nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

MacOS 27 - learn from the past

adissu

Announcement to NI!

Apple released public beta of macOS 27, so this is a good chance for NI to make things finally better than in the past and start testing their products now, so they will officially support macOS 27 end of September instead of several months later or early 2027.

Find more posts tagged with

Comments

mykejb

The problem with doing that is that Apple are very clear that anything that's in the beta can change before the official release. Some advance testing would be good and probably happens, but there's no point doing a complete test until the release version is available.

Vocalpoint

https://community.native-instruments.com/discussion/comment/265458#Comment_265458

This.

There hasn't been a Mac OS release in years that didn't change something - even after the usual September "official" drop.

Who wants to retest everything just because Apple decides to change a bunch of stuff AFTER major Q&A has concluded?

The smart play is wait until the OS has firmed up over a number of months and then declare official support.

Matthew_NI

I was at iZotope for 10 years, and Native Instruments for the past 5 years.

It has often been the case that a release version of an OS contains changes that prior beta versions did not. Sometimes, those changes even come after launch in minor version updates.

More than once, such changes have caused significant and disruptive breakages to various software or hardware products on which I've worked. This was all the more frustrating if we'd been testing against a beta version, and confirmed stability or communicated support based on that version.

Furthermore, those changes are not always documented, and the OS manufacturer may not even be aware of the implications a particular change may have. Deprecating or changing certain underlying frameworks, directories, libraries, etc, can completely break 3rd party software.

It is for this reason that we, and many other companies, may begin testing on beta versions, but follow good development practice and avoid declaring final compatibility until after the stable release has arrived, and we've undergone a full test effort against that release.

Vocalpoint

"It is for this reason that we, and many other companies, may begin testing on beta versions, but follow good development practice and avoid declaring final compatibility until after the stable release has arrived, and we've undergone a full test effort against that release."

This. 100%

adissu

https://community.native-instruments.com/discussion/comment/265512#Comment_265512

OK, got that. But this still does not answer why it takes months to run these tests while in the meantime a lot of new stuff from NI gets released. I'd expect that you have a high level of test automation to run these test quickly. At my company, we also create safety-critical software and testing our whole release takes just serveral ours.

Please keep also in mind that the months of not being able to update comes along with trade offs for the users. If I have to get a new Mac and it has already macOS 27 preinstalled, I am trapped and for others, they cannot benefit from new features coming along with the new version of macOS.

maniacintosh81

Any Mac model that was released prior to macOS 27 will still run macOS 26 - even if it was delivered with macOS 27 already after its release. Only new models are trapped on macOS 27. In any other case: There is no urgency in upgrading to macOS 27. My essential Macs are in fact still running on macOS 15 Sequoia - I am just starting to consider an upgrade to macOS 26 now. Especially if you really rely on your Mac as production gear (and don't have a second one as fallback) I would recommend to wait instead of upgrading instantly after the red 1 in system settings appears - even with x.x.y-Updates. You never know if something breaks. I learned this lesson the hard way back with macOS 10.12 Sierra. Suddenly my audio interface - sold as class compliant without dedicated drivers - stopped working. Just doesn't work with macOS since 10.12. Period.

Even with small system updates I usually wait a few days and check for reports about compatibility issue online before I install them myself.

Vocalpoint

https://community.native-instruments.com/discussion/comment/265540#Comment_265540

"But this still does not answer why it takes months to run these tests while in the meantime a lot of new stuff from NI gets released"

What does this have to do with a new Mac OS release? Are you saying NI should put their entire business on hold (and offer nothing new at all) until Apple gets their act together for a new OS? This is a business and business takes priority - while a new OS can (and will) wait.

"If I have to get a new Mac and it has already macOS 27 preinstalled, I am trapped and for others, they cannot benefit from new features coming along with the new version of macOS"

@maniacintosh81 nailed it. There is no urgency for anything.

I am not a Mac user but I still cannot figure out why anyone would ever "need" to get a new Mac (short of maybe theft or properly loss due to a fire/hurricane) especially with the new price increases starting to land due to RAM etc. If anything - I think more and more Mac users are going actually roll right into "preservation" mode and use what they have for much longer than ever before.

That said - it is your choice to do whatever you want to - but jumping on Mac OS 27 prior to full Q&A and certification - is a tactical error you should avoid anyway possible.

NI has been very clear - over years and years of using their identical testing schedule - every new Mac OS is usually certified in January the following year.

Maciej Repetowski

All true and I agree with most of what has been said, however when it comes to compliance with next MacOS version, NI seems to be always one of the slowest companies. They have lots of products, sure - but some improvement in this regard would be welcomed.

mickaa

I always wait a year before upgrading to the next version. I started with a dual-processor G4 in 2004, and I’ve been doing it this way since 2010.

Matthew_NI

https://community.native-instruments.com/discussion/comment/265458#Comment_265458

Indeed. It's this that isn't well understood en masse.

adissu

@Vocalpoint
I won't get into the debate about whether Macs have become more expensive (What does that have to do with my question, exactly?) or the usual "that's how we've always done it" argument—as I mentioned in my original post, there's room for improvement.

The core issue remains: why does this take months? Other developers prove that it can be done much faster, and automation is key here. Nobody needs to put their entire business on hold for this. So far, the only thing explained is why testing starts after the official release, not why it takes so long after that point.

Besides that: And then there’s that tired old debate about waiting for period X before updating your OS. Maybe some people actually want to benefit from new features right away instead of waiting a year. As for anyone who holds off on updates within a major OS version, I sincerely hope their machine isn't connected to the internet, happily roaming around with known security vulnerabilities. And the "nothing has ever happened to me" argument is just naive—it simply means you’ve been extremely lucky so far.

Coming back: So far, people have only explained why testing isn't even started until after the official release, ok, got that. But not why it takes months after that. The original question of why this takes months remains almost completely unanswered.

Vocalpoint

"Coming back: So far, people have only explained why testing isn't even started until after the official release, ok, got that. But not why it takes months after that. The original question of why this takes months remains almost completely unanswered."

Respectfully - you are misunderstanding the answer provided by NI is all. You believe that the version that is "released" on Sept 12 (or whatever) is the "real" one (and the "final" one). It is not. It is the one that Apple believes is good enough for Sept 12 and that is all.

Then on Sept 30 - an update is released - one that could (and has) changed major things for vendors like NI. Then another update comes on Nov 19 - which could make more changes and on and on.

Then - also respectfully - I believe you are misunderstanding the effort that is required by a vendor like NI - to actually perform meaningful Q&A on this stuff.

It requires a substantial outlay of time, talent and resource to do proper end to end testing on a catalog of this size. (especially with the thousands of third party libraries that load into something like Kontakt) EVERY one of these is a potential pitfall.

And as an application support leader myself - I can tell you that automation is no substitute for a good Q&A team.

This is not something NI wants to do - 3 times. They want to do it one time - on a release that is considered stable, mature and not one that is going to be updated AGAIN - in 2 weeks.

Finally - and this is the area that is never talked about - I firmly believe that all vendors - big or small - view this entire annual Apple OS update as a complete PITA. No one WANTS to do this, no one wants to waste time on it or deal with it in any way.

Vendors want to focus on their business, doing their thing - not burning time, talent and resource on non-revenue generating 6-8-10 week painful detour EVERY FALL - just to bow down to Apple and their annoying OS update schedule.

Now - of course if you want to sell Mac based products - you have no choice by to accept this annual finger in the eye. And it's super easy to do if you sell just 5 products.

But ramp up the catalog into the hundreds of products and you will seen see why this becomes a non-stop hassle - and in NIs case - they will only declare compatibility - when they are uber-convinced that the version of the OS they choose to test against - is the right candidate to do so.

Matthew_NI

We have hundreds of products - that much is true.

We also support thousands of 3rd party products, primarily via NKS and secondarily any independent non NKS plug-ins users may host within our software(s) - that much is sometimes overlooked.

When qualifying Komplete Kontrol or Maschine for instance (each of which are plug-in hosts), we endeavor to test the stability of that plug-in hosting as well, and have at times spent days or weeks in conversations with those manufacturers to find workarounds to / solutions for OS related regressions.

nightjar

What I have historically experienced is that NI products mostly work fine with a new macOS well before they are officially approved.

I usually jump right in and manage things as they are in the moment.

This is part of the "fun" for me. I don't depend on this stuff working for my livelihood.

Living close to the leading edge of music technology is what I enjoy as much as creating music.

To each their own….

LostInFoundation

As somebody else already said, “there is room for improvement”

A simple “yes, we should improve” (followed by action) would be very welcome.

Always stating the amount of products NI has isn’t very reassuring…for 2 reasons:

First: having hundreds of products that must ALL be eventually corrected means one thing: these products have all been realized without following a scheme. Without a proper, professional programming. By that I mean that saying “we must test them all” means that they don’t have a common basis, which would allow to, once that an issue with a new OS gets discovered, “correct” what is not working anymore in various products. If you have to test and correct differently every product, the problem is at the roots of how the products have been realized. Maybe it would be useful to address that, before saying “NI have hundreds of products”.

Second: having hundreds of products means selling hundreds of products. Selling means cashing in. Cashing in means taking responsibilities (with part of the money earned) for what has been sold.
Selling isn’t a one way road. “I sell things and take the money” would be too easy. By taking the money, also a part of the aforementioned responsibilities must be taken in consideration. Otherwise… it is too easy. We could all take an employment and think “Now I will cash my salary” without also thinking “…and I have to offer my work in exchange”.
In the case of softwares (specially if you decide to sell also for Mac), part of that “work” consist in ensuring that the people who paid you for the software would be able to use it. Even when new OSes mess up the thing.

And… about plug-ins hosts… yeah… that’s what all plug-ins hosts (DAWs or what else) have to do.
What people are saying is “Why is everyone else doing it better/faster?” And… once again… the “we have more products” explanation isn’t standing: more products = bigger company and more revenues to invest in doing it. Bigger companies (which should have the right amount of employees to cover all the aspects of the work to be done… and therefore also the fact that programs must be “corrected” when needed… and in a reasonable amount of time) should be FASTER than small companies, which are maybe doing everything in 3-4 guys. Not SLOWER.

Otherwise… the solution is easy: if NI can’t manage that “amount of products”… release less, make them better, test them better, support them better.

One should always do what he can manage. Otherwise… it enters the “irresponsibility” ballpark

Quick links

Unanswered questions