Beta

×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

Intel On Track For 32 nm Manufacturing

samzenpus posted more than 5 years ago | from the wafer-thin dept.

Intel 139

yaksha writes "Intel said on Wednesday that it has completed the development phase of its next manufacturing process that will shrink chip circuits to 32 nanometers. The milestone means that Intel will be able to push faster, more efficient chips starting in the fourth quarter. In a statement, Intel said it will provide more technical details at the International Electron Devices Meeting next week in San Francisco. Bottom line: Shrinking to a 32 nanometer is one more step in its 'tick tock' strategy, which aims to create a new architecture with new manufacturing process every 12 months. Intel is obviously betting that its rapid-fire advancements will produce performance gains so jaw dropping that customers can't resist."

cancel ×

139 comments

Sorry! There are no comments related to the filter you selected.

The new ones are impressive (1)

FlyByPC (841016) | more than 5 years ago | (#26070991)

I'm just finishing a rebuild of my system, going from an Athlon64X2 to a Core i7. 3DMark06 is downloading now; can't wait to see how well it does on that and Flight Simulator X.

...Now if they could only make some progress on coordinating RAID implementations across motherboards, so a MB swap doesn't have to mean that the path-of-least-resistance is a complete reinstall...

Re:The new ones are impressive (3, Informative)

afidel (530433) | more than 5 years ago | (#26071065)

I can't wait for the multichip Xeon's based on Corei7, Intel might finally have a chip that can compete with AMD in the database space next year. Oh and for your raid problem, use HP, a RAID array is portal across all systems and controllers that use the same generation HDD's. I have picked up an array out of a server, put it into a MSA and mounted it through an HBA with no problems then expanded the array online with additional disks to grow capacity =)

Re:The new ones are impressive (2, Informative)

Anonymous Coward | more than 5 years ago | (#26071073)

Two words: software raid. You have 4 cores, chances are you will usually be IO bound, so the performance will be better than HW raid.

Re:The new ones are impressive (0)

earnest murderer (888716) | more than 5 years ago | (#26071159)

Now if only Windows supported RAID in software.

Re:The new ones are impressive (4, Informative)

jaxtherat (1165473) | more than 5 years ago | (#26071221)

It does, here is a RAID 5 example: http://support.microsoft.com/kb/323434 [microsoft.com]

Re:The new ones are impressive (2, Insightful)

Firethorn (177587) | more than 5 years ago | (#26074223)

Yes, but unless they've changed stuff lately, he can't use RAID 5 on his boot disk - only mirroring is supported, and only sorta at that.

Though with the way SSDs are going, I'd seriously consider putting the OS on a SSD, then going with the RAID array.

And have things really changed so much that true hardware RAID is slower? I'm aware that there are RAID devices that depend on the CPU much like winmodems did, but surely a good RAID card still beats software?

Re:The new ones are impressive (0)

Anonymous Coward | more than 5 years ago | (#26071291)

Is this a joke? software RAID is available on Windows from when, NT4? Where have you been all these years?

Re:The new ones are impressive (2)

PiSkyHi (1049584) | more than 5 years ago | (#26072069)

Know of any recovery CD/DVD for a Windows RAID 5 system when it won't boot anymore ? It happened to me on a system I did not set up.

Linux has recovery CDs to the hilt - many with RAID XX support, so you can recover data even when your system won't boot. Under Linux, you can't use RAID5 for a boot device anyway, so it will boot. I thought this was also true of Windows, but this machine had it.

Note: Hardware RAID is dead, long live RAID!

Never use a motherboards SATA for RAID, buy a cheap SiI 3132 or SiI 3124 card.

Re:The new ones are impressive (1)

KingMotley (944240) | more than 5 years ago | (#26072347)

You are mistaken, windows does not support RAID-5 on the boot partition or where the OS is either. If it truly did have RAID-5 and the boot or OS partition was on it, then it was via hardware.

Re:The new ones are impressive (2, Informative)

PiSkyHi (1049584) | more than 5 years ago | (#26072413)

... if you could call motherboard RAID hardware, then yes.

As far as I can tell, its the worst kind of RAID and it has given software RAID a bad name.

The motherboard doesn't have parity chips, its just a flag to Windows to handle the RAID5.

This one went bad and not only marked it as degraded, but windows would not boot and the only tool we could find to get access to the data was a DOS boot floppy with the RAID drivers installed - but then, it didn't have permission to read the files, and the USB tools for moving the data somewhere from a DOS boot disk caused the system to hang.

Re:The new ones are impressive (-1, Troll)

Anonymous Coward | more than 5 years ago | (#26072447)

Then quit using that shitty OS. If your bike keeps falling over when you're making turns, its time to take off the fucking training wheels.

Re:The new ones are impressive (1)

Kane Devaid (1339253) | more than 5 years ago | (#26073733)

RAID Kills Installs Dead.

Re:The new ones are impressive (1, Informative)

Anonymous Coward | more than 5 years ago | (#26071211)

Assuming you're on Linux, buy a processor with more cores, and use softraid. Autodetect = painless movement.

Re:The new ones are impressive (1)

afidel (530433) | more than 5 years ago | (#26071543)

If only software raid had battery backed write cache it might be an option for real production systems....

Re:The new ones are impressive (0)

Anonymous Coward | more than 5 years ago | (#26071885)

battery backed write cache ? sure. its called a UPS.

Re:The new ones are impressive (1)

PiSkyHi (1049584) | more than 5 years ago | (#26072081)

Use a big fat journal, no need for a big fat wallet.

Normal people don't need faster computers (3, Insightful)

EmotionToilet (1083453) | more than 5 years ago | (#26071701)

Faster computers are going to be generally irrelevant to about 85% of the population. They only really use computers for surfing the internet, checking e-mail, MS Office, iTunes, organizing photos, and playing The Sims occasionally. Most people play video games on consoles (PS3, WII, Xbox 360). There are few things that 90% of the population regularly do that require a faster computer. These advancements are going to affect businesses and scientists who need super computers to perform large amounts of computations, or servers that need to respond to heavy demands. The only thing, I think, that needs to be improved is the hard drive. Right now they're just way too slow.

Re:Normal people don't need faster computers (0)

Anonymous Coward | more than 5 years ago | (#26072207)

640 kib should be enough to prove anyone wrong when they make stupid remarks like this.

Sure, YOU may not need a "really really" fast computer. But you'll still buy something running a newer intel cpu the next time you need to update your sims gaming machine, because "why not? It runs better"

Re:Normal people don't need faster computers (1)

EmotionToilet (1083453) | more than 5 years ago | (#26072411)

Actually I think even this statement is wrong. If I go to buy a new computer and I can buy a new model with a super fast processor for $1900, or a refurbished older model for $1300 that is slower, but more than fast enough for my needs, then I'll get the cheaper one and save myself $600. In fact, I did just that 4 months ago and completely love my iMac. :) A 2.4 ghz processor and 2GB ram is just fine with me. I don't need the 3.0 ghz one. It's not worth the extra $600. I use it mainly for internet and programming in Xcode and it does these things just fine. I may not need to upgrade my computer for another 5 to 7 years. If I do, it will only be so that I can get the Mac Pro with the 30" screen or for some new feature that Apple comes out with. :)

Re:Normal people don't need faster computers (2, Funny)

Anonymous Coward | more than 5 years ago | (#26072607)

If I go to buy a new computer and I can buy a new model with a super fast processor for $1900, or a refurbished older model for $1300 that is slower, but more than fast enough for my needs, then I'll get the cheaper one and save myself $600. In fact, I did just that 4 months ago and completely love my iMac.

You got an Apple product cheaper? Amazing.

Re:Normal people don't need faster computers (0)

Anonymous Coward | more than 5 years ago | (#26075187)

I was at a loss as to what refurbished computer could possible cost $1300, especially once I glanced 2.4 GHz. Then I saw that it was a Mac, and it all made sense.

Re:Normal people don't need faster computers (3, Insightful)

smilindog2000 (907665) | more than 5 years ago | (#26072249)

Good point. With solid-state drives coming down the pipe, even that bottle-neck will be somewhat relieved for what most people do (lot's of disk reads, few writes). I write programs to help designers place and route chips. The problem size scales with Moore's Law, so we never have enough CPU power. I'm part of a shrinking population that remains focused on squeezing a bit more power out of their code. I wrote the DataDraw [sourceforge.net] CASE tool to dramatically improve overall place-and-route performance, but few programmers care all that much now days. On routing-graph traversal benchmarks, it sped up C-code 7X while cutting memory required by 40%. But what's a factor of 7 now days?

Re:Normal people don't need faster computers (3, Informative)

EmotionToilet (1083453) | more than 5 years ago | (#26072451)

In my programming classes at UW-Milwaukee the professors emphasize that we should design our code to be easy to read/edit even if that means using up more computation cycles. This makes editing the code easier in the future, which is appreciated by future programmers who have to learn your code and can save the company some time and money. And since computation resources have become so cheap (practically unlimited for most applications) it doesn't really affect the performance of the program to a noticeable degree.

Re:Normal people don't need faster computers (2, Insightful)

theheadlessrabbit (1022587) | more than 5 years ago | (#26072745)

was this professor involved with the design of vista at all?

there is this thing called 'documentation' that you add to your code so other people can understand it.

ignore your instructor. as a user, i very much appreciate whatever gains in efficiency i can get.

Re:Normal people don't need faster computers (2, Insightful)

Peter La Casse (3992) | more than 5 years ago | (#26074227)

In my programming classes at UW-Milwaukee the professors emphasize that we should design our code to be easy to read/edit even if that means using up more computation cycles. This makes editing the code easier in the future, which is appreciated by future programmers who have to learn your code and can save the company some time and money.

was this professor involved with the design of vista at all? there is this thing called 'documentation' that you add to your code so other people can understand it. ignore your instructor. as a user, i very much appreciate whatever gains in efficiency i can get.

This is bad advice. Clear design and coding are extremely important in a nontrivial program. Once it's written, you can profile it to find out exactly where you need to improve performance. Documentation is necessary but not sufficient, and premature optimization makes programs less efficient not more.

Re:Normal people don't need faster computers (2, Interesting)

smilindog2000 (907665) | more than 5 years ago | (#26072831)

The sad part is that improved runtime speed and code readability can be had together at the same time. The reason the DataDraw based code ran 7x faster was simple: cache performance. C, C++, D, and C# all specify the layout of objects in memory, making it impossible for the compiler to optimize cache hit rates. If we simply go to a slightly more readable higher level of coding, and let the compiler muck with the individual bits and bytes, huge performance gains can be had. The reason DataDraw saved 40% in memory was that it uses 32-bit integers to reference graph objects rather than 64-bit pointers. Again, C, C++, and most languages specify a common pointer size for all class types. If the compiler were allowed to take over that task, life would be easier for the programmer, and we'd save a ton of memory.

But then again... what's a mere factor of 7X runtime with today's computers? With the low price of DRAM, who cares about 40%? It's easier to stick with the crud we've used since 1970 (C, and it's offspring) than to bother building more efficient languages. Language research has abandoned efficiency as a goal.

Re:Normal people don't need faster computers (1)

jandrese (485) | more than 5 years ago | (#26074093)

Huh? The cache only contains a set of what the processor already things it is likely to need. It's not like it's loading a fixed window of memory over the entire cache space. How C and other such languages organize their own memory space shouldn't matter much at all. Switching to 32 bit offsets instead of 64 bit pointers is fine so long as you never need to reference more than 4 billion records, but one application does not a whole industry make.

Re:Normal people don't need faster computers (2, Interesting)

smilindog2000 (907665) | more than 5 years ago | (#26074585)

Check out the benchmark table at this informative link [sourceforge.net] . On every cache miss, the CPU loads an entire cache line, typically 64 or more bytes. Cache miss rates are massively dependent on the probability that those extra bytes will soon be accessed. Since typical structures and objects are 64 bytes or more, the cache line typically gets filled with fields of just one object. Typical inner loops may access two of those object's fields, but rarely three, meaning that the cache is loaded with useless junk. By keeping data of like fields together in arrays, the cache line will be filled with the same field, but from different objects, often objects that will soon be accessed. This, plus the 32 vs 64 bit object references, and cache-sensitive memory organization (unlike malloc), leads to a 7X speedup in DataDraw backed graph traversals vs plain C code.

Understanding cache performance is critical for fast code, yet most programmers are virtually clueless about it. Just run the benchmarks yourself if you want to see the impact.

sometimes, yes, sometimes no (1)

HelloKitty (71619) | more than 5 years ago | (#26074867)

premature optimization is sometimes as you say, bad. however there is an idea of mature optimization where you know something needs to be written in such a way as to be fast.

say your task has to run in realtime, and it involves iterating over most of the machine's memory. if it doesn't run fast, you have a real problem.

always choose the correct read/write patterns, the correct architecture, and then make that code as clear as possible...

Re:Normal people don't need faster computers (3, Funny)

Singularitarian2048 (1068276) | more than 5 years ago | (#26072365)

Soon enough people will have robots in their homes, doing chores. Very fast computers will be needed for that.

Re:Normal people don't need faster computers (4, Insightful)

NerveGas (168686) | more than 5 years ago | (#26072375)

A surprising number of people that I know - and not just tech-savvy people - do video compression, either for converting camcorder movies into DVDs, creating slideshows, or using DVDshrink. And those are apps where more CPU is always good...

Just wait until HD camcorders are more prevalent, and you have people that want to convert their home movies into X.264 Bluray discs...

Re:Normal people don't need faster computers (3, Interesting)

TheRaven64 (641858) | more than 5 years ago | (#26073877)

General purpose CPUs are quite bad for video compression. A DSP or GPU is generally laid out in a way that maps more closely to the algorithms. I'd be interested to see what performance ffmpeg gets once they've finished optimising it for the DSP in the OMAP3530 (for reference, the entire BeagleBoard system built around one of these uses 1.8W - less than just the CPU of Intel's 'low power' systems and include the ARM Cortex A8 core, an OpenGL ES 2.0 GPU and a DSP).

Re:Normal people don't need faster computers (2, Insightful)

repvik (96666) | more than 5 years ago | (#26073671)

Until the next version of Windows is out...

Seriously though. Of course the top-of-the-line, state-of-the-art, bleeding-edge PC's are irrelevant for the general populace when they are released. That doesn't mean that they're irrelevant to the general populace in a year or two.
When the next Windows is released, some new fancy games are released, websites are even more riddled with flash, java and whatever new tech they come up with to use more resources.

Re:The new ones are impressive (1)

KingMotley (944240) | more than 5 years ago | (#26072367)

BTW, you can move your drives from one motherboard to the next so long as the raid is/was done via an intel raid controller. I've moved my complete OS from other motherboard to another with a different chipset with no problems, and that was on a 4-drive raid-0.

It was from a ICH6R to ICH8R I believe. Of course if you went from an nvidia/amd chipset to an intel one, then you can't. Unless the raid was done via an add-in card, of course.

Re:The new ones are impressive (1)

skolima (1159779) | more than 5 years ago | (#26072775)

You are using RAID0, and on four drives at once? Is you data really that disposable?

Not surprising. (3, Interesting)

pclminion (145572) | more than 5 years ago | (#26070997)

At WinHEC 2008 the Intel speakers continued to hint at the fact that they had operating, packaged cores at this size. On track for manufacturing? More like they've been making it for 9-12 months already. At any rate, it's cool, though not surprising.

Re:Not surprising. (4, Insightful)

Devil's BSD (562630) | more than 5 years ago | (#26071417)

I think they meant more that they're on track to scale it up for mass production at volumes that will hopefully meet the demand. I'm glad they're on target, I'm looking forward to Westmere (the 32nm Core i7 that will hopefully make it to mobile platforms by the end of next year).

Nm (5, Funny)

Anonymous Coward | more than 5 years ago | (#26071021)

Newton-metres? You mean Joules?

What could possibly make you confuse N which is a symbol for Newton with n which is a prefix for nano.

You're definitely not geeky enough.

Captain Metric to the rescue (3, Informative)

Anonymous Coward | more than 5 years ago | (#26071123)

For this reason the SI standard dictates that metric units such as "km" or "nm" are never capitalized, even on a sign that is written ALL-CAPS [ltsa.govt.nz] .

Re:Captain Metric to the rescue (1)

Stormx2 (1003260) | more than 5 years ago | (#26072033)

Of course. Because capitalisation changes the meaning in some cases, e.g. nm -> Nm

Re:Captain Metric to the rescue (1)

LiLWiP (918943) | more than 5 years ago | (#26074395)

You know what they say... Capitalization is the difference between "I helped my uncle Jack off a horse." and "I helped my uncle jack off a horse."

Re:Captain Metric to the rescue (0)

Anonymous Coward | more than 5 years ago | (#26072057)

What you mean to say is that capitalism is unacceptable with the French system of measurements...

Re:Nm (5, Funny)

Anonymous Coward | more than 5 years ago | (#26071241)

It's nautical miles. The chips are gigantic. Marvels of engineering.

Re:Nm (1)

n3tcat (664243) | more than 5 years ago | (#26072801)

So quantum computing would be giving your bytes to a group of people and sending them across the chips without a routing map. They may or may not arrive at the end point before starving to death!

Re:Nm (1)

McWilde (643703) | more than 5 years ago | (#26072717)

As a measure of torque, the Newton meter is more common than the Joule.

And yet. (1)

mail2345 (1201389) | more than 5 years ago | (#26071031)

At some point, it will stop getting smaller.

Re:And yet. (4, Funny)

Anonymous Coward | more than 5 years ago | (#26071187)

At some point, it will stop getting smaller.

As opposed to the more common problem where it stops getting bigger.

Re:And yet. (1, Funny)

Panspechi (948400) | more than 5 years ago | (#26071351)

That's what she said!

Re:And yet. (0)

andersa (687550) | more than 5 years ago | (#26072141)

At some point, it will stop getting smaller.

That's the point where you have achieved 'significant shrinkage'.

Re:And yet. (1)

FTWinston (1332785) | more than 5 years ago | (#26073195)

Indeed. They probably won't be able to get more than a few generations smaller before quantum fluctuations become too signficant. By the time your transistor is on the scale of only a few dozen atoms, its going to stop being a transistor some of the time, and do its own damn thing.

42 32 nm.. (-1, Offtopic)

deathguppie (768263) | more than 5 years ago | (#26071049)

ooooooh... it says nm, not m&m.. my bad ..

Re:42 32 nm.. (3, Funny)

Anonymous Coward | more than 5 years ago | (#26071425)

Intel: I'm a chip company. I make chips, that's all I'm programmed to do.
AC: Were you any good?
Intel: Are you kidding? I was a star. I could make a chip to any size. 30 nm, 32 nm, you name it. 31... But I couldn't go on living once I found out what the chips were for.
AC: What for?
Intel: MacBooks.

Chipsets (5, Interesting)

lobiusmoop (305328) | more than 5 years ago | (#26071061)

It's great that Intel are working on die shrinks for their processors, but I wish they would do the same for their support chipsets. It's annoying that on most laptops the northbridge for Atom processors uses more power than the processor does.

Re:Chipsets (5, Interesting)

Anonymous Coward | more than 5 years ago | (#26071243)

This should be partially alleviated once the i7 architecture is fully adopted. Pretty much no more north bridge. That's probably why they're neglecting the current chip set technology with more aggressive updates.

And who knows, if a better chip interconnect comes around in the next generation (unlikely, but possible), Intel could start putting more and more in the CPU package. Things like a Larrabee GPU and south bridge functionality (audio, networking, general I/O). System on a chip is common place in embedded systems now. If Intel wants to eat ARM's lunch they're going to have to adopt some of the same techniques.

Re:Chipsets (1)

TheRaven64 (641858) | more than 5 years ago | (#26073947)

The divide between north and south bridges has not really existed for a few years, but there are quite a few things that are in the supporting chipset and not in the CPU for Intel systems. Compare this with a real low-power chip, like the OMAP3530, which has USB, and disk / flash controllers, a GPU and a DSP on die, and RAM and flash stacked on the package, so you don't need much else to make a complete system with a power dissipation of 1.8W (and faster than the computer my mother uses).

Re:Chipsets (3, Insightful)

zonker (1158) | more than 5 years ago | (#26071339)

Very true. The problem is that chipsets don't sell computers like processors do. Joe Shopper at WalMart doesn't know what a northbridge is but he has some understanding of what a Core 2 Duo is.

Re:Chipsets (5, Insightful)

Anonymous Coward | more than 5 years ago | (#26071673)

That's entirely a marketing issue.

Joe shopper doesn't know what a core 2 duo is any more than he knows what a northbridge is. The only difference between the two is there are millions of dollars poured into making sure Joe recognized the term "core 2 duo". He still doesn't know a damn thing about it.

Computers are funny from a marketing standpoint. They are purchased by people that don't know anything about them. Sold by people that don't know much about them and supported by people that don't even speak the same language. (often literally).

Even more interesting, they are the only consumer device I know of where there is very little difference between first and third party parts. Obviously the technical specs change, but the average computer buyer wouldn't know the difference if you highlighted it in red.

Selling computers therefore is a the most perfect example of marketing at work. Your customer doesn't know ANYTHING about the product in question, and so wants the one that he's heard the most about. So the customer buys what is best advertised.

Re:Chipsets (1)

Yarhj (1305397) | more than 5 years ago | (#26072167)

Interestingly enough, the primary goal of die shrinks is not better performance, but lower cost. If a given die can be shrunk by a factor of k, we can fit roughly k^2 devices on a wafer of the same size. If the smaller chips work just as well as the larger chips we can then turn around and sell them for exactly the same price. It's like printing money(Step 3: PROFIT!). Of course, there's the expense in R&D and equipment to consider as well (Step 2: ????), but the basic reasoning is sound. If our competitors are stuck with a bigger chip, we win. The largest semiconductor companies are pushing for 450mm Silicon wafers for the same reason.

Of course, smaller dimensions do enable certain performance enhancements (maximum device frequency comes to mind), but it takes a lot of work to get these smaller devices working.

Re:Chipsets (0)

Anonymous Coward | more than 5 years ago | (#26074831)

Blue tooth and 802.11 are radio things
Here's why the property areas cant be smaller
. Radio requires capacitors , inductors crystals anbd internal wire routing and lengths necessary for the radio frequency band used .
  Making them physically smaller or larger changes their characteristics and renders them useless for the task. . This is not the case with Chips which are semiconductors. So making radio stuff smaller can be impossible as Inductors and capacitors and wiring and Crystals may need to be external to the device or be of a specific size inside the chip.

It's a question of manufacturing capacity (1)

brucmack (572780) | more than 5 years ago | (#26075197)

Really, this is just a matter of having limited manufacturing capacity. Every time they create a new manufacturing process, they have to upgrade a factory to use it. This puts the factory out of service for however long it takes to roll out the new tech, and costs billions of dollars in the process. In other words, even Intel doesn't have the resources to upgrade all of their factories at once.

Instead, they take one or two factories running the oldest tech, and upgrade them. Once they are ready, they start manufacturing the high-end processors. The last-generation tech manufactures lower-end processors. The generation before that manufactures chipsets, graphics chips, etc. The generation before that manufactures DRAM / flash / whatever else is needed. This is just an example, I have no idea what the split is in reality.

So from Intel's perspective, they are always using the newest manufacturing tech for their most important products (high performance, high profit margin). This in turn gives them the capital needed to develop their next manufacturing process, and the cycle continues...

GET DOWN! (-1, Offtopic)

Anonymous Coward | more than 5 years ago | (#26071071)

£00|{, Ð00Ð2, 1n73£ 1$ 7h3 £337

Point of Diminishing Returns? (3, Interesting)

Anonymous Coward | more than 5 years ago | (#26071095)

Am I the only one feeling we might have reached the point of diminishing returns, at least for desktops, in the last 2-3 years. All the shrinkage past 90 nanometers just feels underwhelming. Stuff beyond Pentium 3 has not been revolutionary, performance wise, for a desktop.

Re:Point of Diminishing Returns? (3, Interesting)

sunami (751539) | more than 5 years ago | (#26071113)

Yea, there's a pretty big wall that's been hit in terms of clock speed, which is why multiple core processors is the direction instead of ramping up speeds.

Re:Point of Diminishing Returns? (1)

Spatial (1235392) | more than 5 years ago | (#26073413)

"Speed" as in performance? No. A 3Ghz P4 is a shitload slower than a 3Ghz Athlon X2, which in turn is a shitload slower than a 3GHz Core 2 Duo. The per-core speed of Desktop CPUs has never stopped increasing.

Re:Point of Diminishing Returns? (2, Funny)

Anonymous Coward | more than 5 years ago | (#26071229)

Tee-he-he-he, you said "shrinkage". (nothing to see here)

Re:Point of Diminishing Returns? (1, Informative)

ColdWetDog (752185) | more than 5 years ago | (#26071271)

Am I the only one feeling we might have reached the point of diminishing returns, at least for desktops, in the last 2-3 years. All the shrinkage past 90 nanometers just feels underwhelming. Stuff beyond Pentium 3 has not been revolutionary, performance wise, for a desktop.

I see we haven't been using Adobe software. Or Windows. Or Crysis. Or Slashdot's CSS 'implementation'.

But if browsing Usenet with Lynx is where you're out, more power to you.

Re:Point of Diminishing Returns? (1)

Corporate Troll (537873) | more than 5 years ago | (#26073269)

Slashdot's CSS 'implementation'.

I know it's offtopic, but I don't think it's their "CSS implementation", but rather their Web 2.0 implementation with all the JavaScript that comes with it.

On my Asus EEE PC 701, I pretty much always get the warning that a script is running and offers to kill it. I always do. On my wifes 5 year old PC, which is plently fast for pretty much anything we do, slashdot freezes the browser about 2 seconds for a reload. Annoying, but not that problematic. The behaviour with the EEE PC pisses me off though

Re:Point of Diminishing Returns? (0)

Anonymous Coward | more than 5 years ago | (#26071353)

Negative ghostrider, shrinking the minimum feature size reduces the required die size. Reducing the die size means more die per wafer for approximately the same processing cost. End result is more for less. Unfortunately the required immersion lithography is crazy tough, very expensive, and very low throughput. Good luck Intel.

Re:Point of Diminishing Returns? (2, Interesting)

NerveGas (168686) | more than 5 years ago | (#26072337)

Anything past the P3 may not have been revolutionary, but it's steadily progressed quite nicely.

I have a dual 1.4GHz P3 system, and a 1.6GHz Core Duo. The Core Duo is *much* faster, and that chip is already outdated. Not to mention the fact that it's comparing the fastest P3s made to the lowest of the Core Duo lineup.

People also forget about things that can't be measured in nanometers or gigahertz, like the advances that have greatly lowered leakage current. Without them, something like 85% of the power used in the 32nM chips would be leakage, and liquid cooling would be an absolute necessity.

Also, these advances allow Intel to make modest chips VERY cheaply... like the Atom. I've got a micro-atx board with one on it, and considering that the entire board+cpu only cost $65, it is an AMAZING performer.

Re:Point of Diminishing Returns? (1)

KingMotley (944240) | more than 5 years ago | (#26072427)

Actually, I've felt that way ever since the 486. When the first 486's came out, they ran at 25MHz. You could keep your entire system for years, and then upgrade it to the 100MHz 486DX4. That's a 4 fold speed increase without needing to change motherboards. Today, we are lucky if we see a 20% speed increase (above the inital released speeds) before the next bump requires a different chipset and motherboard.

We just aren't seeing the large speed bumps like we used to. Instead of going from say 25Mhz to 33 (a 32% increase), we see 2.66 to 2.93 (a 10% increase).

Re:Point of Diminishing Returns? (1)

Daengbo (523424) | more than 5 years ago | (#26073337)

Really, though, that 486 to 486DX4 conversion was the point where multipliers came into the scene, and suddenly 4X the processor speed didn't equate to 4X the system speed because suddenly all the other components weren't keeping up and became the bottleneck. Do you remember how disappointing that was?

Re:Point of Diminishing Returns? (1)

TheRaven64 (641858) | more than 5 years ago | (#26074001)

Intel have a test system where they can simulate new architectures. It lets them adjust every parameter, so they can simulate things that aren't even remotely feasible. A few years ago, they simulated an infinitely fast CPU in it, and ran some benchmarks. The total system speed was around 3-4 times faster than it was with the original design. For any given computer, doubling the CPU speed means that, unless you upgrade a lot more than the CPU, all you've done is moved the bottleneck somewhere else. This is why the 486 was the last system to do that, and why there wasn't a DX5.

Re:Point of Diminishing Returns? (0)

Anonymous Coward | more than 5 years ago | (#26072823)

Stuff beyond Pentium 3 has not been revolutionary, performance wise, for a desktop.

Dual Core + 5x speed increase per core compared to typical PIII speeds makes a hell of a difference when you're doing something like rendering a DVD iso (e.g. music video compilation - fairly basic requirement) AND want to play some music AND have your web browser responsive. And there's still enough CPU left to rip a CD. On my previous PIII equivalent I could only do one or two of these things at once (and it took about 8 hours to render a 30minute VCD as opposed to 1-2 hours to render a 3 hours DVD now).
This sort of requirement isn't particularly nerdy or specialist these days. Lots of people do basic video processing.

Re:Point of Diminishing Returns? (1)

darkwhite (139802) | more than 5 years ago | (#26075125)

Stuff beyond Pentium 3 has not been revolutionary, performance wise, for a desktop.

It has. You've been living under a rock.

What about AMD? (1)

blackyottabyte (1422241) | more than 5 years ago | (#26071233)

If Intel is able to shrink its die size every 12 months AMD is in trouble. A more efficient design is usually beaten by a less efficient design fabricated in less space. That is if you think AMD's design is still more efficient.

Re:What about AMD? (5, Informative)

vsage3 (718267) | more than 5 years ago | (#26071285)

If Intel is able to shrink its die size every 12 months AMD is in trouble.

For what it's worth "tick-tock" is actually alternating between a new architecture and a process shrink every 12 months. "Q4" in the summary means Q4 2009.

Am I the only one feeling we might have reached the point of diminishing returns, at least for desktops, in the last 2-3 years. All the shrinkage past 90 nanometers just feels underwhelming. Stuff beyond Pentium 3 has not been revolutionary, performance wise, for a desktop.

I hate to be snarky but you sound like one of those people who bought the crap about the "Megahertz Myth". Processor clock rate has little to do with performance. I'll agree that pentium 4 was underwhelming, but Core was a huge hit and saw huge performance, especially toward the ones that were released in early this year that used the high k dielectric.

Re:What about AMD? (1)

benthurston27 (1220268) | more than 5 years ago | (#26071529)

The "megahertz myth" is that processor clock rate has a lot to do with performance, It seems to me like his post suggests that he didn't buy into it not being impressed with the pentium 4. But the shrinkage is definitely important, as in being able to fit more than one of a modified older design like the pentium III on one chip.

Re:What about AMD? (4, Interesting)

afidel (530433) | more than 5 years ago | (#26071573)

Actually I think the biggest post P3 improvement has been the move to dual core as standard on the desktop in the last couple years. At least on Windows the non-blocking nature with a stalled thread is huge for overall system performance and UI snapiness. It's great to be able to get those benefits without a $200 motherboard and two CPU's =)

Re:What about AMD? (1)

Firethorn (177587) | more than 5 years ago | (#26074379)

Being that I have a tendency to run a few pieces of software that'll peg a CPU to 100% today, going to a dual core processor was a 'I LOVE THIS!!!' moment.

I went with a dual core for the higher individual core speed and that games were, on the whole, still not optimized for using multiple cores, so the best you could get is the game on one core and everything else on the second, which STILL wouldn't be strained. Of course, prices come down, performance goes up, software advances, I'd consider a quad today.

Obligatory Shrinkage Comment (1)

blackyottabyte (1422241) | more than 5 years ago | (#26071609)

This is one case where shrinkage is damn good.

Don't take that out of context.

Oh, boy. Here comes the nostalgia again. (1)

jcr (53032) | more than 5 years ago | (#26071659)

In my day, getting to one micron feature sizes was a big deal. And we were grateful!

You kids get off my lawn!

-jcr

Re:Oh, boy. Here comes the nostalgia again. (0)

Anonymous Coward | more than 5 years ago | (#26072513)

What's a 'micron', USian?

Re:Oh, boy. Here comes the nostalgia again. (1)

jcr (53032) | more than 5 years ago | (#26072671)

What's a 'micron'

It's a term used in semiconductor manufacturing to refer to one millionth of a meter.

-jcr

Re:Oh, boy. Here comes the nostalgia again. (1)

TheRaven64 (641858) | more than 5 years ago | (#26074033)

I have an issue of Byte up in my attic where the process shrink to 1 micron is the cover story. I read it a couple of years ago, around the time of the 65nm process shrink. It really gives you a sense of the speed at which process technology is improving.

Intel (3, Funny)

IDKmyBFFJill (1428815) | more than 5 years ago | (#26071781)

It's all about splitting hair nowadays

So long to the competition... (4, Insightful)

NerveGas (168686) | more than 5 years ago | (#26072245)

Intel has always enjoyed a much better manufacturing technology than AMD. But, Intel made some stupid architectural decisions with the P4 architecture.

Once Intel came out with the Core series, then the combination of a decent architecture and terrific fab capabilities really started eating away at AMD. This will only continue the rally.

The sad thing is that this will actually be a step back in pricing... it's getting back to where AMD simply cannot touch the higher-end Intel territory, and so Intel is back to enjoying terrific profit margins on those chips.

So long to the competition...Race...relations. (1)

Ostracus (1354233) | more than 5 years ago | (#26072673)

I think AMDs strategy is overclocking and lots of it. Look at what it's introducing in it's latest and upcoming hardware. Features that make overclocking easier. Also I wouldn't count AMD out too soon. Amd is just one design correction away from having perfect hardware for HTPCs And their IGP is still better than Intels.

Re:So long to the competition... (1)

maxume (22995) | more than 5 years ago | (#26074983)

As long as Intel is willing to enjoy a lower profit margin on their middle of the road chips, who cares?

I'm not a gamer, so my 2+ year old computer is fast enough about 98% of the time. I'm not willing to pay a great deal of money to bump that to 98.5%, or even 98.1%. I get that there are plenty of people who just need more speed, but that groups keeps getting smaller.

Smaller better? (0)

Anonymous Coward | more than 5 years ago | (#26072305)

Could someone explain to me hoe smaller chips are better? Doesn't that decrease the surface area and make it even harder to dissipates heat from the chip?

Re:Smaller better? (1)

Kane Devaid (1339253) | more than 5 years ago | (#26074401)

The smaller the feature size, the faster the chip can be pushed. Alternatively, the chip can run at the same speed with less power. Die shrinks are the reason for the large speed increases all through the nineties. The problem is, the smaller you go the harder they are to make.

Amazing...Grace. (1)

Ostracus (1354233) | more than 5 years ago | (#26072727)

"Intel is obviously betting that its rapid-fire advancements will produce performance gains so jaw dropping that customers can't resist.""

Two things. One it doesn't matter how awesome your hardware is. If the majority can't afford it then it doesn't matter? Second as Microsoft is learning prior success can be a barrier to future growth. How many are going to throw out their Core 2 Duos in order to have the most amazing hardware from Intel?

So can someone summarize the current state? (1)

DG (989) | more than 5 years ago | (#26072961)

I found out from my wife that our home server died and won't reboot. AMD Athlon 3200+ running Fedora.

It is almost certainly a hardware problem, and that server has been running 24/7 for years now... time to upgrade.

My hardware philosophy has been to buy big and milk it for a long time. You pay more up front for that power, but the fact that it has power means it doesn't get obsoleted immediately either.

So then, cut through the marketing crap. Assume a desktop PC purchase in the May-ish time frame, to run Linux. What is likely to be the way to go from a hardware perspective?

DG

Re:So can someone summarize the current state? (0)

Anonymous Coward | more than 5 years ago | (#26073463)

If you wanted something now, I'd say grab a Q6600. Undervolt and/or underclock it if you don't need so much power, but the quad-core and 8MB L2 gives some potential for next several years.

Re:So can someone summarize the current state? (1)

TheRaven64 (641858) | more than 5 years ago | (#26074157)

If it's left on all of the time, you might want to consider low power, rather than high speed. A 45W chip will use 395 kilowatt hours [google.com] per year, which will cost me around around £60 (around $90, now the pound's collapsed against the dollar, around $120 two months ago) per year to run. Something like the BeagleBoard consumes only 1.8W. For a home server, you'd want something in the middle (actually, I'd like a BeagleBoard with 4 SATA connectors, but, sadly, it doesn't exist). Depending on how much you're taxing it, I'd be inclined to recommend the AMD Geode line - they're an order of magnitude more power-thirsty than ARM or MIPS offerings, but they're x86 and well-supported by commodity software.

Re:So can someone summarize the current state? (1)

jps25 (1286898) | more than 5 years ago | (#26074225)

If it's running 24/7 and your old Athlon 3200+ was good enough, then pick any current dual or quad desktop CPU with the lowest energy usage.
Pick a motherboard with an IGP and plenty of SATA2.
Throw in 8 or 16GB RAM and a couple of hdds.
Most importantly, check silentpcreview.com so you know which case to buy and how to silence it.

Re:So can someone summarize the current state? (1)

sssssss27 (1117705) | more than 5 years ago | (#26074851)

I just rebuilt my server using an Intel Atom dual core and a 4 port SATA PCI card from my old server. I'm running Linux with three 750 GB HDDs in RAID-5. Have an 80$+ PSU and a case with a 240mm top fan so it's pretty quiet as well. I just rip all my media to it so I can access it from anywhere in my house or through the Internet. It's worked well for me.

nm, not Nm (1)

the_other_chewey (1119125) | more than 5 years ago | (#26073933)

Stupidly uppercasing everything in a headline will regularly backfire if scientific units are used.

What's meant here is nanometer, not Newtonmeter - which, by the way, is equal to Joule.

And now here I am, unable to think of a good pun about a 32 Joule chip...

Intel chip (1)

Andypcguy (1052300) | more than 5 years ago | (#26073961)

What about not only putting the more cores on die bit also RAM and the chipset. As said the motherboard would be different from anything we've seen so far. Would this result in any meaninful gains compaired to the cost and limites on options?

Jaw Dropping? (1)

frenchgates (531731) | more than 5 years ago | (#26075163)

Can anyone remember the last time an incremental advance in chip speed was anything close to "jaw dropping?" Having been in this industry a while I can't count the number of times people like Steve Jobs and Andy Grove claimed speed increases of more than double with almost no apparent effect on anything but benchmarks. The early days of 3D accelerators was about the only time I really went "wow!"
Load More Comments
Slashdot Login

Need an Account?

Forgot your password?
or Connect with...

Don't worry, we never post anything without your permission.

Submission Text Formatting Tips

We support a small subset of HTML, namely these tags:

  • b
  • i
  • p
  • br
  • a
  • ol
  • ul
  • li
  • dl
  • dt
  • dd
  • em
  • strong
  • tt
  • blockquote
  • div
  • quote
  • ecode

"ecode" can be used for code snippets, for example:

<ecode>    while(1) { do_something(); } </ecode>