nospam <nos...@nospam.invalid> wrote:
> In article <
1n8bmav.15lzhfjl55e0bN%dem...@actrix.gen.nz>, David Empson
> <
dem...@actrix.gen.nz> wrote:
>
> > Nasty to also require 64-bit with this version, even though it still
> > supports iOS 9, because that artificially cuts off everyone with an
> > iPhone 4S, iPhone 5, iPhone 5c, iPad 2, iPad 3, iPad 4, original iPad
> > mini and 5th gen iPod Touch.
>
> not enough to matter.
Roughly half of all iPads ever sold is not enough to matter?
Between April 2011 and September 2013, Apple sold 149.7 million iPads.
Apart from a lingering few original iPads, all of them will be iPad 2
through 4, and original iPad mini.
Total iPad sales between October 2013 and March 2017 was 190.4 million.
Many of those will also be iPad 2, 4 or original mini which Apple was
still selling until 2014 or 2015 depending on the model.
It may be small numbers compared to the iPhone, but there are a lot of
people who bought an iPad, still have one of the above models and are
not inclined to replace it with a new model without a compelling reason.
The iPhone situation is going to affect proportionally fewer people but
there still a reasonable number hanging on to the affected models.
That's irrelevant - it is talking about sales of models during a
particular period, not installed base.
That might be relevant depending on what they mean by "market share",
but the numbers seem to be the right ballpark.
That might be relevant but only considers the US, not the rest of the
world.
This at least explicitly states what they mean:
https://www.statista.com/statistics/606147/iphone-model-device-market-share-worldwide/
Total of 13% of all iPhone models in use worldwide as of August 2016
were iPhone 4S, 5 or 5C. That may be a relatively small percentage but
it is still in the order of 80 million iPhones (based on other estimates
of installed base rising from about 600 to 700 million during 2016).
> > I can't think of any reasonable software explanation for requiring a
> > 64-bit device but still supporting iOS 9.
>
> less to test and support. expect it to be ios 10 only by the fall, if
> not sooner.
Then why bother releasing an update now which supports an almost unused
combination?
It looks like a mistake which they are at least investigating and will
hopefully fix ASAP.
https://answers.microsoft.com/en-us/skype/forum/skype_ios-skype_startms/unable-to-download-update-skype-on-older-iphones/1cbe3b2a-96fc-4978-9260-d81a7404b9e1
Allen[S] from Microsoft:
"This was not the intended behaviour, our team are investigating what
has happened with Apple and I will post an update as soon as one is
available."
My guess is that Microsoft forgot to tick the box in Xcode to enable
32-bit devices to be supported, and they weren't using any such device
in their internal testing. The list of supported models is derived from
those settings.
--
David Empson
dem...@actrix.gen.nz