Slashdot Mirror
Intel Makes 45nm Chip
dolphinlover writes "Intel announced today that it created its first microchip using the 45 nanometer manufacturing process that it says will go into its processors in the second half of 2007. Intel said that this development provides it with a 'considerable lead over our competitors in the 45-nanometer generation'."
Intel said that this development provides it with a 'considerable lead over our competitors in the 45-nanometer generation'."
Which means, what?
Predicitons for the next 18 months:
i think it's somehow related to moore's law
A feeling of having made the same mistake before: Deja Foobar
Perhaps this what Steve Jobs referred to when he talked about the efficiency of future chips in Intel's roadmap?
Yes, by announcing that we have made one chip at 45nm, we obviously win! ...nevermind that it probably doesn't actually run anything. We haven't made a motherboard for it yet.
Whaaa? n.m.? Nano Meee....whaaa??
Oopps! Sorry!
I'll believe it when they start yielding these things at greater numbers than one, on chips with a high SRAM and logic density.
Unless I misplaced a decimal point or misunderstand physics, isn't 45 nm only a very few generations from needing connections only one molecule thick?
Not really. Creating a >2000 mm^2 chip without any flaws, with the expectation of being able to eventually make a profit on them in the consumer market, would be quite an accomplishment. Such a large die area would not only result in low yields, but present serious obstacles in power consumption and heat dissipation.
For comparison, the Pentium IV 600 series has a 135 mm^2 die area. If I'm not mistaken, yields fall exponentially, so no only would they be able to produce only 6% as many chips on a wafer, but could also expect a greatly lower number of working ones from the total.
"National Security is the chief cause of national insecurity." - Celine's First Law
It just goes to show that design does play a part in making a chip, and not trying to cram as many transistors as one can onto a die.
My 3D Texturing Skinning work (under construction)
Intel's marketing campaign: Smaller than AMD!
Wait...
TO START
PRESS ANY KEY
Where's the 'ANY' key? I see Esk, Kitarl, and Pig-Up...
- AMD's new fab, Fab 36, supports 300mm wafers (like Intels have for some time).
- It uses a 90nm process (Intel and IBM have been on 65nm for some time).
- It will transition to 65nm by the end of 2006.
- It will use 45nm and 32nm processes by the end of the decade.
It doesn't really sound like Intel is playing catch-up here.I am TheRaven on Soylent News
molecule? This is a crystal we are talking about, so the entire wafer is a "molecule". An atom of Si is about .3nm across.
"Hey, Intel's making 45nm chips!"
"Yum, what flavour?"
"Er... Internets?"
Seriously though, I know this is a step forward, but someone tell me when either vendor starts actual production on these chips
Most molecules are a few to a few dozen angstroms thick (from here), and 45 nm is 450 angstrom. So there is about another factor of 10 till we get down to the size of complex molecules. However, I do believe that most of the "stuff" used in the manufacture of chips are either pure elements or simple molecules, which are much smaller (varying from 1 to 5 angstrom)..
Name: Mr. Anon E Mouse; SSN: 555-55-5555
It seams to make sense that because Intel has the most money, that they can spend money on developing better manufacturing and engineering techniques than their competition. But with all of this extra money, and seamingly having better technological capabilities, AMD is still beating out Intel as far as performance.
Looks like Intel basically does all of the hard work figuring out how to do things for the first time, and AMD just has to wait until Intel is finished and then just learn from them. I of course know nothing about how to make processors, but it seams that this is the most plausible reason why Intel has trouble making chips that are as good as AMD.
This news about the 45nm manufacturing looks very bad for AMD, but I doubt it will matter very much. If Intel is doing it by the end of 2007, AMD will probably be doing it by first or second quarter 2008. And if history is any indicator, they will probably be doing it better. But I guess time will tell, maybe this 45nm technique really is too hard for a company without endless money to figure out.
--
-- All that is necessary for the triumph of evil is that good men do nothing. -- Edmund Burke
From one molecule thick? We're far from that.
But we ARE only a few more generations from hitting a rather thick wall: at the 5nm, electrons begin jumping _through_ the insulators to a nearby circuit. So while we're far away from the molecular level, we're still getting closer and closer every day to a very real limit. We should be able to push it down to 4nm with a little extra engineering....but as far a I know, thats going to be it. Anyone else want to comment?
Intel's logic development is striving for a two-year cycle for each new process technology. This announcement of functional first-silicon (who knows how long they've actually had it) is part of that natural progression. Here's a table showing this announcement along with previous SRAM test chip announcements:
... it's not a table...
Process
Litho
Size
Date
P860
130 nm
18 Mbit
Mar 2000
P862
90 nm
50 Mbit
Feb 2002
P1264
65 nm
70 Mbit
Apr 2004
P1266
45 nm
153 Mbit
Jan 2006
Okay
Mr. Scott. To put it another way -- how big would one of your gates have to be, with a 300 mm wafer, to resolve properly using your current method of lithography?
Intel Exec. That's easy. Six molecules. We have stuff that big in stock.
Mr. Scott. Well, suppose I could show you a way to build a gate that could do the same job -- but be only one molecule thick. Would that be worth somethin' to ye?
Intel Exec. You must be joking.
Dr. McCoy. Perhaps the professor could use your computer...
[Later]
Dr. McCoy. [Whispering] You realize that by giving him the formula we're altering the future.
Mr. Scott. How do we know he didn't invent the thing?
Dr. McCoy. [Smiling] Yeah.
Just to make sure, this is not a CPU chip using 45nm technology. This is a test vehicle which contained SRAM (static RAM) and some control logic. SRAM arrays are regular and don't have the same complexity as ALU (arithmetic logic unit) and other control circuits found in CPU. So yes this is a big step because it is gives some indication about how complicated will it be to get a good yield in this process. Also note that SRAM arrays can be easily made defect tolerant by using spare rows/columns. Same is not true for CPU cores. So there is still al long way to go before an efficient working CPU with production acceptable yield is available.
Seriously? Can't I have a chip that runs relatively fast, does everything a modern computer is used for, sans games, and I *don't* have to water-cool? Something like what the VIA Epia series does, but with Intel's backing?
Is it just me, or is web-browsing and document writing fast enough? It seems like 99% of the time these days I just want something smaller and quieter. If I want pretty shiny games, I'll play them on my xbox390 or sumsuch. Sure you can make bunches of chips for gamers, but give me a slimline chip and I, like many others would flock to it.
I'm writing this on my 733Mhz laptop, bought for college way back when, and my typing fingers really don't recognize the lack of dual cores.
-- I have fans? Wow.
I made a 45nm chip meself, but I sneezed and I haven't been able to find it since.
I once had a conversation with someone who was doing developmental research for an even smaller process for some very large semiconductor manufacturer. According to him, they were one day running some measurements on the first prototype wafers. From experience with every previous process (65, 90, 130, etc.), they were expecting this particular measurement to yield a nice bell curve. Instead, they got a strongly quantized bell curve: it looked more like a histogram. The reason, they realized, was because the gate oxide was becoming only a handful of atoms thick. The quantization between two steps in that bell curve was the difference between, say, a gate oxide 5 atoms thick and 6 atoms thick.
So, yes, they are indeed coming up against some real physical limitations for CMOS technology. Of course, people have been saying that for years.
Why does the media insist on using the term "Moore's Law"???
Since when do self-fufilling prophesies become law?
Self-fufilling prophesies tend to restict one's actions rather than sustain them.... Which is why superstition is harmful....
If every PHB believes in Moore's quip, then do people get fired for not doubling # transistors every 18 months? Do they get a bonus for doubling the # of transistors in 17 months?
Perhaps if they weren't so darn busy cramming more transistors on the chip, they could better improve their compiler or come up with truly innovative architectural techiques that *work*....
How big is the bowl of dip? Can't have chips without dip.
today is spelling optional day.
The last thing you want is to be bringing the 45nm fab online as 65nm is reaching it's limits, only to find you need another 9 months of working the bugs out before you get useful yeilds.