Testing the speed of iOS 11 versus iOS 12 on the iPhone 6 and iPad Mini 2

2»

Comments

  • Reply 21 of 40
    ivanhivanh Posts: 179member
    The attempt would be more appreciated if you concentrated on testing the 70% faster camera. Not late yet, you can do it when the iOS 12 public release is available.  Do it this way: 
    Load 1,000 photos into the Photos app with iCloud synced on both iPhone 6 using the same Apple ID. Quit all apps and turn both iPhone 6 on, launch the Camera app and see how many seconds later they take the first picture and the second one. (Use a stop watch). It’s not feasible to get an iPhone 6 with iOS 8 installed, so iOS 11 vs iOS 12 will be a viable alternative. Redo the test with 3,000 photos.
  • Reply 22 of 40
    propodpropod Posts: 51member
    entropys said:
    It would be interesting to see a comparison with an ios10 iPhone 6 when ios12 reaches final, but it might be a bit hard to find one.  Mrs Entropy has kept her 6 plus on ios10 because she has heard ios11 slows it down too much.

    iOS 10 made my iPhone 5 very laggy and before that iOS 9 crippled my iPhone 4s. Apple shouldn’t allow iOS  versions on hardware that can’t run smooth imo. So I understand that people in general are skeptical about new iOS releases.

    If it was possible I would pay to downgrade these iPhones to iOS 6.

    So far I’m happy with iOS 11 on 7 +.

    williamlondonMisterKitmuthuk_vanalingamdysamoriaglynh
  • Reply 23 of 40
    eriamjheriamjh Posts: 1,061member
    I wish every last iOS update for a supported device gave a speed boost.   

    Did Apple do this by boosting processors speed more quickly to get the launch times up?   This has to have some affect on battery life.    

    My wife’s iPad mini may last her another year (if she’d stop dropping it).   
    muthuk_vanalingamdysamoria
  • Reply 24 of 40
    hagarhagar Posts: 100member
    This test and article are so misguided it borders on click bate. 

    A beta OS is full of debugging code so running benchmarks and doing performance tests and comparisons are completely pointless. 

    Secondly, an improvement of 1 second on a 40 second loading time is not a huge improvement but rather within the margin of error. Your measurements are rounded to the second so 1 second difference has no significance whatsoever. 

    Dissapointing article. 


    williamlondon
  • Reply 25 of 40
    Mike WuertheleMike Wuerthele Posts: 3,139administrator
    hagar said:
    This test and article are so misguided it borders on click bate. 

    A beta OS is full of debugging code so running benchmarks and doing performance tests and comparisons are completely pointless. 

    Secondly, an improvement of 1 second on a 40 second loading time is not a huge improvement but rather within the margin of error. Your measurements are rounded to the second so 1 second difference has no significance whatsoever. 

    Dissapointing article. 


    Read further up in the comments.

    We were specifically asked to do this, and do it now. We also thought it was a good idea, so we did it.

    BTW, while this is not that, the term is "Click bait"
    edited June 11 williamlondonGeorgeBMac
  • Reply 26 of 40
    dewmedewme Posts: 1,511member
    This article is fully reasonable and well intentioned since Apple made a big deal in the WWDC keynote about improving the performance of existing iPhones with iOS 12. Apple made this a release-defining feature.

    One of the primary purposes of a Beta release of a product is to demonstrate to prospective customers actual progress towards the release-defining features of a product. Despite all the caveats associated with Beta releases they must absolutely demonstrate progress that indicates the product is on a trajectory to deliver what the product owner claimed the product will deliver. In this case Apple said this product will deliver improved performance and even the Beta releases must show at least incremental progress towards this claim. Proclaiming that "this is a Beta and you never evaluate performance with a Beta" does not apply in this case. When a Beta is intended to show progress towards a new feature would anyone expect Apple not to include an early incarnation of the feature in the Beta? In this case the absolute numbers are far less important than the trajectory and I fully expect that subsequent Beta versions will be evaluated for performance improvements as well to see how well Apple is progressing towards their claims as the product nears completion. 

    As far as the absolute improvements are concerned I wouldn't expect any magic to occur that will somehow transform the performance of an iPhone 6 to that of an iPhone X. It's not going to happen unless all those laws of physics that we learned in school were actually just suggestions. Improving the performance of an existing product isn't rocket science, it's sound engineering methodology. You profile the product in-use to determine where and how each operation the processing chain contributes to the overall performance. You identify bottlenecks and come up with ways to remove or reduce the bottleneck. Once you do that the bottleneck moves to another part of the processing chain. Ultimately you reach a point where the last remaining bottleneck cannot be reduced any further or removed. In Apple's case they can redesign their apps and services to change the processing chain to be less impacted by the remaining bottlenecks, but this will move the bottleneck once again. Typically, every optimization exercise reaches a point of diminishing return for the additional cost, and the costs can quickly skyrocket, e.g. NASCAR teams spending close to $1 million to develop faster pneumatic air wrenches to save a couple tenths of a second off the time needed to change a tire. This is all theory of constraints 101 stuff.

    I'm thrilled that Apple is taking steps to make running iOS 12 on older devices more tolerable, but my expectations are realistically modest. Not making matters worse is in-fact a win. Small improvement over iOS 11 - total gravy.
    muthuk_vanalingampscooter63watto_cobra
  • Reply 27 of 40
    frantisekfrantisek Posts: 325member
    frantisek said:
    Hello guys,

    with all the respect to your work I watch daily, why have you made me to video which is full YouTube. 
    Bay qq are still waiting for someone to do comparison with iOS 10. 

    Thanhs
    Or, you know, you could read the article, which is literally the shooting script.
    I do not understand your comment but I must quote myself.

    Dam, that is horrible comment. On phone o overlook some typos sometime or autocorrect but this one is hardcore lol.

     with all the respect to your work I watch daily Why have you made "me too" video which is YouTube full off?
    But we are still waiting for someone to do comparison with iOS 10.

    And I will reply myself. Actually few videos already appeared on Youtube.
    https://www.youtube.com/results?search_query=%22ios+12%22+%22ios+10%22

  • Reply 28 of 40
    avon b7avon b7 Posts: 2,475member
    foggyhill said:
    ...one of the biggest slow downs I noticed from iOS10 was in address book searches which seem to hang after a third letter for varying amounts of time... This, rather than game performance is work related ie. 'pro'. This and the loss of many free and paid apps used frequently due to them not being upgraded has me wishing for more user choice and easier restoration of older iOS...
    Security concerns means this not a good idea, people could just downgrade your OS to get more security exploits. 
    Doing qa also would become much harder to do with an explosion of possible confins. There is a reason why people are moving towards apples update model.
    This depends on how you see things.

    One way of seeing it is that any problems existing in an OS should be resolved in that same OS and that a major upgrade with its everyday downsides - this article describes the iPad Mini 2 as almost unbearable under iOS 11 - should not be the solution.

    My Mini 2 is currently running iOS 8 and rarely touches the internet. It is set in stone with content moving between a 2009 Snow Leopard iMac (which also rarely touches the internet) with fried graphics card.

    If I had updated the Mini 2 for security reasons I would have received a seemingly unbearable user experience in return and that would be banging me over the head every single day I used it.

    Solutions?

    Slow down iOS releases themselves, iron out more bugs before release, let APIs mature before adopting them, introduce interim OS roll outs so that you get an official iOS 12i (which would be equivalent to a normal iOS 12 release today) and one year later follow up with a much more mature, optimised iOS 12 system. If support for older systems makes them virtually unbearable to use, don't support those devices or give the user the option to roll back to a version they consider acceptable.

    Or, as I said at the start, apply fixes to the OS that shipped with those issues in the first place.




    muthuk_vanalingam
  • Reply 29 of 40
    Please provide future results in tables and charts. 

    Much easier to digest. 

    Thank you. 
    watto_cobra
  • Reply 30 of 40
    DAalsethDAalseth Posts: 226member
    Is there a list somewhere saying what will and won't be able to run iOS12? 
    watto_cobra
  • Reply 31 of 40
    dysamoriadysamoria Posts: 1,740member
    I'd like to see comparisons between iOS 12 and iOS 10, made on an iPhone 6s and iPad Pro 12.9" first generation. I've refused to put iOS 11 on either device due to how newer OS versions historically make existing devices slower/laggy.
    williamlondon
  • Reply 32 of 40
    Mike WuertheleMike Wuerthele Posts: 3,139administrator
    DAalseth said:
    Is there a list somewhere saying what will and won't be able to run iOS12? 
    Not yet.

    Wait - you mean hardware? if it runs iOS 11, it'll run iOS 12.
    edited June 11 GeorgeBMacwatto_cobra
  • Reply 33 of 40
    foggyhillfoggyhill Posts: 4,726member
    dewme said:
    This article is fully reasonable and well intentioned since Apple made a big deal in the WWDC keynote about improving the performance of existing iPhones with iOS 12. Apple made this a release-defining feature.

    One of the primary purposes of a Beta release of a product is to demonstrate to prospective customers actual progress towards the release-defining features of a product. Despite all the caveats associated with Beta releases they must absolutely demonstrate progress that indicates the product is on a trajectory to deliver what the product owner claimed the product will deliver. In this case Apple said this product will deliver improved performance and even the Beta releases must show at least incremental progress towards this claim. Proclaiming that "this is a Beta and you never evaluate performance with a Beta" does not apply in this case. When a Beta is intended to show progress towards a new feature would anyone expect Apple not to include an early incarnation of the feature in the Beta? In this case the absolute numbers are far less important than the trajectory and I fully expect that subsequent Beta versions will be evaluated for performance improvements as well to see how well Apple is progressing towards their claims as the product nears completion. 

    As far as the absolute improvements are concerned I wouldn't expect any magic to occur that will somehow transform the performance of an iPhone 6 to that of an iPhone X. It's not going to happen unless all those laws of physics that we learned in school were actually just suggestions. Improving the performance of an existing product isn't rocket science, it's sound engineering methodology. You profile the product in-use to determine where and how each operation the processing chain contributes to the overall performance. You identify bottlenecks and come up with ways to remove or reduce the bottleneck. Once you do that the bottleneck moves to another part of the processing chain. Ultimately you reach a point where the last remaining bottleneck cannot be reduced any further or removed. In Apple's case they can redesign their apps and services to change the processing chain to be less impacted by the remaining bottlenecks, but this will move the bottleneck once again. Typically, every optimization exercise reaches a point of diminishing return for the additional cost, and the costs can quickly skyrocket, e.g. NASCAR teams spending close to $1 million to develop faster pneumatic air wrenches to save a couple tenths of a second off the time needed to change a tire. This is all theory of constraints 101 stuff.

    I'm thrilled that Apple is taking steps to make running iOS 12 on older devices more tolerable, but my expectations are realistically modest. Not making matters worse is in-fact a win. Small improvement over iOS 11 - total gravy.
    Actually, it's harder than rocket science when you have a small device with a broad amounts of use cases with a hell of a lot of things competing for attention. It's the scheduling of tasks, especially the UI ones, that are the main reason people see lag on older devices. Space  have a very narrower scope of use and thus the software part can be dev, tested and tuned to a much greater degree

    Apple can only take care of their end, tune the APIs and the system code. The Apps themselves when a phone is released have been tuned to work the best on 1-2 versions back. So, the 4s when the 5s was released. Those apps which were running on devices with a lot less performance, obviously ran great on the brand new 5s. But, with time, the app's sweet spot advances and right now, it is probably the 6s or even 7. That means that even if Apple does everything in its power to speed up old phones, those phones won't run modern apps well because they're not made to run well on those phones.  If you want the same perf with your apps as when you got your phone you need to keep the same apps. That's often not possible when apps have an online component.

    With Apps being more and more beefy, even the dumbest ones (and web pages being some of the biggest culprits) and the SOC becoming more complex, scheduling becomes a much bigger issue if you want to keep a fluid UI.

    watto_cobra
  • Reply 34 of 40
    claire1claire1 Posts: 205unconfirmed, member
    "Planned obsolescence my a**!!"

    Where was this line at WWDC?

    Oh well, the iKnockoffers will still scream this while praising their 6 month old abandonware. 
    williamlondonwatto_cobra
  • Reply 35 of 40
    GeorgeBMacGeorgeBMac Posts: 2,520member
    entropys said:
    It would be interesting to see a comparison with an ios10 iPhone 6 when ios12 reaches final, but it might be a bit hard to find one.  Mrs Entropy has kept her 6 plus on ios10 because she has heard ios11 slows it down too much.
    Why would that be "interesting"?
    ...  To provide (potential) fodder for the Apple Haters?
    .......... To "prove" that Apple "Intentionally slows down older phones"?

    Bullshit.
    watto_cobra
  • Reply 36 of 40
    GeorgeBMacGeorgeBMac Posts: 2,520member
    igohmmm said:
    Please provide future results in tables and charts. 

    Much easier to digest. 

    Thank you. 
    That depends on what is being tested....
    If they are testing opening speed, then yes, that can be put into a table or graph.
    If they are testing lag time during operations (say on the keyboard while typing), then a more subjective, text based review is better....

    As others have pointed out in these comments, it is not the time to open an app that bothers most people, it is the slow response once it is open.   For me that is best illustrated with Apple Music on my 6Plus:  The app opens almost immediately -- but loading content (especially immediately after opening the app) is painfully slow...  It is so slow, I suspect it would push most users back to FM radio which has no lag...
    watto_cobra
  • Reply 37 of 40
    GeorgeBMacGeorgeBMac Posts: 2,520member
    DAalseth said:
    Is there a list somewhere saying what will and won't be able to run iOS12? 
    Apple already said during the keynote:  If it runs iOS11 it will run iOS12.
    watto_cobra
  • Reply 38 of 40
    GeorgeBMacGeorgeBMac Posts: 2,520member
    dysamoria said:
    I'd like to see comparisons between iOS 12 and iOS 10, made on an iPhone 6s and iPad Pro 12.9" first generation. I've refused to put iOS 11 on either device due to how newer OS versions historically make existing devices slower/laggy.
    Why not compare an HP desktop running DOS to Windows 10?   Wouldn't that be just as informative -- and give you  bullets to criticize Microsoft for "intentionally slowing down older devices"?

    Like DOS, iOS10 is an outdated operating system no longer relevant.   There is no legitimate reason to run a comparison with it.

    Or, another analogy:   When I started in IT in the 80's, we were instructed to use extreme measures to increase program efficiency and reduce storage (particularly DASD storage) because CPU power and disk storage were so limited -- to the extent of using 2 digit years knowing that it would blow up the system at the upcoming turn of the century.   But, hardware resources were limited so software was adjusted accordingly -- despite the limitations that imposed on the software...

    Today, hardware power and resources are much greater than they were just a few years ago -- and software is adjusted to take advantage of those resources.   iOS 12 is an exception to that rule.   Be grateful!
    watto_cobra
  • Reply 39 of 40
    Very nice for older Apple devices, iOS 12 should speed things up a bit after a few more builds before public launch.
    watto_cobra
  • Reply 40 of 40
    mycophilemycophile Posts: 1unconfirmed, member
    "launched much quicker under iOS 12, taking 40 seconds compared to 41" Since when is this "much quicker? How many users would even notice the difference without a stopwatch?
Sign In or Register to comment.