Moore's Law Blowout Sale Is Ending, Says Broadcom CTO 267
itwbennett writes "Broadcom Chairman and CTO Henry Samueli has some bad news for you: Moore's Law isn't making chips cheaper anymore because it now requires complicated manufacturing techniques that are so expensive they cancel out the cost savings. Instead of getting more speed, less power consumption and lower cost with each generation, chip makers now have to choose two out of three, Samueli said. He pointed to new techniques such as High-K Metal Gate and FinFET, which have been used in recent years to achieve new so-called process nodes. The most advanced process node on the market, defined by the size of the features on a chip, is due to reach 14 nanometers next year. At levels like that, chip makers need more than traditional manufacturing techniques to achieve the high density, Samueli said. The more dense chips get, the more expensive it will be to make them, he said."
On schedule (Score:5, Interesting)
About ten years ago, I went to a talk at Stanford where someone showed that the increasing costs of wafer fabs would make this happen around 2013. We're right on schedule.
Storage can still get cheaper. We can look forward to a few more generations of flash devices. Those don't have to go faster.
Re:A decade long product cycle sounds good to me (Score:5, Interesting)
No, you are still upgrading at the same rate. Except now, because more and more stuff is being pushed out onto the web, it is the servers that are being upgraded. So it is transparent to you. Oh, and phones too.
Re:Just in time too. (Score:5, Interesting)
Erh... no.
As an "old" programmer who happens to know a few languages, ASM for a few different machines among them, I can reassure you that you do NOT want to return to the good ol' days of Assembler hacking. For more than one reason.
The most obvious one is maintenance. I still write ASM for embedded applications where size does matter because you're measuring your available space in Bytes. Not even kBytes. Where it matters that your code takes exactly this or that many cycles, none more, none less. But these are very, very specific routines with a "write once, never touch again" policy in mind. You do not want to be the poor bastard who gets to maintain ASM code. Even less so if it's not your own (which is already anything but trivial). ASM is often a very ugly mess of processor side effects being used for some kind of hack because you simply didn't have the time and/or space to do it "right".
C is probably the closst you should get today to the "metal" anymore. Unless of course you have a VERY good reason to go lower, but I can not really think of anything that doesn't deal with the OS itself.
Re:350mm (18inch) wafer (Score:5, Interesting)
Re:Just in time too. (Score:5, Interesting)
As an addendum to the parent (I, too, have a background in ASM programming): You're working at such low level of detail that any application of non-trivial size becomes extremely difficult to write truly effectively. You just can't keep so many details in mind at once. And when you need to work as a team, not alone, interfacing code becomes a nightmare.
So of course you abstract your assembler code. You define interfaces, develop and use libraries of common application tasks, and just generally structure your code at small and large scales.
But at that point, you are starting to lose the advantage of ASM. A good, modern C compiler is a lot better than you to find serendipitous optimization points in structured code, and it is not constrained by human memory and understanding so it doesn't need to structure the final code in a readable (but slower) way.
Small, time-critical sections, fine. Small embedded apps on tiny hardware, no problem. But ASM as a general-purpose application language? That stopped making sense decades ago.
Re:350mm (18inch) wafer (Score:5, Interesting)
This hits the nail on the head. For decades, software developers have been able to play fast and loose, while counting on the ever-faster hardware to make up for bloated, inefficient programs. Those days are ending. Programmers will need to be a lot more disciplined, and really engineer their programs, in order to get as much performance as possible out of the hardware. In a lot of ways, it will be similar to the early days of computing.
Or, y'know.... (Score:5, Interesting)
Encourage inventors rather than patent troll them into oblivion.
Just a thought, I know it would destroy much of the current economic model, but maybe - just maybe - those expensive techniques are merely the product of insufficient brains. Does the semiconductor world forget so soon that "cutting edge" in the 1970s was to melt silicon and scrape off the scum on top? Does it eve r occur to anyone that, just as we use reduction techniques to obtain silicon today because older methods were crap, there exists the potential that the expensive, low-quality techniques of today could be the rejects of tomorrow?
There are no inventors any more because silicon is a bloody expensive field to get kicked out of by patent trolls. Mind you, it's also a difficult area to get into, what with TARP being used to fund golden parachutes, bonuses and doubtless a few ladies of the night rather than business loans and venture capital. There's probably a few tens of thousands of mad scientists on Slashdot, and I'm probably one of the maddest. Give each of us 15 million and I guarantee the semiconductor market will never be the same.
(P.S. For the NSA regulars on Slashdot, and if you don't know who you are, you can look it up, feel free to post on your journals or as an article all the nifty chip ideas you've intercepted that have never been used. After all, you're either for us or for the terrorists.)
Re:Herb Sutter wrote about this 8 years ago (Score:4, Interesting)
8 years ago I was rocking a single core pc with two gigs of memory and a phone with ~320mhz cpu....
and when did they attach to moores law "lower power use" - if that were it then the law would have been out of the window 1985 and athlons would never have been either..
Re:A decade long product cycle sounds good to me (Score:4, Interesting)
Re:350mm (18inch) wafer (Score:5, Interesting)
It's not cheap to get rid of that much processor power without improving anything.
Office XP system requirements: Single core processor at 0.133 GHz minimum. 0.4 GHz recommended. RAM 0.024 GB (OS) + 0.008 GB (Office). Storage 0.21 to 0.26 GB.