If you start doing synth modeling (which I think is one of the reasons Rudess has problems with the Kronos, which models some of its sounds) on any CPU, you can drag polyphony down quite quickly. It really doesn't matter what CPU you have in there, a few stacked, modeled synth sounds, and you are going to reach your limit.

TBH, there are still considerable drawbacks to CPU generation of sounds, especially live, as voice count is dynamic depending on what VSTi's you use, and when the ceiling is approached, FAR more nasty things happen than mere note (or voice within a note) dropout. Your entire CPU can drop out..!

At least arrangers haven't generally gone this route so far. I KNOW that, no matter what sound I am using, I get that full 128. You can't say that with certainty on a CPU generated instrument.

Hi Diki

Most VST manufactures realised what you mentioned a number of years ago, which is why most these days have been optimised to do more with less power and have a setting which activates the algorithm to prevent overloads from happening. (As most VST are designed to be used live these days, it gave them the incentive to solve the problem)

NOTE: It is not normally activated by default, as most players’ test everything out before playing with it live, but it does help the less experienced.

BTW: I can guarantee that no matter how or what you played on the Wersi Pegasus Wing you would never reach its polyphony limit, and that is a CPU based instrument.

Technology and software has come a long way since those early days when the problems you mentioned were rampant.

Bill

You should try some of the Arturia stuff, Bill. Absolutely spot on emulations of most of the great synth classics (Prophet 5, CS-80, Minimoog, Moog Modular, ARP 2600, Jupiter-8 etc.) all done in modeling. These will pretty much bring anything to its knees, especially if you do the same insane layering and stacking that can choke a hardware synth, too.

Bottom line is, it is down to the player.

But unless you install these sorts of things, generally, most arrangers rely on simple sample playback, which taxes a CPU hardly at all. Things like Gigasampler can give you incredibly high voice counts, which might make you think you are invincible! But trust me, start running a bunch of modeled synth stuff. Then start stacking.

Somebody somewhere will ALWAYS find a way to stack too much for any system, then complain that polyphony isn't sufficient. Until there are unlimited resources, this will always be the case..!

If you are a great player a monophonic synth will sound awesome in your hands .... if you aren't a 128 voice polyphonic synth will sound like a mess. Stacking sounds to the ceiling aint gonna help but will probably make things just sound even messier.

Just learn to play and don't get so hung up on polyphony. Musicality beats technicality hands down every time. Personally I prefer simpler pure sounding voices. Sounds that are over stacked just sound fake and don't blend as well.

Yep, could not agree more.

Dennis

The thing that makes modern arrangers sound so great isn't really a stacking thing. SA/DNC sounds are so good because of sample switching, not stacking and layering, generally. In other words, you get different voices (or oscillators, or samples, whatever you want to call them) sounding depending on how you play. Play staccato, get one sound. Play legato INSTEAD of the staccato, you get the legato sample. But only ONE voice is being used...

Drum sounds, same thing. Many modern arranger's drum sounds have four or so different samples, for gradually increasing velocities. But only one plays at a time. Some of the big drum VSTi's have as many as 64 samples per drum (or more), but you are only hearing the one per note.

I agree with Dennis and Nigel... there are FAR more important things to worry about than maxing out a 128 voice arranger!