1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

Apple

Discussion in 'Computer Hardware' started by amusicsite, 25 Apr 2020.

  1. amusicsite

    amusicsite dn ʎɐʍ sᴉɥ┴ Staff Member

    Location:
    UK
    Well if the rumours are right at the moment then next year Apple could be moving one of it's Macs over to it's own chips. My guess would be the Mac Air laptop and the Mac Mini. I'd imagine they will run on a modified version of the phone OS and is probably the beginning of the end of OS X. I can't see this being one of those things they drop on an 'unsuspecting' crowd. They will need to get the developers on board to migrate their apps over to the new OS.

    The reason I picked those two Apple products is that I am willing to bet that the people who use them don't run many apps. For the consumers there is probably already an iPhone app that will work on the new machines. Maybe a few tweaks to how the UI works but probably not much more work than upgrading the app to work on the latest phone OS. For the commercial customers they may not run any extra apps at all or a few bespoke apps for their needs. I'm sure all these companies will roll out updates so it's seamless between the different hardware versions.

    They are not really powerful enough to seriously do heavy work on anyway and mostly used for light repetitive work. A business laptop that connects to cloud services, a laptop to upload photos to Instagram. A low cost mini server. The big selling points on these machines are energy. For the Air it's all about how long it goes for before needing a charge and for the Mini it's about the power consumption of a data centre with hundreds of these running. My bet is that these machines will basically be phone hardware with more cooling space and a few tweaks for the different interface.

    It reminds me of the last time they changed chips. The big switch to Intel along with the birth of OS X. One of the key drivers for the switch was Intel had a better roadmap for getting more performance per watt of power consumed. They had a growing laptop market and realised the importance of that, not to forget that a lot of their nicely designed computers are terrible at dissipating heat! I've seen people jump on the bandwagon and compare this move to the last switch but there are a few key differences as I see it.

    The move from OS 9 to OS X was a painful one. Quite a few of the high end OS 9 machines worked fine for years to come but the more middle of the road computers where a bit hit an miss. You could upgrade some OS 9 machines to OS X with a performance hit. Great if you have the processing power to burn, not so great if your machine only just about has enough power. Then there was the software switchover. As the early versions on OS X were very close to Unix there was at least some good codebase and libraries kicking around for developers to use. Though it was a big codebase shift and some apps got abandoned, passed on the cost of migrating or some other way were affected by the whole hardware and OS change.

    The move from an x86 chip to a phone chip I think is less of an issue these days. There is a thriving industry that churns out tons of apps for iOS. A lot of the apps people will use on these low end machines probably already have an iOS app, if not the one they are used to at least an alternative that does the same job. My guess is there may be an emulator on chip to speed up running legacy apps but you will want to run the native ones as much as possible.

    I guess the big question is how quickly they can move this up the Mac family. How well can it scale? The advantage of having the low end ones in the field early is it gives all the developers something to play with. You can start getting those Photoshop type apps working on these early machines. With a good roadmap Apple could make it easier for themselves and developers. For example, say the first machines are X cpu cores, Y A.I. cores, Z gpu cores. Then they say later chips will have doubles of these for higher end systems, or multiple CPUs or whatever. The developers can get their code working on the entry level machines while building in the ability to utilise the extra resources when they become available.

    Makes me wonder how Google is getting on with their combined Android and Chrome OS platform and there are some promising developments in the 'getting Linux on a phone' space. I think over the next few years the difference between a phone OS and a desktop one is going to get less and less. Maybe even the death of x86 as we seem to be going down the road of simple instruction sets, multiple cores and libraries instead of complex instruction sets.