The End is Near

Discussion in 'Apple/iOS' started by dstrauss, Jun 25, 2020.

  1. desertlap

    desertlap Scribbler - Standard Member Senior Member

    Messages:
    2,662
    Likes Received:
    3,354
    Trophy Points:
    181
    I don't think that is as much of a barrier as people think and one of the benefits (from a manufacturers standpoint) of ARM in general.

    In other words you would see a similar slide for a Snap Dragon 855 which would look quite different from an 8CX slide, which would look different than the slide for Fujitsu's super computer chip.

    One way to think of it from an x86 perspective is that ARM chips are based on intel type extensions on steroids (think MMX for instance)

    The danger/problem is if developers take the easy route and build to lowest common denominator instead of optimizing. That even occurs to some degree on IOS with garden variety apps, but most do take advantage of the newer chips (such as Pro Create, Lightroom etc.)
     
    Last edited: Jun 29, 2020
    dstrauss likes this.
  2. sonichedgehog360

    sonichedgehog360 AKA Hifihedgehog Senior Member

    Messages:
    2,197
    Likes Received:
    1,766
    Trophy Points:
    181
    Sorry to be such a negative Nancy and to say the end is NOT nigh, but it is NOT. Apple's Rosetta 2 isn't all it's cracked up to be. It is fast, just not iOS fast. iOS results of the A12Z:

    [​IMG]

    [​IMG]

    Now, judging now by the A12Z's results below in macOS, there is a 30%-40% drop in multi-threaded performance and a 25-35% drop in single-threaded performance. As I figured, not so stellar as they made it out to be:

    [​IMG]

    Where does the Surface Pro X stand in all of this? It actually does very well for a year-old product and a mere Qualcomm Snapdragon, which gets derided a forever-second-rate competitor:

    [​IMG]


    Surface Pro X--granted, it is unemulated--is behind by a hair for most of single-threaded results (it is ahead of two of results) and it is ahead of all the multi-threaded results. Not at all at death's doorstep, especially since their custom silicon was a relatively standard, last gen smartphone processor (basically the 8cx, which is really a last-gen 855 and no extra cores).

    Sources:
    https://browser.geekbench.com/ios-benchmarks
    https://wccftech.com/a12z-bionic-rosetta-beats-microsoft-surface-pro-x/
     

    Attached Files:

    Last edited: Jun 29, 2020
    Marty likes this.
  3. desertlap

    desertlap Scribbler - Standard Member Senior Member

    Messages:
    2,662
    Likes Received:
    3,354
    Trophy Points:
    181

    Two things,

    1.Apple has always prioritized single core performance in their A series chips and because few apps take advantage of even dual cores let alone umpty zillion cores like on Snap Dragon

    2. Apple will move their base and their developers far faster to native apps than MS could. Did you see the number Apps that will be native ready at launch?
    2a. Apples solution to non native apps is much more clever than MS was which is to essentially put a wrapper around each x86 app at launch. It potentially uses more disk space if you have a lot of apps but creates a much more efficient runtime than the bloated X86 emulation engine in WOA which is a kitchen sink of emulation cross functions none of which seem to be very optimized

    @sonichedgehog360 I know from your posts you are at last somewhat of an x86 advocate though possibly more towards AMD (whose latest Ryzens are very impressive), but what apple is actually executing is a comprehensive and cohesive (though not completely seamless) suite that allows both users and developers much more flexibility to scale their needs and applications.

    And they just took another leap today towards enterprise with this acquisition. If there is anything that big IT loves more than just about anything is management tools and "a single pane of glass" to manage them.
    https://arstechnica.com/gadgets/202...leetsmith-a-remote-device-management-startup/

    Last but not least, they have already said that the released Macs will have chips that are optimized for laptops and desktops and while they will be part of the same family as in the iPhone/Ipad it will likely be in the same way that a 5 watt core y chip is like a 45 watt core I chip.

    Of course the proof will be in the delivery, but honestly they almost couldn't do worse than MS did with the Pro X. And it's not because it's a bad device per se. It is easily the nicest product from a pure hardware standpoint that MS has ever released. But the fact that there are virtually no 3rd party major apps available over a year out; not to mention that they just in the last month or so released native Edge and Office is still not fully native is just inexcusable.

    I really wanted it to be great too and I put my money where my mouth was by ordering the Pro X, but the delivery was an utter failure and I ended up returning mine.

    There are some happy users including here like jhoff, but he's a somewhat niche case I'd argue. And to me the proof is in the reality that most of out major customers piloted the Pro X but only one actually deployed any and that was to a small subset of their global sales force that is the definition of road warrior and is never in the office.


    Ok, so now that I've thoroughly trashed the Pro X, come back in fall 2021 when I rip on Apple for the new ARM MBP. :)
     
    Last edited: Jun 29, 2020
    dellaster, Chris_Kez, JoeS and 2 others like this.
  4. sonichedgehog360

    sonichedgehog360 AKA Hifihedgehog Senior Member

    Messages:
    2,197
    Likes Received:
    1,766
    Trophy Points:
    181
    As much as Apple did snatch up a company, that doesn't necessarily translate to a working solution for everyone. iOS and macOS still lags behind in offering the complete accounting and EHR tools that a rehabilitation company like the one I work for would need.

    Apple has a large gapping whole that makes them a no-go for HIPAA compliance in all cases and that is their insistence on by-user Apple ID ownership model. They do not allow devices which the company owns to claim full access to the device data without the assigned user's express permission.

    Suppose the employee has stored clients information that is protected but that person unexpectedly left the company (they die or go missing?). Now, they are unreachable or otherwise unable to physically unlock it for us to pull the data for investigative purposes and that is absolutely unacceptable.

    That is also, for example, why medical providers like us are not using the technological marvel of Apple Watches in field. Even with cloud solutions somewhat getting up to speed, direct device access is an absolute must and we cannot be given direct, global administrative access to the data of the Apple devices we own because they are assigned on an Apple ID-basis.

    Even an outfit like Garmin surprisingly is far ahead in this regard for the health care smartwatch situation that I just brought up. Apple still has a lot to learn about catering to the diverse needs of corporations which is one area Microsoft has a compelling advantage in since it has been their niche from the very beginning.

    In fact, if there were a major push for in-field, thin-and-light tablets again in the company which is possible with WFH having become the norm here for many employees, we would most likely go with the Surface Pro X or the Go 2. The Surface Pro X or Go 2 actually meets our specific needs unlike Apple, who approaches things from a nonsensical consumer-first, single-user approach.
     
    Last edited: Jun 29, 2020
    lovelaptops likes this.
  5. desertlap

    desertlap Scribbler - Standard Member Senior Member

    Messages:
    2,662
    Likes Received:
    3,354
    Trophy Points:
    181
    Agreed that they have a ways to go and for instance are not even in the same ball park as say someone like Dell with their management suite. OTOH they are the class leader (pun intended) in the education market with the raw frameworks used by companies like JAMF or Airwatch, so I expect it they will ramp up quickly.

    PS: You should research the MDM software market for both IOS and Android a bit, they have in places lotions for all of the scenarios you mention and many more you may not have thought of
     
    Last edited: Jun 29, 2020
    Chris_Kez and sonichedgehog360 like this.
  6. sonichedgehog360

    sonichedgehog360 AKA Hifihedgehog Senior Member

    Messages:
    2,197
    Likes Received:
    1,766
    Trophy Points:
    181
    That said, I am not taking anything for granted here or underestimating them because tech companies have turned on a dime before. Apple could pull off something out of nowhere that is absolutely amazing that neither Apple nor Google could have foreseen or envisioned.

    We do widely use Citrix that we use to deliver some applications so we could provide a tablet-driven, dumb terminal-style experience for part of the time, but Internet is never 100% reliable as even many modernized regions of the world observed during the epidemic experience.

    So what that means for us it will always be a base requirement for us to be able to operate totally independent of the Internet at a moment's notice, off-the-grid without the cloud. This is for the sake of our clients' health and safety many of whom suffer from quadriplegia (or full body paralysis) and need things monitored and recorded constantly.

    EDIT: I will note that Epic is largely considered the leader and pioneer in electronic health record (EHR) software. They had something as early as 2010 but that doesn't say much. Note the harsh reviews (https://apps.apple.com/us/app/epic-canto/id395395172) (https://apps.apple.com/us/app/epic-haiku-limerick/id348308661).

    They describe the lack of quality and support of Epic's iOS software, observing how actually they have fallen in quality over the years if anything. In my experience, they are truthfully the high watermark or the bar of excellence of the market, which speaks unfavorably of the alternatives which are frankly unusable.

    With this in mind, Apple is primarily consumer-oriented with immense experience in K-12 education, but between their decades of unparalleled, rapid innovation in these two key markets, they still fall massively short for us in meeting the specific, specialized needs of health care.
     
    Last edited: Jun 29, 2020
  7. desertlap

    desertlap Scribbler - Standard Member Senior Member

    Messages:
    2,662
    Likes Received:
    3,354
    Trophy Points:
    181
    Ahh, didn't know you were in the healthcare field and agreed that the market there is somewhat unique. We sell a couple of health care adjacent custom devices (for air quality and temperature assessment) primarily to government agencies like CDC.

    I do think Apple has a huge short term learning curve ahead of them if they try to seriously go after the enterprise market again. And the core user base of creatives that they have made some inroads with are not much like the typical productivity worker that is the majority of the business market.

    Interring times regardless, and if nothing else this should light a fire under intel unlike nothing so far has. (eg. where are the 10nm chips when ARM is fully in 7nm ?)
     
    sonichedgehog360 likes this.
  8. Chris_Kez

    Chris_Kez Reformed Lumia Fan

    Messages:
    354
    Likes Received:
    366
    Trophy Points:
    76
    [mention]sonichedgehog360 [/mention] are those benchmarks running emulated or native? I assume emulated for the A12Z, but I thought Geekbench had native support for the Pro X.


    Sent from my iPhone using Tapatalk
     
  9. sonichedgehog360

    sonichedgehog360 AKA Hifihedgehog Senior Member

    Messages:
    2,197
    Likes Received:
    1,766
    Trophy Points:
    181
    Edited. It is unemulated. That does not discount the fact that it was last-gen at release (855 derived) with the same core count as a smartphone processor. I am figuring Microsoft will be looking to increase performance from improved silicon and better emulation.
     
  10. Marty

    Marty Pen Pro - Senior Member Senior Member

    Messages:
    3,436
    Likes Received:
    3,386
    Trophy Points:
    231
    I've found that Apple has the uncanny ability to hang out in the industry's blind spots for many years, leading to the paradox of the industry experts constantly praising them, yet constantly underestimating them.

    Imagine walking into Mark Lazaridis' office in 2006, and saying: "you know Apple is going to kill the Blackberry in 5 years...RIM in 10."

    "Apple? GTFO, they make overpriced mp3 players for *** sake...what could they possibly do to us?"

    The market could never grasp the iPhone until it was literally in their hands, because they already deemed the Newton a failure ("Apple's done with that..."), and never thought of the iPod as anything more than a music player.

    Except Apple never stopped working on the Newton. As Jobs said, the iPad came first, before the iPhone, before the mobile hype....a natural evolution of Apple's constant search for the best computer. And the iPod, was not an end in itself, but a form-factor milestone that Apple deemed ready for the masses, as was every iOS device after it.

    Apple products are deliberately missing key features (that 'any smart company' would implement on hardware as capable), because Apple isn't interested in that: they are laser-focused on delivering little slices of perfection, that customers will bond with, not as tools, but like pets, whose short-comings are always forgivable (and even lovable).


    So coming back to the glaring lack of enterprise support, one little thought experiment that might be worth thinking through is, what kind of enterprise management do you think Apple uses internally? They must use something, and their pride would never let them use 3rd party hardware, so you know it's running on iPhone, iPad and Mac already. Centralized data storage, thin clients, remote deployment, collaboration tools, they've got this all already...they just aren't showing it.

    Like any good magician, they show you what they want you to see, while they prepare the final astonishing act, "The Prestige," for the end, leaving everyone amazed and applauding (and the Ballmers and Lazaridis' flabbergasted :p).

    I suspect Apple won't bring any sort of enterprise management features to iOS for a good long while. Unlike certain companies, they know how to build a foundation thoroughly before chasing new markets. They'll be scaling up and solidifying their A-series architecture, transitioning their mobile developer base, and of course, continually feeding the tech press with beatific visions of an 'Apple silicon' future.

    But all the while, they will be looking for the perfect moment to pull another jujitsu tachi-waza and suddenly that silly, single-user toy people are holding, will become their fully-fledged, integrated client device.
     
    Last edited: Jun 29, 2020
    dellaster, Chris_Kez and WillAdams like this.
Loading...
Similar Threads - Near
  1. Bronsky
    Replies:
    1
    Views:
    921

Share This Page