrade. The more voices the merrier is where I stand, which is why I used to own the sixteen voice REV2, not the eight voice.

I’ll take that as a compliment.

Dog bless ye! 

[AlainHubert]

One good reason to elect to get the 16 voice version is the fact that you can layer A+B and get 4 oscillators per voice (+ 2X sub), with 8 voices polyphony ! Of course you can also do that with the 8 voice version, but you'll end up with only 4 voices polyphony, and won't be able to create long pads without voice stealing.