More consolidation: Image-Line acquires UVI

[Michael Canavan]

As for us, Apple's current approach might make our lives easier in the long run. Right now, Silicon and Ventura is a big mess as you can see from the threads on this forum. It's easy to blame the developers, but I think Apple took a big risk if they can't bring the developers, others in the industry, and regulators along with them. Some clever people in Redmond must wondering if they have some opportunities.

Great post, I don't see the end the same as you though. We've been through this a least twice before in the last 20 years, with Os 9 to OS X, with PPC to X86, and now with x86 to Apple Silicon. Looking at the bigger picture a couple things come to mind, first off the timeline for those changes hasn't left any breathing room. I'm 100% certain at least with PPC to Intel and with Intel to Apple Silicon what we've seen with each OS tweak that sent developers into fits are adjustments to the OS to make the final transition easy for those that followed the guidelines. The problem is always the same though, developers are humans who want to work on the fun stuff, making an analog emulation or modern digital plug in that they think people will love and make them bank. So they do what most people do and look for ways to quickly port a GUI to both Windows and Mac. This has resulted in the past in wildly unstable plug ins on Mac OS, I mean this is essentially why AU was developed. Under OS 9 using VSTs was like rolling the dice on whether you would have to restart your computer. So, IMO Apple actively crushes third party cross platform GUI frameworks. The evidence is there if you pay attention to developers, they will mention it on occasion.

All this becomes painfully obvious with plug in developers who had updated their GUI frameworks for resizability and VST3 etc. in the last couple years previous to Apple Silicon being announced, they were Catalina ready out the door, and it took a year or less to port their entire library over to Apple Silicon. Then a company like NI are caught a bit with their pants down because they relied on frameworks from 10-15 years ago by their own admission. [Kudos to them for getting it done almost now.] NI are a lot like Apple in one major way though, they do not bother attempting to string together emulation layers for legacy products, or to rewrite them, Absynth is dead. And as we all know Apple kills things all the time. Alchemy VST AU the Camel Audio version which I unfortunately never got a lifetime unlock for, it's GUI just went blank white a year ago with some security update for Mojave, a simple update for an OS more than four OS's ago kills the GUI on a plug in.

Point being Apple isn't going to lose any large swath of developers over this, the ones that don't want to play their game have gone away a long time ago, Melda are whingers to the nth degree, it's not the only subject he loses his mind over. I mean if they were to be thinking fairly about this, they would realize exactly why Apple does at least 90% of the things they do, and if they're trying to tack on things that work but have been marked deprecated they should be happy they got away with using it for as long as they did instead of being angry it was deprecated in the first place.

[Michael Canavan]

One more thing, if Windows is so fantastic then why do MOTU struggle with stability on that platform? I think we can all agree MOTU know what they're doing, and it's been very obvious that there are issues getting DP stable on Windows. Developers who started on Windows like Reaper etc. seem to do alright, but even Cubase has it's versions that have major issues on Windows.

I think we see a microcosm of the audio world where plug in developers start and are well versed in Windows, then the porting process along with Apple cutthroat OS pace drives them nuts, they come up with solutions that cut corners, and that leads to issues later on. 100% the impression I get is that the few developers that started on Mac OS do not have any of the issues that Melda talk about. U-He, Plasmonic etc. are not losing their mind about it.

[bayswater]

Great post, I don't see the end the same as you though.
[/quote]
I’m not arguing that this current Apple strategy will fail, just that they take a risk in alienating developers. They don’t really have a choice. They have to go down this path, and must know that not everyone is going to following them.

But I would argue (a la George Harrison) that all things must pass.

[bayswater]

One more thing, if Windows is so fantastic then why do MOTU struggle with stability on that platform? I think we can all agree MOTU know what they're doing, and it's been very obvious that there are issues getting DP stable on Windows.

I don’t think anyone is arguing that Windows is fantastic. But as 80% of the market will tell you, it’s not totally useless. And the gap between macOS and Windows has certainly narrowed. The days of “just works” are long gone, and some of the things that pushed me away from Windows are a part of macOS now.

[Michael Canavan]

I’m not arguing that this current Apple strategy will fail, just that they take a risk in alienating developers. They don’t really have a choice. They have to go down this path, and must know that not everyone is going to following them.

But I would argue (a la George Harrison) that all things must pass.

My novel was to say that Apple have gone down this path multiple times in the past, the arguments that developers have are IMO always generated by their workarounds to port to both platforms with the least amount of effort, and Apples incremental changes to the OS to prepare for the next chip exposes any shortcuts they've used.

They've threatened to leave the Mac OS ecosystem in the past, PPC to Intel really didn't have a lot of developer losses, although Intel made it easier of a port for developers. I don't know what the answer is, developers including Melda here are angry that Apple changes things too much, but it's logical to switch to their own chips, and it's rational not to tell developers that it's because of chips no one knows about that are being developed.

[James Steele]

Innovation can be painful. I don’t always understand or agree with everything Apple does, but there is no doubt that they help drive forward innovation in the industry.

[bayswater]

My novel was to say that Apple have gone down this path multiple times in the past

Yes, they have, and they’ve been very successful, and I hope they are again, and that the current leadership has the foresight that earlier leaders did. I hope they don’t take anything for granted.

[David Polich]

Well I don't care who owns UVI. I would just like to see UVI address the glaring problem with Falcon and UVI Workstation..namely, there is no multi-core threading for either of those products. So no matter how many cores u have, their products can only access one of them.

This causes a number of their products to almost cripple an Intel Mac, and slow down an M1/M2 Mac. For example. I can load up any of the Quadra expansions, but most of the presets bring DP to its knees, just by playing ONE note, on my 2019 8-core Mac Pro..not the fastest machine in the universe, but certainly not a slouch. Omnisphere barely tickles the CPU in DP on my machine. Same for Superior Drummer 3, EZ Bass, any Arturia synth.

[bayswater]

Well I don't care who owns UVI. I would just like to see UVI address the glaring problem with Falcon and UVI Workstation..namely, there is no multi-core threading for either of those products. So no matter how many cores u have, their products can only access one of them.

This causes a number of their products to almost cripple an Intel Mac, and slow down an M1/M2 Mac. For example. I can load up any of the Quadra expansions, but most of the presets bring DP to its knees, just by playing ONE note, on my 2019 8-core Mac Pro..not the fastest machine in the universe, but certainly not a slouch. Omnisphere barely tickles the CPU in DP on my machine. Same for Superior Drummer 3, EZ Bass, any Arturia synth.

I'd heard this about Falcon before, but never paid attention because I haven't had any problems using it as my main VI. I'm surprised to see that it performs so poorly on your system. You can see in my signature, I use a 2018 Mini with an Intel i7, 6 cores, 32G RAM, but presumably older and less powerful than yours.

So here's a couple of tests.

These have either 4 or 8 instances of Falcon running. Each instance of Falcon has four of the original MachFive presets loaded. There are four MIDI tracks playing these four presets on each instance of Falcon. So, either 16 tracks, or 32 tracks playing Falcon at once.

In both cases the buffer is at 512.

The pics below show Activity Monitor with the process graphs, and the DP Audio Monitor. The first shows the 4 instance test, and the second shows the 8 instance test. I tried to embed the images, but it didn't work, but the links should be fine.

In the 4 instance test, the overall system is 82% idle. User CPU is about 12.5%. In the DP Audio Monitor, there is some Pre-Gen activity, and a small amount of real time activity.

In the 8 instance test, user CPU goes up to about 15%. There's a little more showing in the DP monitor, but still nothing to be concerned about.

Also, note that in the 8 instance test, you can see in the processor charts, the secondary process threads have kicked in.

There are a load of other processes going on, many of them related to Apple's incessant updates of Photos libraries, and this accounts for a lot of the CPU load.

No doubt if Falcon supported multiple processors properly, I'd see some improvement, but it's hard to see there would be any practical improvements.

www.heavyethics.com/DPAV/Falcon%204%20Instances.png

www.heavyethics.com/DPAV/Falcon%208%20Instances.png