KVR Audio

2DaT · Post by **2DaT** » Fri Mar 07, 2014 7:01 pm

FLWrd wrote:Did you ever hear about bit rot? Changing os'es, changing daw's, new plugin formats, etc., make that each software product has a limited life. Try to run 20 year old software and see how far you get.

Open source live much longer

fluffy_little_something · Fri Mar 07, 2014 7:09 pm

FLWrd wrote:Did you ever hear about bit rot? Changing os'es, changing daw's, new plugin formats, etc., make that each software product has a limited life. Try to run 20 year old software and see how far you get.

I don't think anyone will still give a crap about the OBXD 20 years from now. If progress continues at this pace, the OBXD will be considered primitive by the end of the decade and much better Oberheim emulations will have replaced it.

The oldest software I still use is Winamp 2.95. which is from 2002 or so. It has not been updated since then if I am not mistaken. So, either it still works because computer and OS suppliers still support it, or it will no longer work.

2DaT · Post by **2DaT** » Fri Mar 07, 2014 7:38 pm

If progress continues at this pace

It will decay at some time. Technological limits and cargo of backward compatibility already being felt on modern systems.

Breeze · Post by **Breeze** » Fri Mar 07, 2014 7:42 pm

Breeze wrote:Actually: you can already in fact just automate the voice count to reduce CPU use for instances in idle.

@2DaT: will this work without affecting the sound? There seems a "settling" time happens with some sounds (they don't sound right until they've played a few times); is adding voices going to have an identical effect on the Obxd as initial instantiation?

Tricky-Loops · Post by **Tricky-Loops** » Fri Mar 07, 2014 7:47 pm

2DaT wrote:
If progress continues at this pace
It will decay at some time. Technological limits and cargo of backward compatibility already being felt on modern systems.

Not only this. The limited resources will be a big problem in the next decades. Everything will get more expensive...

Touch The Universe · Post by **Touch The Universe** » Fri Mar 07, 2014 7:52 pm

Some good patches released for this yet?

fluffy_little_something · Fri Mar 07, 2014 7:57 pm

mysticvibes wrote:Some good patches released for this yet?

Good one

Spitfire31 · Post by **Spitfire31** » Fri Mar 07, 2014 8:01 pm

mysticvibes wrote:Some good patches released for this yet?

You haven't bothered to browse the thread, I gather?

Breeze · Post by **Breeze** » Fri Mar 07, 2014 8:02 pm

mysticvibes wrote:Some good patches released for this yet?

Nope. Nothing.

fluffy_little_something · Fri Mar 07, 2014 8:06 pm

2DaT wrote:
If progress continues at this pace
It will decay at some time. Technological limits and cargo of backward compatibility already being felt on modern systems.

I am not so sure, they are already working on processors that are no longer subject to the physical limitations of conventional semiconductors. To the contrary, I think there will be a new explosion once that technology is ready. Developers will no longer have to even think about the CPU efficiency of their software.

Yes, compatibility will continue to be an issue, especially as new software platforms will emerge in addition to the ones that already exist today.

Touch The Universe · Post by **Touch The Universe** » Fri Mar 07, 2014 8:07 pm

sorry, just opened up obxd and seems I already had a few good banks, was looking for any newer ones to use in the osc challenge. will just peruse the thread instead. Great synth

2DaT · Post by **2DaT** » Fri Mar 07, 2014 8:07 pm

@Breeze:
Yes, toogling voice numbers produces artifacts, and of course non-active voices remain "cold".
However i cant reproduce that "warming" effect after bugfix.

Maybe post a quick audio example ?(make sure it isnt natural beating of slightly detuned oscs).

fmr · Post by **fmr** » Fri Mar 07, 2014 8:30 pm

Breeze wrote:
2DaT wrote:
layzer wrote: Matrix-12 / Obie Xpander perhaps?
Waay too expensive now. Lucky for those who bought it before analog boom
That would be a tall order! What's really interesting with the 12/Xpander is the variety of filters, advanced modulation options and the live modulation matrix. The core seems pretty much what the Obxd already does.

IMO, if someone added most of the important feature requests in this thread,it would significantly expand the flexibility of this already remarkable machine.

As Breeze told, the core of the sound is already there, and your filters already have some funtionalities that a normal Obie user would only be able to find in a Matrix-12/Xpander. So, whats missing, basically, is a 20x20 matrix modulation, and enough modulation sources and destinations.
Basically, you'd have to implement mixers in several parts of the engine (the signal routing of tme Matrix-12 is a little complicated), and implement more envelopes (five), more LFOs (also five), ramps (kind of simple two segment envelopes), create some laggers, and some complex keymap followers.
Then, you would need a matrix of 20 points where the user could route all the modulation sources to several destinantions.
AFAIK, you did the OB-Xd without the real think right? Based on what you already have, I believe you could accomplish a Matrix-12, with some support. I will volunteer for support. I worked with a matrix-12 in college, and still have a copy of the manual, and some material I have collected about it over the years.

2DaT · Post by **2DaT** » Fri Mar 07, 2014 8:34 pm

@FLS:

I am not so sure, they are already working on processors that are no longer subject to the physical limitations of conventional semiconductors. To the contrary, I think there will be a new explosion once that technology is ready. Developers will no longer have to even think about the CPU efficiency of their software.

Doubt it.

Yes, compatibility will continue to be an issue, especially as new software platforms will emerge in addition to the ones that already exist today

Not only software, but hardware too. Specialized hardware nowadays can produce quite good results but cant interact fast with something like IA-32.

Breeze · Post by **Breeze** » Fri Mar 07, 2014 8:59 pm

fmr wrote:Basically, you'd have to implement mixers in several parts of the engine (the signal routing of tme Matrix-12 is a little complicated), and implement more envelopes (five), more LFOs (also five), ramps (kind of simple two segment envelopes), create some laggers, and some complex keymap followers.
Then, you would need a matrix of 20 points where the user could route all the modulation sources to several destinantions.

I've actually started experimenting with some of this in Bidule. Anything that doesn't involve inserting something in the audio path can be done. Within Bidule MIDI is processed at audio rates so adding modulators, scaled modulators, envelopes and LFO's works well. Just this week I tried using 8 instances set to one or two voices controlled by a Note Splitter sending different notes to each instance, and that opens up some other doors like per instance key tracking (I can do partial with this) and polyphonic modulation. At 1-voice per instance, it consumes marginally more than the basic Obxd @ 8-voices.

It certainly isn't elegant, especially GUI-wise (Bidule doesn't win any beauty contests) and sometimes I'm concerned about stability, so having all this stuff built-in would be ideal. I wanted to put this out there but first you need Bidule, and second the Creative Commons license doesn't allow me to distribute the Bidule with the Obxd; without it you'd have to partially rebuild the setup, which would be real dog to step-by-step. I might do a video at some point as a proof of concept though, to inspire anyone that might want to pick up the Obxd flame.

You can also do a lot of this in hosts like MUX and others. I just happen to have Bidule; it's pretty flexible and can go pretty low-level.

Obxd synthesizer