IBM Beats The Rest of the World To 7nm Chips, But You'll Need to Wait For Them

Mickeycaskill writes: IBM's research division has successfully produced test chips containing 7nm transistors, potentially paving the way for slimmer, more powerful devices. The advance was made possible by using silicon-germanium instead of pure silicon in key regions of the molecular-size switches, making transistor switching faster and meaning the chips need less power. Most current smartphones use processors containing 14nm technology, with Qualcomm, Nvidia and MediaTek looking for ways to create slimmer chips. However, despite its evident pride, IBM is not saying when the 7nm technology will become commercially available. Also at ComputerWorld and The Register.

Comments

They didn't "beat" anybody

[CajunArson]

A test chip made in a lab is not proof of being first to anything with the possible exception of being first to put out an advertising announcement. That goes double for IBM who recently paid GloFo $1.5Billion just to takeover its actual production fabs that make the real chips.

Re:They didn't "beat" anybody

[Anonymous Coward]

Yes they did. They beat everyone in the technology, but they have not yet beat anyone for actual commercial products.

Where is our 350GHz room temp CPU?

[suso]

In 2006 they developed a 350GHz room temperature capable silicon gallium CPU. Where is that?

Re:Where is our 350GHz room temp CPU?

[Anonymous Coward]

The world isn't ready for it.

Christ, - you'd actually be able to run Crysis.

Re:Where is our 350GHz room temp CPU?

[Anonymous Coward]

Withheld by the US intelligence community to maintain their technological gap. If this was declassified in 2006, then the classified stuff was running at a couple of THz even then.

It also allows these companies to drip feed the market minor incremental improvements to maxmise their profit margins. Unless someone forces their hand and drops a really fast chip onto the market.

Re:Where is our 350GHz room temp CPU?

[Anonymous Coward]

By 'developed' you mean, mocked-up-one-for-patent-purposes....

To deliver 7nm devices you need to solve real world production problems, but to mock up a basic chip and simulate how it would run if you ever solved those problems....for the purposes of writing a patent, none of that actual work needs to be done.

The output of this work is a patent not a working silicon gallium processor. The market there is to hijack profits from any company that actually intends to make silicon gallium processors.

Re:Where is our 350GHz room temp CPU?

[kamapuaa]

No they didn't. They developed a 350 GHz room temperature transistor.

Re:Where is our 350GHz room temp CPU?

[Anonymous Coward]

And where is THAT?

Re:Where is our 350GHz room temp CPU?

[suso]

No they didn't. They developed a 350 GHz room temperature transistor.

According to this article it was a CPU:

http://www.techspot.com/news/2...

Maybe the article is wrong?

Re:Where is our 350GHz room temp CPU?

[kamapuaa]

It wasn't a CPU in the sense that you could actually process things with it and make it the central design element of a computer.

Re:Where is our 350GHz room temp CPU?

[Anonymous Coward]

In 2006 they developed a 350GHz room temperature capable silicon gallium CPU. Where is that?

All records were altered to show it as a dead-end technology, once they figured out that it would lead to Skynet.

Re:Where is our 350GHz room temp CPU?

[Anonymous Coward]

I ate it.

Re:Where is our 350GHz room temp CPU?

[suso]

It wasn't a CPU in the sense that you could actually process things with it and make it the central design element of a computer.

That's funny I thought CPU meant Central Processing Unit. Unless you can provide some reference to backup your claim I'm more inclined to believe the article I referenced. Back in the mid 90s there was a Scientific American article talking about how IBM was trying to use gallium arsenide and other materials to make higher frequency CPUs. I would think they would already have a transistor that fast back then, so it makes sense that after 10 years they may have some basic general purpose CPU. So the timeline makes sense, which raises the question, where are the 100GHz CPUs for consumers?

Re:Where is our 350GHz room temp CPU?

[Anonymous Coward]

In 2006 they developed a 350GHz room temperature capable silicon gallium CPU. Where is that?

I would look for it on the front of an air to air missile where cost is no factor at all.

Re:Where is our 350GHz room temp CPU?

[Art3x]

In 2006 they developed a 350GHz room temperature capable silicon gallium CPU. Where is that?

No they didn't. They developed a 350 GHz room temperature transistor.

According to this article it was a CPU:

http://www.techspot.com/news/2...

Maybe the article is wrong?

In 2002 they developed a 350 GHz silicon-germanium transistor.
In 2006 they developed a silicon-germanium processor that reached 350 GHz at room temperature and 500 GHz when supercooled with liquid helium.

Re:Where is our 350GHz room temp CPU?

[ExekielS]

Theoretically the maximum we tap out at is around 10 THz before raising the speed is just not practically doable. Even at 1 THz you need a clock for every core at a small scale. The big problem is power/heat that comes from capacitors. Since their power draw is theoretically infinite the first second and gradually falls, the shorter you run them as switches, the more time they spend at this peak theoretical power, the hotter they get. With non-capacitive transistors, there is no power increase with frequency, so 1Hz and 1 THz use the same power, generate the same heat. It is an experiment every junior in EE does themselves, if not sooner in their academic career. But not all transistors are capacitive. We can put TTL non-capacitive transistors onto silicon even at 14 nm technology level on regular silicon, which means that with the appropriate clocks, virtually all modern computers could be as much as 1000x faster than they presently run while using as much power as if we ran them at 250 MHz. There is no practical reason in any textbook I've found or any chip architecture designer or physicist I've talked to why we can't have processors running at least 500x faster than we currently run them at lower power usage. It is purely to drip feed the market small improvements to keep making tons of money overcharging people for marginally improved hardware, if you give them that much more than they need, there would be no demand for new improvements in architecture and structure, functionality, core design, not in consumer markets anyways, so they would have to get by on a much lower budget, and they wouldn't be able to get away with as many tiers of products. If somebody had a fat wad of cash, hired some good hackers to steal the chip designs from nVidia, AMD, Intel, and hired some timing architecture engineers to get the chips to still communicate at those speeds, they could flatten the computer hardware market for a very long time.

Re:Where is our 350GHz room temp CPU?

[raftpeople]

Those high frequency transistors (like the 350ghz one) are used for communications, it wasn't for a cpu.

Re:Where is our 350GHz room temp CPU?

[Anonymous Coward]

Butter-fingered...

Re:Where is our 350GHz room temp CPU?

[Black Dragon]

In 2006 they developed a 350GHz room temperature capable silicon gallium CPU. Where is that?

Ahem:

http://arstechnica.com/uncategorized/2006/06/7117-2/

Re:Where is our 350GHz room temp CPU?

[waferbuster]

"There is no practical reason in any textbook I've found or any chip architecture designer or physicist I've talked to why we can't have processors running at least 500x faster than we currently run them at lower power usage." You need to read different textbooks, and talk with more knowledgeable chip architecture designers and physicists, preferably ones who actually work on CPU design. Not only is capacitance inherent in transistor design, it also impacts the interconnect layers and the substrate to which the chip is attached (you know... conductive structures with an dielectric insulating material separating them...)

Of course, it's a giant conspiracy to keep reducing the gate width, and making fin-fet transisters, and changing to EUV (and trying to design high power lasers to feed these machines) because that's sooo much easier than changing to a non-capacitive transister. Google must be in on the conspiracy, since a search for "non-capacitive transistor" doesn't return any meaningful results.

Sorry, your post is wrong in so many ways, that I don't know where to start... Maybe this will help: https://xkcd.com/386/

Re:Where is our 350GHz room temp CPU?

[delt0r]

That was not a "switching" transistor. They are also physically quite bit.

Re:Where is our 350GHz room temp CPU?

[Anonymous Coward]

Newsflash. You don't even have to mock something up to patent it. You can patent complete bullshit these days. All a patent is is an idea.

Re:Where is our 350GHz room temp CPU?

[ExekielS]

Are you being serious? Capacitance is NOT inherent to transistor design. Anybody who has taken even their first circuits lab knows this. I still have printouts from the oscilloscope for the lab in quesiton. CMOS the power usage increased exponentially with frequency we switched it at, and heated up rapidly, but BJT had constant power draw all the way from 1 Hz to 100 THz with no change in power draw. Course, beyond 1 THz it couldn't switch fully before switching again, so our readout of the signal was less square, but those were both large transistors. Both BJT and CMOS can be integrated at the same scale on modern chips. We could easily replace CMOS with BJT and ramp up the speed by a factor of a thousand. When I asked the professor why we used CMOS, his only answer was that in the early days of computing at very low frequencies the BJT used more power, and had higher leakage, but we are clearly in the opposite situation today And it isn't exactly unreasonable, TTL based on BJT used to do all of our logic circuits, so it isn't as though it is unsuitable for the task.

Re:Where is our 350GHz room temp CPU?

[ExekielS]

Now if you have a real reason it won't work aside from angry "if we aren't doing it now then clearly it doesn't work" kind of quackery, I'd love to hear it. Really, I'd hate to have my hopes up for no reason. But I'm not going to roll over at such a stupid post with no real information in it.

Re:Where is our 350GHz room temp CPU?

[ExekielS]

It was one of the labs we did in this course, so I've seen it first hand: http://classes.engineering.wus...

Re:Where is our 350GHz room temp CPU?

[bws111]

Good grief. Bipolar? Are you kidding?

In the mid-90s (hardly the 'early days of computing'), IBM made the decision to switch from bipolar to CMOS for its mainframes. This was quite painful, as CMOS did not have near the performance of bipolar, and they risked losing customers. So why did they do it? Because the fastest machine they offered, with 10 CPUs, used 165KVA and created 475,000 BTUs/hr of heat, and the roadmap showed ever increasing rates of energy usage. By the time CMOS caught up in performance (about 8 years later), those numbers were 0.6KVA and 2,000 BTUs/hr. Todays mainframes, with about 200x that performance, only use 27KVA.

You might consider the fact that your little school experiments bear no relation to the real world of processor design.

Re:Where is our 350GHz room temp CPU?

[ExekielS]

And back then the fastest processors were running in the low MHz range, and the thermal runaway and leakage was high in large BJT transistors. But when you scale them down much of that thermal runaway disappears, as does the leakage, and, since they are not capacitors in their function, you can run them at speeds 2-3 orders of magnitude faster. I'm not ignorant of the problems of early BJTs, but CMOS can not be pushed faster due to it's reliance on capacitance for switching, that is a fundamental upper limit that we have hit and why we are pushing out sideways, it is why we could no longer frequence scale and started adding cores and building wider pipelines that could do more functions per clock, etc. Modern BJTs on silicon can be made in the same way we make CMOS, with extremely low leakage, no thermal runaway, the damping techniques have gotten a lot more advanced and efficient. The switch back to BJT development would allow us to break through that frequency barrier that so strongly limits our current technology. With today's technology, there is no reason at all we can't make that switch and the gains would be unfathomably high. But if you have a real reason it can't be done, a technical reason, not an appeal to a past situation or an offhand dismissal, I would *really, really, really* like to hear it. Honestly. I'm not in the same camp as many slashdotters, I'm here to learn.

Re:Where is our 350GHz room temp CPU?

[bws111]

No, the fastest processors were certainly not in the 'low MHz' range. As I recall, they ran at about 700MHz. Not as fast as today, but far from low MHz.

I do not know exactly why BJT is not used, I am not a semiconductor expert. But I do know that there are specialized applications where they mix BJT and CMOS on the same chip, which is way harder than just making a BJT chip, so there must be some reason they do it. And I know that there are very smart people all over the world looking for any way to increase performance while keeping power usage low. And I highly doubt that every one of them has overlooked the 'obvious' solution of just using BJT.

Re:Where is our 350GHz room temp CPU?

[Anonymous Coward]

What do you do for a living? Because in virtually every place, business, organization, etc I've seen, there are always glaring problems with trivial solutions that nobody has bothered to check our or verify. It is always "somebody else's job" or they are "too busy" to look into it, or some other problem. Most people in most US brain industries are working 70+ hours a week and completely drained doing their jobs, much less trying to figure out a full, new approach to a problem. The whole "everything that could be done has been done" mentality is the mentality of failures, it is virtually never true, and often is about as far from accurate as it is possible to be.

Design with BJT is similar to CMOS but you have to make a few adjustments, and in order to ramp up frequency you need to add more clocks because propagation delay becomes an issue, you need to add damping and signal clamping on top of the transition to BJT, so it is slightly more complicated, and it is a different style of design than when you are working with CMOS, so those who are experts in CMOS chip design aren't going to be that deep in BJT design. It would be a lot of work to make the transition. Back in the days of 145+ nm tech the switching time of a BJT was too long to get bigger, but in the days of 15nm tech the switching time is substantially faster, so it is very well and easily possible to make the transition.

All that is needed is a few experts to rework designs (one in BJT development, one in high frequency timing), a good chunk of start up capital, and some designs to derive from.

If there is a reason we don't use BJTs, I've searched very hard and I haven't found it, for the past several years it has been on the back of my mind. The only reasons I can ever find are ones that haven't been applicable in about a decade

Re:Where is our 350GHz room temp CPU?

[Keybounce]

I was taught that very high speeds came with very short wires, and getting chips faster than about 5 GHz meant that you could not even have external support chips, never mind the constant wait for memory to respond and sitting around idle at the inevitable cache failures if you went faster.

Hence, the push from Intel, etc., to massive numbers of cores rather than faster cores as transistors shrank farther.

Re:Where is our 350GHz room temp CPU?

[suso]

The article is titled:

"IBM produces 500GHz silicon-germanium CPU"

Re:Where is our 350GHz room temp CPU?

[benthurston27]

I think the problem is the higher the clock the less time the signals have to get to other parts of the chip, so the number of transistors you can have in lockstep gets smaller and smaller for a given size of transistor... so even if everything was switching at 1 thz most of the time they'd be waiting for signals to reach them from other parts of the chip...

Re:Where is our 350GHz room temp CPU?

[ExekielS]

The big issue is clocks. Right now with current size, 1 THz, the propagation is a big issue around 500 GHz. With smaller technologies like 15nm, 7nm, it gets even smaller so propagation is even less of an issue. With a clock per core in a typical nVidia chip, you could get up to about 5 THz before propagation delay becomes an overriding factor. And the power efficiency gains would be amazing. Even if we just ran them at 5 GHz to quintouple speed, or 50 Ghz even, it would just mean no increase on power usage.

Re:Where is our 350GHz room temp CPU?

[benthurston27]

Just some back of the envelope calculation but at light speed at 1THZ a signal could travel 3000000/1000000000000 = .0003 meters and i5 processors are about 17x17 mm so a signal could only get about 20% of the way across the chip on that clock at light speed, I don't know how much slower than light speed they actually move though...

Re:Where is our 350GHz room temp CPU?

[Anonymous Coward]

Still, even half that speed, 500 GHz, is about 500x faster than we run many chips today, and 100x faster than our best processors out while using as much power as the chips we are using now that are running under 500 MHz. And there are ways to account for it. For example, separate clock for each core and accounting for propagation delay in circuit design, you can lower the total distance required to about 2x2mm, and the signalts travel within 5% of light speed.

But it is certainly possibly to plan for propagation delay in chip design, to know that the signals are coming out x clocks behind where expected with 0-time propagation. It isn't exactly a difficult hurdle, we do it all the time in designing chips for high frequency communication systems.

It's about yield

[cerberusss]

IBM is not saying when the 7nm technology will become commercially available

No, because a big hurdle is of course lithography on 7 nanos, but the even bigger hurdle is using it with a high enough yield to make it commercially viable.

Re:It's about yield

[Anonymous Coward]

So much BS coming out of IBM these days. Can I remind you of their 'worlds fastest CPU' (using clock speed of 5.2GHZ as the 'fastest' claim), actually is as powerful as a 1990s pentium, about 650 MIPS! But if you dare benchmark it they sue you under the license you signed when you paid millions for it.

http://www.osnews.com/thread?438597

You are right about yield, the idea of making a 7nm or 6nm or 5nm chip is there, but actually fabricating chips in a production run is the problem to be solved. Which they haven't done here.

Perhaps Watson can help them, the way its helped nobody.

Re:It's about yield

[Anonymous Coward]

because a big hurdle is of course lithography on 7 nanos

That and the cost of germanium. GE will never be common in consumer volume VLSI applications. It's a "rare earth" the is actually rare, in addition to being costly to refine. It appears in discrete devices where nothing else will suffice and that's it; need solar panel for your Mars rover and have a NASA size budget? Use Germanium! Everyone else gets Silicon.

Re:It's about yield

[bws111]

What kind of imbecile actually believes that the z12 has the performance of a 1990s pentium? The kind that thinks a mainframe MIPS has ANY relation at all to an Intel MIPS.

Re:It's about yield

[amalcolm]

huh? Germanium ranks near fiftieth in relative abundance of the elements in the Earth's crust. Was/is used in transistors - before silcon came along.

Re: It's about yield

[bws111]

Any evidence at all for that no benchmarking claim? Thought not.

AS400s are not mainframes and never have been. You seem to know very little about a subject you are talking about.

Re:It's about yield

[Anonymous Coward]

And yet also nearly 4 times the cost of silver despite that being more than a dozen places down the relative abundance list. The price isn't just about relative abundance, but also how easy it is to extract and how often you find it concentrated. Some relatively common elements are difficult to extract because they don't concentrate in ways conductive to a simple mining operation. There is also the costs of purifying it enough for vsli use. And the comment you replied to even mentions that germanium was used in discrete devices...

Re:It's about yield

[Anonymous Coward]

ranks near fiftieth

Since there are only 92 naturally occurring elements 50th is pretty far down the histogram. Pure Germanium is about $2000/kg right now. Purified Silicon is under $14/kg.

Was/is used in transistors

It was used before Silicon. Cost is the reason Silicon replaced Germanium when it appeared. That reason hasn't changed.

Germanium is used today where it is essential to the performance of a discrete device. Where it is not essential you get Silicon. Germanium is used today where cost is no object. Where cost is as object you get Silicon.

You're not going to be buying Germanium i7's off Newegg anytime soon.

Re:It's about yield

[Anonymous Coward]

“...but the even bigger hurdle is using it with a high enough yield to make it commercially viable.”

The biggest hurdles to making it commercially viable are corporate politics and management fiefdoms.

Re: It's about yield

[Anonymous Coward]

IBM mainframe CRAP from the 90s was a joke. "AS400s are great boxes! Trust us! They have great workflow processing. "

LOL

I bet both the IBM mainframe and also the AS400 still run faster than your Toshiba laptop mainframe* running Windows ME.

*(Using a term incorrectly so that you might comprehend my statement, since you clearly use the word "mainframe" to mean "any possible type of computer varing wildly from mainframe to miniframe to minicomputer to desktop to laptop to adruino")

Re:It's about yield

[Big Hairy Ian]

high enough yield

You mean they've got to make weapons grade Silicon Galanium?

Incorrect...

[Junta]

Most current smartphones use processors containing 14nm technology

Only a few use 14nm today. It's still relatively scarce.

Also, a company that no longer had a fab did a proof of concept in a lab. This is not what the headline suggests. It's nice to know that we have a proven hypothetical to get down to 7, but the practical side of things has a tenuous relation to research.

Re:Incorrect...

[gstoddart]

Meh ... years ago when they spelled IBM by dragging around individual molecules hasn't really turned into much in the way of practical.

Nonetheless, IBM does a lot of basic research into things, because some of it will eventually pay off.

At least someone is doing it.

Not sure..

[Virtucon]

Yeah because IBM sold their FAB so they don't know when anybody will produce chips based on this 7nm technology. They'll be happy to license it to chip manufacturers, they just won't produce it themselves.

Re:Not sure..

[Anonymous Coward]

http://arstechnica.com/gadgets/2015/07/ibm-unveils-industrys-first-7nm-chip-moving-beyond-silicon

so they don't know when anybody will produce chips based on this 7nm technology

If you ignore "GlobalFoundries, Samsung, SUNY, and various equipment suppliers".

My guess is 2-3 years from now is when we start to see this tech. 10nm is just coming online with most of the 'big guys' having a bit of trouble with it. 10nm was basically 'labware' 2-3 years ago.

Re:Not sure..

[Anonymous Coward]

Yeah because IBM sold their FAB so they don't know when anybody will produce chips based on this 7nm technology. .

Um.. the article clearly points out that this was research was accomplished with the help of GlobalFoundries.. which is a fab, and they will probably produce this. Yes IBM will license it but I think you are incorrect when you say 'they dont know when anybody will produce chips" ...

Re:Not sure..

[Anonymous Coward]

What a strange definition of selling: normally you sell something and receive money. IBM gave the foundry and money to the company (I can't call it the buyer) that ended up running the foundry.
That said, IBM has committed a large amount of money to research in this area. And $3 billion go a long way for lab research and even manufacturability studies, it is nothing for mass production when a state of the art semiconductor processing line which will start 2-3 years from now needs an investement to the tune of $15 billions.

Re:Not sure..

[phantomfive]

My guess is 2-3 years from now is when we start to see this tech. 10nm is just coming online with most of the 'big guys' having a bit of trouble with it. 10nm was basically 'labware' 2-3 years ago.

On the roadmap, it's predicted to be 2017-2018.

As for me, I really hope we get another clock-speed doubling or two before we get to 5nm. Not sure if it'll happen, though.

Re:Not sure..

[Virtucon]

True, yeah it was pretty much "here, take it. just take it away" but now they come up with new technology and they'll either have to partner or license to get it to market. It's worked out so well for AMD after all.

Re:Not sure..

[Virtucon]

I do ignore them because GF isn't going to build something without a market or orders standing out the door waiting to be filled. I can see this as a boon to mobile first, so yes maybe Samsung but they're no slouches either and may just do their on 7nm process. The bet for IBM is that they can get somebody to license it and turn it into hard bottom line profit. Either that or in year Ginny will axe the chip R&D group because they're not "profitable"

Re:Not sure..

[Anonymous Coward]

Yeah sure if you ignore Broadcom and Qualcomm and TI and AMD and all the ARM dudes out there. But your way may work too.

Slimmer devices

[hsa]

Most likely not.

The CPU/GPU is not the bottleneck anymore. The screen and wireless consume more power. The sad truth is, everything else has advanced, but battery technology is still in the last decade.

Re:Slimmer devices

[Anonymous Coward]

but battery technology is still in the last decade.

Battery technology is always in the last decade.

Re:Slimmer devices

[Anonymous Coward]

No, but smaller chip scales => less power which leads to longer life with the same battery. Or same life with a smaller battery. Hey look, smaller things are thinner!

Re:Slimmer devices

[Anonymous Coward]

I'm sorry, but this is absolutely not true. You think the batteries in your devices are merely smaller because of reduced CPU consumption? The truth of the matter is that battery technology is improving at a constant rate, and the amount of R&D being poured into the industry is impressive. Just because you don't know that it's taking place doesn't mean it's not happening. And seriously, you can confirm this for yourself with some simple google searches.

Re:Slimmer devices

[Anonymous Coward]

This, batteries are horrible.
Here is hoping the lithium air (I think it was that) experiments are successful. They might have found out what gives its stability and are doing further tests.

Not to mention I simply DO NOT want a thinner anything.
I don't event want bloody computerised paper, flexi-computer or whatever other hundred names they have had over the years.
I want something with functionality. I actually want features. Not less.
How the hell am I going to plug things in to paper?
Wireless? Bluetooth? Fuck wireless. Wireless this, wireless that, NO.

I'd rather have a nice beefy tablet over something thinner. A beefy laptop over literally the worst piece of crap ever, MacBook Air. Fuck you Apple, you don't get to decide optical is dead. Or Flash. Or anything.

Re:Slimmer devices

[Anonymous Coward]

Google searches take up too much battery power.

Re:Slimmer devices

[Anonymous Coward]

Then why is the most common mobile device complaint still "The battery dies too damn fast!"?

Until we can go weeks without recharging or sacrificing current levels of performance/functionality, the assertion that battery technology is still in the last decade compared to all the advances in this one is true. R&D is well and good, but the results aren't there yet.

Re:Slimmer devices

[PRMan]

Just because YOU don't see 1 day as enough for a phone, most people do. Therefore manufacturers target "1 day" when determining how large the battery needs to be and how efficient or fast the CPU and GPU need to be. Batteries have come a long way, but manufacturers would rather eat them up with a faster phone with more pixels.

Re:Slimmer devices

[monkeyxpress]

Most likely not. The CPU/GPU is not the bottleneck anymore. The screen and wireless consume more power. The sad truth is, everything else has advanced, but battery technology is still in the last decade.

This is very true. However, the battery doesn't need to be the bottleneck for significant improvements. At the moment much of the energy is wasted in the screen backlight on LCD devices, and as more displays move towards new technologies like quantum dot or AMOLED this will be reduced significantly. On the wireless side it is a bit harder to get gains, but as smaller mobile cells gets built out this will allow for a reduction in power in the situations most people use their phones.

With these developments it is not improbable for us to move towards phones with half the battery size (or twice the battery life) of current products without needing any sort of big leap in battery technology. Given that there is still scope for the main board/sim card/audio jack to shrink, I don't see any reason why our current technology trajectory won't allow us to have 4-5mm thick iPhone in a couple of years. Whether having phones this thin is actually useful is a different question of course.

Re:Slimmer devices

[DigiShaman]

As of June 28th, it was announced that Samsung effectively doubled the capacity of Lithium Ion battery technology.

http://gadgets.ndtv.com/mobile...

Re:Slimmer devices

[meta-monkey]

The results are absolutely there. Your phone today has bigger battery with smarter management technology than last year's model. But your phone is also faster, more powerful, and doing more stuff.

You can get a phone that last for days. But not one running iOS8 or Android 5.

Re:Slimmer devices

[Anonymous Coward]

I'd rather have a nice beefy tablet over something thinner. A beefy laptop over literally the worst piece of crap ever, MacBook Air. Fuck you Apple, you don't get to decide optical is dead. Or Flash. Or anything.

Newegg has over 2000 laptops with optical drives, and 100 laptops over 8 lbs, so I think you're safe for now. Deep breath.

Some other people might find a use for a MacBook Air. Options are good.

If you don't think options are good, maybe you can get Bernie to look at tablets and laptops after he's fixed our deodorant problem:
http://www.washingtonpost.com/blogs/wonkblog/wp/2015/05/26/sorry-bernie-sanders-deodorant-isnt-starving-americas-children/

Re:Slimmer devices

[Zeromous]

Shrinking GPU and CPU die is one of the major reasons battery power has been able to keep up until recently.

Re:Slimmer devices

[BitZtream]

The chip is a tiny sliver of silicon, it has made no significant contribution to device "thickness" anytime in my 40 years of life.

It effects power usage, and therefore battery size, and through that device thickness.

Going from 14 to 7 is nothing about physical size and everything about electron flows.

They can't actually make the dies smaller even on the new process because you still have to dissipate the heat it generates. AND bring in hundreds of electrical connections to the package so it can actually connect to things other than itself.

The whole "thinner devices" comment in the summary just lets you know the submitter doesn't understand why size matters.

His wife probably agrees

Re:Slimmer devices

[alvinrod]

The SoC in modern devices doesn't take anywhere near as much power as the screen or the cellular/wireless connection, so improvements to the fabrication process aren't going to add a significant amount of battery life. Here's a link to an analysis of the power consumption of the Openmoko phone (PDF warning) that shows that for most use cases, the SoC is only a minor part of the total power draw of the system.

Thinness is mostly a by-product of being able to cram more circuits in a smaller space which cuts down on the size of the logic board more so than battery life improvements have allowed for smaller batteries.

Re:Slimmer devices

[spamchang]

And the problem of heat dissipation still exists. I remember a graph comparing various materials' energy densities--the P4 at full-bore was putting out more energy per volume than a nuclear reactor (sub-critical, of course).

Re:Slimmer devices

[ogdenk]

I'd rather have a nice beefy tablet over something thinner. A beefy laptop over literally the worst piece of crap ever, MacBook Air. Fuck you Apple, you don't get to decide optical is dead. Or Flash. Or anything.

I get a lot of use out of my Macbook Air. I don't need optical, I have an SDXC slot and thumb drives. If I really want an optical drive I have a USB one in my bag. Apple hasn't said that Flash is dead. It works fine on a Macbook Air. They just don't support it on iOS.

In the end, Apple tech works great for a lot of people. If you don't like it, don't buy it. Personally, I actually LIKE having a commercially supported desktop/mobile UNIX system with a standard suite of off-the-shelf familiar desktop productivity and graphics apps. In a system with only one moving part to fail. And I don't even have to dick around with X-Windows unless I want to. Works out great.

Who cares what they did in the lab?

[Anonymous Coward]

Intel will continue to manufacture hundreds of millions of chips a year.

IBM will try to use their legal department to file patents and extort money from genuine chip manufacturers. Sadly, that's the only reason IBM have to do hardware R&D these days.

Pull the Other One

[Jahoda]

Given the challenges Intel faced with yields at 14nm.... and indication they face the same challenges with 10nm, evidenced by the push back to 2017 for the technology - I'm pretty goddamned skeptical that IBM has "beat" anyone to anything. Could I go to an Intel laboratory today and see a proof-of-concept 7nm chip? 5nm? Probably using all manner of interesting silicon replacements? I bet that I could.

No, as you can see from the market today, this is merely an attempt by IBM to resurrect their flagging stock prices (which has worked).

Re:Pull the Other One

[rubycodez]

You are silly, IBM is showing a working device that proves 7nm is possible if using heavier atom in alloy, which is a first step

Mass production lithography and multipatterning issues are another step that needs to be addressed.

Re:Pull the Other One

[Anonymous Coward]

The announcement contains the mentioning of finally, almost, possibly working EUV, which is the really interesting part in my opinion. Hopefully that's not all fluff.

Re:It's about yield -- and leakage

[harrkev]

And no mention on the leakage power. Curious. Smaller transistors have less dynamic power, but higher static power.

'Possible' != 'Practical'

[kheldan]

If a huge percentage of and entire wafer ends up unusable due to defects, then it doesn't really matter how tiny you've made the transistors because it'll be too expensive to be marketable. What we've got here is at best a proof-of-concept. At some future date I'm sure the process will be refined to the point where it's mass-producable enough to be practical, profitable, and affordable, but who can say how long that will take?

probably NSA only

[Anonymous Coward]

Since these would probably require an entirely new fab to be built, (probably around 10 Billion $), and they won't have the volume either, (remember how successful GaAs chips despite their power?), they will probably be relegated to fab dedicated to ASIC circuitry for encryption busting for the NSA.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Cost of germanium? Per chip?

[jeffb (2.718)]

Sure, germanium is rarer and more expensive than silicon. And given the mass of germanium required to add the necessary layer to a single chip, that might raise the per-unit material cost by multiple cents -- but probably not.

SiGe devices are more expensive because they're harder to make and (so far) don't enjoy the same economies of scale as silicon. As I understand it, material cost, especially raw material cost, is a vanishingly small contributor.

It does, it was benchmarked

[Anonymous Coward]

MIPS = millions of instructions per second.
IBM uses the same multiplies, moves and so on as everyone else, so yes they are the same basic ops.
2*3=6 is the same operations on Intel or IBM.

And yes it really is that slow, which is why they won't let you benchmark it but instead put out 50 PDFs from 50 partners making vague performance claims in relation to older Mainframe chips. Pathetic.

http://www.theregister.co.uk/2013/07/23/ibm_system_z_bc12_mainframe/

"There were two models of the z114 system, the M05 and the M10. Both machines had a 3.8GHz z11 engine that was rated at a raw 782 MIPS, with the entry performance band set at 26 MIPS. The single-drawer z114 had up to five configurable engines and could scale main memory to 120GB, while the two-drawer M10 had up to 248GB of main memory and up to ten configurable engines. With all ten engines fired up, that came to 3,139 MIPS of aggregate performance after SMP overhead was taken off the top."

So the top of the line chip (costing millions) is a fraction of the power in your desktop PC.

Re:It does, it was benchmarked

[bws111]

The fact that you could even write that shows just how clueless you are. MIPS has not actually meant Million Instruction Per Second for about 3 decades.

Even if we assume it is a meaningful count of instructions, what instructions are they? Are they register-register adds, decimal floating-point divides, a move of 100MB of data, encyption of a block of data? All of those can be done in a single instruction on Z, yet they all require different amounts of time to execute.

And even if you can choose the instructions, where is the data? Is it in L1, L2, L3, or L4? Does some other processor have to cast it out of cache before you can access it? All of that affects execution speed and thus widely changes the 'MIPS' count.

The Z114 uses the z196 processor. That processor runs at 5.2GHz, is superscalar and out-of-order. Each core has 64KB of instruction L1 cache and 128KB data L1 cache and 1.5MB L2 cache. Each processor has 2 integer units, 2 load-store units, 1 binary floating point and 1 decimal floating point unit. It can decode 3 instructions and execute 5 instructions in a single cycle. If you actually believe all that adds up to 'a fraction of the power of a pc', based of some entirely meaningless MIPS number, you are seriously deluding yourself.

I beat the world to making a time machine...

[Joolz50]

...but I don't know when it will be commercially available

Also covered on CRN

[MoarSauce123]

www.crn.com/news/components-peripherals/300077387/why-ibms-breakthrough-7-nanometer-chip-matters-to-partners.htm

7's the key number here. Think about it.

[jpellino]

"7-Elevens. 7 dwarves. 7, man, that's the number. 7 chipmunks twirlin' on a branch, eatin' lots of sunflowers on my uncle's ranch. You know that old children's tale from the sea. It's like you're dreamin' about Gorgonzola cheese when it's clearly Brie time, baby. Step into my office." Sorry. It could not be helped.