Slashdot Mirror
To Keep Pace With Moore's Law, Chipmakers Turn to 'Chiplets' (wired.com)
As chipmakers struggle to keep up with Moore's law, they are increasingly looking for alternatives to boost computers' performance. "We're seeing Moore's law slowing," says Mark Papermaster, chief technology officer at chip designer AMD. "You're still getting more density but it costs more and takes longer. It's a fundamental change." Wired has a feature story which looks at those alternatives and the progress chipmakers have been able to make with them so far. From a report: AMD's Papermaster is part of an industry-wide effort around a new doctrine of chip design that Intel, AMD, and the Pentagon all say can help keep computers improving at the pace Moore's law has conditioned society to expect. The new approach comes with a snappy name: chiplets. You can think of them as something like high-tech Lego blocks. Instead of carving new processors from silicon as single chips, semiconductor companies assemble them from multiple smaller pieces of silicon -- known as chiplets. "I think the whole industry is going to be moving in this direction," Papermaster says. Ramune Nagisetty, a senior principal engineer at Intel, agrees. She calls it "an evolution of Moore's law."
Chip chiefs say chiplets will enable their silicon architects to ship more powerful processors more quickly. One reason is that it's quicker to mix and match modular pieces linked by short data connections than to painstakingly graft and redesign them into a single new chip. That makes it easier to serve customer demand, for example for chips customized to machine learning, says Nagisetty. New artificial-intelligence-powered services such as Google's Duplex bot that makes phone calls are enabled in part by chips specialized for running AI algorithms.
Chiplets also provide a way to minimize the challenges of building with cutting-edge transistor technology. The latest, greatest, and smallest transistors are also the trickiest and most expensive to design and manufacture with. In processors made up of chiplets, that cutting-edge technology can be reserved for the pieces of a design where the investment will most pay off. Other chiplets can be made using more reliable, established, and cheaper techniques. Smaller pieces of silicon are also inherently less prone to manufacturing defects.
Chip chiefs say chiplets will enable their silicon architects to ship more powerful processors more quickly. One reason is that it's quicker to mix and match modular pieces linked by short data connections than to painstakingly graft and redesign them into a single new chip. That makes it easier to serve customer demand, for example for chips customized to machine learning, says Nagisetty. New artificial-intelligence-powered services such as Google's Duplex bot that makes phone calls are enabled in part by chips specialized for running AI algorithms.
Chiplets also provide a way to minimize the challenges of building with cutting-edge transistor technology. The latest, greatest, and smallest transistors are also the trickiest and most expensive to design and manufacture with. In processors made up of chiplets, that cutting-edge technology can be reserved for the pieces of a design where the investment will most pay off. Other chiplets can be made using more reliable, established, and cheaper techniques. Smaller pieces of silicon are also inherently less prone to manufacturing defects.
Our of all places on the Internet I want at least /. to admit that there has never been Moore's law - it was a mere observation": from Wikipedia, "Moore's law is the observation that the number of transistors in a dense integrated circuit doubles about every two years".
Whoever decided to call it a "law" was a moron and now we have this idiocy repeated every news story. And since it's not a law, we could simple move on and realize that physics simply doesn't allow it to exist.
Not saying it can't work but this sort of thing seems to crop up every few years. At least at chip level it seems to do better than the ever recurring schemes for code reuse.
Bicycle reinvented. These used to be called co-processors.
Did anyone read the headline as being about Chiclets? The chewing gum, not the keyboard.
It seems that the semiconductor industry goes through these cycles periodically. Whenever they run up against limits to single chip integration, they go back to this strategy of wiring together separate chips together. Ultimately, this proves to be inefficient and once technology improves, they return to putting everything on a single chip.
I don't read your sig. Why are you reading mine?
Hitting up against Moore's law is probably the best thing for the chip industry. It's going to force them to innovate.
Each chiplet highly optimized at one task, with enough different tasks to make a general purpose system.
The chipmakers do not follow Moore's Law. Moore's Law follows the chipmakers.
The do not do it "to keep up" to that law. There is no "Moore's Judge" that will spite them if they do not do it.
Don't fight for your country, if your country does not fight for you.
They just use much smaller sockets now. :)
Looks like they've reinvented the IBM 3081 mainframe from 40 years ago:
The elimination of a layer of packaging was achieved through the development of the Thermal Conduction Module (TCM), a flat ceramic module containing about 30,000 logic circuits on up to 118 chips.
I'm pretty sure I saw something similar, almost three decades ago.
#DeleteFacebook
intel still wants massive die sizes to charge insane prices for overheating chips
Anyone remember AMT on Intel?
Back in the dark ages we used to put floating point math on a special chip. Intel had the x87 co-processor, and most computers didn't have them since computers didn't need high-speed floating point math, and when you did, you just the operation in software. This started to go away with the 486, which (mostly) had an integrated FPU on chip, while a few cheaper models disabled it. The Pentium was the first chip where every model had an FPU.
So the future seems to be de-integrating circuits? Possibly. But the whole point of putting everything together on one chip was cost saving. Now it seems like the future is more flexibility, and specialization.
What made the co-processor era crummy was you could rely on it being their. IIRC the linux kernel had some funky Floating point emulation stuff in it to perform these operations if on the integer unit instead.
Also, as I recall one of the problems of this era was non-generic chips. Intel had it's own FPU, the 287 and 387. But there was also a competing, and incompatible FPU you could also have who's name I forget.
i believe she has this already figured out. https://bit.ly/2DuzzPg see memory alpha -- ;-) some group of nanites are going to form a union if we aren't careful.
Wikipedia:MCM
It's like technology companies are starting to behave like Hollywood. Come out with a rehash of what they did a few years ago instead of any new revolutionary ideas.
It sounds like what AMD was doing with thread ripper. With Intel claiming they were "gluing" together chips. Cant make a single piece of silicon cheaply with 16 cores? "glue" four 4-core chips to the same CPU substrate. BINGO!
Chiplets! Imagine a Beowulf cluster of those. On a chip!
Chiplets are good, but what's wrong with wafer scale integration? You could even combine them, chiplets as daughter boards in a 2+1 dimensional arrangement.
It's a small world and it smells funny; I'd buy another if it wasn't for the money; Take back what I paid (SoM)
Its deja vu all over again. We have rediscovered ICs and co-processors and want to call it something new. One of the most irritating aspects of IT is the need to keep the salesforce entertained with some brand new overcooked old steak. Bet it will end up there with the renaming of timesharing as 'the cloud'... The cloud is the BS surrounding an old and solid implementation strategy...
Making larger chips from smaller "chiplets" will not improve the speed nor efficiency of the final combined chip. This is purely a money-saving trick around the age-old problem of having lower yields every subsequent process technology. If anything, the added circuitry necessary to facilitate interconnects between these chiplets only add to power usage and transmission delays.
Chiplets also do nothing for speeding up chip development as monolithic chips are already built from various templates stamped over and over on a large work sheet. These chiplets simply contain a collection of templates, so no gains there.
The summary clearly was aimed towards investors, lots of buzzwords, not much ham.
I wouldn't mind seeing both ASIC chiplets, dedicated for a specific task, like AES array shifting, RSA exponentiation and multiplication, and other tasks a computer commonly does. From there, it would be nice to have FPGAs for most anything else. This can easily allow a hypervisor to run x86 code as well as ARM. Done right, this could also improve security between VMs.
Of course, if someone wants to grind cryptocurrency, next to dedicated ASIC boards, FPGAs are not bad.
This is a 1960s idea, that was replaced by... integrated circuits. At the time, yields were not adequate to put all functions on a chip; it made more sense to mix and match smaller chips with better yield. In addition, at the time it was difficult to do everything at once with silicon in a single process, so other materials, such as germanium, were also used.
Chip makers are currently doing 7 nm lithography. Copper atoms are 0.2 nm wide. We may not have reached the limits for lithography, but we have to be getting real close.
[Insert pithy quote here]
I love their gum!
To communicate 100 light years away instantaneously all it takes is a long stiff rod where very small movement back and forth can communicate. Of course this is just a non practical idea but maybe its not a matter of trying to continue moores law at the hardware level but one of a mindset change on how much speed do you really need, or is it a tortoise and hare race in how you approach getting something done. Sometime I feel that a commodore 64 is more responsive than a today system running some overcomplexifabulocated OS and applications.
Need a Quantum Computer? Well you are in Luck, you have one sitting on your shoulders. You just need to learn to efficiently and effectively use it. Oh wait, the tech/software industry has dumb-down your user ability to comprehend this. And watch as I'm downgraded for this post.
sure sounds like it to me
GENERATION 26: The first time you see this, copy it into your sig on any forum and add 1 to the generation.
Hitting up against Moore's law is probably the best thing for the chip industry. It's going to force them to innovate.
Someone's going to invent...MAGIC!
Every couple of years somebody says that "Moore's Law is ending" and then it doesn't.
I propose we call this Borehd's Law.
Intel's Law: For every advancement in semiconductor technology there will be equal and opposite CPU flaw.
It's the cloud all over again. Isn't it?
is this long (light travels "THIS" far in a nano-second, holding up two hands "THIS" far apart). Chiplets solve nothing. Unless you only have more cores in mind. Then you still solve nothing but interconnect problems. Moore is spinning in his grave, trying to get out and beat the shit out of these chiplet people.
Any attorney without expertise in Bird Law cannot call themselves a legal eagle.
https://en.wikipedia.org/wiki/Slot_1
This is what happens when "quotas" mean that engineers have to pander to less skilled women and negros on their team. They get annoyed from the pandering and burnt out from having to take on workload their team mates cant perform. Then you get brain drain and are left with the lowest common denominator
Are there going to be open source chiplets? It seems to be likely.
Some great videos on chiplets
https://www.youtube.com/watch?v=G3kGSbWFig4&t=1195s
and
https://www.youtube.com/watch?v=d3RVwLa3EmM
by adoredTV
INTEL is fucked.