Affinity 2 Requirements

[Bran Cardello]

Applesupports the current major version and two previous versions - with macOS 13 out now, that means that macOS 11 is the oldest supported version (the oldest version which will continue to get security patches and the like).

Serif is already supporting one version older than Apple is, which imposes limits on what they as developers can do as newer APIs become available and older ones are deprecated by Apple. To support newer versions of macOS sometimes developers need to switch to newer APIs which don't exist on the older versions, because the older APIs don't work on the newer macOS.

It also increases the testing effort and the need to keep the older versions available (which is risky to them as well with the lack of security updates) in order to provide support and handle bugs and the like.

There is a limit to how far they can reasonably go, and a major version release is a good opportunity to prune the need to maintain extra code to support older versions. The more recent the minimum requirements are, the longer the major version can last without needing to further drop support for the older macOS versions later on, so even supporting 10.15 seems reasonably generous at this point.

Yes, you're right. From developer perspective, I do agree. From customer perspective, we had Affinity v1 supporting macOS 12 Monterey all the way down to OS X 10.9 Mavericks. That are altogether 9 (nine!) supported different macOS versions from Affinity apps, including around 7 operating systems that Apple didn't support anymore. Not so long ago even OS X 10.8 was still supported by Affinity. With the release of Affinity v2, we do have a huge jump of minimum system requirement from OS X 10.9 to macOS 10.15. That are 6 system versions difference and that's a lot.

I do agree with your security concerns, but there are many ways to get a reasonable security even on older OS versions, like using modern third-party browsers, mail clients and firewalls, etc. There are good reasons for keep using older system releases, too. If you want to be productive, you'll need a system that just works. You don't have the time to spend every weekend in installing latest OS versions, fixing important broken workflows or dealing with latest OS bugs. I'm still using old macOS versions as I'm still using Adobe Creative Suite in parallel to Affinity apps. Adobe CS doesn't run anymore on macOS 10.15 and I don't want to be forced to Adobe subscriptions. Indeed, that was a main reason for my transition to Affinity apps. I know there are more people, that want to keep their Adobe CS AND want to use the latest Affinity apps until the transition is complete.

I like all the Affinity apps and I'm sure that those will become my main tools in the near future. Meanwhile I don't ask for supporting OS X 10.9. Supporting macOS 10.12 would be my wish and would be a really great thing. At least supporting macOS 10.14 would do me a big favor, making my personal transition from Adobe to Serif more comfortable and smooth.

If the idea is to avoid dropping support for macOS versions within a major version of the Affinity suite, then cutting back to four supported versions now means that the suite would be back to supporting nine major versions five years from now.

The question then becomes, what is the lifespan of a major version of the suite? If it lasts another five years before v3 comes out, then they will be supporting the same number of major versions of macOS that they were supporting for v1 up until now.

I'm also on 10.14.6 and thus locked out for the time being. Obviously would love to see Mojave making a comeback on the support list. In my book Catalina is a climb down and one to definitely stay far away from, Big Sur perhaps a minor improvement on that and whatever the current one is called only recently got its firmware issues for my hardware worked out, otherwise a somewhat mixed bag.

So from a productivity perspective they're all unattractive choices to me. Especially when you factor in that upgrading the OS and especially when jumping a few version numbers at once usually also means upgrading (or replacing) other software for compatible versions.

However, I feel (note the subjective word) that Serif must be aware they will be missing out on a lot of sales because of the dropped support of comparatively recent systems, so I conclude that they either had no choice in view of what they are trying to achieve, or else that they estimated that the cost of whatever complicated workaround or complexity of conditional coding would outweigh the extra sales by extending support.

As far as I know, the latest version of Xcode for Ventura lets you compile only as far back as High Sierra compability. My impression is that Affinity 2 relies on underlying technologies the were not available before Catalina. Perhaps it doesn't necessarily affect features we're seeing in v2.0.0 yet, but it will surely be a factor for new features coming with 2.x updates.

So I'm almost as sad as anyone else who doesn't want or can't switch to Catalina or higher, but I can understand the move. Maintaining legacy compatibility comes at high costs for developers, and not many have the resources for that. (One recent example is 2manyrobots.com, developer of the brilliant Yate app for audio files tagging. He is currently providing two separately compiled versions to ensure compatibility with Intel Macs that cannot be officially upgraded beyond El Capitan, e.g. like my spare MacBook Pro 2008. Amazing!)

I have a Catalina partition more or less in sync with my main El Capitan partition (using a lot of symbolic link trickery, haha), and I will install Affinity 2 later today. But for the forseeable future I will stick to El Capitan as my primary workspace. My main MacBook Pro is 10 years old, so inevitably I will have to get a new one rather sooner than later anyway, I guess.

I believe Affinity did the right thing here possibly updating the Apple framework in the new branch of the Affinity line (v2) - basing v2 on the latest base technology they can making future development easier (and possible) (to integrate with future updates of macOS and ipadOS). From what I have collected and understand Affinity is based on quite a lot of third party products including as much as they can from the OS. Backwards compatibility with older versions of macOS is simply not part of many of these updated frameworks. And sticking to older versions of frameworks and libraries is a losing game.

3) I was here to report bugs and submit improvement requests for professional work professionally in a large setup and to bring a lot of knowledge from the world, i.e. professional product development, web- and software development, usability, user experience design and accessibility. I actually know what I am talking about!

I have not used any surface devices but you can try with the trial version for 30 days and if the suite works on that device you can then buy the full license and activate the software there is no limits on the trial other than the 30 day limit.

Thinking of Adobe, one was able to upgrade from one version to the next but one in the past. That were around 2 years, before one lost the value of a license. As there are no upgrade options for Affinity right now, there'll probably be another sales model. A shorter lifespan of major releases would minimize the need to support older systems thus minimize developing efforts and generate a new income. Of course the requirements could also be raised within the same major version number, as we have seen before with OS X 10.8, but as you mentioned, a new major release is a better reason to drop support.

The question is, what are customers willing to accept to get the latest features and how often. Would be a new major release every 2 or 3 years accepted by most customers or is a 5 year frequency generating enough income for Serif to continue with their developing? Personally, I'm really willing to pay for the Affinity v2 update, but for me it means to buy a new Mac. I'm not sure, if I can afford upgrading to a next major release of Affinity, if it always means to buy a new Mac to get this done.

Apple does a good job to release Swift frameworks (formerly Objective-C) that make developers' lives easier. As a developer, it's not only a big time saver, but also makes difficult things possible by just using some advanced high level code. The bad thing is, Apple lost the reliability and endurance of their frameworks long time ago when they started with a yearly system release cycle. That's good for progress of software evolution, but it's also making developers live more complex to keep track of every API change. Apple announced around 4 years ago, that they'll stop short release cycles in favor of returning to software quality. Unfortunately that was not the case and nothing changed. I understand that software needs progress, but it's a pity if quantity comes before quality. I'd prefer a matured rock solid software base over a bunch of features that I don't really need. The only ways for developers to escape from Apple's 'pushing to new framework versions' is to have a broad base of lower level code in their codebase or to use third-party frameworks with slower release cycles. That makes software development slower and more complicated, but it's easier for maintaining long-term support for legacy systems or to port software to other platforms. In contrast, the most developers tend to use more short term solutions as they are faster with publishing and monetizing their ideas. Mac OS X 10.6 Snow Leopard had a release cycle from 2009 to 2011 with 8 minor releases. That was one of the most stable macOS of all time. I'd like to see Apple returning to longer software release cycles and fixing their code, too.

3a8082e126