Editorial: Apple's move to ARM is possible because most users want power more than compati...

1234568»

Comments

  • Reply 141 of 154
    docno42docno42 Posts: 3,755member
    Also when it comes to emulation, what's to say Apple can't work with VMware/parralles, identify the x86 instructions causing the greatest performance issues in software and address them in hardware.  They are in control of the ARM chips they use. 

    People need to get outside the box and live a little.  
    Soli
  • Reply 142 of 154
    docno42docno42 Posts: 3,755member
    wallym said:
    The problem is that emulating one cpu architecture on another cpu has horrific performance.
    Today.  But remember Apple owns their ARM architecture.  Why couldn't they put in instructions to help with emulation if it was warranted?

    A lot of assumptions flying around in here and hard limits that in reality aren't necessarily limits at all. 

    Heck Apple could license x86 from Intel and flat out embed it if it was really important.  But I don't think it is.  Hell many x86 instructions aren't hardcoded in silicon but executed by microcode on Intel processors today. 

    This isn't the 90's, so assumptions around 90's virtualization aren't very useful. 
    Soli
  • Reply 143 of 154
    Mike WuertheleMike Wuerthele Posts: 6,858administrator
    melgross said:
    melgross said:
    melgross said:
    nht said:
    nht said:
    nht said:
    nht said:
    nht said:

    nht said:
    wallym said:
    As a developer, I need both mac and windows support.  To openly campaign to remove Windows compat is to be irresponsible to the marketplace.  If users don't need Windows, that's fine.  Don't penalize me for your lack of needs.
    I don't think you, nor FredFref read the article.
    Why does a dissenting opinion mean they didn't read the article?  Maybe they read, disagreed with the basic premise "cross-platform software compatibility is now mostly irrelevant to the wider user base" and everything that follows.  Especially since you had a poll, found 35% that said, yes they needed windows and then proceeded to hand wave that away as AI readers aren't a representative sample.  Which begs the question of WHY RUN THE POLL IN THE FIRST PLACE?

    The next assertion "for Apple's biggest user base, the need for Windows compatibility isn't the same as it is for the main readers of this site" is fabricated out of thin air and has zero supporting data.  Whether true or not it's based on nothing but speculation.

    If the primary uses of the Macs are Pro and everyone else migrates to iPads then a significant fraction of Mac users (dare I say 35%) will want x86 compatibility.

    But, nope...because they disagree they didn't read the article.
    That's not why I said that, and you know it. And, there's a lot more to this quote of mine than what you clipped out. And, I didn't even say anything about the ludicrous assumption that this article is a "campaign" to remove Windows compatibility.

    It wasn't handwaved away. What it is, is that 35% of the user base that reads AI doesn't need it, which is an overly conservative estimate of what the larger user base needs and does with their machines, and you know this as well, based on your own interactions with the rest of the AI readership. And, even if you translate it literally, it does mean that the majority doesn't care about Windows on the Mac.
    Why run the poll and then disregard it?  So what if 35% isn't the majority?  It's still a large part of the user base.   

    And how do you know that it is "an overly conservative estimate of what the larger user base needs"?  On what data is this assertion based on?  Why do you assume that the majority of your readers are pros?  Why did you not include in your survey to self identify if they were pros or just general users?  Never mind that these polls are generally horridly misleading anyway.

    The article, and you, would like to make it seem like it's 0.35% of the user base to sell the idea that x86 compatibility is no longer needed.  Apple may have a good idea as to the number but you don't.  Moreover you ignored the entire enterprise market because it's inconvenient.  Does IBM and other major Mac deployments believe x86 compatibility is irrelevant?  I have no idea and neither do you.  It would have been fairly easy to reach out to IT folks highlighted in past articles and ask "hey, is x86 compatibility important to your Mac enterprise deployment?"

    Nope.

    But hey...35% is an overly conservative estimate of what the larger user base needs...
    Regarding the bolded section, we are, and your own supposition of how that is going so far is wrong because what they care about so far is iOS development and general productivity. We'll see how it goes in total when we're done.

    In regards to our audience, exactly who do you think AppleInsider is read by far, far more? College grads with advanced degrees, industry folk, designers and whatnot, or the "new Apple user" which is iOS centric, where the iPhone is a halo for the Mac and not the other way around?
    And the demographics for Mac users are what?  Gee maybe folks who are "college grads with advanced degrees, industry folk, designers and whatnot"?  

    Nah.

    I will assert, based on personal experience, that there are very few enterprise iOS developers that don't care about MS project, DOORS and a bevy of windows/x86 corporate tools...still dependent on Excel with macros.  People send me a lot of stuff in Visio to boot.  Also, most of us aren't iOS developers but enterprise developers and the docker tool chain is a significant part of devops.
    Yeah. we're not done. So far, we've spoken to IBM, Cisco, and Deloitte. There are about eight more on the docket. The point of this article, stands, though, that there is a line, where below it, the need for Windows is non-existent.
    And you have failed to show where that line is.  Again, is the AI demographic you just stated significantly different than that of Mac users?  Or have most of the "new Apple users" that don't frequent AI already moved to the iPad or never bought a Mac in the first place and have a windows laptop somewhere?
    I'm not really sure what you're asking, here.

    We didn't set out to draw precisely where the line is, so there is no failure to show something that we didn't set out to show. The piece is more to remind folks that there is a line, even though that there is the assumption that Windows compatibility is everything to everybody. We were pretty clear in the end of the piece in regards to the Mac Pro maybe never shifting.

    Who do you think reads AI? Do you not think it's primarily Apple devout for decades? William addresses this in the piece, somewhat, in regards to who reads AI. Who reads AI  should be apparent from the forums at least. Based on what we know, the "average" AI reader has been in the Apple ecosystem for well over a decade, is pretty heavily technologically savvy, has many Apple devices and has for ages, well before the iPhone 3gs, iPad, and iPhone 6 explosions in Apple user volume.

    If we could tap into 1% of the "new" Apple customer, we'd be sitting on a gold mine. Most of the new Apple users bought an iPhone and have just that so aren't relevant to this particular conversation, or got an iPhone or iPad and said "hey, this Mac thing might be pretty great" rather than the other way around like it was a decade ago.
    This is the point.  The article (and you) argues that the poll results are irrelevant (ie "overly conservative") because it does not represent the "larger user base needs".  I argue that the AI demographic more closely matches that of Mac users than the larger "new Apple users" and therefore not necessarily "overly conservative".

    If around a third of the prospective user base needs a feature it sure as hell isn't a minor feature.  There isn't any "assumption that Windows compatibility is everything to everybody" but that a significant part of the Mac user base (say closer to 35% than 3.5%) wants that feature.

    There is nothing an ARM based Mac does that an ARM based iPad Pro couldn't do with a couple further tweaks to iOS.

    So why go through the disruption of a significant processor change and leave the Mac lineup half Intel and half ARM?
    We didn't say they were irrelevant, and I'm not sure why you keep saying that. What they are is overly tilted in favor of virtualization and windows compatibility given what we know about who reads and interacts with AppleInsider. And, even given that, the majority still isn't doing it, and it isn't close.

    At no point are we saying that there won't be disruption.

    And as far as why do it? You just have to look to the last two major Mac processor shifts -- Vendors not delivering what they promise. 
    Again, how do you know the poll is “overly tilted in favor of virtualization and windows compatibility“ when those that read AI tend NOT to be the “new Apple” crowd and more of the “old Apple” crowd that owns more than just iOS devices?  In other words the AI demographic more closely matches the Mac demographic than it does the “new Apple” demographic.

    You keep harping on “majority” and trying to avoid admitting that 35% is significant enough percentage to warrant keeping that feature.  Which is why you want to call 35% results to be “overly tilted”.  

    While intel faltered on 10nm it looks like Ice Lake will finally fulfill those promises and Intel has steadily improved power per watt on 14nm.  Further it seems that Intel has been fairly responsive to Apple and my guess is that the customer that requested lakefiejld processor (bigLittle) from Intel was Apple.  Or they will apply Foveros to stack something like T2 with Intel cores to reduce footprint.

    Intel had a bad few years...just like with Itanium.  It’s fashionable to bag on Intel right now and who knows maybe the new Ice Lakes won’t hit 18% IPC improvements.  I’m guessing Intel is back on track
    I've already addressed this in some depth, but here it is again: The AI demo is the high end, tech savvy part of the Mac market. You know, the part of the Mac market that knows that the feature exists, and has the aptitude and/or need to do so. The numbers are right in front of you. If 2/3 of the tech savvy users don't use it, do you honestly think that the new Apple users do in a higher percentage, or even the same? 100% of the Mac using population used USB-A when the 2016 MacBook Pro came out, and Apple shifted anyway. 100% of Mac users used ADB before the iMac, and Apple shifted anyway. Do you think that 35% is going to slow them down on a shift that they want to make to not be beholden to Intel?

    Intel hasn't had a "bad few years." It's had a bad 2011-2019 and 10nm is literally, three years late. For the last eight years, it hasn't set a deadline for release that it's made. It was less time in 94, and in 06 where deadlines weren't being met when the other shifts were made.

    We don't get to say what Apple keeps and what doesn't, and this still isn't some kind of pitch for Apple to do so. It is an observation about why it probably will. There are enough signposts on the road that an ARM Mac is coming. You're welcome to ignore them at your leisure.

    Look, I get that you don't like it. I get that you don't want to see it. I get that you use it, and 35% of the AI demo does. I use it. But, none of that matters to Apple one bit, and you know it.
    You’re ignoring a number of things. As I mentioned in a post that hasn’t been answered, a lot of Mac users boot into Windows because of games. That’s not a trivial number. It can easily account for a large part of that 35%. Not everyone who does something is tech savvy. Installing Windows isn’t that hard.

    maybe there are signposts, and maybe it’s just a desire to see signposts. We don’t know.

    if Apple doesn’t care, then why do they do it, and why do they maintain it?
    They don't do a lot of maintenance on Boot Camp, because the thermal characteristics on the MacBook Pro running Windows are pretty bad. Also, until the hardware changed, the 13-inch MacBook Pro couldn't use an eGPU in it either.

    Apple cares about what it wants to care about, The whims change, and have always done so.

    And no, we aren't ignoring that. It isn't hard to do so, but there are technical hurdles, not the least of which is Windows not being free.
    Still, they always do what they have to to keep it compatible, and a lot of people use it. People who game don’t care about the cost of Windows.
    People who game don't buy a MacBook, MBA, or MacBook Pro to do so.
    I cannot believe that you can’t seem to understand what I’ve been saying. It’s not that difficult. There are people who use Macs. They also like to game. The state of gaming on the Mac is poor. To overcome that, they install Windows into Bootcamp, and play their games there. When they’re through with gaming, they reboot into macOS, and use their computer as a. Mac. I’ve never even hinted that people buy Macs to game. They buy Macs because they want a Mac for most of what they do. But when it comes to games, quite frankly, the Mac sucks, so they need Windows. But they’re not going to buy a Windows machine just to play games. Buying the Windows OS is a lot cheaper, and convenient, so they do that instead.
    One more time -- I do understand what you're saying. What you're saying does happen. This is at least the third time that I've acknowledged that what you're saying does happen to some extent, so I'm not sure why you keep beating the drum.

    However, all indications point to the fact that your supposition that it is a large percentage is in error. This is likely because PCs are cheap enough that it's easier to buy one of those instead, and you get better gaming performance for the dollar. And, if gaming is a big priority, you don't buy a Mac in the first place.
    edited June 2019
  • Reply 144 of 154
    nhtnht Posts: 4,522member
    nht said:
     
    Again, how do you know the poll is “overly tilted in favor of virtualization and windows compatibility“ when those that read AI tend NOT to be the “new Apple” crowd and more of the “old Apple” crowd that owns more than just iOS devices?  In other words the AI demographic more closely matches the Mac demographic than it does the “new Apple” demographic.

    You keep harping on “majority” and trying to avoid admitting that 35% is significant enough percentage to warrant keeping that feature.  Which is why you want to call 35% results to be “overly tilted”.  

    While intel faltered on 10nm it looks like Ice Lake will finally fulfill those promises and Intel has steadily improved power per watt on 14nm.  Further it seems that Intel has been fairly responsive to Apple and my guess is that the customer that requested lakefiejld processor (bigLittle) from Intel was Apple.  Or they will apply Foveros to stack something like T2 with Intel cores to reduce footprint.

    Intel had a bad few years...just like with Itanium.  It’s fashionable to bag on Intel right now and who knows maybe the new Ice Lakes won’t hit 18% IPC improvements.  I’m guessing Intel is back on track
    I've already addressed this in some depth, but here it is again: The AI demo is the high end, tech savvy part of the Mac market. You know, the part of the Mac market that knows that the feature exists, and has the aptitude and/or need to do so. The numbers are right in front of you. If 2/3 of the tech savvy users don't use it, do you honestly think that the new Apple users do in a higher percentage, or even the same? 100% of the Mac using population used USB-A when the 2016 MacBook Pro came out, and Apple shifted anyway. 100% of Mac users used ADB before the iMac, and Apple shifted anyway. Do you think that 35% is going to slow them down on a shift that they want to make to not be beholden to Intel?

    Intel hasn't had a "bad few years." It's had a bad 2011-2019 and 10nm is literally, three years late. For the last eight years, it hasn't set a deadline for release that it's made. It was less time in 94, and in 06 where deadlines weren't being met when the other shifts were made.

    We don't get to say what Apple keeps and what doesn't, and this still isn't some kind of pitch for Apple to do so. It is an observation about why it probably will. There are enough signposts on the road that an ARM Mac is coming. You're welcome to ignore them at your leisure.

    Look, I get that you don't like it. I get that you don't want to see it. I get that you use it, and 35% of the AI demo does. I use it. But, none of that matters to Apple one bit, and you know it.
    This I what you wrote before:


    "In regards to our audience, exactly who do you think AppleInsider is read by far, far more? College grads with advanced degrees, industry folk, designers and whatnot, or the "new Apple user" which is iOS centric, where the iPhone is a halo for the Mac and not the other way around?"

    "If we could tap into 1% of the "new" Apple customer, we'd be sitting on a gold mine. Most of the new Apple users bought an iPhone and have just that so aren't relevant to this particular conversation, or got an iPhone or iPad and said "hey, this Mac thing might be pretty great" rather than the other way around like it was a decade ago."

    Your "new Apple users" demographics...by your own statements...are largely iOS users.

    2011 was bad how? By releasing 32nm Sandy Bridge?
    2012 they went to 22nm and Ivy Bridge.  That sure sucked.
    2013 sucked too when they released Haswell...with DDR4, and TB2
    2014 Broadwell was obviously a complete disaster with 3.5W TDP processors and 14nm
    2015 Skylake was their last Tock for 14nm and gave us TB3.

    2016 Kaby Lake was the first missed Tick.  3 years ago.

    2011-2015 was all stuff Apple wanted:  Thunderbolt, IGP and improved Performance per Watt etc.


  • Reply 145 of 154
    Mike WuertheleMike Wuerthele Posts: 6,858administrator
    nht said:
    nht said:
     
    Again, how do you know the poll is “overly tilted in favor of virtualization and windows compatibility“ when those that read AI tend NOT to be the “new Apple” crowd and more of the “old Apple” crowd that owns more than just iOS devices?  In other words the AI demographic more closely matches the Mac demographic than it does the “new Apple” demographic.

    You keep harping on “majority” and trying to avoid admitting that 35% is significant enough percentage to warrant keeping that feature.  Which is why you want to call 35% results to be “overly tilted”.  

    While intel faltered on 10nm it looks like Ice Lake will finally fulfill those promises and Intel has steadily improved power per watt on 14nm.  Further it seems that Intel has been fairly responsive to Apple and my guess is that the customer that requested lakefiejld processor (bigLittle) from Intel was Apple.  Or they will apply Foveros to stack something like T2 with Intel cores to reduce footprint.

    Intel had a bad few years...just like with Itanium.  It’s fashionable to bag on Intel right now and who knows maybe the new Ice Lakes won’t hit 18% IPC improvements.  I’m guessing Intel is back on track
    I've already addressed this in some depth, but here it is again: The AI demo is the high end, tech savvy part of the Mac market. You know, the part of the Mac market that knows that the feature exists, and has the aptitude and/or need to do so. The numbers are right in front of you. If 2/3 of the tech savvy users don't use it, do you honestly think that the new Apple users do in a higher percentage, or even the same? 100% of the Mac using population used USB-A when the 2016 MacBook Pro came out, and Apple shifted anyway. 100% of Mac users used ADB before the iMac, and Apple shifted anyway. Do you think that 35% is going to slow them down on a shift that they want to make to not be beholden to Intel?

    Intel hasn't had a "bad few years." It's had a bad 2011-2019 and 10nm is literally, three years late. For the last eight years, it hasn't set a deadline for release that it's made. It was less time in 94, and in 06 where deadlines weren't being met when the other shifts were made.

    We don't get to say what Apple keeps and what doesn't, and this still isn't some kind of pitch for Apple to do so. It is an observation about why it probably will. There are enough signposts on the road that an ARM Mac is coming. You're welcome to ignore them at your leisure.

    Look, I get that you don't like it. I get that you don't want to see it. I get that you use it, and 35% of the AI demo does. I use it. But, none of that matters to Apple one bit, and you know it.
    This I what you wrote before:


    "In regards to our audience, exactly who do you think AppleInsider is read by far, far more? College grads with advanced degrees, industry folk, designers and whatnot, or the "new Apple user" which is iOS centric, where the iPhone is a halo for the Mac and not the other way around?"

    "If we could tap into 1% of the "new" Apple customer, we'd be sitting on a gold mine. Most of the new Apple users bought an iPhone and have just that so aren't relevant to this particular conversation, or got an iPhone or iPad and said "hey, this Mac thing might be pretty great" rather than the other way around like it was a decade ago."

    Your "new Apple users" demographics...by your own statements...are largely iOS users.

    2011 was bad how? By releasing 32nm Sandy Bridge?
    2012 they went to 22nm and Ivy Bridge.  That sure sucked.
    2013 sucked too when they released Haswell...with DDR4, and TB2
    2014 Broadwell was obviously a complete disaster with 3.5W TDP processors and 14nm
    2015 Skylake was their last Tock for 14nm and gave us TB3.

    2016 Kaby Lake was the first missed Tick.  3 years ago.

    2011-2015 was all stuff Apple wanted:  Thunderbolt, IGP and improved Performance per Watt etc.


    I'm not sure what you're going for in the bolded part, but admittedly, I am very busy this morning. I've also lost track of your overall point, to be honest -- and that's on me.
     
    You know full well that I wasn't talking about the quality of the processors, and I never said word one about quality aspects. All of these 2011-2019 releases were late, and in some cases, multiple years later than Intel said that they'd ship. Just because it went tick/tock/tick/tock didn't mean that it was on schedule. It just meant that they hit their die shrink/optimization and repeat goal.

    Literally, from what you quoted: "For the last eight years, it hasn't set a deadline for release that it's made. It was less time in 94, and in 06 where deadlines weren't being met when the other shifts were made."

    Full disclosure, I'm not sure when I'm returning to this post.
    edited June 2019
  • Reply 146 of 154
    nhtnht Posts: 4,522member

    As I understand it, Java runs most of the servers world-wide...  If Java is such a target of malware, how do these servers protect themselves from getting and passing on viruses?
     
    It, likely, is Java that exposed my Mac to malware.  I had a very old version of Java installed on this Mac but updated to version 8 to run UGSPlatform -- an open source Java app:  https://github.com/winder/Universal-G-Code-Sender.   I first installed this on the Mac so I could control the UNC natively, while designing under Virtual Windows 10.  I am having trouble with it it on the Mac, so I installed it and am currently running it under virtual Windows 10.  I did not install any anti-virus software on the Windows 10 and am not aware of any Windows viruses.  I don't use Windows for mail -- only CNC stuff with an occasional use of Edge for CNC-related downloads and Help.

    The most active virus is apparently not detected:  About 50% of the time I follow a link to open a new Safari window, the linked window opens, then another window opens right on top of it.  The virus window usually contains ads for someone like Amazon:  

    Left: Original Window -- clicked on diamond ring.  Center: correct window target of link.  Right: Virus window -- opens over center (target) window -- moved aside to display all 3 windows.

    My Virus scanner detects viruses, mostly adware, as shown below -- I get rid of them, but they keep coming back.

    Those are likely ad server redirects and not java related.

    Java may still run most servers world-wide but node is probably more popular for newer sites and has a big fraction of the market share.  Java isn't any more insecure as a language than any other.  Java in the browser had a large attack surface since there was a lot of places for bugs...in the browser, in the VM, in the sandbox implementation, etc. 
    edited June 2019
  • Reply 147 of 154
    SoliSoli Posts: 10,035member
    mattinoz said:
    Soli said:
    I've read the back and forth among respected members/veterans of the AI community.   I, also go way back with Apple: 1978.   I've watched Apple release, then deprecate too many things to count -- No Sure Shitlock!  

    But, my gut tells me that Apple will release a non-Intel Mac (in addition to their Intel Macs) -- because it takes them in a direction (total control) they want to go.   Apple can walk, chew gum and juggle at the same time, so they're capable of it.

    If they fail, they will have pissed away several 10s or hundreds of millions of dollars -- but, even that won't be a total loss.

    As for the non-Intel chip, I don't think it will be an A chip...  There are several reasons:  1) They can produce a chip customized to what Macs do/need (and then some);  2) From a PR perspective it is important to release a non-Intel Mac with a serious commitment and raison d'être.
    I find it weird that so many people on this forum feel that for an ARM-based Mac to exist that an Intel-based Mac can't -or- that so many can only see it within the scope of an A-series chip, which often gets argued as not having enough RAM or memory bandwidth compared be in a traditional PC.
    How are they going to cope if Apple put both ARM and Intel in one machine?
    I mean more than the current T series and it's tiny strip of interface.
    I don't know what you mean by "cope" but several people have noted that they could build a chip that executes x86_64 instruction. I don't think that is the most likely scenario and I suspect that you're taking my comment and assuming that I mean in the same Mac when I'm referring to the Mac line up. Despite certain posters who are claiming that every Mac user needs to a gaming Bootcamp/Windows user before Apple will consider it, that's simply not the case and in the lower-end of the Mac base that's not even a consideration because (in my experience) most don't even know it exists.

    Your Intel-based Mac is going to be fine for the foreseeable future. I bet that when Apple finally does pull the finally plug on Intel-based Macs everyone here who is ardently against Apple offering an ARM-based Mac on the lower-end will be saying "it's about damn time Apple finally shed that dead weight."
    docno42
  • Reply 148 of 154
    hmlongcohmlongco Posts: 533member
    Apple could easily reboot the MacBook Air this fall as an ultra-thin and light ARM-based notebook with great battery life, almost just by taking an iPad Pro and slapping it into a case. 

    And with no need for "fat binaries". Just upload BitCode enabled applications to the App Store and Apple will download the appropriate binary to the device, be it ARM or Intel.

    Boom. We now have a "Mac" running macOS and running "Mac" applications, that (like a ChromeBook) is suitable for the needs of quite a few people.

    Then, maybe next year, the MacBook gets the same treatment.

    Then, in 2021, the low-end MacBook Pro's start seeing ARM-based versions. Heck, or maybe even high-end versions. A 7nm process lets you put a lot of transistors on silicon. How about a dozen high performance cores, eight low performance cores, and 64 neural engine cores?

    People talk about losing the engineering/cad crowd, but they could snag a lot of people with a machine optimized for AI and ML.
    Solidocno42
  • Reply 149 of 154
    dick applebaumdick applebaum Posts: 12,527member
    nht said:

    As I understand it, Java runs most of the servers world-wide...  If Java is such a target of malware, how do these servers protect themselves from getting and passing on viruses?
     
    It, likely, is Java that exposed my Mac to malware.  I had a very old version of Java installed on this Mac but updated to version 8 to run UGSPlatform -- an open source Java app:  https://github.com/winder/Universal-G-Code-Sender.   I first installed this on the Mac so I could control the UNC natively, while designing under Virtual Windows 10.  I am having trouble with it it on the Mac, so I installed it and am currently running it under virtual Windows 10.  I did not install any anti-virus software on the Windows 10 and am not aware of any Windows viruses.  I don't use Windows for mail -- only CNC stuff with an occasional use of Edge for CNC-related downloads and Help.

    The most active virus is apparently not detected:  About 50% of the time I follow a link to open a new Safari window, the linked window opens, then another window opens right on top of it.  The virus window usually contains ads for someone like Amazon:  

    Left: Original Window -- clicked on diamond ring.  Center: correct window target of link.  Right: Virus window -- opens over center (target) window -- moved aside to display all 3 windows.

    My Virus scanner detects viruses, mostly adware, as shown below -- I get rid of them, but they keep coming back.

    Those are likely ad server redirects and not java related.

    Java may still run most servers world-wide but node is probably more popular for newer sites and has a big fraction of the market share.  Java isn't any more insecure as a language than any other.  Java in the browser had a large attack surface since there was a lot of places for bugs...in the browser, in the VM, in the sandbox implementation, etc. 
    FWIW, I uninstalled Java 8 on the machine and the problem with the extra webpages no-longer happens
    roundaboutnow
  • Reply 150 of 154
    viqsiviqsi Posts: 15member
    Soli said:
    I find it weird that so many people on this forum feel that for an ARM-based Mac to exist that an Intel-based Mac can't -or- that so many can only see it within the scope of an A-series chip, which often gets argued as not having enough RAM or memory bandwidth compared be in a traditional PC.
    docno42 said:
    I always find it amusing that people assume a transition would affect all Macs.  
    Well, let's see. How many 68k based Macs were made after PowerPC Macs came out? I'm going to have to go with "none".

    How many PowerPC based Macs were made after Intel Macs came out? Again, pretty sure we'd be looking at "none".

    We've seen this movie before. It's not "can't", it's "Apple won't bother because it never has and has not shown any indication it ever will."

  • Reply 151 of 154
    melgrossmelgross Posts: 33,510member
    melgross said:
    melgross said:
    melgross said:
    nht said:
    nht said:
    nht said:
    nht said:
    nht said:

    nht said:
    wallym said:
    As a developer, I need both mac and windows support.  To openly campaign to remove Windows compat is to be irresponsible to the marketplace.  If users don't need Windows, that's fine.  Don't penalize me for your lack of needs.
    I don't think you, nor FredFref read the article.
    Why does a dissenting opinion mean they didn't read the article?  Maybe they read, disagreed with the basic premise "cross-platform software compatibility is now mostly irrelevant to the wider user base" and everything that follows.  Especially since you had a poll, found 35% that said, yes they needed windows and then proceeded to hand wave that away as AI readers aren't a representative sample.  Which begs the question of WHY RUN THE POLL IN THE FIRST PLACE?

    The next assertion "for Apple's biggest user base, the need for Windows compatibility isn't the same as it is for the main readers of this site" is fabricated out of thin air and has zero supporting data.  Whether true or not it's based on nothing but speculation.

    If the primary uses of the Macs are Pro and everyone else migrates to iPads then a significant fraction of Mac users (dare I say 35%) will want x86 compatibility.

    But, nope...because they disagree they didn't read the article.
    That's not why I said that, and you know it. And, there's a lot more to this quote of mine than what you clipped out. And, I didn't even say anything about the ludicrous assumption that this article is a "campaign" to remove Windows compatibility.

    It wasn't handwaved away. What it is, is that 35% of the user base that reads AI doesn't need it, which is an overly conservative estimate of what the larger user base needs and does with their machines, and you know this as well, based on your own interactions with the rest of the AI readership. And, even if you translate it literally, it does mean that the majority doesn't care about Windows on the Mac.
    Why run the poll and then disregard it?  So what if 35% isn't the majority?  It's still a large part of the user base.   

    And how do you know that it is "an overly conservative estimate of what the larger user base needs"?  On what data is this assertion based on?  Why do you assume that the majority of your readers are pros?  Why did you not include in your survey to self identify if they were pros or just general users?  Never mind that these polls are generally horridly misleading anyway.

    The article, and you, would like to make it seem like it's 0.35% of the user base to sell the idea that x86 compatibility is no longer needed.  Apple may have a good idea as to the number but you don't.  Moreover you ignored the entire enterprise market because it's inconvenient.  Does IBM and other major Mac deployments believe x86 compatibility is irrelevant?  I have no idea and neither do you.  It would have been fairly easy to reach out to IT folks highlighted in past articles and ask "hey, is x86 compatibility important to your Mac enterprise deployment?"

    Nope.

    But hey...35% is an overly conservative estimate of what the larger user base needs...
    Regarding the bolded section, we are, and your own supposition of how that is going so far is wrong because what they care about so far is iOS development and general productivity. We'll see how it goes in total when we're done.

    In regards to our audience, exactly who do you think AppleInsider is read by far, far more? College grads with advanced degrees, industry folk, designers and whatnot, or the "new Apple user" which is iOS centric, where the iPhone is a halo for the Mac and not the other way around?
    And the demographics for Mac users are what?  Gee maybe folks who are "college grads with advanced degrees, industry folk, designers and whatnot"?  

    Nah.

    I will assert, based on personal experience, that there are very few enterprise iOS developers that don't care about MS project, DOORS and a bevy of windows/x86 corporate tools...still dependent on Excel with macros.  People send me a lot of stuff in Visio to boot.  Also, most of us aren't iOS developers but enterprise developers and the docker tool chain is a significant part of devops.
    Yeah. we're not done. So far, we've spoken to IBM, Cisco, and Deloitte. There are about eight more on the docket. The point of this article, stands, though, that there is a line, where below it, the need for Windows is non-existent.
    And you have failed to show where that line is.  Again, is the AI demographic you just stated significantly different than that of Mac users?  Or have most of the "new Apple users" that don't frequent AI already moved to the iPad or never bought a Mac in the first place and have a windows laptop somewhere?
    I'm not really sure what you're asking, here.

    We didn't set out to draw precisely where the line is, so there is no failure to show something that we didn't set out to show. The piece is more to remind folks that there is a line, even though that there is the assumption that Windows compatibility is everything to everybody. We were pretty clear in the end of the piece in regards to the Mac Pro maybe never shifting.

    Who do you think reads AI? Do you not think it's primarily Apple devout for decades? William addresses this in the piece, somewhat, in regards to who reads AI. Who reads AI  should be apparent from the forums at least. Based on what we know, the "average" AI reader has been in the Apple ecosystem for well over a decade, is pretty heavily technologically savvy, has many Apple devices and has for ages, well before the iPhone 3gs, iPad, and iPhone 6 explosions in Apple user volume.

    If we could tap into 1% of the "new" Apple customer, we'd be sitting on a gold mine. Most of the new Apple users bought an iPhone and have just that so aren't relevant to this particular conversation, or got an iPhone or iPad and said "hey, this Mac thing might be pretty great" rather than the other way around like it was a decade ago.
    This is the point.  The article (and you) argues that the poll results are irrelevant (ie "overly conservative") because it does not represent the "larger user base needs".  I argue that the AI demographic more closely matches that of Mac users than the larger "new Apple users" and therefore not necessarily "overly conservative".

    If around a third of the prospective user base needs a feature it sure as hell isn't a minor feature.  There isn't any "assumption that Windows compatibility is everything to everybody" but that a significant part of the Mac user base (say closer to 35% than 3.5%) wants that feature.

    There is nothing an ARM based Mac does that an ARM based iPad Pro couldn't do with a couple further tweaks to iOS.

    So why go through the disruption of a significant processor change and leave the Mac lineup half Intel and half ARM?
    We didn't say they were irrelevant, and I'm not sure why you keep saying that. What they are is overly tilted in favor of virtualization and windows compatibility given what we know about who reads and interacts with AppleInsider. And, even given that, the majority still isn't doing it, and it isn't close.

    At no point are we saying that there won't be disruption.

    And as far as why do it? You just have to look to the last two major Mac processor shifts -- Vendors not delivering what they promise. 
    Again, how do you know the poll is “overly tilted in favor of virtualization and windows compatibility“ when those that read AI tend NOT to be the “new Apple” crowd and more of the “old Apple” crowd that owns more than just iOS devices?  In other words the AI demographic more closely matches the Mac demographic than it does the “new Apple” demographic.

    You keep harping on “majority” and trying to avoid admitting that 35% is significant enough percentage to warrant keeping that feature.  Which is why you want to call 35% results to be “overly tilted”.  

    While intel faltered on 10nm it looks like Ice Lake will finally fulfill those promises and Intel has steadily improved power per watt on 14nm.  Further it seems that Intel has been fairly responsive to Apple and my guess is that the customer that requested lakefiejld processor (bigLittle) from Intel was Apple.  Or they will apply Foveros to stack something like T2 with Intel cores to reduce footprint.

    Intel had a bad few years...just like with Itanium.  It’s fashionable to bag on Intel right now and who knows maybe the new Ice Lakes won’t hit 18% IPC improvements.  I’m guessing Intel is back on track
    I've already addressed this in some depth, but here it is again: The AI demo is the high end, tech savvy part of the Mac market. You know, the part of the Mac market that knows that the feature exists, and has the aptitude and/or need to do so. The numbers are right in front of you. If 2/3 of the tech savvy users don't use it, do you honestly think that the new Apple users do in a higher percentage, or even the same? 100% of the Mac using population used USB-A when the 2016 MacBook Pro came out, and Apple shifted anyway. 100% of Mac users used ADB before the iMac, and Apple shifted anyway. Do you think that 35% is going to slow them down on a shift that they want to make to not be beholden to Intel?

    Intel hasn't had a "bad few years." It's had a bad 2011-2019 and 10nm is literally, three years late. For the last eight years, it hasn't set a deadline for release that it's made. It was less time in 94, and in 06 where deadlines weren't being met when the other shifts were made.

    We don't get to say what Apple keeps and what doesn't, and this still isn't some kind of pitch for Apple to do so. It is an observation about why it probably will. There are enough signposts on the road that an ARM Mac is coming. You're welcome to ignore them at your leisure.

    Look, I get that you don't like it. I get that you don't want to see it. I get that you use it, and 35% of the AI demo does. I use it. But, none of that matters to Apple one bit, and you know it.
    You’re ignoring a number of things. As I mentioned in a post that hasn’t been answered, a lot of Mac users boot into Windows because of games. That’s not a trivial number. It can easily account for a large part of that 35%. Not everyone who does something is tech savvy. Installing Windows isn’t that hard.

    maybe there are signposts, and maybe it’s just a desire to see signposts. We don’t know.

    if Apple doesn’t care, then why do they do it, and why do they maintain it?
    They don't do a lot of maintenance on Boot Camp, because the thermal characteristics on the MacBook Pro running Windows are pretty bad. Also, until the hardware changed, the 13-inch MacBook Pro couldn't use an eGPU in it either.

    Apple cares about what it wants to care about, The whims change, and have always done so.

    And no, we aren't ignoring that. It isn't hard to do so, but there are technical hurdles, not the least of which is Windows not being free.
    Still, they always do what they have to to keep it compatible, and a lot of people use it. People who game don’t care about the cost of Windows.
    People who game don't buy a MacBook, MBA, or MacBook Pro to do so.
    I cannot believe that you can’t seem to understand what I’ve been saying. It’s not that difficult. There are people who use Macs. They also like to game. The state of gaming on the Mac is poor. To overcome that, they install Windows into Bootcamp, and play their games there. When they’re through with gaming, they reboot into macOS, and use their computer as a. Mac. I’ve never even hinted that people buy Macs to game. They buy Macs because they want a Mac for most of what they do. But when it comes to games, quite frankly, the Mac sucks, so they need Windows. But they’re not going to buy a Windows machine just to play games. Buying the Windows OS is a lot cheaper, and convenient, so they do that instead.
    One more time -- I do understand what you're saying. What you're saying does happen. This is at least the third time that I've acknowledged that what you're saying does happen to some extent, so I'm not sure why you keep beating the drum.

    However, all indications point to the fact that your supposition that it is a large percentage is in error. This is likely because PCs are cheap enough that it's easier to buy one of those instead, and you get better gaming performance for the dollar. And, if gaming is a big priority, you don't buy a Mac in the first place.
    Because you keep saying that people don’t buy MacBooks to game. That’s saying that I’m saying that people are buying MacBooks specifically to game, and I’m not saying that. I’d also like to see some evidence, if there is any, as to percentages of those using Windows for business vs gaming. You’re also denying the overall 35% number. Maybe that number is correct, and it’s not mostly for business, outside of those you mention, such as those coming here, though I doubt most coming here are necessarily using Macs for business with Windows.

    its also not cheaper for Mac users to buy a Windows laptop to game. It’s cheaper to just get windows and install it.

    at any rate, it doesn’t matter. Apple will do whatever they want to.
    edited June 2019
  • Reply 152 of 154
    crowleycrowley Posts: 10,453member
    viqsi said:

    Well, let's see. How many 68k based Macs were made after PowerPC Macs came out? I'm going to have to go with "none".

    How many PowerPC based Macs were made after Intel Macs came out? Again, pretty sure we'd be looking at "none".
    If you watch the Intel transition announcement keynote, after talking about moving to Intel and talking about universal binaries, Steve mentions that they still have great PowerPC products to come.

    But you’re right, they never did.
  • Reply 153 of 154
    docno42docno42 Posts: 3,755member
    viqsi said:
    How many PowerPC based Macs were made after Intel Macs came out? Again, pretty sure we'd be looking at "none".

    We've seen this movie before. It's not "can't", it's "Apple won't bother because it never has and has not shown any indication it ever will."

    lol - 68K to PowerPC and PowerPC to Intel were due to performance.  Primarily mobile performance.

    ARM is performing BETTER than Intel mobile performance TODAY.  

    So leverage ARM for where it is superior and leave Intel on the desktop.  Apples software ecosystem has been co-existing on ARM and Intel ever since the iPhone debuted so it's not like this is a huge technical challenge for them. 

    I don't see Apple focusing on high end ARM desktop chips - unless they have to.  I think they would switch to AMD from Intel before they release an all ARM Mac lineup.  I have no doubt the MB Air will one day sport an ARM CPU and the vast (VAST!) majority of people won't care; aside from the dramatically better performance, battery life and reduced heat.  

    I'm more than ready for a MBA with ARM and a non-butterfly keyboard next year.  C'mon Apple!
    Soli
Sign In or Register to comment.