Intel reveals its first 450mm wafer
- Thread starter fellix
- Start date
Cost prohibitive at this point. Besides, this is an increase of - edited - (from 300 mm to 450 mm) which is quite impressive, although wafer prices will go up. I doubt they will upgrade all their fabs any time soon, as it would basically give them capacity issues in the sense that they would have too much wafer capacity.
No equipment exists for so large wafer sizes (larger than 450 mm), Intel had to basically drive the technology needed forward (and spend heaps of cash in that process) in collaboration with 3 other companies.
Last edited by a moderator:
OpenGL guy
Veteran
Actually, a 450mm wafer has more than twice the area of a 300mm wafer.
It's impressive to think that 99.7% of the features on the chip are within a 1.2nm range of their target.
It's going to be an interesting transition to 450mm wafers, since so few players can afford it. If the 300mm transition is a guide, likely fewer of that set are going to transition without significant problems with yields or variability.
It's going to be an interesting transition to 450mm wafers, since so few players can afford it. If the 300mm transition is a guide, likely fewer of that set are going to transition without significant problems with yields or variability.
No wonder Intel's cap ex. was through the roof last year at $12 billion and another $13 billion planned for 2013
Reminds me of this scene . Intel personified by Conan the Barbarian.
Cheers
Reminds me of this scene . Intel personified by Conan the Barbarian.
Cheers
Last edited by a moderator:
I won't open a new thread and there is no "general" thread here dedicated to Intel so I use that one.
I 've read quite a few papers from financial analysts about Intel and they are more than often quite conservative if not slightly pessimistic. I don't get it. I don't get how for example Qualcomm market capitalization managed to exceed Intel one.
Is it me that is biased or the financial people that fail to see the whole picture?
My POV is that Intel technological lead has never been that overwhelming:
*They are a node ahead of everyone and just added eDRAM to their port folio, not only that they did it somehow with panache beating IBM to it and deploying it in a part that could address sane volume. They should hurt Nvidia and AMD pretty badly raising their margins while doing so.
*They should be ahead of everyone when it comes to 450mm wafer.
*They have now tech in RF that will allow them to put their fab capacity to use (like the eDRAM) and They should have the silicon budget to integrate that tech in their chip (if not now at 14nm).
*They are ahead of everybody when it comes to do anything with silicon. From the CPU cores, to the "uncore" to the memory controllers, caches, nobody can touch what they are doing. Now they even have great GPUs.
I do get that technical merit goes only that far and that they have tough battles ahead on the mobile market though I do not get why the "market" is so conservative in its forecast.
I simply don't get it
from my POV there lead is overwhelming and looking at their track record I would be surprised if they can make something out of it. To me analysts are wary at the very moment (or close) their offering is about to come altogether (say in the next couple of years).
I 've read quite a few papers from financial analysts about Intel and they are more than often quite conservative if not slightly pessimistic. I don't get it. I don't get how for example Qualcomm market capitalization managed to exceed Intel one.
Is it me that is biased or the financial people that fail to see the whole picture?
My POV is that Intel technological lead has never been that overwhelming:
*They are a node ahead of everyone and just added eDRAM to their port folio, not only that they did it somehow with panache beating IBM to it and deploying it in a part that could address sane volume. They should hurt Nvidia and AMD pretty badly raising their margins while doing so.
*They should be ahead of everyone when it comes to 450mm wafer.
*They have now tech in RF that will allow them to put their fab capacity to use (like the eDRAM) and They should have the silicon budget to integrate that tech in their chip (if not now at 14nm).
*They are ahead of everybody when it comes to do anything with silicon. From the CPU cores, to the "uncore" to the memory controllers, caches, nobody can touch what they are doing. Now they even have great GPUs.
I do get that technical merit goes only that far and that they have tough battles ahead on the mobile market though I do not get why the "market" is so conservative in its forecast.
I simply don't get it
from my POV there lead is overwhelming and looking at their track record I would be surprised if they can make something out of it. To me analysts are wary at the very moment (or close) their offering is about to come altogether (say in the next couple of years).
Yes, if I were loaded I would be buying the shit out of Intel stock. They will continue to dominate their current markets, and they are in a great position for huge expansion in the mobile sector in the coming years. Atom is just not quite there yet.
By the way, clicking through to that link in the OP and seeing that enormous 450mm wafer made me realize just how far we've come since the early 1980s. Didn't they use like 60mm wafers back then? You got a couple dozen CPUs out of each at best... Whereas on this one at today's integration level, each sqmm of that giant wafer could hold over 100 MC68000s. Holy crab! (Not that you could ever hope to wirebond them though... 
)
)I do get that technical merit goes only that far and that they have tough battles ahead on the mobile market though I do not get why the "market" is so conservative in its forecast.
I simply don't get it
Right now QCOM's market cap is a bit lower than INTC, although they are on comparable levels. If you consider their P/E ratio, I'd say that they are all too low. Heck, considering that TSM also has a similar market cap, that's really weird.
However, I also understand why the market is so conservative about INTC. Intel is now still heavily invested into x86. With PC's market share declining, Intel will need to successfully transform at least some of its core business in order to sustain its current technology advantages.
A great buying opportunity, IMO.
Once Intel integrates their LTE modem into Silvermont derived SOCs, share price is likely to rise sharply.
Cheers
Mintmaster
Veteran
I think the problem is that while Intel is doing fantastic work with its process technology, the PC market and demand for high performance in personal computing processors is falling. That's going to prevent them from exceeding expectations by much, which is what's needed for a stock to go up in value.I do get that technical merit goes only that far and that they have tough battles ahead on the mobile market though I do not get why the "market" is so conservative in its forecast.
I simply don't get it
I don't know if Qualcomm's price is justified, but if Intel executes well on mobile, I'd bet more on Qualcomm's profit sliding than Intel's going up. There's too much competition in the mobile sector for Intel to make a bundle there. Remember when ATI came out with the HD4000 series, leaping past NVidia in perf/$? That wound up killing NVidia's margins instead of boosting ATI's.
Looks like they will introduce 450mm wafers at the same time as 10nm.
450mm wafers in the 10nm node -- the technology required to pull that off properly is simply astounding. Imagine how many quadrillions of transistors you could cram onto a 450mm wafer at 10nm gate length.
Not more astounding than the amount of money required.