Hive preview (April 1 update)
-
- Banned
- 511 posts since 5 Jul, 2014
May have been answered already but is u-he going to be working on Hive and releasing it somewhere around late March? I know that it was between that and pushing it out as soon as possible. I don't know about everyone else but I am personally most fond of the first option. Better to have it nice and complete instead of aiming to simply update it later IMO.
-
- u-he
- Topic Starter
- 30215 posts since 8 Aug, 2002 from Berlin
Waiting for the build server to build it - unfortunately a bugfix for Linux has killed the Mac build. This should be fixed by tonight, so a new build may happen tomorrow. Including a Factory Preview bank.2ZrgE wrote:As I'm very short of time I'd like a version with presets asap.
-
- KVRian
- 535 posts since 11 Sep, 2004 from inne Büchs
I think this is so cool. Will you release this tool to the public?dumbledog wrote:
Project status: Good God, what have I done.
Andy is a support ninja.
-
- KVRAF
- 4141 posts since 11 Aug, 2006 from Texas
You're not alone here, I really am enjoying dumbledog's updates too.uselessmind wrote:I think this is so cool. Will you release this tool to the public?dumbledog wrote: Project status: Good God, what have I done.
-
- Banned
- 6129 posts since 9 Oct, 2007 from an inharmonious society
Urs wrote:Waiting for the build server to build it - unfortunately a bugfix for Linux has killed the Mac build. This should be fixed by tonight, so a new build may happen tomorrow. Including a Factory Preview bank.2ZrgE wrote:As I'm very short of time I'd like a version with presets asap.
-
- KVRian
- 948 posts since 25 Sep, 2014
Well I'm gonna have to know what the heck I'm doing first, arrgh.
As it stands I can get two of VCF Frequency, Envelope amount or Sustain amount correct. Once I change things around to nail down two of them, the third slips out from under me and everything goes haywire.
Sendy's comment a couple pages back about thinking of "pre-cooked" synths as modular devices hit home with me and I'm more or less starting to grasp what happens when the three combine in terms of voltages, before the conversion to frequency. So for example (and correct me if I'm wrong here), when you hit a key the cutoff starts at the VCF cutoff voltage, maxes out at the combined voltage of VCF and whatever the envelope is set at, then decays to the combined voltage of VCF and sustain.
But I haven't been able to put the pieces together and went to bed pretty damn frustrated last night.
I'm still at work (posting during lunch, natch) but this morning I noticed something about Hive that I didn't have time to investigate fully. When the resonance is maxed out with both VCF and the envelope set above zero, rather than one super-high frequency I was getting *two* freq peaks, one at the base VCF and another at (VCF+Env). I'll post a screenshot of this when I get home as it's pretty weird. I think this is what was throwing me off, and from a quick look it seems that the Dirty engine doesn't do this. When I get home from work and taking care of the dog I'll see if my theory is correct, though it'll mean recalibrating the frequency and envelopes. Stay tuned.
As for releasing the script... yeah probably. The only misgiving I have is that I've basically had to reverse engineer U-NO-LX (I just got what the LX means last night, lol) so it feels a bit weird, but it's not work that can't be done with the, a frequency analyzer/oscilloscope and Excel (oh my god so much Excel). Maybe when this is done I can do a writeup on Hive's inner workings as well, like why the mod envelope goes from 0-120 and such, as I think this carries over across u-he synths.
As it stands I can get two of VCF Frequency, Envelope amount or Sustain amount correct. Once I change things around to nail down two of them, the third slips out from under me and everything goes haywire.
Sendy's comment a couple pages back about thinking of "pre-cooked" synths as modular devices hit home with me and I'm more or less starting to grasp what happens when the three combine in terms of voltages, before the conversion to frequency. So for example (and correct me if I'm wrong here), when you hit a key the cutoff starts at the VCF cutoff voltage, maxes out at the combined voltage of VCF and whatever the envelope is set at, then decays to the combined voltage of VCF and sustain.
But I haven't been able to put the pieces together and went to bed pretty damn frustrated last night.
I'm still at work (posting during lunch, natch) but this morning I noticed something about Hive that I didn't have time to investigate fully. When the resonance is maxed out with both VCF and the envelope set above zero, rather than one super-high frequency I was getting *two* freq peaks, one at the base VCF and another at (VCF+Env). I'll post a screenshot of this when I get home as it's pretty weird. I think this is what was throwing me off, and from a quick look it seems that the Dirty engine doesn't do this. When I get home from work and taking care of the dog I'll see if my theory is correct, though it'll mean recalibrating the frequency and envelopes. Stay tuned.
As for releasing the script... yeah probably. The only misgiving I have is that I've basically had to reverse engineer U-NO-LX (I just got what the LX means last night, lol) so it feels a bit weird, but it's not work that can't be done with the, a frequency analyzer/oscilloscope and Excel (oh my god so much Excel). Maybe when this is done I can do a writeup on Hive's inner workings as well, like why the mod envelope goes from 0-120 and such, as I think this carries over across u-he synths.
-
- KVRAF
- 6466 posts since 18 Jul, 2008 from New York
Did you manually convert a few patches from synth A to B as a proof of concept before investing 8 million hours attempting to build the Deluxe Automatic Patch Convert-o-tron?dumbledog wrote: Project status: Good God, what have I done.
-
- KVRian
- 948 posts since 25 Sep, 2014
Haha, well I did do a couple Juno patches yes, along with twenty or so patches from other synths, mostly Diva since I was counting on the scales more or less matching up.
Cool thing about the Juno tho is that its topology fits neatly inside of Hive without the use of any switches. I mean here's the Juno in a nutshell:
Just turn up the oscillators you need and set the dials right and in theory you're done. I have a demo patch upthread, maybe a week or so ago. (Note that I had the hi/lo pass swapped here, I changed them back after seeing a schematic)
And since nothing really needs to be toggled, if you set up the mod matrix right you can create morphs between pretty much any Jumo timbre using the mod wheel, second LFO, two spare envelopes or even a combination of them all.
And once THIS is done, I don't think the SH-101's architecture is terribly different and whaddya know, there's already a soft synth by the same guy, with the same patch format...
Cool thing about the Juno tho is that its topology fits neatly inside of Hive without the use of any switches. I mean here's the Juno in a nutshell:
Just turn up the oscillators you need and set the dials right and in theory you're done. I have a demo patch upthread, maybe a week or so ago. (Note that I had the hi/lo pass swapped here, I changed them back after seeing a schematic)
And since nothing really needs to be toggled, if you set up the mod matrix right you can create morphs between pretty much any Jumo timbre using the mod wheel, second LFO, two spare envelopes or even a combination of them all.
And once THIS is done, I don't think the SH-101's architecture is terribly different and whaddya know, there's already a soft synth by the same guy, with the same patch format...
-
- KVRAF
- 6466 posts since 18 Jul, 2008 from New York
The filters in Juno and Hive are not identical. The envelopes, even if you can match the times, will probably have different curves - linear or different exponential shapes. The oscillators are different. And so on.dumbledog wrote: Just turn up the oscillators you need and set the dials right and in theory you're done.
While I'm sure you can get very close with some patches, I would be surprised if it could be done for all patches. I think it would be quicker programming new patches from scratch in Hive instead of proceeding with this approach.
-
- KVRian
- 948 posts since 25 Sep, 2014
This is more fun tho
These definitely aren't going to be perfectly phase-cancelling matchups. As you say, the filters certainly are different:
But I think that's okay. If you want the Juno character and likely a lower CPU hit, go with TAL. If you want to throw in some supersauce, effects or some resonance on the high-pass there's Hive. Or double 'em up for a weird chorus. Think of it as a few hundred new starting points for a new synth
These definitely aren't going to be perfectly phase-cancelling matchups. As you say, the filters certainly are different:
But I think that's okay. If you want the Juno character and likely a lower CPU hit, go with TAL. If you want to throw in some supersauce, effects or some resonance on the high-pass there's Hive. Or double 'em up for a weird chorus. Think of it as a few hundred new starting points for a new synth
-
- KVRist
- 130 posts since 31 Jan, 2014
Hive owner here.
Does anyone else find the detune knob a bit touchy in the lower ranges (below 5.0)?
I realize you can hold shift for finer control, but for a lot of the music I make, only the first quarter or third of the knob is useful - after that, the detune is too great to be useful (generally speaking).
There may be a simple explanation behind the "100" value of the detune knob, and if so, I'm all ears. It just seems like maybe the first quarter of the knob could be "stretched" over a larger range to accommodate more fine control (without having to click the shift key).
Does anyone else find the detune knob a bit touchy in the lower ranges (below 5.0)?
I realize you can hold shift for finer control, but for a lot of the music I make, only the first quarter or third of the knob is useful - after that, the detune is too great to be useful (generally speaking).
There may be a simple explanation behind the "100" value of the detune knob, and if so, I'm all ears. It just seems like maybe the first quarter of the knob could be "stretched" over a larger range to accommodate more fine control (without having to click the shift key).
-
- Banned
- 6129 posts since 9 Oct, 2007 from an inharmonious society
I agree.fourthmanfire wrote:Hive owner here.
Does anyone else find the detune knob a bit touchy in the lower ranges (below 5.0)?
I realize you can hold shift for finer control, but for a lot of the music I make, only the first quarter or third of the knob is useful - after that, the detune is too great to be useful (generally speaking).
There may be a simple explanation behind the "100" value of the detune knob, and if so, I'm all ears. It just seems like maybe the first quarter of the knob could be "stretched" over a larger range to accommodate more fine control (without having to click the shift key).
I noticed something weird with detune.
Sometimes, I had the detune at 0 setting. Then moved it up to any random value. Then back down to 0-zero. The tone of the sound would not revert back to as it was before, but instead start phasing and getting all strange. I could only get it to correct by either saving, and reloading the sound, or just switching to a different sound, then back. The sound would then be correct with a detune of 0, and without the phasing type thing. I think it only happens with Pulse wave sounds, but will have to check more on this. Note: The phase type tone, is as if the phaser in the effects was on, or like modulating a pulse wave, with an Lfo.
edit: Although with what your suggesting, almost all the sounds people made using detune now, would become not like their intended original sounds, if they changed the detune now.
-
- KVRAF
- 6466 posts since 18 Jul, 2008 from New York
OK, suit yourself.dumbledog wrote:This is more fun tho