Quote from: 8&8 on July 27, 2021, 14:19:31ARM must have their industry chip makers so this another bipolarism between TSMC/Intel must end! GF will be inglobed by Intel but MT&Qc use TSMC node.

ARM has revolutionized with scaling vector pc users market so why not to complete their work with their company and why not start use carbon nanotubes 10nm nodes.

Huh?! 

For Gods sake, how do you people bypass [defacto] spelling/grammar checks that is built into nearly everything?!

That aside, and while this is an old article, let's not forget this:

"Critical Feature Size"—-and a Foundry's "node size"—-does NOT  in and of itself dictate "who is gonna have the 'fastest/best' chip" here.

You can bet Apple's SOC will continue to kick any Android 'processor' into the dirt, as it has been doing by orders of magnitude, and will continue to do.

Fact, not opinion.

Such an "article" like this is always indicative of someone's (lack of) knowledge of *why* what I'm saying is absolute fact.

Remember kids: Apple is in control of both its hardware, firmware, and software... always yielding the highest performing and most polished experience.

Ever notice no idiots are running around calling people "Android fanboys"?!

Sincerely,
One f*** of an experienced wafer fab process- and equipment-engineering geek and guru 🤘

I don't think the premise of this article is sound. The implications of a H2 2022 release of 3nm processor tech by TSMC are less easy to pin down than the article pretends. Today, at the premium end of the smartphone market the annual cycle and timing of product releases depends mostly on a) competitive marketing strategies and b) the close alignment of those marketing strategies with the concerted action of strategic partners to lay the groundwork to ensure a generous provisioning of as yet unreleased advanced processors that are expected to be used in future products. While marketing strategies can be readily changed when the moment demands product development is a process that contains more balast. Product development starts long before we see any marketable products. Making changes to a product development life cycle for the sake of a marketing opportunity that has presented itself is trickier than it sounds. For instance, there have to be new chips that are ready to be manufactured and clients that can use those chips in products they are ready to manufacture for the scenario set out in this article to amount to anything.

September is recognizably the month to expect new iPhone wares and March the month to watch out to see what the Android manufacturers have in store for us is a convenience for all of the players in this industry. It is not fixed in stone. It is a convention only, albeit one that tends to help smartphone manufacturers to realise their commercial goals. So, contrary to the reasoning informing the article it wouldn't be a dangerous interruption to Apple's standard annual marketing cycle to delay release of its iPhone in order to take advantage of TSMCs 3nm tech for several months. Indeed, Apple's standing with TSMC is so good that it would be unlikely that any other company could jump the queue to get ahead of Apple if Apple saw 3nm silicon as essential to its cause, i.e. to prove the performance superiority of the A16 and its credentials as a highly energy efficient part.

So, no, TSMC not being able to reach high volume manufacture for 3nm silicon before H2 2022 doesn't really present Apple with a lot of hard business choices. And, no, it does not leave Apple at a disadvantage. If Apple sticks with its standard release schedule in 2022 and foregoes 3nm until 2023 that would be a sign that Apple isn't particularly worried about the other ARM chip manufacturers getting to use 3nm silicon first which shows the confidence they have in their own Apple microarchitecture.

It is worth noting that should Apple elect not to use 3nm in 2022 only some very woolly thinking could conclude from such a scenario that Android phone manufacturers and their ARM chip suppliers would therefore be in a position to pounce to drive home a competition advantage that has propitiously presented itself. Nonsense. Even if we allow that the superseding part to Qualcomm's SD895, say, could be technically quite good why is it being supposed that Qualcomm will have that chip ready to put into finished consumer products before the end of 2022? Of course the plan for ARM chip suppliers and their Android phone manufacturer partners would always have been to introduce 3nm parts in early 2023. And, that is still the best strategy from a marketing point of view - makers of premium Android phones want to get 'clear air' free from the buzz around Apple products, that takes some time to die away, before making their own product introductions. Also, to reiterate, Qualcomm probably won't be in a position to ship its processor in high volumes until then.

Could the availability of TSMC's 3nm silicon in 2022 without Apple scooping up the whole supply be relevant to the scheduled NUVIA (Qualcomm) laptop chip release late next year? It can't be ruled out, but I wouldn't be betting on it.

ARM must have their industry chip makers so this another bipolarism between TSMC/Intel must end! GF will be inglobed by Intel but MT&Qc use TSMC node.

ARM has revolutionized with scaling vector pc users market so why not to complete their work with their company and why not start use carbon nanotubes 10nm nodes.