Slashdot Mirror
Intel Says Chips To Become Slower But More Energy Efficient (thestack.com)
An anonymous reader writes: William Holt, Executive Vice President and General Manager of Intel's Technology and Manufacturing Group, has said at a conference that chips will become slower after industry re-tools for new technologies such as spintronics and tunneling transistors. "The best pure technology improvements we can make will bring improvements in power consumption but will reduce speed." If true, it's not just the end of Moore's Law, but a rolling back of the progress it made over the last fifty years.
Hopefully if this does happen they will keep making the existing products, at least until they *do* manage performance improvements that catch up / exceed older stuff. Where I work we have lots of customers that *need* more processing power, and efficiency be damned.
William George
Probably not. Going beyond 5Ghz limit has been a problem for the last decade or so. This is why we have multicore processors. It's easier to add more cores than go to plaid.
Moore's law says nothing about power or speed. It's strictly about the number of transistors on a chip.
"National Security is the chief cause of national insecurity." - Celine's First Law
Optical
:T:R:A:N:S:
A flight from London to New York takes as long today as it did about 50 years ago. But the current planes achieve that more efficiently, with slightly larger windows, and some more pressure and humidity in the cabin. How depressing to think that the computing world might be about to enter a similarly dismal stage as well.
So the plan to make transistors tolerate higher clock speeds by using better materials is not going to happen?
Yet another restating of Moore's Law? The thing gets revised to whatever the latest growth area is.
The original 1965 article it was about "component counts", then it was revised in a later talk to be "circuit density", then revised in 1975 to be "semiconductor complexity", then revised in the later '70s to be "circuit and device cleverness", has been restated yet again when serial devices flatlined in favor of highly parallel chips.
Assuming this goes through the chipset, it will likely be restated again in terms of whatever other factor on the chips continues to grow.
//TODO: Think of witty sig statement
Intel's so far ahead of AMD, they have to roll back the clocks in order to stay competitive.
AMD isn't Intel's competition. Intel needs AMD to prevent Anti-Trust litigation. Intel's competition is ARM and all the OEM's who use ARM based chips. Especially if Microsoft ports full Windows 10 to the ARM. The big draw of ARM is performance/price per watt which is exactly what Intel is shooting for.
"A person is smart. People are dumb, panicky dangerous animals and you know it." - K
Intel's been shooting itself in the foot with power vs performance for years. AMD was better, Intel reversed course and then beat AMD down. Now Intel's gunning for ARM because ARM is becoming a real threat to their core business. How many phones have Intel chips? How many tablets? Notebooks are moving towards ARM as well. Imagine an ARM based server farm. ARM is moving up the food chain into Intel's core business, and doing so with a class of processors Intel can't match.
The cesspool just got a check and balance.
No, a lot of applications don't scale well across multiple cores / CPUs.
William George
MS are playing a very long game because they can afford to. Despite it's well-publicized problems, I find Windows 10 is fast and rock-solid on a desktop and on a Lumia phone. They already have Windows compiled for ARM and they have Office desktop apps compiled for ARM. OK it's a kludged version on the RT platform, but most of the work is done. They are making it easy and attractive (at least in a 'hell, why not?' sense) for new app development to compile for both x86 and ARM. I think one of the reasons why Windows 10 Mobile ('Phone') still exists is because it keeps the ARM branch current and that has sufficient value for MS that they don't even care if the phones never sell.
"Our opponent is an alien starship packed with atomic bombs," I said. "we have a protractor"
Contrary to popular belief, Moore's Law doesn't say that processors will double in speed every 18~24 months. It says that the number of transistors that can economically be put on a single chip will double every 18~24 months. Up until recently, that has translated into a doubling of speed for two reasons: 1) more transistors can be used to optimize the processing of instructions through a variety of techniques and 2) the distances signals have to travel is lessened as the transistors shrink. More transistors contribute not only to power consumption but also more heat, which is another problem with high performance processors. This was partially dealt with by putting multiple cores on a die running at less than max clock rates, thereby distributing the heat and making it easier to deal with. It still may be economical to put more and more transistors on a die, but maybe we don't want to. More transistors consume more power. What's your priority, raw speed or power consumption. Maybe you can't optimize for both at the same time.
Except that Intel has been a licensor of ARM for a very long time, so even if there was some magical shift to ARM in non-mobile ultra-low-voltage devices, Intel would still be able to apply what they know about advancing the state of the art.
Don't worry about Intel, they'll be just fine.
Slashdot still doesnâ(TM)t support Unicode after it was added to the HTML standard in 1997.
Exactly. Apple kept a secret x86 / x64 version of Mac OS X in the closet for 5 years as a hedge against IBM screwing them over on PowerPC. Turns out to be one of the best decisions that they ever made.
Slashdot still doesnâ(TM)t support Unicode after it was added to the HTML standard in 1997.
> Going beyond 5Ghz limit has been a problem for the last decade or so.
Last decade? Uhm, try the last ~40 years. A close friend of mine worked with the military running GaAs CPUs at ~4.7 GHz in late 70's. He also worked on GaAs devices operating up to ~100 GHz. Hey, when you have a nearly unlimited tech budget you can do all sorts of things that the commercial sector won't have access to until decades later.
Anyways, the problem with Silicon is that it needs to be < 110 degrees C. In contradistinction GaAs only need < 175 degrees C.
Hardware designers have known about alternatives for years -- Silicon is just plentiful, dirt cheap, and "good enough." No one wants to pay $100,000 for a 10 GHz GaAs CPU, when you could buy 2,000x Silicon chips instead for the same amount of money.
Of course, the fundamental problem this presents is that it does *not* automatically result in improved performance.
Architectural changes require that performance code be tuned or re-tuned, which means every at-scale application has to be somewhere between rejiggered and given a huge dedicated rewrite effort (The DOE's upcoming 300 petaflop GPU machine will have exactly ten applications that can run at full scale, each of which will have an entire dedicated team rewriting it to do so). And, of course, Amdahl's Law puts an ironclad limit on the effect that more parallel hardware can have on performance, and some problems simply cannot be parallelized no matter how much we wish otherwise.
Contrast with the effect of improving the serial performance of hardware: All else being equal, double the CPU and memory clock rates and absolutely every program will run twice as fast, full stop. That was the desktop miracle from 1990 to 2003 or so - the same exact code screamed twice as fast every year.
But as processors trend towards slower and wider, everything becomes an exercise in parallel programming. OpenMP parallel, MPI parallel, SSE simd instructions, GPU simd parallel... It's harder to do at all, and harder yet to do *right*, and historically the average programmer has enough trouble working with a runtime that's sequentially consistent.
Rant aside though, I agree you're right - until we move to diamond substrates & heatsinks, we've hit the thermal brick wall (actually we hit it circa 2003) and there will not be any further increases in serial processing speed. Plus, AFAICT, there's a similar brick wall with access rates to DRAM and the fact that it requires a microwave-frequency bus with literally hundreds of pins extending for entire centimeters... so forget that too.
1024 cores will make it possible to get ten steps in, assume each step is a binary choice. The software I work with is way more complex than that. Not to mention, cache coherence is going to be a big problem, and multiplying the power draw and heat production by a thousand may be inconvenient.
There are ways to make problems more parallelizable, but they aren't going to work on all problems. Some problems are just really, really difficult to split up efficiently.
"When you have eliminated the unacceptable, whatever is left, however improbable, must be the truthiness" - Holmes
You can have strong AI in ~20W, because that's what our brain uses. Each neuron is really, really slow like 100Hz and below, but when you have absurdly many it works. The problem is understanding the programming model, because it's nothing like our one list of instructions.
Live today, because you never know what tomorrow brings