Hive preview (April 1 update)

[Urs]

@ V-GER: It's too late for me to read that (going to bed) but Hive doesn't exhibit the behaviour I mentioned in Sylenth. If you play a low note, it cuts off at the same point it's higher octave brethren do. In Sylenth, there's a hard split point where it goes from bandlimited to full frequency range.

Though in all fairness, all of the bandlimiting is high enough that it never really comes into the noticable range of human hearing, unless you went out of your way to reveal it.

This was in Hive until a day before the public alpha. The non-bandlimited mode uses about four times the CPU of the bandlimited one because it isn't vecotrized yet. So I decided for a little trade off now instead of a CPU panic. The final version will switch, and it'll be as CPU efficient.

That said, there's a good chance that single oscillator patches will be much faster as well. I saw someone pointing out that Hive uses more or as much CPU as its brethren - this is only true when unison is set to 1 or 2 for each oscillator and Subosc waveform set to "like osc". From 3 on Hive uses less CPU, and from about 6 it uses less than any of the synths we've checked out. Some however forgo of oversampling the filter distortion and the effects - which is a compromise we didn't like. The main CPU usage however is the oscillators, no matter what. Hive's synth part with a lot of unsion and "Clean" filter should be as fast or faster than any synth out there with the same setup - there's not much room left for optimization.

[Urs]

Btw i really hope that the final version won't change the preset. I've been designing sound like crazy with this synth, and i'd be dissapointed if the final version will render my preset unusable.

No worries, no change will be big enough to render presets useless.

If you work at 44.1/48 kHz you won't hear much of a change from this version to final. We have yet to figure out if we need to adjust filter tuning for 88.2/96 kHz or higher. But that should only involve cutoff changes of, say +/- 2.00 or so.

The things we add (waveforms maybe, Matrix targets, stuff for the hexagon) won't change existing sounds.

However, what might change to a degree is the order of automatable parameters. We'll point that out before the final release, possibly with a table of changes. This way one can update the automation in existing projects.

[tehlord]

Here's quick demo of the presets I've come up with in few days. I think HIVE is my favourite U-he synth because it's so fast to program. I want to see oscillator sync though in the future and maybe some more mod targets, like another mod slot knobs...

Great demo

[xamido]

No worries, no change will be big enough to render presets useless.

If you work at 44.1/48 kHz you won't hear much of a change from this version to final. We have yet to figure out if we need to adjust filter tuning for 88.2/96 kHz or higher. But that should only involve cutoff changes of, say +/- 2.00 or so.

The things we add (waveforms maybe, Matrix targets, stuff for the hexagon) won't change existing sounds.

Thanks for clearing that. And i'm hoping there will be sync in the future, or did i miss something? Is it possible to make sync sound with the mod matrix?

[Urs]

No worries, no change will be big enough to render presets useless.

If you work at 44.1/48 kHz you won't hear much of a change from this version to final. We have yet to figure out if we need to adjust filter tuning for 88.2/96 kHz or higher. But that should only involve cutoff changes of, say +/- 2.00 or so.

The things we add (waveforms maybe, Matrix targets, stuff for the hexagon) won't change existing sounds.

Thanks for clearing that. And i'm hoping there will be sync in the future, or did i miss something? Is it possible to make sync sound with the mod matrix?

I'll be evaluating hard sync shortly. Those are decisions we need to make carefully. There are multiple concepts on adding hard sync, which can't be changed once added. We need to get this right from the beginning.

[tehlord]

I know you can adjust oscillator phase via the mod matrix, but it would be nice to have a dedicated button next to each main oscillator just like you know what.

[fourthmanfire]

Awesome! Are you going to do tutorial videos or sell the sounds or something?

I was thinking of selling the soundbank. 100 sounds for about 10-15 euros... But I am curious if they will add for example oscillator sync or other features so I of course would like to make use of them

Nice sounds! I'd probably buy them..

[Sendy]

@ V-GER: It's too late for me to read that (going to bed) but Hive doesn't exhibit the behaviour I mentioned in Sylenth. If you play a low note, it cuts off at the same point it's higher octave brethren do. In Sylenth, there's a hard split point where it goes from bandlimited to full frequency range.

Though in all fairness, all of the bandlimiting is high enough that it never really comes into the noticable range of human hearing, unless you went out of your way to reveal it.

This was in Hive until a day before the public alpha. The non-bandlimited mode uses about four times the CPU of the bandlimited one because it isn't vecotrized yet. So I decided for a little trade off now instead of a CPU panic. The final version will switch, and it'll be as CPU efficient.

That said, there's a good chance that single oscillator patches will be much faster as well. I saw someone pointing out that Hive uses more or as much CPU as its brethren - this is only true when unison is set to 1 or 2 for each oscillator and Subosc waveform set to "like osc". From 3 on Hive uses less CPU, and from about 6 it uses less than any of the synths we've checked out. Some however forgo of oversampling the filter distortion and the effects - which is a compromise we didn't like. The main CPU usage however is the oscillators, no matter what. Hive's synth part with a lot of unsion and "Clean" filter should be as fast or faster than any synth out there with the same setup - there's not much room left for optimization.

Good to know

Also good to know some form of sync is planned. Might be good to go down the self-sync route, to save oscillators and increase ease of use.

I'm guessing nobody thought much of my suggestion for setting the initial phase offsets of key-phase-synced oscillator stacks? Plus an option to lock the sub in a certain phase relationship to the main oscillator? These kinds of features are good for dubstep-type sounds The former is actually in the MicroKorg XL and it's good fun to mess about with. With detune set to zero you can even use it to create a new family of waveforms.

I think the hoover waveform from the Alpha Juno might help sell it as well. Tell them it's PWM on a saw wave and blow their minds

[dwozzle]

Is the 3284 link on the first page of this thread still the current beta?

Thanks.

[Ameyah Audio]

Is the 3284 link on the first page of this thread still the current beta?

Thanks.

yes it is, there is no new beta version. i am sure the whole u-he team enjoys christmas too and the days off from work.

[dwozzle]

Thanks, understood about the holidays, and development time, and life in general, just wanting to make sure I've got latest when I check it out.

[GeorgeZ]

[rod_zero]

@Urs:

I made a Lemur template to control HIve, but it's only possible to map the first 127 parameters. Is there a way around this limitation?

I am trying to map another 127 parameters using another midi chanel but hive seems to only recognize CC's by number and not specific midi channels.

Thanks!

[faun2500]

anyone tell me where a master PAN is?

[Urs]

I'm guessing nobody thought much of my suggestion for setting the initial phase offsets of key-phase-synced oscillator stacks? Plus an option to lock the sub in a certain phase relationship to the main oscillator? These kinds of features are good for dubstep-type sounds The former is actually in the MicroKorg XL and it's good fun to mess about with. With detune set to zero you can even use it to create a new family of waveforms.

You're totally describing a potential Zebra feature here. I had to read twice before I understood it

The sub btw. always starts in a distinct phase relation to the main osc. If you tune statically in octaves you have phase them locked (if not - file bug report!). Pulse and Square start at different offsets (Square being true square wave, Pulse being a sawtooth and a delayed inverted sawtooth).