ups, top right there it is Sorry Urs.Urs wrote:Then use ite-modic wrote:I need an undo function in Bazille, I often lose my patch cable and forgot where it was.
I must say, your FM, phase distortion Modular VST is amazingly great.
ups, top right there it is Sorry Urs.Urs wrote:Then use ite-modic wrote:I need an undo function in Bazille, I often lose my patch cable and forgot where it was.
I suspect the same thing. I suspect that current compilers automatically optimize for current technologies.Slarti wrote:I wonder if u-he synths are really well optimized enough to work with older (2-core) processors? i suspect they are not!
Yes. I guess that "everything new" does not just relate to our stuff but to anything from developers that are up to date.I still get beautiful sounds out of synths of ten years of age with almost no cpu-load! Nowadays everything 'new' is breaking my cpu! Is that progress or just lazy programming?
Urs wrote:I suspect the same thing. I suspect that current compilers automatically optimize for current technologies.Slarti wrote:I wonder if u-he synths are really well optimized enough to work with older (2-core) processors? i suspect they are not!Yes. I guess that "everything new" does not just relate to our stuff but to anything from developers that are up to date.I still get beautiful sounds out of synths of ten years of age with almost no cpu-load! Nowadays everything 'new' is breaking my cpu! Is that progress or just lazy programming?
Things change. Everything has gone multithreaded, favours branch prediction over outdated paradigms and respects the move to faster memory. At the same time operating systems have adopted malware paranoia that require developers to rethink communication strategies. I don't think it's "lazy" to move on from old concepts to optimize (or develop in general), it is indeed a necessity.
Example: Until 2005 or so, using an if() was a no-no in DSP programming. Because CPUs had bad branch prediction. This has changed, our code nowadays is full of if() statements which makes it much faster. Because branch prediction has become really great.
If you want to use stuff that's optimized for a 6 year old CPU, you need a build that's from the same era. Newer stuff will certainly run, but it won't ignore paradigm changes - and you can't really expect us to write the same code twice, once with ifs and once with a gazillion lookup tables whatsoever.
Cheers,
- Urs
Interesting! I'm seeing different, but also broken, behavior. If I set a modulator knob to Breath or Xpress modulation never works when I send Breath, Xpress, Modwheel. When I set a knob to Modwheel everything works fine. If do direct Midi-learn of Breath or Xpress it will modulate a parameter.EvilDragon wrote:Yep, in Zeebs I noticed if I set a mod to BC or EXP, it will respond to modwheel, breath or expression pedal at the same time. If I set a mod to MW, it responds only to MW. So that's good. But others are borked.
OK, I'll bite. I don't see anything different about ACE in that screenshot...Urs wrote:http://www.u-he.com/img/ace_tease2.png
Curious, do you know if the rececentUrs wrote:Yes it is!bmrzycki wrote:EDIT: Wait...that's a Linux console....
© KVR Audio, Inc. 2000-2024
Submit: News, Plugins, Hosts & Apps | Advertise @ KVR | Developer Account | About KVR / Contact Us | Privacy Statement