Intel's Tick-Tock Cycle Skips a Beat 84
New submitter Ramze writes: Several outlets are reporting on Intel's confirmation that it will make three generations of 14nm processors, delaying the switch to 10nm. The planned 14nm Kaby Lake processor marks the first time Intel has skipped the "tick" of a die shrink on its regular "tick/tock" cycle. Production of Cannonlake processors on 10nm has been pushed back to the second half of 2017 — likely due to manufacturing difficulties. Intel reported earlier this year that it may have to switch away from silicon to exotic materials such as indium gallium arsenide to make the next shrink to 7nm.
Re: Boring. (Score:5, Funny)
I know - the transistor count should have enabled us to build neural nets to filter out inane AC comments by now.
Re: (Score:1)
Back when I actually cared enough to be a signed in user I would have my prefs treat a Funny comment downward. Far. Funny comments are or were the scourge of /. - now I just don't care. /. is the scourge of /. Now I use alterslash as another layer of filtering and post as AC.
Re: (Score:2, Informative)
I know - the transistor count should have enabled us to build neural nets to filter out inane AC comments by now.
Hardly. The brain has ~100 billion neurons and 100-500 trillion synapses, of which the latter is closest to a transistor. Leading CPU/GPUs have 5-9 billion transistors or less than 0.01% of that. Remember, we are approaching atom size but only in an extremely thin 2D slice. Current processors are about 100k*100k transistors big, if we could have the same density in three dimensions we'd have 100k^3 = 1000 trillion transistors in a 2.5 cm cube, comparable or even beyond the brain in density. I wouldn't try c
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Synapse is NOT like a transistor, it is a schmitt-trigger circuit with hysteresis. Multiply your transistor count by order of magnitude or more per neuron
Re: (Score:2)
Actually in a GOP politician they function as a biode (e.g. length of wire)
Re: (Score:3)
If you can tolerate latency, you at least h
Re: (Score:2)
Re: (Score:2)
Even better the brain is 3d where chips are two 2d flat landers.
Re: (Score:2)
Of course it runs much slower. With the kind of density the brain has if you increased the clockspeed a lot you couldn't even cool down the brain properly.
Re: Boring. (Score:5, Funny)
Re: (Score:2)
ACTUALLY, it is getting us into a fickle afterlife & fantasy world where the sum of your desires make the walls of your sepulchre.
MAYBE you are already in your afterlife, eternal and alone with your base ego.
Re: (Score:1)
Re: (Score:2)
personally, i make a habit of blaming developers
Re: (Score:2)
When I think about how much faster (at a hardware level) my current computer is relative to my old 486SX 33MHz, and how I still have to spend time waiting for the damn thing to catch up to what I'm trying to do, I just have to shake my head.
Part of the blame goes to slow mechanical hard disks, though.
The Valley will never be the same... (Score:4, Funny)
Re: (Score:2)
http://www.jobapplicationform.... [jobapplicationform.us]
Re: (Score:3)
It's not worth it any more (Score:5, Insightful)
Look at Intel's recent earning and revenues. Business is so bad it doesn't justify investing money in a new engineering shrink.
Re: (Score:2)
Re: (Score:2)
Maybe you and the GP are both right. It looks like they think spending money on architecture will yield more bang-for-the-buck of performance at the moment than yet another geometry shrink. You'd think they would have played all the architectural games possible by now. Bu now that everybody already has more cores than they can use, maybe more can be done at the architecture level to make better use of the same number of cores. (Just a guess.)
Re: (Score:3)
I definitely do not have more cores than I can use. Intel just needs to get off their butt and start shipping processors that measure cores in multiples of dozens.
Re: (Score:1)
They already ship them for quite some time. Look up Xeon Phi. 60 cores per die and 4x HT making it 260 or so execution threads.
Re: (Score:2)
Even in the situation of many-core CPUs with lots of simple cores, many cores do no scale well for most work loads. If your workload has lots of random memory access, a multi-core large CPU can save more power.
Re: (Score:2)
Manufacturers make very little money on low power processors as well. The semiconductor industry has been dog-eat-dog low margin business for a very long time and it looks like Intel's ability to make abnormal profits has come to an end.
Re: (Score:2)
Intel has had a process lead, but that's not the only reason they make abnormal margins.
- Lock in. AMD's the only viable x86 vendor and they've been off in the weeds after their glory days that culminated in x86-64 and NUMA x86 architecture
-Ecosystem. Intel invests heavily in things like compilers and standard libraries and so forth. They pay to have good software developers enrich an ecosystem that favors their processors
-Microarchitecture. They frankly have very good hardware engineers.
ARM came to pre
Re: (Score:2)
Re: (Score:2)
Why? Do engineers have less behavior and mental processes problems these days?
Re: (Score:2)
They are going to need one when Intel lays them of ;-)
Re: (Score:3)
Also, I'm pretty sure that AMD is still at 28 nm, so there really isn't much point in pushing hard for 10 nm, when the competition still uses a process that's twice as big as what you already have.
Re: (Score:2)
Re: (Score:1)
This is the thought I had. Intel has enough of a lead at this point to skip a die shrink. Samsung just got their 14 nm Exynos SOC out a couple months ago. The fabs AMD uses are far behind. And that's about it for fab competition.
Microarchitecture is more important anyhow. When AMD made big gains against Intel with K8 (Athlon 64/X2) it wasn't feature size that did it; Intel had competitive fab tech at the time. The difference was microarchitecture. And today Intel is ahead in that dimension as well.
Th
Toxic metals and metalloids (Score:2)
Silicon-on-Silicon CMOS: non-toxic.
SOD-CMOS: non-toxic.
Indium-Gallium-Arsenide: toxic heavy metals combined with toxic metalloids. Holy fuck. If ROHS doesn't lift their ban on lead after this, they've got their heads up their asses. Mercury is a little worse.
Re: (Score:3)
Sodium and Chlorine being dumped on our roads when it snows? The horror.
Re: (Score:2)
Re: (Score:3)
Not to mention dihydrogen monoxide - which is still used in large quantities in many residential and commercial settings, despite its many dangers. [dhmo.org]
Re: (Score:2)
Oh shit, I think my coffee actual has dihydrogen monoxide in it!
I better drink another cup, just to be sure!
Re: (Score:2)
It also contains a chemical that is toxic to birds, dogs, and cats [wikipedia.org]. Do you really think you should be drinking that...?
Re: (Score:2)
Look, Mitt Romney, I know you think corporations are people, but birds, dogs, and cats are not people either. Now go make me some coffee!
Re: (Score:2)
Don't joke around, Hydrogen is very flammable!
Oxygen is a known oxidizer, so be sure to eat lots of antioxidant's when living or working around sources of oxygen.
All joking aside, Gallium-Arsenide as a compound is not all that dangerous. I wouldn't go eating it, but I have handled it many times and it does not require any special safety procedures. You already have GaAs parts inside of almost every cell phone. Virtually all output amplifiers to date are GaAs, with a minority of the low end phones using C
Re: (Score:2)
Actually, that is horrible. There's a reason conquering armies used to literally salt the earth if they really hated you. Plus it eats cars.
Re: (Score:2)
Feel free to "dispose" of all the indium and gallium you want by sending it to me! As for the arsenic, easily removed.
Moving to InGaAs will make "scrap" chips practically a form of bullion storage.
Re: (Score:3)
Getting changes made when there are at least partial alternatives and the material is distributed in substantial quantities through
Re: (Score:2)
CMOS non-toxic? p type is boron doped, n type is either arsenic or phosphorous doped.
Re: (Score:2)
A quick Wikipedia
Elemental boron, boron oxide, boric acid, borates, and many organoboron compounds are nontoxic to humans and animals (with toxicity similar to that of table salt)
Phosphorus I'll just claim is an essential mineral for life, which is why we put trisodium phosphate in cereal (although if you eat four boxes of it at once, you start nearing the toxicity threshold).
Leap Seconds, Years (Score:1)
Intel Marketing slogan (Score:5, Funny)
"Vee have vays of making you tock!"
Re: (Score:2)
Re: (Score:1)
Re: (Score:2)
I *knew* I shouldn't have partnered with Intel on my line of computer-chip lollipops!
They're adding a tock, not skipping a tick (Score:1)
They've not skipped anything - they're adding a second 14nm tock and delaying the Cannonlake 10nm "tick" by almost a year, in the same way they delayed the Broadwell 14nm "tick" by a year and filled in the gap with 22nm Haswell Refresh.
Given they've not called it Skylake Refresh, we can only assume Intel plan a more substantial change in architecture compared to
Tomatoes (Score:2)
You say ta-may-toh, I say ta-mah-toh (they say, "holy-fuck-this-die-shrink-is-a-bitch")
Re: (Score:2, Insightful)
Broadwell was delayed by a quarter due to defects, not a year. 3 months is a mere glitch in the usual 12 to 18 months between Ticks and Tocks. This new delay pushes a regular part of the cycle out a year or more to when we'd expect another "Tick" -- effectively skipping that part of the cycle entirely.
If one were to say that they're "Adding a new tick, not skipping a tock", then we'd expect the usual shrink every 24 to 36 months with 2 architecture improvements instead of 1 in-between the shrinks.
As an a
Stagnation as far as the eye can see (Score:2, Interesting)
I am not sure why there is anything more on wasted on desktop processors given the last 6-7 years of only ~10 percent gains? We are expecting almost zero improvement in desktop performance with Skylake over the 4790K processors, and barely a power reduction. Billions were spent to get us almost nothing tangible.
Laptop machines have come a long way, but the desktop is stuck at 4 cores and no hint at anything but maybe 10% performance gains per year for the foreseeable future.
We are instead getting integrat
Re: (Score:2)
Meant to say 10% per year, my bad.
Re: (Score:2)
Performance gains 10% over 6 years? They're waaay higher. 10-15% per tick maybe.
Why is this modded up?
The fastest Intel CPUs currently on the market are two years old. Current trend for the last few years is 0% per year, with no gains at all in the latest generation, but there were big improvements before that but mostly with Sandy Bridge, but that is close to being 5-6 years ago now.
Re: (Score:2)
So why don't you buy an Xeon then? Instead of complaining that one product segment (which isn't designed for your unusual requirements) doesn't have the features you want?
And both power levels and performance per power have improved strongly the latest generations so I don't know why you are complaining - if you were right you could just continue running your old machine instead of upgrading so...
Re: (Score:1)
>So why don't you buy an Xeon then?
The markup on those things is absolutely absurd since Intel has no real competition at the moment.
Re: (Score:3)
Equivalent Xeon's get roughly a 2x multiplier for cost, as do the motherboards for them. My work machine is a 6 core Xeon (E5-1650 v2) that can be bought for about $650 compared to about $330 for an i7-4790k, which is also what is roughly expected for the i7-6700k when it arrives with its piddly little 4 cores sitting next to a vast wasteland of third rate GPU.
So either I would like a cheaper i7 without an on-die GPU, or more cores and cache in an i7 in place of the GPU.
With AMD continuing to gasp for life
Re: (Score:3)
We are instead getting integrated crappy GPU's in flagship processors that will mostly never get utilized
Actually I think you'll find Intel is the number one graphics vendor in the world. The number of business machines with dedicated video card sits so close to zero that it disappears in a rounding error. They are also the largest market for computers in the world. That's not even taking into account low end laptops, sometimes high end laptops, tablets, and most low-power devices on the market.
Better on-board / integrated graphics processors have been sought after for a long time. You won't run Crysis on it,
Intel is behind (Score:5, Interesting)
Intel is stalling at 14nm. Everyone else stalled at 28nm. 28nm is still the cheapest node in per transistor terms. Since most chip makers are driven by cost rather than transistor performance, there have been few takers for 20nm and 14nm.
Re: (Score:3)
28nm is still the cheapest node in per transistor terms.
That's not really true anymore. 14nm is cheaper for Intel to manufacture than 22nm (but Intel is the only company thus far with a mature, cost effective node at 14nm.) Remember that all the problems Intel had with ramping 14nm to high volume every other silicon fab will also experience.
Really what this tells us is that if you look at Intel's past two nodes (22nm and 14nm) they both have had about a 2 year, 6 month development cycle instead of the 2 year cycle we are used to. I think this is more just Int