Apple's claims about M1 Mac speed 'shocking,' but 'extremely plausible'

135

Comments

  • Reply 41 of 84
    docno42docno42 Posts: 3,755member

    ddawson100 said:
    Just because you can buy a $250 Windows computer doesn't mean everyone should!

    lol - remember when Apple was doomed because of Netbooks?  

    Their only redeeming quality was they were cheap.  So when you recycled them into e-waste it didn't sting as bad.  That the same people are wrong on a continual basis but lack the self awareness to be embarrassed and thus maybe breaking the cycle...  At least it provides some entertainment :)
    williamlondonmagman1979jdb8167watto_cobra
  • Reply 42 of 84
    docno42docno42 Posts: 3,755member
    kkqd1337 said:
    whats the best selling windows computer its meant to be so much better than? some $200 education machine.
    The perennial problem for Apple haters - facts don't care about your feelings!
    williamlondonmagman1979jdb8167equality72521watto_cobra
  • Reply 43 of 84
    Xed said:
    Sarkany said:
    Sarkany said:
    Comparing mobile chips to the likes of 10900K...that is extremely laughable.
    Yes, laughable...


    We got desktop computers (with hefty cooling solutions) running at 80C+ during CPU/GPU intensive tasks, but we’re supposed to believe iPhones/iPads/Mac Laptops are able to run the same programs without undergoing a full meltdown. Yeah, okay. I love my Apple devices, but don’t get ahead of yourself there, buddy.
    That's why AnandTech is a respected web publication that does deep dives into processors and you're just a guy on the internet with absolutely nothing but irrational hate to back up your claims.

    You know that you can go to AnandTech's website to see why the Apple A13 scored a 63.3 on the SPECiint2006 test, right? Of course, that takes a little effort on your part so I guess I shouldn't hold my breath.
    I build my own systems, I benchmark my own systems, and I run real world tests on those systems. I have enough years of experience to back up my claims. All you’re doing is employing your appeal to authority fallacy and nothing else.
    williamlondon
  • Reply 44 of 84
    XedXed Posts: 2,519member
    Sarkany said:
    Xed said:
    Sarkany said:
    Sarkany said:
    Comparing mobile chips to the likes of 10900K...that is extremely laughable.
    Yes, laughable...


    We got desktop computers (with hefty cooling solutions) running at 80C+ during CPU/GPU intensive tasks, but we’re supposed to believe iPhones/iPads/Mac Laptops are able to run the same programs without undergoing a full meltdown. Yeah, okay. I love my Apple devices, but don’t get ahead of yourself there, buddy.
    That's why AnandTech is a respected web publication that does deep dives into processors and you're just a guy on the internet with absolutely nothing but irrational hate to back up your claims.

    You know that you can go to AnandTech's website to see why the Apple A13 scored a 63.3 on the SPECiint2006 test, right? Of course, that takes a little effort on your part so I guess I shouldn't hold my breath.
    I build my own systems, I benchmark my own systems, and I run real world tests on those systems. I have enough years of experience to back up my claims. All you’re doing is employing your appeal to authority fallacy and nothing else.
    Then show us where AnandTech has misled anyone. Redo all their tests and get back to us with the new results.

    PS: Checking frame rate on a video game isn't what I'd define as benchmarking, but I guess it technically counts.
    edited November 2020 rezwitstmaywilliamlondonmagman1979jdb8167watto_cobra
  • Reply 45 of 84
    Xed said:
    Sarkany said:
    Xed said:
    Sarkany said:
    Sarkany said:
    Comparing mobile chips to the likes of 10900K...that is extremely laughable.
    Yes, laughable...


    We got desktop computers (with hefty cooling solutions) running at 80C+ during CPU/GPU intensive tasks, but we’re supposed to believe iPhones/iPads/Mac Laptops are able to run the same programs without undergoing a full meltdown. Yeah, okay. I love my Apple devices, but don’t get ahead of yourself there, buddy.
    That's why AnandTech is a respected web publication that does deep dives into processors and you're just a guy on the internet with absolutely nothing but irrational hate to back up your claims.

    You know that you can go to AnandTech's website to see why the Apple A13 scored a 63.3 on the SPECiint2006 test, right? Of course, that takes a little effort on your part so I guess I shouldn't hold my breath.
    I build my own systems, I benchmark my own systems, and I run real world tests on those systems. I have enough years of experience to back up my claims. All you’re doing is employing your appeal to authority fallacy and nothing else.
    Then show us where AnandTech has misled anyone. Redo all their tests and get back to us with the new results.

    PS: Checking frame rate on a video game isn't what I'd define as benchmarking, but I guess it technically counts.
    P.S. I don’t play video games.
    williamlondon
  • Reply 46 of 84
    docno42 said:
    kkqd1337 said:
    whats the best selling windows computer its meant to be so much better than? some $200 education machine.
    The perennial problem for Apple haters - facts don't care about your feelings!
    I have gin, campari and vermouth rosso for my feelings
    williamlondonwatto_cobra
  • Reply 47 of 84
    dewmedewme Posts: 5,333member
    Pjs said:
    The future of the Ipad pro is bleak.  Prices are higher.  And they still use the Ios base ....
    I wouldn't go as far as saying that. It really depends on your use cases and whether you need a touch based device and support for Apple Pencil, which is a big plus for iPad Pro and totally missing from MacBook Air. However, if you are leaving your iPad Pro permanently parked in a keyboard doc, don't use Apple Pencil, and don't really care about touch, the fanless M1 MacBook Air is a very attractive option for some folks.

    I've yet to find a detachable keyboard with iPad or MS Surface that was enjoyable to use when sitting on my lap. On the other hand, I've always considered the MacBook Air to be the most enjoyable laptop/notebook computer ever from an ergonomics perspective. The thin front edge, great keyboard, glorious trackpad, and lap stability is unmatched by any "composite" device.

    The least enjoyable Apple product from an ergonomic standpoint, for me, is the 12.9" iPad Pro. It's too tall, too wide, too heavy, and just too clumsy to hold and use just as a tablet. I feel like I'm holding a big screen TV in my hands when I'm cozied up on the sofa with it. But yes, it is pretty good as a small countertop TV in the kitchen when it's docked. I've had one for 3+ years and I really dislike it a lot. I'll either trade it in for a new iPad Air or a new M1 MacBook Air. I'm leaning much more towards the latter option because the MBA is such a joy to use and I have a smaller iPad Pro to fill my tablet needs. If the iPad Mini wasn't so far behind the technology and design curve I'd also get one of those for tablet-only usage, but I think my next Apple purchase will be an M1 MacBook Air. 
    watto_cobra
  • Reply 48 of 84
    docno42 said:
    h4y3s said:
    Don’t overlook the unified memory architecture that Apple can deploy, (as they own the whole stack) this will save 2x on a lot of common functions! 
    I suspect that the RAM being closely coupled to the SOC package is where a good chunk of their speed increase is coming from, not from the unified memory architecture itself.  Being physically closer to the CPU cores should not be overlooked.

    Which means it will be interesting to see how they handle larger desktop macs with larger memory demands.  If they support off SOC RAM with those chips, we will get to see just how much of an impact that integrating memory directly onto the SOC makes.
    There are other possibilities for off-SOC memory, beyond traditional RAM — think persistent memory like what Intel is doing with Optane. 
    watto_cobra
  • Reply 49 of 84
    Sarkany said:
    Xed said:
    Sarkany said:
    Sarkany said:
    Comparing mobile chips to the likes of 10900K...that is extremely laughable.
    Yes, laughable...


    We got desktop computers (with hefty cooling solutions) running at 80C+ during CPU/GPU intensive tasks, but we’re supposed to believe iPhones/iPads/Mac Laptops are able to run the same programs without undergoing a full meltdown. Yeah, okay. I love my Apple devices, but don’t get ahead of yourself there, buddy.
    That's why AnandTech is a respected web publication that does deep dives into processors and you're just a guy on the internet with absolutely nothing but irrational hate to back up your claims.

    You know that you can go to AnandTech's website to see why the Apple A13 scored a 63.3 on the SPECiint2006 test, right? Of course, that takes a little effort on your part so I guess I shouldn't hold my breath.
    I build my own systems, I benchmark my own systems, and I run real world tests on those systems. I have enough years of experience to back up my claims. All you’re doing is employing your appeal to authority fallacy and nothing else.
    And you think that your one-man-show approach is better than Anandtech with an entire STAFF of professionals and a full LAB of testing gear?

    My GOD your arrogance and stupidity just showcased themselves with a damned spotlight! What a freaking idiot...
    Xedwilliamlondonericthehalfbeejdb8167anonconformistwatto_cobra
  • Reply 50 of 84
    wizard69wizard69 Posts: 13,377member
    Pjs said:
    The future of the Ipad pro is bleak.  Prices are higher.  And they still use the Ios base ....
    On the contrary, put this class of chip in an iPad and you have the most powerful table going.    IPad OS is a mixed bag and I really hope that they address the short comings there.   It wold be interesting to see Mac OS on a tablet like iPad but I'm not sure that is a reasonable long term play for Apple.    Rather I see Apple working far more AI techniques into both Mac OS and iPad OS to drive these operating systems forward.
  • Reply 51 of 84
    wizard69wizard69 Posts: 13,377member
    kkqd1337 said:
    I'm dubious of their claims. But i'm all for it if its true. I'll probably wait for M2 tho as I dont need anything right now.

    One thought I had....

    if these chips are so amazing.... why are they not making servers? using them in data centres, thats where power per watt really matters isnt it?
    Apple has never been successful selling servers.    In fact I'm not even sure they understand the market for small systems.  

    As for power per watt that is exactly where Apple has been most competitive.   Performance per watt is what makes the iPad great, they have great functionality in a thin device with performance nobody can match.   This is even more obvious when you consider that the MBA is now fan free.   Performance per watt is far more importnat in the portable world than in the data center.    In the data center it is about the electrical bill, in the portable world it is about getting work done without the battery dying.   
    watto_cobra
  • Reply 52 of 84
    wizard69wizard69 Posts: 13,377member
    flydog said:
    kkqd1337 said:
    I'm dubious of their claims. But i'm all for it if its true. I'll probably wait for M2 tho as I dont need anything right now.

    One thought I had....

    if these chips are so amazing.... why are they not making servers? using them in data centres, thats where power per watt really matters isnt it?
    You're dubious of Apple's claims?  So your position until you see evidence is that Apple is committing fraud by misrepresenting the capability of its hardware. 
    It isn't fraud but Apple has a long history of focusing on parameters that they excel at to take the public gaze off the shortcomings.   For example Apple as a fanless machine - this is very good by the way - but until the machine is tested we don't know what its performance will be like without testing.    Back in history with Power PC Apple focused its dog and pony shows on floating point performance while PPC absolutely sucked at integer performance.    It isn't fraud but it isn't exactly being upfront.   

    so while I might not completely dismiss Apples claims, they are often true in very fixed contexts, I'd rather would like to see real world testing.    This especially in the case of that fanless MBA because we have no way to judge it as it is a completely new offering.   So the question then becomes how much does it loose compared the new MBP.    We don't know and Apples BS did nothing to help us understand the performance differences.  
  • Reply 53 of 84
    wizard69wizard69 Posts: 13,377member
    mjtomlin said:
    People seem to overlook where Apple thinks this type of performance belongs; at the very bottom. This is and will be their least performant SoC. I think in the Spring we'll see a variant of the M1 that adds 2 more high performance CPU cores, 4 more GPU cores and doubles on board memory. This new SoC will be an upgrade option for the mini and PowerBook 13" and may replace the low-end 16" model and even appear in the smaller 21" iMac. Then in the Fall the M2 will increase performance, double the PCI lanes and allow for off board memory expansion (along with other improvements across the board).
    Yes exactly.   Apple is pushing ARM from the bottom up and is doing so with pretty incredible performance specs (which I hope get verified real soon now).   

    As for higher performance chips it isn't clear to me which way they will go.   Adding a couple of more cores might work for some work loads but what Apple really needs is to remain competitive in workloads demanding high core counts.   So I'm thinking Apple will look for an economical way to add lots of cores so that they can hit at least 24 cores in the high end MBP laptop.   This would likely mean a chiplet like approach.    

    Now given what you have described as the M2, I suspect hat such a chip isn't far off but it will be a direct replacement for M1 next year.    The goal being to beef up the GPU which frankly kinda sucks on M1.   More CPU cores might happen but it could be as little as one more core.   Next year will still mean 5nm and frankly the die will end up larger and the best use of that area will be to fix the GPU.  
  • Reply 54 of 84
    It now appears the IPad Pro is an isolated island stuck with iPadOS and cut down versions of apps which need to be full blown versions.  The Mac can now run iPad apps and no sign or possibility of running the iPad Pro running Mac apps.  If you are an owner of an extremely expensive 2020 iPad Pro you have a right to feel badly let down and ripped off.  Apple needs to do a lot better.  Perhaps its time for a trade up deal for iPad Pro owners, there is only so much frustration one can take.
    williamlondon
  • Reply 55 of 84
    dewmedewme Posts: 5,333member
    wizard69 said:
    Pjs said:
    The future of the Ipad pro is bleak.  Prices are higher.  And they still use the Ios base ....
    On the contrary, put this class of chip in an iPad and you have the most powerful table going.    IPad OS is a mixed bag and I really hope that they address the short comings there.   It wold be interesting to see Mac OS on a tablet like iPad but I'm not sure that is a reasonable long term play for Apple.    Rather I see Apple working far more AI techniques into both Mac OS and iPad OS to drive these operating systems forward.
    I’m waiting to see how iOS/iPadOS apps work out on M1 Macs. Running them is one thing, but running them well and with a good user experience is quite another. My expectations are high because the mouse and keyboard experience in iPadOS is actually quite good, but I’m probably forgetting the few times I’ve had to (or decided to) revert to touch only because it was still an option. Things like window sizing is still a mystery. Do you have to pick a reference iPad screen like you do in XCode?

    We’ll see. 
    edited November 2020 williamlondonwatto_cobra
  • Reply 56 of 84
    wizard69wizard69 Posts: 13,377member
    Sarkany said:
    Sarkany said:
    Comparing mobile chips to the likes of 10900K...that is extremely laughable.
    Yes, laughable...


    We got desktop computers (with hefty cooling solutions) running at 80C+ during CPU/GPU intensive tasks, but we’re supposed to believe iPhones/iPads/Mac Laptops are able to run the same programs without undergoing a full meltdown. Yeah, okay. I love my Apple devices, but don’t get ahead of yourself there, buddy.
    It is possible and in large part due to x86 being a terrible architecture.    Now do I believe Apples hype, well lets put it this way I believe they are accurate in the very precise context that they have laid out.   That means very little for the average user as we need to know how these machines perform under various workloads.   In the case of MBA we need to know how badly it throttles compared to the MBP.   You see Apple doesn't so much lie as they only focus on parameters that make the product look good - classic marketing.   What we need is a board selection of benchmarking from various sources that give us a better idea of what is happening with this hardware.   In effect before we dismiss the products we need all the information that Apple left out.

    To put it another way we already have a good idea of how A14 works in a thermally constrained environment.   What we have now is a slightly enhanced version of A14 that has far fewer thermal constraints.   This is going to be real interesting hardware no matter what your point of view.
    muthuk_vanalingamwatto_cobra
  • Reply 57 of 84
    entropysentropys Posts: 4,152member
    wizard69 said:
    Sarkany said:
    Sarkany said:
    Comparing mobile chips to the likes of 10900K...that is extremely laughable.
    Yes, laughable...


    We got desktop computers (with hefty cooling solutions) running at 80C+ during CPU/GPU intensive tasks, but we’re supposed to believe iPhones/iPads/Mac Laptops are able to run the same programs without undergoing a full meltdown. Yeah, okay. I love my Apple devices, but don’t get ahead of yourself there, buddy.
    It is possible and in large part due to x86 being a terrible architecture.    Now do I believe Apples hype, well lets put it this way I believe they are accurate in the very precise context that they have laid out.   That means very little for the average user as we need to know how these machines perform under various workloads.   In the case of MBA we need to know how badly it throttles compared to the MBP.   You see Apple doesn't so much lie as they only focus on parameters that make the product look good - classic marketing.   What we need is a board selection of benchmarking from various sources that give us a better idea of what is happening with this hardware.   In effect before we dismiss the products we need all the information that Apple left out.

    To put it another way we already have a good idea of how A14 works in a thermally constrained environment.   What we have now is a slightly enhanced version of A14 that has far fewer thermal constraints.   This is going to be real interesting hardware no matter what your point of view.
    Quite so. And even more interesting, the base level configuration of the chip that goes in the lowest level consumer product Apple offers. What will be available in six months’ time?
    watto_cobra
  • Reply 58 of 84
    wizard69wizard69 Posts: 13,377member
    dewme said:
    wizard69 said:
    Pjs said:
    The future of the Ipad pro is bleak.  Prices are higher.  And they still use the Ios base ....
    On the contrary, put this class of chip in an iPad and you have the most powerful table going.    IPad OS is a mixed bag and I really hope that they address the short comings there.   It wold be interesting to see Mac OS on a tablet like iPad but I'm not sure that is a reasonable long term play for Apple.    Rather I see Apple working far more AI techniques into both Mac OS and iPad OS to drive these operating systems forward.
    I’m waiting to see how iOS/iPadOS apps work out on M1 Macs. Running them is one thing, but running them well and with a good user experience is quite another. My expectations are high because the mouse and keyboard experience in iPadOS is actually quite good, but I’m probably forgetting the few times I’ve had to (or decided to) revert to touch only because it was still an option. Things like window sizing is still a mystery. Do you have to pick a reference iPad screen like you do in XCode?

    We’ll see. 
    I will be interesting to see how iPad software on the Mac works out.   I'd be surprised if Apples first effort at the support remains around for a long time.    That is running iPad apps will be refined over time.

    I'm not too convinced that running ipad apps will be a long term thing on the ARM Macs, what it will be is a great way to get native apps early in the development cycle.   If Apple performance specs reflect the real life experience I'm expecting developers to transition to these machine pretty fast.   In a year or two the old x86 Macs will not even be getting app updates.
  • Reply 59 of 84
    I have two questions: 
    1. Is it theoretically possible using any virtualization software that I will be able to run two copies of Big Sur on the same Apple Silicon computer?
    2. People talk about being able to run binaries for iOS/iPadOS on MacOS, because they are the same binaries, but does that also mean that I will be able to run macOS binaries on iPadOS (since it supports a mouse and keyboard)?
    I haven't noticed anyone talking about these issues. But I may have missed it. These are important questions for me.

    williamlondon
  • Reply 60 of 84
    Sarkany said:
    Xed said:
    Sarkany said:
    Sarkany said:
    Comparing mobile chips to the likes of 10900K...that is extremely laughable.
    Yes, laughable...


    We got desktop computers (with hefty cooling solutions) running at 80C+ during CPU/GPU intensive tasks, but we’re supposed to believe iPhones/iPads/Mac Laptops are able to run the same programs without undergoing a full meltdown. Yeah, okay. I love my Apple devices, but don’t get ahead of yourself there, buddy.
    That's why AnandTech is a respected web publication that does deep dives into processors and you're just a guy on the internet with absolutely nothing but irrational hate to back up your claims.

    You know that you can go to AnandTech's website to see why the Apple A13 scored a 63.3 on the SPECiint2006 test, right? Of course, that takes a little effort on your part so I guess I shouldn't hold my breath.
    I build my own systems, I benchmark my own systems, and I run real world tests on those systems. I have enough years of experience to back up my claims. All you’re doing is employing your appeal to authority fallacy and nothing else.

    The village idiot complains about "appeal to authority", then turns around and does the same thing by talking about how they build their own systems to add credibility to their lies.

    Andrei from Anand is an actual chip designer who worked for Imagination at one point. Ian has a doctorate in Computational Chemistry from Oxford. Curious what your background relating to processor design is.
    thtwilliamlondonjdb8167tmayanonconformistequality72521watto_cobra
Sign In or Register to comment.