I've never been clear on how much Intel couldn't make cooler chips, and how much they just can't admit to themselves that it matters (and thus didn't try very hard).
Having worked there under BK and Bobby it is very clear why. They didn't invest anything into engineering. Even today their comp is peanuts compared to any other company in the tech industry.
They pay TSMC loads of money to have the first chips for each process node. They were first to 5nm, they'll be first to 3nm next year too. This is because they sell whole products rather than chips like Intel and AMD, so Apple's profit margins are astronomical compared to them, and so they can afford to pay TSMC so much.
Also their products include iPhones and iPads. Intel nor AMD are putting their chips into 10s of millions of phones, especially phones people will pay top dollar for, per quarter.
They know it matters, they just don’t have the competence to do it since they promoted a bunch of toadies and charlatans into their technical leadership, and also outsourced a ton of technical work to “low cost geo” so managers can brag about cutting costs.
How will the gap widen? 8 instruction parallel decoder will give Apple single core performance per watt, but other than that I don't see what Apple does differently. M1 Pro 10 core is their best performance-per-watt, and Ryzen 6800U is just 6% behind [0].
Ryzen is only 6% behind for multi-core PPW, but The M1 appears to still have a huge advantage in single-core PPW.
I have a half dozen thoughts on this, but the foremost is;
Apple keeping memory on chip, is likely providing a huge memory latency advantage, (1) as well as some power benefits. It would not surprise me if this is a large part of (if not the majority) of the advantage here outside of the bigger decoder.
Think about a single threaded vs multithreaded benchmark; In the case of ST, there's only one thread the prefetcher can deal with, that one thread is going to be waiting for data. In the case of MT, there's a much greater likelyhood that you'll have multiple threads making memory requests, and the latency can be amortized by having the threads do other work (i.e. Thread 1 can start working on data it got back while thread 2's request is already in-flight from controller to DRAM.)
This is one of those moments I miss David Kanter (RealWorldTech, [0]) doing CPU arch breakdowns.
(1) - Back in the day, one of the 'better' things you could do for a SDRAM P3/Athlon/Duron system, especially if you were replacing all modules for an upgrade anyway, was to hunt for CL2 memory.
> The trace length has a negligible impact on latency.
Uhm, that is not true. If that was true, then your typical motherboard kept slots where they were 10 years ago, but they're getting closer to combat the increase in latency in other places. This is why HEDT motherboards have such layouts.
But just getting it close is not enough: M1 has higher memory latency than AMD and Intel SKUs.
Single threaded perf per watt has never been a benchmark anyone cares about. Perf per watt for the whole die is what counts in power-constrained scenarios. One core on even a 4-core Intel chip (the most power hungry beasts around) cannot hit die-level power or thermal limits.
They were never in trouble on the manufacturing part. In fact, they’ve been the best at it for 40 years.
I think it’s way easier to dig yourself out of a whole by picking up a previous architecture and updating it (ditching the Pentium 4 and using the 3 as the basis for the Core architecture) than it is by regaining manufacturing lead, especially when there’s only one company that’s able to do it these days. Lots of companies are able to make compelling architectures with different instruction sets. Actual chip making, however, is TSCM.
How Intel lost their lead is probably the greatest business case to be studied in our industry’s recent history.
Right now I would take AMD over Intel if the rumours are true. They are about to leap frog each other Intel (now) > AMD(zen4) > Intel(13th Gen) > AMD(zen4 3d ace up it’s sleeve).
Apple Silicon will have the title of world's best designed silicon for the foreseeable future. The only way for AMD and Intel to be competitive is if they can match the performance/watt, thin-ness, fan-less and battery life of the M2 MacBook Air. Forget having all of these things at once. AMD and Intel chips can't even function in fan-less enclosures!
And the operating system too, so they can fully leverage all the special features on the chip to improve UX.
Once you have critical mass being able to have enough experienced engineers up and down the whole stack, and you’ve figured out an organizational structure that allows them to work together efficiently, I don’t see how you can beat the vertical integration in a world where the fabs Apple can use are actually better than Intel’s own fab.
Correct me if I’m wrong, but a decade ago Intel would probably have said their fab/process node was their moat?
I know nothing about chip design, but like most things, I feel like one exceptionally brilliant tech lead surrounded by a group of say ~100 merely very smart collaborators can make a world class ARM based processor-design. A mere $100mm annually in hiring and overhead?
The rest of Apple’s advantage comes from being able to actually hire the best, and being their own customer at scale, which means being able to buy first place in line at the fab with billions of dollars of cash.
This is perhaps less true now that the chip has so many specialized areas on the die? Like, does the neural engine get allocated a certain mm^2 and certain number of bus lanes, and then a fully separate team of 100 designs it? I suspect the neural engine part of the chip is actually super simple to design, it’s the tight coupling with the OS and getting apps to properly leverage it which is tricky.
That’s not fair. Some people care about different things than you do. I think op’s description is more aligned with the broader market, but that’s just, like, my opinion.
Can we please have ONE. SINGLE. SILICON DISCUSSION ON HN. stay on the topic at hand without fanboys spamming "hurr durr, my M2 is da best, Intel suxxx, X86 is dead!!!111one", while not bringing anything useful or relevant to the discussion?
Seeing your comment at the top makes me not want to ever open any silicon topic here again as I'm sure it will be full of these kind of low effort comments vomiting marketing garble on how AS is the best and everything else is doomed to failure, while not bringing any useful info or arguments on-topic.
I think a suitable counter argument is that Apple has grown too large and needs to be split into multiple smaller companies to better aid competition.
Between what they're doing to silicon, the outrageous App Store behavior, and how they flaunt that they're a quazi-government entity, I think this could find broad support in Congress and the DOJ.
Nobody can compete with Apple, and that's a bad thing for everyone.
So Apple brought competitiveness to an industry that was badly needing it and stalling in the most recent years.. and your “solution” is that we should prevent them from doing so?
It's a strawman. Apple has the capacity to do immense good, but only because they're the single largest company in the world! We should scrutinize concentration of power heavily, and so far Apple has done nothing to suggest their benevolence to the rest of the market. They're blowing off Dutch regulators like it's a middle school homework assignment, and refusing to loose their asinine monopoly over software distribution on iPhone. They're behaving childishly, and everyone knows they're not a child. They're a company with hundreds of billions of dollars, and they're demonstrating organizational failure to address the demand of the market. On top of that, they're largest revenue sources are rent-collection and unibody aluminum computers made by political prisoners in Chinese concentration camps.
They made a chip that's vastly better than anything out there at a given power consumption level. They are not attempting to use this advantage to corner the silicon market; indeed, they are neither licensing the design, nor selling the chips outside their own end-user hardware.
Using their enormous lead on cellphones and their incredible negotiation power and playing that into mobile business computing and supply chain / process monopoly.
The same goes for Google. They have higher market share in many categories and cover many more of them. Near-monopoly on search, ads, online video, email, and at least half of the smartphone OS market...
No comment on the other stuff, but the “silicon” part of this argument boils down to “anyone who makes something way better than the competition must be shut down because that’s unfair”.
We’d still be in the Stone Age if everybody had that mentality
Apple successfully monopolized the 5nm node by buying out all of TSMC's manufacturing capability at that node size. Characterizing it as "anyone who makes something way better than the competition" is a strawman portrayal of the underlying issue at hand.
Apple Silicon is the most exiting thing to happen in the field in decades. Apple handles platform transitions really well so it may seem less of a tectonic shift than it actually is.
It’s normal for people to be enthusiastic amongst such facts.
Then why not be enthusiastic about it on AS threads. This thread is about Intel chiplet design, it ahs nothing to do with AS so why pollute every silicone thread with this offtopic.
Because Apple only sells Apple devices, not chips. AS parts can only be found in the Apple ecosystem and can only be bought exclusively within devices Apple sells, while the other part (X86 from Intel and AMD) can be bough by anyone, run almost any OS and SW to date, and be used in any design/application you can dream of from laptops, PCs, servers, consoles, mainframes, cloud datacenters, robots, planes, even inside Tesla's head unit, while AS is just in a few Apple devices that don't overlap with most for the segments I mentioned before.
One silicone part has an impact over the entire computing sphere, while the other is relevant only within the Apple ecosystem. So bringing up AS in all threads about generic computing chips that anyone can buy directly is pointless as Apple does not cater to that market.
That's like me bringing up how fast my Ferrari is in threads about utilitarian vans and pickup trucks. Sure, they're both cars, but they don't compete in the same segments, so Ferrari making an even faster car has no impact or threat on the market of vans and pickups.
People buy laptops, mostly. They don’t even buy individual chips, aside from niche markets. So comparing MacBooks to Lenovos and the likes seem pretty distant from the Ferrari vs trucks analogy.
I also seems undeniable that Apple's move to custom silicon shook up the entire market, challenged preconceptions. It was a pivotal moment for the industry as a whole, not Macs only.
The compute chip market is much, much bigger than just laptops, iPhones and MacBooks. Like I said, it' also servers, FPGAs, cloud, datacenters, consoles, GPUs, AI-ML, cars, vision, etc. and Apple's silicone is absent form most of those sectors while these are Nvidia's, Intel and AMD's bread and butter. So comparing a chip that can only be found in Apple laptops to chips from Intel and AMD that are sold and used in hundreds of other industries and thousands of different devices is completely pointless as they're not competing in the same sectors.
When Apple will sell its chips to consumers, PC OEMs, datacenters, console manufacturers, AI and car companies, with full documentation and Linux support, in order to compete on the same turf with Intel and AMD, then we can talk about an industry wide revolution. But until then, this "revolution" will not extend beyond the Apple/Mac ecosystem where Apple has no competitors, and who's market share in the general PC space is still tiny.
It would be exciting if you could add more memory to their machines, and maybe gpus, on systems that cost less than 6000 dollars (and even those haven't transitioned to apple's in house cpu architecture).
For now, Apple Silicon is just sorta a blip because most people buy them for coding environments and editing pictures, and apple decides to dedicate all of those performance per watts on having Apple Music boot up as fast as possible on boot.
For me, I get the popcorn out for all the ongoing GPU drama between AMD, Nvidia and now Intel. All of that seems to be the sequel to Pirates of Silicon Valley.
1st paragraph: SOC are not upgradable, that’s the whole point of the trade of and most of what makes this tradition exciting. What will Apple do with the Mac Pro is the most important question in this industry for the last couple of years.
2nd: what? Apple Music what?
3rd: what does that movie, which was about Apple and Microsoft, had to do with any of the current chip industry and its many players?
I keep wondering, did people always use the word ‘silicon’ when talking about CPUs & SoCs? Somehow I never noticed it before Apple released their ‘Apple Silicon’.
Just echoing what others are saying, no, we called them Intel chips/CPUs. What I don't get is why people go along with it. I personally prefer not being a miniature speakerphone for the marketing department at Apple.
This line of reasoning is just dumb. Every single chip on the mother board is made from Silicon. (There maybe some gallium or germanium parts but those are insignificant and irrelevant).
The original thread comment asked if this was a thing before "Apple Silicon", that's what I was referring to; I've seen it before, and not just years before but over a decade before, possibly even two decades.
As such it's not something Apple PR invented, but rather hijacked.
It used to be slang, though. Something you'd use to punch up the first paragraph of an article, not the way people actually talked. For some reason calling it "Apple Silicon" really grates on me, too. But such are the whims of megacorps.
LOL. Anyone who works in chip design would know much M1/M2 changed the hardware game. It is actually the dabbler enthusiast talking about how power/perf isn’t important because his LED-laden shitbox has a wall plug (muh absolute performance) that doesn’t grasp how utterly irrelevant DIY builders are in the market. Just look at the relative sales of servers, laptops and desktop and see what we care about.
Not a single second of thought is ever spent by the architects/designers on optimizing “absolute performance”. We only care about perf/area and perf/watt. It is the marketing teams that try to hype up gamer performance. Overclocking/high voltage performance requires the engineering knowledge of a freshman intern: go raise the voltage/freq, run the test program, make a SKU.
Source: worked on CPU/GPU arch/design for 20 years, including at Intel.
Intel fanboys whine about Apple/Intel CPU comparisons, here is their answer.
The takeaway from the story: connecting a bunch of shitty chiplets together makes a shitty SoC. Except this time, Intel paid TSMC a buttload of money to make their shitty design.
Silicon as a term for semiconductors and the things we make from them is common in my experience. But I've never seen such a concerted effort to pronounce it "silly kin" as Apple Silicon marketing.
Cosign, and thank you: any CPU discussion past "Apple ARM great" has been impossible to have since launch.
List of places I tried taking conversation over that time, but it was ignored or read as complaints about Apple. (see Disclaimers in footer if you read these and think 'Wow, he just wanted to talk about why Apple was bad')
- M1 was the first processor on a particular node; so there was a short term opportunity to do an apples to apples comparison by taking down M1 numbers and waiting for upcoming launches
- it wasn't as trivial as "manufacturing on the improved node" for AMD, but it was for Qualcomm
- performance of ARM vs. X86 could be teased out by tracking tuple of node x manufactor of chips and being patient; projections and tracking of Qualcomm & AMD chips performance
- the initial M1 was beaten by Tiger Lake in desktop & sustained performance cases, which was two(!) nodes behind
- performance and noise issues from Apple optimizing for absolute fan silence always, leading to them only kicking on at extremely high speeds far into the performance workload, that had already been throttled
== DISCLAIMERS ==
1. I am a happy M2 MBP owner and think its the best chip.
2. My more nuanced view, summarized is that it is almost restrained in that the hardware got bigger, somehow, and in software there are growing pains as drivers adopt from iPhone use case to mostly-plugged-in use case. To wit, throttling seems optimized for ad copy around fan noise at low workloads than the user.
3. If you feel these discussions were focused on denigrating Apple, please recommend curious thoughts to have about chips that aren't denigrating Apple
This is US site. Apple is US company. The biggest company in the world, source of so much pride. So tons and tons and tons of fanboyism here is inevitable. I'd say most of it is damn well earned.
That said, fanboyism makes people blind and uncritical, and (at least to me) its apparent company like Apple needs some good old criticism, rather than blind worship. Otherwise they will fall (if not fallen) into "we know whats best for you and you have no say in it" like with cough cough "child porn" filters or battery-gate.
I truly honestly don't trust their "we are more secure" marketing pitch, especially as non-US person.
At the end, its just another corporation driven by huge army of managers with main focus on salaries and bonuses. The idea that they are somehow morally better than everybody else when they keep hiring from companies like Facebook is pretty dangerous and goes back to beginning of my post.
Why are you so triggered that M2 is the best chip on the market? I'm pretty sure that is relevant in a discussion about the CPU market. AMD did a similar thing to Intel with the Ryzen launch. Intel is currently stagnating. They need a miracle at this point.
Why are you so excited? Did you design the M2? Do you manufacture the M2? Did you fund the M2? If so, feel free to be proud of it. You made a technological advance happen. But if you just walked into a store and bought one, I dunno, I think you're arriving pretty late in the evolution to take a personal interest in its success.
Not sure about others but I personally am not a big fan of generalising that much: I'd prefer it if people would ideally say that it was for example the "most power efficient general purpose CPU/SoC" or at least something in that regard, not just "the best".
For example, have a look at https://openbenchmarking.org/vs/Processor/AMD%20Ryzen%20Thre... (user benchmarks of M1 and Threadripper). Compiling Linux on the 2990WX appears to be about 4 times faster than on the M2. (There are lots of other examples of one of the two CPUs being faster than the other but compiling Linux is the most time-expensive task I regularly do on my 2990WX. The energy usage in this task on the 2990WX is almost certainly a lot higher of course; this will be true for most tasks. However, the 2990WX is also 4 years older of course, manufactured in a different node, not very optimized for power saving and not operated in a very power saving mode.)
There is no such thing as a "best chip on the market". Best chip for what? You're confusing the word SoC/CPU with "chip" which is a very generic word.
The best "chip" is the one that best suits your individual application or business needs, but there is no such thing as a best chip on the market. That's why Apple is only a tiny fraction of the computing market share and so many other chip vendors are still in business, because every application requires different chips.
M2 doesn't solve every needs neither as a CPU (since you can't buy it outside the Apple ecosystem), neither a a generic "chip". Why can't you accept that?
While I could agree with the general sentiment, I think it's hard to understate how much of a role Apple played in the background of all of this.
But in any case, there's plenty of things to be said about this article. About one year ago (random link with relevant quotes : https://www.pcgamer.com/intels-3d-chip-tech-is-perfect-so-it... ), Intel was mocking AMD for using a chiplet approach, before announcing today that it was - clickbaity title aside - going to change everything.
The sad truth is, both Intel and AMD are in the exact situation. AMD went chiplet in order to make their performance cores at TSMC, and their less critical cores ("IO") at GloFo.
In both cases, this is just a question of using a very limited resource (TSMC's best in class process) the more effectively that you can (by throwing extra engineering at making a chiplet design that works).
And it's supremely relevant to the discussion to talk about how Apple, by throwing capital at a company (TSMC) that was, for the couple of decades I used to cover this, at best 2 years behind the best in class, today where they are (far far in front).
We could definitely have a long discussion about the hubris that led 2015 Intel where they are today (completely stuck with a 7 year old "+ paint coatings" aging 14nm "performance" process), or how Gelsinger is trying to make the best out of the situation (I personally think he's immensely qualified and Intel's best hope, though that may not be enough to bring Intel back to where it was), but at the end of the day, Apple threw a wrench in what seemed like an unshakable performance lead from Intel by spewing a bit of money left and right (they didn't only bet on TSMC early on, they threw money at GF for example, and it wasn't that massive early on from my understanding), and the silicon world hasn't been the same since.
>a role Apple played in the background of all of this
All of what? This topic is about Intel chilpets, which many of those will end up in datacenters where most Intel chips go, and that's not where Apple sells chips for.
Not every chip made and sold in the world revolves around laptops, tablets, smartphones or the apple ecosystem.
So can we please talk about Intel's chiplets impact on the industry and less about Apple silicone which has nothing to do with this?
Maybe I wasn't clear or went too fast on some things, I'm not a native speaker.
This is all about semi manufacturing, and the position that TSMC now has in the fab space, thanks to Apple (yes, really, that's what I went on about in the previous comment). No part of my comment referred to arm, architectures, anything of the sort, just that Apple's money, applied broadly at first in the semi manufacturing space, then in a very very targeted way, took TSMC from tier 2 manufaturer to the best in class.
If you look back a few years, only x86 chips were having volume at the bleeding edge of manufacturing process. Gpus, Smartphone, everything else was one node back at the very least. Apple threw money and orders with a massive volume (iPhone + iPad is pretty close in units to the x86 cpu market, above 300M roughly off the top of my heard) at TSMC and that early + continous investment helped them fast forward their processes while Intel is still stalled in 2015.
Apple is using TSMC today (the best bits), AMD is using TSMC today (the second best bits) for the performance part of their chiplets and so will Intel tomorrow for the exact same reason. This is the relevant bit that I was pointing at.
That's a fair assessment although perhaps not giving AMD enough credit.
IMO GloFo's spin-off worked out very poorly for AMD in the short term, but long term it let them partially-leapfrog Intel much as they had done 20 years prior with the K6/K7.
There's two main things that IMO give the M1/M2 their 'magic';
- Dram on die (helping their PPW, especially single threaded PPW)
- Tight integration between OS and CPU.
This is, perhaps, where the x86 consortium has fallen into a challenge in the face of tight integration; The majority of that group would likely shriek at the idea of a DRAM on CPU, "here you go that's all you get" idea. I saw it a lot when I slung PC hardware; Folks who would insist on having as much upgrade-ability as possible, but never actually bought the upgrades between PC purchases. Even still, RAM is the main thing I personally find myself still upgrading on either purchased or older PCs.
That being said, It would be interesting to see if they try doing DRAM chiplets for these; I'm sure some 'ideal state' would be where DRAM chiplets + slotted RAM cause the chiplets to be dedicated to integrated GPU resources, or act as a form of L4 cache for one or more banks of DRAM.
> This is, perhaps, where the x86 consortium has fallen into a challenge in the face of tight integration; The majority of that group would likely shriek at the idea of a DRAM on CPU, "here you go that's all you get" idea. I saw it a lot when I slung PC hardware; Folks who would insist on having as much upgrade-ability as possible, but never actually bought the upgrades between PC purchases. Even still, RAM is the main thing I personally find myself still upgrading on either purchased or older PCs.
Memory people usually either buy as much as they need, or buy some and then add more. I've done it just about every PC build of mine, sometimes completely swap memory.
GPU generally gets upgraded, unless you're one of those who buy the best every year.
Storage for sure gets upgraded.
Currently sending this from my skylake i7 that seen 3 different GPUs, 2 different RAM kits, I lost count how many times I've upgraded storage.
I would probably get mad if I couldn't upgrade memory down the line. It maybe makes sense for laptops, but I don't see why do that on desktop. While M1 has memory right there, its latency is higher than intel and amd.
I'd save this judgment call for when there's node parity between Apple's chips, x86 and even other ARM manufacturers. The fact is that Apple's chips are on smaller nodes than the rest of the competition, because they bought up all of manufacturing on those nodes from TSMC. Performance per watt, power draw, thermals, etc are all functions of node size.
Apple Silicon is magnitudes more efficient. It cannot be explained by node size alone. TSMC 5nm is 15%-25% higher performance or 30% lower power compared to TSMC 7nm.
Compare AMD Ryzen 7 5800U [1] with Apple M2 8 Core [2] and you will see that Apple Silicon is not "magnitudes more efficient", but ca. 30% faster in single-thread, but at the same time 30% slower in multi-thread and with 25% higher TDP... This is not "magnitudes more efficient"...
Something about this gives me the under the skin creepy feel of piles of "locked" resources sitting on peoples desk going to complete waste.
Like when they used to sell mainframes with excess processor capacity then you would pay to unlock the processor that was already there if you need it. If not it was simply manufactured to sit unused in a mainframe its entire life, then be thrown in the trash.
I didn't specifically see anything that said this in the article but there is a TON to digest in there though mdular hardware always has that built to waste vibe. Even if they claim the opposite.
Chips always have to be binned. But previously a chip would have to be binned down to it's worst component I guess -- if they had a chip with great CPUs but the GPUs were a little wonky, and they didn't have an appropriate processor line for that combo, they'd have to bin the whole thing down to low-tier. Now they can instead match up the good CPUs and the good CPUs.
Plus they'll be able to satisfy some of their lust for SKUs by mixing and matching tiles, rather than making a bazillion slightly bins.
Core i9 isn't a useful thing to complain about SKUs for. There's probably been thousands of Pentium products at this point.
What you want to look at is how many SKUs for an architecture, like say Alder Lake for Desktop[1]. Do we really need 5 to 7 SKUs at 16, 12, 6, or 4 cores, but only 2 SKUs at 10 cores, and 4 SKUs at 2 cores?
2 cores seems like a niche product at this point, actually 4 SKUs for that is more than I'd expect.
Only 2 SKUs at 10 cores seems a little weird, I wonder if they are 12 core parts with some cores disabled or something like that.
Edit: Note it is just the i5 [...]K's from Q4 '21 that have 10 cores. It isn't that surprising that the early enthusiast parts are a little weird, right?
There are four Alder Lake dies in production, two for desktop (8+8 and 6+0 P/E-cores) and two for mobile (6+8 and 2+8 cores). So the 10-core desktop parts are probably the most cut-down variant of the full 8+8 die, and all the lower core count desktop parts are likely made from the 6-core die that omits the E-cores entirely.
Whether you look at just the desktop segment (29 SKUs from two dies) or the whole family (94 SKUs from four dies), it's an incredibly overcomplicated product stack.
> , I wonder if they are 12 core parts with some cores disabled or something like that.
If demand for that SKU is higher than for the 12 core SKU and margin justifies it, then yes. It happens with every vendor, specially when node been out for some time - yields are much better and there aren't this many down-binned chips to handle the demand.
My understanding is that binning is mostly about core count and core operating frequency.
Intel has a hard-on for segmenting their product lines. To the point they "launch" new products like the i9, which is just what an i7 used to be. They also deliberately cripple products, like selling SKUs with VT-x disabled. Not to mention them keeping ECC memory out of the entire desktop market for basically all of history.
I'm sure part of that is just marketing segmentation, but significant binning does take place. If VT-x is disabled, that may indicate that there were some defects in the location that those circuits inhabit.
Stuff like virtualization features and Hyperthreading are responsible for such a tiny fraction of extra die area and so deeply entwined with critical functionality that it's extremely unlikely for a given part to need to be binned on the basis of whether that feature works or not. Intel would never amass enough broken-HT but otherwise good chips to provide the impetus for an extra SKU.
>> Like when they used to sell mainframes with excess processor capacity then you would pay to unlock the processor that was already there if you need it.
Everyone still does this. The kvetching about artificial segmentation is a consumer phenomenon.
Enterprise actually likes it, because it allows zero-downtime upgrades if needed down the road - you don't have to stop your database for a half hour while the tech upgrades your CPU, you just drop in a new license file that says hey, they paid up, here's a key signature to enable the other cores! and suddenly you have a faster CPU, or more memory, or whatever. This is actually a highly-requested feature on AMD Epyc processors from enterprise customers as well, afaik, so, maybe coming soon.
(and it's really not a good thing even in the consumer space... like the "pay to turn on hyperthreading" processor that people flipped out about, that segmentation never went away, AMD and Intel still artificially disable cores for market segmentation, and sometimes still artificially disable SMT (eg 4700U). But now you aren't allowed to pay to turn them on... so if you change your mind and realize you need SMT, now you have to buy a whole new laptop. It's hugely wasteful and expensive. I bet a lot of people who bought a 6600K or 6700K wish they could have paid another $100 to turn on hyperthreading but nope, their only option was paying $350 on ebay for a used 7700K!)
Consumers don't like it cause they're jealous that there are parts of the chip they can't use yet.
Enterprise LOVES this cause they don't want to pay for parts of the chip they can't use yet, but also forsee wanting to use it in the future. Comes down to a CapEx vs OpEx optimization and having that ability to fine tune that balance is a No 1 requested feature.
I wouldn't call it jealous or mad. Just kinda something close to sad.
As I go though life I keep getting more aware of as silly as it sounds how many things are designed to strip "joy" from our lives. Even if you only wanted 5/10 of the cores or whatever in this thing and sure you paid less. There will always be that nagging feeling in the back of your head that there is a part of what you paid for being intentionally kept away from you. Effectively stripping your ability to have joy/happiness for your purchase. You don't deserve full use of your product because you didn't hustle/work/strive/stress/struggle hard enough to deserve it. Its just anti-human as all things are becoming now day.
meh ... If there is a 5 story apartment building and you rented one apartment that is only a fraction of one of the floors would you be "sad" that you don't get to use all the unrented/empty units in all 5 stories of it?
Of course if you want to rent 2 units, the landlord would be happy to give you the keys if you pay the rent for the second unit!
I think the people who get sad at using 5/10 cores (and paying less for it) are being unreasonable.
It's buffet syndrome, wanting it all and being too greedy.
Now if the manufacturer made you pay all the cost of 10 cores and only letting you use 5, that's a different issue.
It's not "going to complete waste" any more than if you download the MS Office suite and let the installer sit on your computer without installing it because you don't have a license key. Which... nobody cares.
You're licensing the processor capacity. You're not paying for the actual piece of silicon, you're paying your fair share of the R&D and fab investment that went into it. You want to use more, you pay more. The same as software.
The amount of silicon in the chip is, what, a small fraction of the amount of silicon in a single grain of sand? A whole processor is tens of grams of material total. The cardboard boxes a standalone chip comes in probably weigh more. I wouldn't get worried about "waste" here.
From everything I read online, the majority of costs are fixed -- it's building the fab, not producing each chip.
And obviously current chip shortages have nothing to do with it, nobody's including extra unlockable chips in a shortage for that chip.
If a manufacturer calculates it's more profitable to include unlockable chips, the very fact that it's more profitable means it's a more efficient allocation of resources, therefore not wasting anything. We're not running out of silicon generally speaking, and this isn't a case where there are big environmental externalities not being accounted for.
Back in the day, as another comment mentioned, the PPro had a 'chiplet' style configuration where The CPU and Cache were on the same chip but separate dies. The problem with this was the CPU and Cache had to be bonded to the chip first, then tested, and if either was bad, game over. Additionally, at the time die size was at more of a premium, in the case of a PPro, 256Kb of cache was close-ish to 2/3 the size of the CPU die. [0]
The P2, Katmai P3, and the Athlon 'Classic' (Pluto/Orion) used offboard cache; This was far better from a yield standpoint (I'm assuming the cache chips could either be tested before install, or were easier to rework) but limited their speed.
It's crazy to think that the Katmai P3 itself had around 9.5 Million Transistors, but the 512Kb of cache was another 25 Million on it's own!
As of this comment, there are 53 occurrences of "apple" and 4 of "chiplet" in the comments here (three from the same comment).
And to comment on topic, this mix of processes, each optimised to the task at hand and all on the same package sounds perfect for Intel. But I wonder how accessible it is for regular TSMC customers. We've had chiplet presentations, I don't recall anything like this being offered, despite requiring extremely high bandwidth for our application.
> But I wonder how accessible it is for regular TSMC customers.
TSMC is part of the UCIe (Universal Chiplet Interconnect) consortium, so I'd assume they have some capability. But the other members are ARM, Intel, AMD, Qualcomm, and Samsung... so I'm not sure if it's a matter of you have to be 'working in that club' or if TSMC can provide help on custom solutions.
AMD fabs at TSMC (and GloFo for the I/O die before, but this has moved to TSMC too recently) and has been using chiplets. TSMC does have 2.5 and 3D packaging for this. But it's still up to the fabless client of TSMC to design the chip. In other words, TSMC provides the enabling tools, but it's up to their client to use them. For now it's for the big players.
Indeed. And they've got designs using at least two different processes (one for compute, another for IO) but the article has examples using 4 or 5! I just can't see anyone else going as far as that.
Well, AMD has a few "3D V-cache" variants with an extra cache/memory die, so it's 3 chiplets types today. And these chips do not have an integrated GPU, which it would make sense to have on a separate die (what Intel announced).
I wouldn't be surprised that after a period of evolution and leap frogging it ends up in pretty similar spots. They have similar constraints and tools after all.
That's cool. A quick search and I can't see if there is anything special/different about the silicon for that chiplet but it would absolutely make sense that they tweak those wafers to be less leaky and/or denser.
I don't know either if the 3D cache tile uses a different optimized process from the CPU one.
But chiplets are actually first for cost optimization, and second for process optimization IMHO.
The first cost optimization is to leverage the better yield for a small die compared to a monolithic die. The density of defects for a given process means that N small chiplets will be cheaper than one monolithic die with same number of cores: one fault will kill the whole monolithic die where it will kill only one chiplet. A SoC only uses good chiplets ("known good dies" or KGD is the term of the art). That's what has driven AMD too chiplets in the first place.
The process optimization can also be for cost saving more than performance: if an older node is acceptable, it's cheaper.
Then there is the design saving, and this shows up in Intel chiplets presentation: by developing M CPU and N GPU chiplet variants for example, one develop M+N tiles but by mixing and matching can offer MxN SoC. One has to add the chiplets interconnection complexity (extra work vs a monolithic design), but this may still save on design development. And design development costs increase a lot with each new node.
Chiplets may also become a way to extend the life of a design. We may see at some point chips embedding a mix of "old" and "new" chiplets. This could extend the useful life of a chiplet design, and give more time to amortize the increasing design costs.
So I guess a good way to see chiplets is as a cost management tool first, in the face of ever increasing design costs on advanced nodes. With the nice side effect that they also open the possibility to optimize the process per tile, if needed.
With a chiplet standard and the possibility to treat chiplets as silicon IPs today, a chiplet target market may be extended too. For example Intel as a fab may sell its chiplets (or part of their catalog) to their fab customers. This is yet another way to absorb increasing development costs, but we're not there yet.
>But I wonder how accessible it is for regular TSMC customers.
And in case people are not familiar, Intel's version is called Foveros, so it isn't available to TSMC customer. TSMC has a few similar packaging tech on offer, but AFAIK they are not as advanced as Intel, although Intel's Faveros is also a little more costly.
It is interesting we basically cycle back to old computer model with north bridge, south bridge, CPU, GPU etc all linked together.
The article notes some advantages of being able to use different processes for different wafers, but there's not much more on that. It might be helpful if you wanted different fabs for memory and compute, or for flash type memory. Anyone know what they're getting at here?
There are situations when you really need that, but they mostly involve imagers. The Advanced Scientific Concepts flash LIDAR had a two-chip stack, with the light detectors made with InGaAs technology. The counters and timers were ordinary CMOS. This also shows up in some IR sensors.
Memories have their specialized processes indeed, but there are other reasons to specialize.
New nodes are fine for logic, but it takes time for analog IPs to move to new nodes (and some may not). So what AMD did, using an advanced node for compute/logic and a less advanced one for I/Os should be typical. You can also see this in broadband cellular modems, where the baseband part is on an advanced node and the RF on an older one.
When you look at processes offering, you often have variants optimized either for peak performance (frequency) or maximum efficiency. The peak performance would be the natural choice for (big) CPUs, and an efficiency node better suited for a GPU or any massively parallel accelerator where efficiency is more relevant than peak frequency (on this, I think Intel planned to use TSMC for their HPC GPU, could be related: they can focus on high perf for their CPUs).
> The article notes some advantages of being able to use different processes for different wafers, but there's not much more on that. It might be helpful if you wanted different fabs for memory and compute, or for flash type memory. Anyone know what they're getting at here?
Well, one advantage is that it can be a cost/efficiency savings on a few levels.
For example, in the case of Zen2/3, the main CPU chiplet is at 7nm, but the I/O die is at 12nm. If I had to guess, generally speaking it is useful in cases where parts of the final module would benefit from higher transistor density versus others; Memory controller tech generally gets fewer updates than the CPU/APU itself, so it allows faster design cycles; you already know the existing I/O controller works, one less portion to re-qualify.
My understanding is that one of the key benefits is that if you have an imperfection on a wafer, you lose one small chiplet instead of losing a big monolithic CPU. So for equivalent imperfections you get a better yield.
Different functions scale differently. With the latest processes, logic scales best, while analog scales worse: SRAM (which is internally pretty analog) still scales decently, but less than logic; and I/O scales very little at all (think of it this way: the size of transistors that drive outputs is pretty much determined by the current you need to drive, and the current is determined by e.g. the PCIe spec, which is itself subject to the physical constraints of relatively long wires).
As a consequence, if your design has CPU dies and I/O dies, using a smaller process only for the CPU dies is likely to be a good trade-off.
It is just cost reduction and speed of iteration in silicon. You can now mass manufacture a new CPU Core in a new Node, while your Media Engine, the Hardware Video Encode and Decode stays the same on an old node. Your Media Engine is the same across the entire product line up, so you get economy of scale with Fabs just printing it, you can also iterate on your Media Engine for new functions or errata individually without stopping progress on the larger SoC Design.
The same on I/O where there are speciality node much better suited for I/O, and their rate of progress is quite slow compared to other part of the system.
But most of these are only good for Desktop and Server. And the world is now largely Smartphone, or Laptop Energy efficiency chips. Where integrated SoC still provides the best results in efficiency.
Has Intel ever released a benchmark or marketing comparison that wasn't the filtering equivalent of spelling and grammar mistakes in spam emails? If you notice it for what it is, you're not their target demographic.
Their whole sour grapes culture is just bizarre to me.
Also, let's be honest. First-gen Epyc was glued-together nonsense. It was very much NUMA, it performed weird, the cause was the weird chiplet design (no IO die at that time). It was fair to criticize it on that basis. It had far less cache and much higher latency than the designs that followed, and it felt far more NUMA as a result.
Hyperscalers agreed and turned their noses up at it after some test installations, and decided to wait for Rome - which was much much better in those areas. But Naples was bad and that shouldn’t be sugarcoated.
I don’t really get why people are always so miffed by Intel saying that, it was a criticism that had been leveled against them, and they were right in applying that criticism to AMD, it was a quirky server design glued together out of consumer dies, it suffered all the same downsides as Intel’s own “glued-together” designs.
Now their CPU lifespan will be same as modern videocards - 3-5 years, same as new ryzen chips. This will surely drive new products adoption and profits.
Are you implying that the silicon somehow has a shorter lifetime due to the presence of an interposer layer? Or is this a way of saying you think cpus are going to become more powerful at a faster rate?
Before ryzen 3000 chiplets, CPU were everlasting, with failure rate of essentially 0 percent. AMD Ryzen 3000 chiplet failure rate was at least 5% first year and they keep degrading at astonishing speed.
Like 90% of mid to high end GPUs from AMD and nvidia produced before 2016 already dead, soon you will have to thoughtfully test CPUs to understand if its condition is still acceptable.
You can buy 15 years old cpu (which is 4 core 64 bit btw) and expect it to work like new.
My understanding of the situation is that Xilinx is the real leader in multichip modules. My impression is that Intel has finally entered the race, years behind AMD, which just bought Xilinx. Probably can't even see 'em in the rear-view. But, ya know, good job Intel, you're doing great.
Altera was also an innovator in that regard. It makes sense for Intel to buy them given that they were struggling with 3D packaging technologies before the acquisition.
However, they appear to have managed Altera into the ground, and time will tell if AMD makes the same mistakes with Xilinx.
Intel is way behind, but at least they know it and are doing something about it. They just need to ship more 10nm processors to float them for a while so they can develop chiplets.
167 comments
[ 3.1 ms ] story [ 220 ms ] threadVery cool
Had a chat with an old timer whl was doing horizontally scaling compute with thr IBM A400.
25 years later it's the starry eyed wonder of the 2010s.
Apple will be ahead, but the gap will not widen.
[0] https://www.notebookcheck.net/AMD-Ryzen-7-6800U-Efficiency-R...
Not an apple zealot by any means, but I think you shouldn’t be so quick to dismiss.
For my own part I’m not sure how to measure performance per watt given all the functional units on die.
Ryzen is only 6% behind for multi-core PPW, but The M1 appears to still have a huge advantage in single-core PPW.
I have a half dozen thoughts on this, but the foremost is;
Apple keeping memory on chip, is likely providing a huge memory latency advantage, (1) as well as some power benefits. It would not surprise me if this is a large part of (if not the majority) of the advantage here outside of the bigger decoder.
Think about a single threaded vs multithreaded benchmark; In the case of ST, there's only one thread the prefetcher can deal with, that one thread is going to be waiting for data. In the case of MT, there's a much greater likelyhood that you'll have multiple threads making memory requests, and the latency can be amortized by having the threads do other work (i.e. Thread 1 can start working on data it got back while thread 2's request is already in-flight from controller to DRAM.)
This is one of those moments I miss David Kanter (RealWorldTech, [0]) doing CPU arch breakdowns.
(1) - Back in the day, one of the 'better' things you could do for a SDRAM P3/Athlon/Duron system, especially if you were replacing all modules for an upgrade anyway, was to hunt for CL2 memory.
[0] - https://www.realworldtech.com/cpu/
How? It's the same LPDDR5 everyone else is using and it's on package not on "chip". The trace length has a negligible impact on latency.
Uhm, that is not true. If that was true, then your typical motherboard kept slots where they were 10 years ago, but they're getting closer to combat the increase in latency in other places. This is why HEDT motherboards have such layouts.
But just getting it close is not enough: M1 has higher memory latency than AMD and Intel SKUs.
They dug themselves out with the Core 2 Duo.
There remains a nonzero chance Intel digs themselves out of the hole again.
I'm strongly considering their stock.
I think it’s way easier to dig yourself out of a whole by picking up a previous architecture and updating it (ditching the Pentium 4 and using the 3 as the basis for the Core architecture) than it is by regaining manufacturing lead, especially when there’s only one company that’s able to do it these days. Lots of companies are able to make compelling architectures with different instruction sets. Actual chip making, however, is TSCM.
How Intel lost their lead is probably the greatest business case to be studied in our industry’s recent history.
Intel and AMD chips go to OEMs
Once you have critical mass being able to have enough experienced engineers up and down the whole stack, and you’ve figured out an organizational structure that allows them to work together efficiently, I don’t see how you can beat the vertical integration in a world where the fabs Apple can use are actually better than Intel’s own fab.
I know nothing about chip design, but like most things, I feel like one exceptionally brilliant tech lead surrounded by a group of say ~100 merely very smart collaborators can make a world class ARM based processor-design. A mere $100mm annually in hiring and overhead?
The rest of Apple’s advantage comes from being able to actually hire the best, and being their own customer at scale, which means being able to buy first place in line at the fab with billions of dollars of cash.
This is perhaps less true now that the chip has so many specialized areas on the die? Like, does the neural engine get allocated a certain mm^2 and certain number of bus lanes, and then a fully separate team of 100 designs it? I suspect the neural engine part of the chip is actually super simple to design, it’s the tight coupling with the OS and getting apps to properly leverage it which is tricky.
I care about openness and performance (watts are irrelevant if reasonable like they are now).
Your list, to me, sounds like marketing. Move the goal posts to these arbitrary points, declare victory.
Seeing your comment at the top makes me not want to ever open any silicon topic here again as I'm sure it will be full of these kind of low effort comments vomiting marketing garble on how AS is the best and everything else is doomed to failure, while not bringing any useful info or arguments on-topic.
Between what they're doing to silicon, the outrageous App Store behavior, and how they flaunt that they're a quazi-government entity, I think this could find broad support in Congress and the DOJ.
Nobody can compete with Apple, and that's a bad thing for everyone.
They need a slap, hard.
They made a chip that's vastly better than anything out there at a given power consumption level. They are not attempting to use this advantage to corner the silicon market; indeed, they are neither licensing the design, nor selling the chips outside their own end-user hardware.
How does any of that say "antitrust" to you?
It would be a healthier ecosystem for startups and competitors and make for faster total sector growth.
We’d still be in the Stone Age if everybody had that mentality
Apple Silicon is the most exiting thing to happen in the field in decades. Apple handles platform transitions really well so it may seem less of a tectonic shift than it actually is.
It’s normal for people to be enthusiastic amongst such facts.
One silicone part has an impact over the entire computing sphere, while the other is relevant only within the Apple ecosystem. So bringing up AS in all threads about generic computing chips that anyone can buy directly is pointless as Apple does not cater to that market.
That's like me bringing up how fast my Ferrari is in threads about utilitarian vans and pickup trucks. Sure, they're both cars, but they don't compete in the same segments, so Ferrari making an even faster car has no impact or threat on the market of vans and pickups.
I also seems undeniable that Apple's move to custom silicon shook up the entire market, challenged preconceptions. It was a pivotal moment for the industry as a whole, not Macs only.
When Apple will sell its chips to consumers, PC OEMs, datacenters, console manufacturers, AI and car companies, with full documentation and Linux support, in order to compete on the same turf with Intel and AMD, then we can talk about an industry wide revolution. But until then, this "revolution" will not extend beyond the Apple/Mac ecosystem where Apple has no competitors, and who's market share in the general PC space is still tiny.
For now, Apple Silicon is just sorta a blip because most people buy them for coding environments and editing pictures, and apple decides to dedicate all of those performance per watts on having Apple Music boot up as fast as possible on boot.
For me, I get the popcorn out for all the ongoing GPU drama between AMD, Nvidia and now Intel. All of that seems to be the sequel to Pirates of Silicon Valley.
1st paragraph: SOC are not upgradable, that’s the whole point of the trade of and most of what makes this tradition exciting. What will Apple do with the Mac Pro is the most important question in this industry for the last couple of years.
2nd: what? Apple Music what?
3rd: what does that movie, which was about Apple and Microsoft, had to do with any of the current chip industry and its many players?
Which has very little to do with Apple PR, but everything with how CPUs/GPUs are overwhelmingly made from silicon.
[1] https://trends.google.com/trends/explore?date=today%205-y&ge...
As such it's not something Apple PR invented, but rather hijacked.
Not a single second of thought is ever spent by the architects/designers on optimizing “absolute performance”. We only care about perf/area and perf/watt. It is the marketing teams that try to hype up gamer performance. Overclocking/high voltage performance requires the engineering knowledge of a freshman intern: go raise the voltage/freq, run the test program, make a SKU.
Source: worked on CPU/GPU arch/design for 20 years, including at Intel.
The takeaway from the story: connecting a bunch of shitty chiplets together makes a shitty SoC. Except this time, Intel paid TSMC a buttload of money to make their shitty design.
EDIT: F this downvoting, people seem to not get the obvious fruit pun.
List of places I tried taking conversation over that time, but it was ignored or read as complaints about Apple. (see Disclaimers in footer if you read these and think 'Wow, he just wanted to talk about why Apple was bad')
- M1 was the first processor on a particular node; so there was a short term opportunity to do an apples to apples comparison by taking down M1 numbers and waiting for upcoming launches
- it wasn't as trivial as "manufacturing on the improved node" for AMD, but it was for Qualcomm
- performance of ARM vs. X86 could be teased out by tracking tuple of node x manufactor of chips and being patient; projections and tracking of Qualcomm & AMD chips performance
- the initial M1 was beaten by Tiger Lake in desktop & sustained performance cases, which was two(!) nodes behind
- performance and noise issues from Apple optimizing for absolute fan silence always, leading to them only kicking on at extremely high speeds far into the performance workload, that had already been throttled
== DISCLAIMERS ==
1. I am a happy M2 MBP owner and think its the best chip.
2. My more nuanced view, summarized is that it is almost restrained in that the hardware got bigger, somehow, and in software there are growing pains as drivers adopt from iPhone use case to mostly-plugged-in use case. To wit, throttling seems optimized for ad copy around fan noise at low workloads than the user.
3. If you feel these discussions were focused on denigrating Apple, please recommend curious thoughts to have about chips that aren't denigrating Apple
That said, fanboyism makes people blind and uncritical, and (at least to me) its apparent company like Apple needs some good old criticism, rather than blind worship. Otherwise they will fall (if not fallen) into "we know whats best for you and you have no say in it" like with cough cough "child porn" filters or battery-gate.
I truly honestly don't trust their "we are more secure" marketing pitch, especially as non-US person.
At the end, its just another corporation driven by huge army of managers with main focus on salaries and bonuses. The idea that they are somehow morally better than everybody else when they keep hiring from companies like Facebook is pretty dangerous and goes back to beginning of my post.
Respect is earned, while fanboyism is rarely a good thing as by definition it's something rather biased.
For example, have a look at https://openbenchmarking.org/vs/Processor/AMD%20Ryzen%20Thre... (user benchmarks of M1 and Threadripper). Compiling Linux on the 2990WX appears to be about 4 times faster than on the M2. (There are lots of other examples of one of the two CPUs being faster than the other but compiling Linux is the most time-expensive task I regularly do on my 2990WX. The energy usage in this task on the 2990WX is almost certainly a lot higher of course; this will be true for most tasks. However, the 2990WX is also 4 years older of course, manufactured in a different node, not very optimized for power saving and not operated in a very power saving mode.)
There is no such thing as a "best chip on the market". Best chip for what? You're confusing the word SoC/CPU with "chip" which is a very generic word.
The best "chip" is the one that best suits your individual application or business needs, but there is no such thing as a best chip on the market. That's why Apple is only a tiny fraction of the computing market share and so many other chip vendors are still in business, because every application requires different chips.
M2 doesn't solve every needs neither as a CPU (since you can't buy it outside the Apple ecosystem), neither a a generic "chip". Why can't you accept that?
But in any case, there's plenty of things to be said about this article. About one year ago (random link with relevant quotes : https://www.pcgamer.com/intels-3d-chip-tech-is-perfect-so-it... ), Intel was mocking AMD for using a chiplet approach, before announcing today that it was - clickbaity title aside - going to change everything.
The sad truth is, both Intel and AMD are in the exact situation. AMD went chiplet in order to make their performance cores at TSMC, and their less critical cores ("IO") at GloFo.
And Intel will be doing the same thing tomorrow (again, random link on the topic: https://www.tomshardware.com/news/intel-ceo-visits-tsmc-agai... ) by producing their performance cores at TSMC and their less critical ones on their own processes.
In both cases, this is just a question of using a very limited resource (TSMC's best in class process) the more effectively that you can (by throwing extra engineering at making a chiplet design that works).
And it's supremely relevant to the discussion to talk about how Apple, by throwing capital at a company (TSMC) that was, for the couple of decades I used to cover this, at best 2 years behind the best in class, today where they are (far far in front).
We could definitely have a long discussion about the hubris that led 2015 Intel where they are today (completely stuck with a 7 year old "+ paint coatings" aging 14nm "performance" process), or how Gelsinger is trying to make the best out of the situation (I personally think he's immensely qualified and Intel's best hope, though that may not be enough to bring Intel back to where it was), but at the end of the day, Apple threw a wrench in what seemed like an unshakable performance lead from Intel by spewing a bit of money left and right (they didn't only bet on TSMC early on, they threw money at GF for example, and it wasn't that massive early on from my understanding), and the silicon world hasn't been the same since.
All of what? This topic is about Intel chilpets, which many of those will end up in datacenters where most Intel chips go, and that's not where Apple sells chips for.
Not every chip made and sold in the world revolves around laptops, tablets, smartphones or the apple ecosystem.
So can we please talk about Intel's chiplets impact on the industry and less about Apple silicone which has nothing to do with this?
This is all about semi manufacturing, and the position that TSMC now has in the fab space, thanks to Apple (yes, really, that's what I went on about in the previous comment). No part of my comment referred to arm, architectures, anything of the sort, just that Apple's money, applied broadly at first in the semi manufacturing space, then in a very very targeted way, took TSMC from tier 2 manufaturer to the best in class.
If you look back a few years, only x86 chips were having volume at the bleeding edge of manufacturing process. Gpus, Smartphone, everything else was one node back at the very least. Apple threw money and orders with a massive volume (iPhone + iPad is pretty close in units to the x86 cpu market, above 300M roughly off the top of my heard) at TSMC and that early + continous investment helped them fast forward their processes while Intel is still stalled in 2015.
Apple is using TSMC today (the best bits), AMD is using TSMC today (the second best bits) for the performance part of their chiplets and so will Intel tomorrow for the exact same reason. This is the relevant bit that I was pointing at.
IMO GloFo's spin-off worked out very poorly for AMD in the short term, but long term it let them partially-leapfrog Intel much as they had done 20 years prior with the K6/K7.
There's two main things that IMO give the M1/M2 their 'magic';
- Dram on die (helping their PPW, especially single threaded PPW) - Tight integration between OS and CPU.
This is, perhaps, where the x86 consortium has fallen into a challenge in the face of tight integration; The majority of that group would likely shriek at the idea of a DRAM on CPU, "here you go that's all you get" idea. I saw it a lot when I slung PC hardware; Folks who would insist on having as much upgrade-ability as possible, but never actually bought the upgrades between PC purchases. Even still, RAM is the main thing I personally find myself still upgrading on either purchased or older PCs.
That being said, It would be interesting to see if they try doing DRAM chiplets for these; I'm sure some 'ideal state' would be where DRAM chiplets + slotted RAM cause the chiplets to be dedicated to integrated GPU resources, or act as a form of L4 cache for one or more banks of DRAM.
Memory people usually either buy as much as they need, or buy some and then add more. I've done it just about every PC build of mine, sometimes completely swap memory.
GPU generally gets upgraded, unless you're one of those who buy the best every year.
Storage for sure gets upgraded.
Currently sending this from my skylake i7 that seen 3 different GPUs, 2 different RAM kits, I lost count how many times I've upgraded storage.
I would probably get mad if I couldn't upgrade memory down the line. It maybe makes sense for laptops, but I don't see why do that on desktop. While M1 has memory right there, its latency is higher than intel and amd.
Also, Apple's 7nm chips outperformed AMD/Intel 7nm chips.
[1] https://www.cpubenchmark.net/cpu.php?cpu=AMD+Ryzen+7+5800U&i...
[2] https://www.cpubenchmark.net/cpu.php?cpu=Apple+M2+8+Core+350...
This Reddit post summarizes the efficiency and speed advantages of the M1: https://www.reddit.com/r/hardware/comments/nii37s/comment/gz...
Like when they used to sell mainframes with excess processor capacity then you would pay to unlock the processor that was already there if you need it. If not it was simply manufactured to sit unused in a mainframe its entire life, then be thrown in the trash.
I didn't specifically see anything that said this in the article but there is a TON to digest in there though mdular hardware always has that built to waste vibe. Even if they claim the opposite.
Chips always have to be binned. But previously a chip would have to be binned down to it's worst component I guess -- if they had a chip with great CPUs but the GPUs were a little wonky, and they didn't have an appropriate processor line for that combo, they'd have to bin the whole thing down to low-tier. Now they can instead match up the good CPUs and the good CPUs.
Plus they'll be able to satisfy some of their lust for SKUs by mixing and matching tiles, rather than making a bazillion slightly bins.
That's so Intel. All 37 variants of the Intel Core i9 CPU: [1]
[1] https://www.intel.com/content/www/us/en/products/details/pro...
What you want to look at is how many SKUs for an architecture, like say Alder Lake for Desktop[1]. Do we really need 5 to 7 SKUs at 16, 12, 6, or 4 cores, but only 2 SKUs at 10 cores, and 4 SKUs at 2 cores?
[1] https://ark.intel.com/content/www/us/en/ark/products/codenam...
Only 2 SKUs at 10 cores seems a little weird, I wonder if they are 12 core parts with some cores disabled or something like that.
Edit: Note it is just the i5 [...]K's from Q4 '21 that have 10 cores. It isn't that surprising that the early enthusiast parts are a little weird, right?
Whether you look at just the desktop segment (29 SKUs from two dies) or the whole family (94 SKUs from four dies), it's an incredibly overcomplicated product stack.
If demand for that SKU is higher than for the 12 core SKU and margin justifies it, then yes. It happens with every vendor, specially when node been out for some time - yields are much better and there aren't this many down-binned chips to handle the demand.
Intel has a hard-on for segmenting their product lines. To the point they "launch" new products like the i9, which is just what an i7 used to be. They also deliberately cripple products, like selling SKUs with VT-x disabled. Not to mention them keeping ECC memory out of the entire desktop market for basically all of history.
Ha, that has been a long time, no? I remember buying a computer with a Q8300 in 2008 or so and there I had to assure a specific revision.
IBM still does this.
Enterprise actually likes it, because it allows zero-downtime upgrades if needed down the road - you don't have to stop your database for a half hour while the tech upgrades your CPU, you just drop in a new license file that says hey, they paid up, here's a key signature to enable the other cores! and suddenly you have a faster CPU, or more memory, or whatever. This is actually a highly-requested feature on AMD Epyc processors from enterprise customers as well, afaik, so, maybe coming soon.
(and it's really not a good thing even in the consumer space... like the "pay to turn on hyperthreading" processor that people flipped out about, that segmentation never went away, AMD and Intel still artificially disable cores for market segmentation, and sometimes still artificially disable SMT (eg 4700U). But now you aren't allowed to pay to turn them on... so if you change your mind and realize you need SMT, now you have to buy a whole new laptop. It's hugely wasteful and expensive. I bet a lot of people who bought a 6600K or 6700K wish they could have paid another $100 to turn on hyperthreading but nope, their only option was paying $350 on ebay for a used 7700K!)
Cores+™ - experience the performance of multiple cores
GHz+™ - awesome speed for the latest games and office spreadsheets
Memory GiB+™ - make room for multiple applications
Note that monthly subscriptions to this Intel Premium Experiences™ will continue regardless of whether re-verification is successfully carried out.
Consumers don't like it cause they're jealous that there are parts of the chip they can't use yet.
Enterprise LOVES this cause they don't want to pay for parts of the chip they can't use yet, but also forsee wanting to use it in the future. Comes down to a CapEx vs OpEx optimization and having that ability to fine tune that balance is a No 1 requested feature.
As I go though life I keep getting more aware of as silly as it sounds how many things are designed to strip "joy" from our lives. Even if you only wanted 5/10 of the cores or whatever in this thing and sure you paid less. There will always be that nagging feeling in the back of your head that there is a part of what you paid for being intentionally kept away from you. Effectively stripping your ability to have joy/happiness for your purchase. You don't deserve full use of your product because you didn't hustle/work/strive/stress/struggle hard enough to deserve it. Its just anti-human as all things are becoming now day.
Of course if you want to rent 2 units, the landlord would be happy to give you the keys if you pay the rent for the second unit!
I think the people who get sad at using 5/10 cores (and paying less for it) are being unreasonable.
It's buffet syndrome, wanting it all and being too greedy.
Now if the manufacturer made you pay all the cost of 10 cores and only letting you use 5, that's a different issue.
People seem fine with software unlocks for software but often confuse the price of hardware with the cost to manufacture the same.
You're licensing the processor capacity. You're not paying for the actual piece of silicon, you're paying your fair share of the R&D and fab investment that went into it. You want to use more, you pay more. The same as software.
The amount of silicon in the chip is, what, a small fraction of the amount of silicon in a single grain of sand? A whole processor is tens of grams of material total. The cardboard boxes a standalone chip comes in probably weigh more. I wouldn't get worried about "waste" here.
And obviously current chip shortages have nothing to do with it, nobody's including extra unlockable chips in a shortage for that chip.
If a manufacturer calculates it's more profitable to include unlockable chips, the very fact that it's more profitable means it's a more efficient allocation of resources, therefore not wasting anything. We're not running out of silicon generally speaking, and this isn't a case where there are big environmental externalities not being accounted for.
Back in the day, as another comment mentioned, the PPro had a 'chiplet' style configuration where The CPU and Cache were on the same chip but separate dies. The problem with this was the CPU and Cache had to be bonded to the chip first, then tested, and if either was bad, game over. Additionally, at the time die size was at more of a premium, in the case of a PPro, 256Kb of cache was close-ish to 2/3 the size of the CPU die. [0]
The P2, Katmai P3, and the Athlon 'Classic' (Pluto/Orion) used offboard cache; This was far better from a yield standpoint (I'm assuming the cache chips could either be tested before install, or were easier to rework) but limited their speed.
It's crazy to think that the Katmai P3 itself had around 9.5 Million Transistors, but the 512Kb of cache was another 25 Million on it's own!
[0] - https://en.wikipedia.org/wiki/Pentium_Pro#/media/File:Pentiu...
I remember when you had to buy an extra processor to get floating point.
* https://en.wikipedia.org/wiki/X87
At one point there were video game(s) with 'extra' functionality that was only available with this hardware 'upgrade':
* https://en.wikipedia.org/wiki/Falcon_3.0
(Get off my lawn.)
And to comment on topic, this mix of processes, each optimised to the task at hand and all on the same package sounds perfect for Intel. But I wonder how accessible it is for regular TSMC customers. We've had chiplet presentations, I don't recall anything like this being offered, despite requiring extremely high bandwidth for our application.
TSMC is part of the UCIe (Universal Chiplet Interconnect) consortium, so I'd assume they have some capability. But the other members are ARM, Intel, AMD, Qualcomm, and Samsung... so I'm not sure if it's a matter of you have to be 'working in that club' or if TSMC can provide help on custom solutions.
I wouldn't be surprised that after a period of evolution and leap frogging it ends up in pretty similar spots. They have similar constraints and tools after all.
But chiplets are actually first for cost optimization, and second for process optimization IMHO.
The first cost optimization is to leverage the better yield for a small die compared to a monolithic die. The density of defects for a given process means that N small chiplets will be cheaper than one monolithic die with same number of cores: one fault will kill the whole monolithic die where it will kill only one chiplet. A SoC only uses good chiplets ("known good dies" or KGD is the term of the art). That's what has driven AMD too chiplets in the first place.
The process optimization can also be for cost saving more than performance: if an older node is acceptable, it's cheaper.
Then there is the design saving, and this shows up in Intel chiplets presentation: by developing M CPU and N GPU chiplet variants for example, one develop M+N tiles but by mixing and matching can offer MxN SoC. One has to add the chiplets interconnection complexity (extra work vs a monolithic design), but this may still save on design development. And design development costs increase a lot with each new node.
Chiplets may also become a way to extend the life of a design. We may see at some point chips embedding a mix of "old" and "new" chiplets. This could extend the useful life of a chiplet design, and give more time to amortize the increasing design costs.
So I guess a good way to see chiplets is as a cost management tool first, in the face of ever increasing design costs on advanced nodes. With the nice side effect that they also open the possibility to optimize the process per tile, if needed.
With a chiplet standard and the possibility to treat chiplets as silicon IPs today, a chiplet target market may be extended too. For example Intel as a fab may sell its chiplets (or part of their catalog) to their fab customers. This is yet another way to absorb increasing development costs, but we're not there yet.
And in case people are not familiar, Intel's version is called Foveros, so it isn't available to TSMC customer. TSMC has a few similar packaging tech on offer, but AFAIK they are not as advanced as Intel, although Intel's Faveros is also a little more costly.
It is interesting we basically cycle back to old computer model with north bridge, south bridge, CPU, GPU etc all linked together.
There are situations when you really need that, but they mostly involve imagers. The Advanced Scientific Concepts flash LIDAR had a two-chip stack, with the light detectors made with InGaAs technology. The counters and timers were ordinary CMOS. This also shows up in some IR sensors.
New nodes are fine for logic, but it takes time for analog IPs to move to new nodes (and some may not). So what AMD did, using an advanced node for compute/logic and a less advanced one for I/Os should be typical. You can also see this in broadband cellular modems, where the baseband part is on an advanced node and the RF on an older one.
When you look at processes offering, you often have variants optimized either for peak performance (frequency) or maximum efficiency. The peak performance would be the natural choice for (big) CPUs, and an efficiency node better suited for a GPU or any massively parallel accelerator where efficiency is more relevant than peak frequency (on this, I think Intel planned to use TSMC for their HPC GPU, could be related: they can focus on high perf for their CPUs).
Well, one advantage is that it can be a cost/efficiency savings on a few levels.
For example, in the case of Zen2/3, the main CPU chiplet is at 7nm, but the I/O die is at 12nm. If I had to guess, generally speaking it is useful in cases where parts of the final module would benefit from higher transistor density versus others; Memory controller tech generally gets fewer updates than the CPU/APU itself, so it allows faster design cycles; you already know the existing I/O controller works, one less portion to re-qualify.
As a consequence, if your design has CPU dies and I/O dies, using a smaller process only for the CPU dies is likely to be a good trade-off.
The same on I/O where there are speciality node much better suited for I/O, and their rate of progress is quite slow compared to other part of the system.
But most of these are only good for Desktop and Server. And the world is now largely Smartphone, or Laptop Energy efficiency chips. Where integrated SoC still provides the best results in efficiency.
Their whole sour grapes culture is just bizarre to me.
https://www.anandtech.com/show/1656/2
https://pcper.com/2006/11/intel-core-2-extreme-qx6700-proces...
Also, let's be honest. First-gen Epyc was glued-together nonsense. It was very much NUMA, it performed weird, the cause was the weird chiplet design (no IO die at that time). It was fair to criticize it on that basis. It had far less cache and much higher latency than the designs that followed, and it felt far more NUMA as a result.
Hyperscalers agreed and turned their noses up at it after some test installations, and decided to wait for Rome - which was much much better in those areas. But Naples was bad and that shouldn’t be sugarcoated.
I don’t really get why people are always so miffed by Intel saying that, it was a criticism that had been leveled against them, and they were right in applying that criticism to AMD, it was a quirky server design glued together out of consumer dies, it suffered all the same downsides as Intel’s own “glued-together” designs.
Didn’t they really just… call a spade a spade?
The new Intel chips would arrive in 2023.
5 years before that Intel shipped Coffee Lake.
5 years before that Intel shipped Haswell.
5 years before that Intel shipped Yorkfield Core 2 Quads.
However, they appear to have managed Altera into the ground, and time will tell if AMD makes the same mistakes with Xilinx.