Apple A14 Bionic Processor (SoC) Specs - Benchmark and Performance

Apple A14 Bionic Processor (SoC) Specs - Benchmark and Performance

Apple A14 Bionic Processor (SoC) Specs - Benchmark and Performance

Apple has just announced the a14 bionic system on a chip for the brand new iPad air and for the new iPhone 12 series yeah you knew there was a reason apple only teased us with the air for now but isn't actually letting us get our hands or our favorite geeky benches on it until next month still we already know quite a few things about the a14 and more than you might think and I'm gonna break all of that right now.

As much as we all may want the new iPhones and want them yesterday if not sooner it is kind of low-key fitting that this wretched was worse than a game of thrones season 8 of a year prompted apple to announce the a14 with the iPad air instead it's nostalgic even given a decade ago this year apple debuted their first-ever branded silicon with the a4 inside the original iPad a full six months before the iphone 4 got it as well and since then apple's vice president of silicon Johnny Srouji has become a senior vice president a series has gone from 32 to 64 bits from licensing arm and imagination designs to licensing architecture and making their own custom cores from secure enclaves and neural engines and literally from zero units to shipping over 2 billion Silicon chips not like barbecue or salt and vinegar or anything though that's probably how intel is feeling right about now. 

But I do think that's largely because apple isn't and doesn't have to act like a merchant silicon provider like intel or even Qualcomm and the a14 in both the new iPad air and upcoming iphone 12 is kind of the perfect example of that and I'll get into all the specifics in a, not at all thermally maxed minute but this part is important to understand apple just doesn't have profit and loss on selling chips they don't have to balance the competing demands of like multiple buyers building very different kinds of devices and they can work multiple years out with the other teams inside apple from the hardware engineering and software teams to industrial and even human interface design and sure marketing and that mean it's not just that they only have one customer they also only have one real job to run ios and its many variants and now mac os and every app on every one of those platforms just faster than anything else in the world. 

And sure some might push back on that and say well with all the new devices apple keeps rolling out doesn't that mean the silicon team effectively now has multiple customers just within apple like the iphone and the iPad, the apple watch, and the airpods and now soon the mac but they're i think apple's done something pretty candy not only have they been working years on scaling ios into everything from watch os to tv os and bringing key characteristics of ios to mac os they've also been working for just as many years maybe more to create a silicon architecture that they can scale from the iphone down into the apple watch and up into not just the ipad pro but now soon the mac as well and i mean the apple watch s6 system in package is getting the iphone 11 A13 system on a chip core architecture this year which is just silly it's silly especially given how non-existent real wearable silicon competition even is at this point and sure the new ipad air might not tickle the A14 ISP or image signal processor anywhere nearly as much as the iphone 12 will because it doesn't have the same camera system and because most people simply don't use cameras the same way on ipads as they do on iphones and equally sure given the ipad air will have a bigger thermal envelope than the iphone which will be better for the kinds of graphically intensive creative apps people tend to use more on the ipad than they do on the iphone things like 4k video editing so yeah absolutely maybe the ISP is a little op for the ipad or the maximum perf a little extra for the iphone and when these chips hit other products like the apple tv or get scaled down for the apple watch there'll be other trade-offs

But in general I think the efficiency the savings on time and talent by having by like having everything so tightly integrated and aligned it seems to really really pay off for apple because they're not doing what other chip companies are doing either not just ramping up power draw or throwing extra cores to brute force performance increases at the expense of efficiency they're increasing performance and efficiency pretty much at the same power and pretty much in the same thermal envelopes and that brings us to how so just like the A12 was the first seven nanometer chip to go into production the A14 is the first 5 nanometer chip in fact rumor has it apple has bought out all just all of taiwan's semiconductor manufacturing companies all of tsmc's five nanometer capacity at least for now like just nobody else is getting any of it not for a while and what that gives apple is a huge advantage when it comes to density see a system in a chip an soc inside an apple device like an ipad especially like an iphone is space constrained but by going from 7 nanometers down to 5 nanometers that lets apple get more transistors into that space up to one point eight times more or around a forty five percent area reduction based on tsmc's claims and yeah we'll have to wait for the inevitable teardowns and die shots to see exactly how apple's balancing everything out but we do already know that the A14 packs in 1.8 billion with a b transistors and that's up from the A13s 8.5 billion or you know just over 3 billion more transistors to spend on existing compute engines and new feature sets starting with the CPU back with the A10 fusion the iphone 7 apple introduced the idea of people get snappish when I call them big. little so I'll just say performance. efficiency cores basically as chips got better and better performance they used more and more power even for tasks that didn't require that much power so the idea was as you increased performance on one side you'd introduce something else to fill the efficiency gap on the other and that's what the 8 and10 did it had smaller zephyr cores that could handle normal tasks without being so power-hungry and bigger I mean larger hurricane cores that could handle more demanding tasks but at the cost of being more power-hungry as well but they were fused together they had to work together hence the fusion name as part of the a11 bionic in the phone's 8 and 10 and yes I said iPhones west wing apple kept the efficiency. performance architecture but ditched the fusion so the quote-unquote big monsoon cores and the little mistral cores could work separately or together as needed hence well not the bionic name just has to come from someone on the silicon team being an obviously huge steve Austin fan and not the stone-cold one the six million dollars one never mind Wikipedia it.

But they've managed to keep that bionic name going on for generations and years now so yeah talk about your architectural efficiencies when it comes to branding as well. The A14 Bionic has a 6 CPU core with 2 high-performance firestorm cores and 4 high-efficiency ice storm cores and yet apple's continuing their habit of making any given generations efficiency cores much more performant and performance cores much more efficient because those things are just inextricably linked apple said the a14 is 40 faster than the a12 and doing some quick back of the kino math the a13's lightning and thunder cores were 20 faster than the a12s vortex and tempest cores though I think Anandtech claimed apple was just underselling that a little bit.

Divide by zero carry the one that should make the a14 again 20 or so faster than the a13 which will, of course, no doubt make Moore's law is deadheaded among us just I roll so hard but hey apple's still kicking away at that more corpse as much as anyone in the industry just using that very wide architecture optimizing it for efficiency and keeping those cache sizes just extra generous while doing it more recently apple started making their own custom GPU as well with the a14 it's again a 4 core graphics engine which they say is 30 percent faster than the a12 and the a13 was 20 faster than the a12 so again ctek astronomy that math and we get 10 over the a13 which legit for sure isn't as much as the CPU gains but I also have a sneaky suspicion apple is leaning even more heavily if anything on the other elements of their custom silicon now.

but first GPU things first the a14 maintains apple's focus on efficiency there as well they want to be able to deliver most tasks most of the time at the lowest voltage in frequency but still be ready to ramp up to spike up even if and when you need it for that performance that lets them preserve as much battery life as possible while also providing really good sustained and peak performance and I think the whole approach is something a lot of people are looking no staring at awkwardly maybe but unabashedly to see how apple scales up GPU for max silicon especially pro and desktop max silicon and of course pro desktop max silicon.

So back in 2017 apple introduced their first apple neural engine as part of the a11 bionic it's sort of big showcase feature back then faced id the ability to scan facial geometry and just figure out if you were you even if the exact way you styled you varied every day even during the day it was something that I think showed how apple silicon interfaces with the rest of the company you know coordinating with the hardware team working on those true depth cameras and the software team working on those adversarial neural networks 2-3 years even before the feature ever hit the stage literally putting A.I into the chip at the silicon level well before people were accusing apple of missing out on the whole buzzword style A.I revolution completely but you know far be it for me to tell the pundits how to do their jobs and that original proto-neural engine in the A11 bionic could do 600 billion operations per second.

This new neural engine in the a14 bionic which seems to be just doubling as fast as all of our shelter-in-place sourdough starters is now up to 16 cores and 11 trillion operations per second and for those calculating along with me at home that's twice as fast as the 8 core apple neural engine found in both the A12 and the A13 but apple has also been optimizing the entire system on a chip the entire SoC for machine learning including the CPU and the GPU and starting last year introducing new custom accelerators called amx blocks for accelerated matrix multiplication something just frequently used in machine learning for these kinds of operations apple said it made the a13 six times faster than the a12. 

This year they're saying it makes the a14 10 times faster than the a12 so plus-minus whatever 4x faster yeah true facts all this math stuff is why my shelter in place sourdough starter just never started anyway again A14 bionic really just means the next generation of pretty much all of apple's silicon IP and yet they use the term IP to refer to cores and blocks in silicon as well and apple's philosophy their mo so to speak seems to be moving the whole chip forward all the IP forward every generation year over year to leave basically no corner untouched and a lot of that is informed by the silicon team staring at what kind of apps people are using both apple apps and app store apps what the os the operating system teams inside apple are planning.

For the next few years out and what sorts of apps and workloads seem to be coming onto the market or at least the ones they expect to become on the market the trends they're anticipating and that results in things like yes the beefy caches but also things like the pro video encode and decode blocks which the series chips use to handle h.265 video and they do it so well that apple actually redirects it away from the intel chips in current max and towards the t2 chips which are variants of the A10 with those blocks also things like the performance controller the secret sauce that figures out what to direct to the CPU to the GPU to the ane and to the other components to get the best performance at the highest efficiency for just any given task and that new ml controller which does the same thing between the apple neural engine the amx and all the main cores.

Also things like the custom storage controller which isn't just about having the solid state chips performing as fast as possible but making sure they're capturing every photo every frame of video right and properly saved which really doesn't sound like much i get it until you hear people with other even recent generation flagship phones complaining they're missing picks or just dropping frames and it's not as flashy i totally get that it's not as flashy as a lot of the more gimmicky things that we see in big launch demos these days but it's the kind of thing that helps real people avoid real problems and just have a better experience day after day same with things like making sure video encode and decode and video capture don't just perform but sustain because if you can't record that 4k60 extended interleaved dynamic range video for as long as you want to again without dropping so much as a single frame that feature just simply won't be as useful for you which is why when we get numbers like 20 more performance but also 40 greater efficiency that's not just the result of a process shrink or some big lever being pulled.

It's the result of touching all those corners of the silicon of working on all those ip of shaving off a tenth of a point here and a fraction of a point there of every tiny advance helping push the overall performance and efficiency forward like never mind death by a thousand cuts life by a thousand cuts to totally just mangle that cliche because again counter-intuitively efficiency is ultimately performance and i know you know i know that you know you know that i know every year someone looks at the numbers apple posts or gets sucked up by geekbench and how the iphone or the ipad air compares to or even beats this or that laptop even macbook and wonders out loud if we really really need this kind of performance on a phone or a mainstream tablet or conversely we get reviewers looking at brand new devices launching with older silicon maybe a year maybe more and saying it's fine everything's fine it does what it needs to do everything is snappy enough so everyone should just quit worrying and complaining about it and yeah both those things are when i start screaming because most people keep their phones for far longer than industry analysts and reviewers for like two or three or even more years and their tablets for much longer than that.

So it's less about how something scrolls or feels snappy on a brand new device the day it comes out and more how it will keep scrolling keep feeling snappy keep performing in general the next year the year after that the year after that when they've used it more and accumulated more craft when the operating system the os has been updated with new features features that make greater demands on the silicon and honestly that they can even keep getting those updates those new features and new security patches both apple i mean has said pretty damn directly that a lot of the overhead they build into the a series chips is not because of what they're doing now this year with like IOS 14 and ipad OS 14 and sure mac os big sur but because of what they're planning to do next year and the year after that for around the five years after that and what they envision apps will be doing and what we'll want to be doing with those apps also the updates for both features and security that we'll still need to be getting and that way we maintain as much of that as much of that day one experience as possible on day 1001 and there'll be bugs for sure and battery drains and all sorts of frustrations and failures along the way and they'll get fixed and they'll get broken again but the overhead in the silicon means we'll have the absolute longest possible runway to get the absolute highest value out of our iphones and ipads possible.

And that's true now with the a14 bionic at five nanometers and it'll be true with the a16 or whatever three nanometers which to be honest is a tiny size that kind of just melts my brain a bit like Pym particles like quantum realm like dr strange t trip scale brain meltage because reckoning with infinity is like something you need major math to deal with but you can start with brilliant and their new infinity course because of course, you can start with brilliant will take you from basic counting and work your way up to the many types of infinity culminating in the profundity of cantor's theorem exploring the stunning beauty of fractals where it's just turtles or I guess chipsets all the way down.

Post a Comment

Please do not enter any spam link in the comment box

Previous Post Next Post