Zebra PC blog

[beej]

I think you wrote before that there were still a lot of optimisation stuff that needed doing - I'm assuming this is one: when holding down notes that have a quick decay/release, the voice is is still used/processed even though the sound has finished.

This can cause processor overloads because Z2 is processing for example 12 voices when it should only be processing 4. You can hold down notes and watch the processor usage climb, even though all the voices are finished and there's no sound coming out...

The sequencer stuff sounds way cool...

[Urs]

Hey beej,

this is another problem that has caused me headaches for years. What if you modulate or automate Sustain? Then it could be absolutely desired to have a not-sounding voice for a few seconds and suddenly it's coming back. The envelopes can not know this, that's why they keep the voice alive.

Maybe I should add a preference for this, like "release unsustained envelopes"... we'll see...

Cheers,

Urs

[beej]

I see what you mean. How do other synths handle this?

(I don't mean the simpler synths that just release voices that are finished, but ones with similarly complex modulation possibilities?)

Is there a way in advance to predict modulation paths that *could* influence how voices are deassigned, and if so, keep the voices active, otherwise release them? Basically, a way to automatically determine the mode, rather than have a preference? Tricky one...

BTW - Moog dinosaur pics sent

[Urs]

I see what you mean. How do other synths handle this?

(I don't mean the simpler synths that just release voices that are finished, but ones with similarly complex modulation possibilities?)

Is there a way in advance to predict modulation paths that *could* influence how voices are deassigned, and if so, keep the voices active, otherwise release them? Basically, a way to automatically determine the mode, rather than have a preference? Tricky one...

Hmmm, I don't know how others do this... however, automation can not be predicted. Everything else could be evaluated by backtracing modulation assignments. But this could involve quite some levels, even circular assignments by the ModMatrix. Thus I'd prefer to leave this stuff up to human cognition rather than risking dead locks

BTW - Moog dinosaur pics sent

Hey, thanks! They look nice! Too bad you already own Z1.5. Remind me to give you the update for free then!

Cheers,

Urs

[beej]

Thus I'd prefer to leave this stuff up to human cognition rather than risking dead locks

Yep - one of the downsides to complexity is it makes some of the "simple" stuff phenominally complex to implement.

I guess a preference would be a good solution then - I would certainly like to have this 'feature' as it really does effect polyphonic playing for me on a Powerbook with limited DSP. Or maybe it could be on a patch-by-patch basis, so you could turn it on or off depending on what the user was doing with the patch?

Hey, thanks! They look nice! Too bad you already own Z1.5. Remind me to give you the update for free then!

That's really generous of you, thanks for the offer , but to be honest I already got 1.5 in the group buy, and $20 is too cheap of an upgrade price anyway , so I think I'd prefer to give you some money (and a nice bank of patches) as a way to help give back something... everyone benefits then...

[Urs]

Heh

Yes, by preference I certainly mean a global switch that would be saved with preset, not globally for all instances. For sake of simplicity it might even enter each envelope in their extended gui (when [+] is active)... just like single/mono/poly trigger functions...

Cheers,

Urs

[beej]

Good stuff

[kodama]

Much fun with combs!

Does the external input work yet?

I sooo wanna run some drum loops, etc... through them!

Also, will be getting the old flat zebra skin soon?


Thanks!

[Urs]

Much fun with combs!

Does the external input work yet?

I sooo wanna run some drum loops, etc... through them!

Also, will be getting the old flat zebra skin soon?


Thanks!

External ins are yet to be done...

I think we'll have skins soon, so there's a good chance for a mute one

Cheers,

Urs

[kodama]

So what's up with the other comb filter circuit in the works?

[Urs]

It's Friday!

Today I'm gonna talk about my personal version of paranoia.

It can't be coincidence that all out of the sudden and shortly after Z2 went public, a couple of other plugin manufacturers come out with products that can almost compete with Z2

What can I say? - If these synths should indeed be answers to Z2, they brought them up too early. Way too early.

It's not as if Z2 was lacking anything to be pretty competitive. However, Z2 has also proved to feature one property that I miss in the others: Extensibility. I dare to say, in common it's the GUI that dictates the feature set. If the GUI doesn't sport any empty space, one can't add anything without redesigning the whole stuff. Not so in Zebra - I can just add another module and boom there it is. Or add another entry in one of the drop down lists, such as new comb filter modes.

Now, when I follow the discussions like "Synth X is better than Y because..." it's almost always about the features that are propelled by the marketing divisions of the manufacturers. The more exotic and vibrant the name of the feature, the more hype around it. I should have made a more comprehensive feature list including all oscillator effects, such as "Waveform Scrambling Synthesis®" and "Waveform Harmonic Distortion®". But that's not the way I think. I prefer it simple, calling it "oscillator fx" and folks who can read between the lines will debunk the marketing bubbles that others come up with.

However, there's one synth that came out (is it out yet?) while I was working on the last milestones for the preview that really scared me: iNetSynth of former Koblo guys. I can't say much about it, except for one thing - it also features a fully stereofied signal flow. This is such a simple sounding feature, but it's totally underrated in the discussions.

Sure, they also brag about stereo capabilities. Who's not? But that's IMHO mostly halfassed stuff when there's a mixer at the end of the signal chain that lets you for instance pan filter outputs or something while everything before that is a mono signal. That's how it was in Z1.5 (panning after the Grid) and it's cool, but it's also not the real thing.

Admittedly, when you look at Z2's modules now, there are only pannings when signals are generated but there's no module that mangles the stereo field. Nevermind, these modules will come soonishly.

Let's put it this way: I've got a lot of stuff buried in my drawer. My paranoia tells me that I should open it.

Nevertheless, I'll work on a new preview version over the holidays that will extend the fully working period until end of January.

Merry Christmas,

Urs

[Mike Janney]

Urs,

I wouldn't worry too much about it as long as the competing products are quality. I'd rather have 5 good quality synths to chose from than two even if they have similar featuresets. What gets me more is self promoting marketing that uses words like 'best', 'finest'. Obviously anyone making a product is doing their 'best' under conditions like skill, budget and so on.

/Mike

[original flipper]

HI

BUT ... there is nothing wrong with a little paranoia!

I look foward to the self scrambling unison oscillators though - assuming they are tempo synced.

Flipper.

[beej]

Ok I'm gonna ramble for a while...

There will always be new stuff coming out, and it's always a lottery as to what comes out when.

Software products are getting increasingly complex, and there's always going to be some features in product X that are lacking in product Y.

There's a place for all products. I mean really, there are *way* too many synths on the market, especially complex ones - they are too cheap, too available and everyone's doing it. Back in the day, when you had to spend thousands on hardware to be able to do basic electronic music, it was cool because you had sacrificed things in order to do it. These days, computers have democratized music for all.

It's both good and bad - we continuously look to the next new thing rather than explore what we have already sufficiently well, and yet there are enough products to appeal to different people so everyone's tastes can be accommodated and we don't all sound the same.

I look to the strengths of a particular product, and see whether it can fulfill a role for me that isn't currently being filled.

On the one hand, I love simple synths for ease of setting up and for making sounds by "feel'. On the other hand, complex synths are great when you want to build sounds without running out of resources - it's annoying when playing your 3-osc Minimoog lead and you want to add vibrato... well, you can't!

I've recently played with the demo's of some products, and despite being great products, they didn't instantly inspire me and connect with me. It's not about being "good", it's what works for you. There is no "better than" - it's a KVR disease. There is what you like, what you don't like, what's unique, what's odd, what works in a given context and what you can or can't afford...

I don't use Zebra to set up standard 2-oscillator patches, I'd probably turn to impOSCar for that. What I *do* do is explore unique Zebra features and see what they can do for me - and I can't do that in any other synth. I love the fact that if I want a new oscillator, I just add it, and all the controls appear. This alone is unique to Zebra - most instruments have a fairly rigid design as you point out.

Is Zebra perfect, the be all and end all of everything? Course not. It is what it is, and what it will be. But it's easily enough to stand alone and to carve its own space in the virtual synth world.

I wouldn't worry too much about what the competition is doing. Sure you gotta keep an eye on other products, but you're doing pretty good from the feedback you are getting. We know you want to blow everyone away, but sooner or later the product has to get out there - if you wait forever until it's got everything we'll never see it!

And there will be people who nick concepts, ideas and implementations for their own products. Well, this has been going on since forever!

So anyway, if I had a point when starting to write this, it's this: I haven't seen InetSynth, but honestly, what a crap name for a synth

[Urs]

I look foward to the self scrambling unison oscillators though - assuming they are tempo synced.

It's already there... listen to the following piece based on a Zebralette preset (1 Z2 oscillator) by Biomechanoid:

http://www.u-he.com/music/BioScramble.mp3



(now gotta find time to convert Zebralette presets to Z2 )

Urs