Do you believe that Intel will lower its Margins for Broadwell ?? Because if they would sell low volume of Broadwell CPUs now at the same price as Haswell, then they would have to significantly lower there Margins to achieve that.
14nm is currently more expensive with lower yields than 22nm. Broadwell Die size is smaller than Haswell but not that much. Broadwell today cost more than Haswell, how Intel will sell at the same price and keep the high margins they have ???
They would/could do that if the market volumes were increasing. The market currently is stagnant at best. Intel Fab utilization is decreasing not increasing. They already mentioned they will close a fab.
But even if they would sell Broadwell at the same prices as Haswell, OEMs would not jump like crazy for High End High cost Laptop CPUs. As i have said before, the market is transitioning to cheaper Lower end laptops and Tablets.
Go learn about "burden of proof," and then come back to me. You just flunked out of Writing Arguments 101.
You're basing all of this off an erroneous belief that Broadwell is more expensive than Haswell. Broadwell is a good 60% of the size of Haswell (GTx for GTx, e.g. BRW DC-GT2 vs HSW DC-GT2), IIRC, and the density of 14nm should be more than double that of 22nm. Yields should be mostly fixed by now.
If it is going to be more expensive, it won't be by much. Regardless, Broadwell will be cheaper for the same level of performance, offer better battery life, and other good stuff.
That's just marketing talk & sales pitch to their investors/shareholders ! Intel was showcasing Haswell based devices IIRC at last year's CES, with the introduction to retail market in Q2 of 2013 & this year there is absolutely no indication of 14nm parts(even Atom) being released anytime during this next 3~6 month period & most rumors hint at the holiday season as being the most likeliest of time for their introduction to retail markets. So we'll see whether my prediction comes true or your bold claims prove me wrong btw I'm about talking real products & not some IDF demos :whiste:It doesn't seem that your doom scenario is going to come true: Intel already stated that volume production of 10nm will start in 2015. TSMC's volume production of 10nm (= Intel's 14nm) will start at least 1 year later.
I see lots of hot air but very little substance in that statement !With 14nm they will likely be able to take an enormous amount of market share in tablets and smartphones, which they will strengthen with 10nm. At that time (2016 and further), TSMC will be more than 3 years behind Intel, so no one is going to want TSMC's silicon.
I don't need no "burden of proof" to prove less(cheap) is selling more to you or to anyone else, if you can't see the long term trend then it;s your loss not mine & I have enough anecdotes to believe in that theory :|Go learn about "burden of proof," and then come back to me. You just flunked out of Writing Arguments 101.
Lenovo K900 is the only real high end mobile based on Intel's chips in the last two years, it retails for as little as 300$ at my place(without any other subsidies) including taxes & it debuted ~10 months back at a price of nearly 600$ ! Other than that there has been nothing, absolutely zero as far as I recall & you can't tell me that they haven't even tried to enter the second largest market in the world or that their efforts were half hearted at best, cause that would be a lie!
Now let's not even get into the tablet realm because no more than a dozen or so Intel based devices, from top manufacturers, have been launched in the past three years, including the current 2014, & the only thing going for them currently is that win8 is gaining traction on mobiles/tablets otherwise they'd have a huge hole in their balance sheets right now !
OR maybe it is the only product that beats the competition in most(if not all) performance based metrics & that Clovertrail(+) based parts were only just competitive against Qulacomm/Tegra/Novathor/TI et al ?Maybe that's because Silvermont is their first serious product for smartphones and tablets, and launched only 6 months ago for tablets and a few weeks ago for smartphones?
And yet you're basically claiming a fifth of market share based on what exactly ?You can't just come into a new market and expect to gain a huge market share within a year. With Bay Trail, Intel will gain a ~20% market share, which they will expand with 14nm Broxton and SoFIA. We'll see how much they can get, but they seem to be very capable products nonetheless.
Intel have been designing SoC for these markets for years, so they've been serious for years.Maybe that's because Silvermont is their first serious product for smartphones and tablets, and launched only 6 months ago for tablets and a few weeks ago for smartphones? You can't just come into a new market and expect to gain a huge market share within a year.
At last now they have a low/mid-range competitive product, and upcoming chips look better, though we'll have to wait what the competition has and how much Intel can get.With Bay Trail, Intel will gain a ~20% market share, which they will expand with 14nm Broxton and SoFIA. We'll see how much they can get, but they seem to be very capable products nonetheless.
They hadn't been updating the underlying CPU architecture for over 5 years. I don't know how you can call that serious.Intel have been designing SoC for these markets for years, so they've been serious for years.
If Intel's products have the best performance for a competitive price, I'm sure there will be a lot of interest. Intel now shows everyone that they're serious about this market: Atom is now on the same level as Core. Margins won't be an issue either, because Intel has 1 node advantage. So they will be able to go for higher margins, or they can have lower prices with the same margins, which is the most likely thing they will do. And once they have a lot of market share, and the market has consolidated, they will go for higher margins/prices.There is no doubt that Intel's products have a future in this market but Intel's deranged history, as far as PC OEM's are concerned, will only work against it in this part of the world, as device makers here won't be tempted by short term gains(subsidies) & that Intel will permanently have to settle for low margins if they want even a foothold in the mobile/tablet industry.
There is no doubt that Intel's products have a future in this market but Intel's deranged history, as far as PC OEM's are concerned, will only work against it in this part of the world, as device makers here won't be tempted by short term gains(subsidies) & that Intel will permanently have to settle for low margins if they want even a foothold in the mobile/tablet industry.
They hadn't been updating the underlying CPU architecture for over 5 years. I don't know how you can call that serious.
Intel plans to sell 40M-60M tablets this year with Intel inside. That's approximately 20% of the total tablet market in 2014.
Of course, this most likely is related to how you define "serious", but to me it was serious business.They hadn't been updating the underlying CPU architecture for over 5 years. I don't know how you can call that serious.
That's exactly what I think! Which also means I missed the "product" part when I previously quoted you. Sorry :$They started to become serious about smartphones around 2010, but it isn't because you're serious (but still, less serious than they are today) that you suddenly also have serious products.
Well, Brian Krzanich thinks different about that: Intel CEO: 'We're on Track' to Supply Chips for 40M Tablets.
Both Intel and TSMC graphs are wrong and misleading, these are just marketing slides.
1. Intel can't know for sure where TSMC stands
2. TSMC can't know for sure where Intel stands.
3. each of them might be measuring density of different kind of cells.
Pseudo-technical marketing BS at his best.
OTOH TSMC have a very poor track record in the comm department (though recently Intel is trying really hard to catch up...).
To sum up, I don't believe either of them.
For 3, are they talking about RAM density? Other kind of cells? Metal layer? Other layer?
Why are they both wrong and misleading? Even if you don't know exact scaling for upcoming nodes, companies can still make predictions based on data that's currently available to them. So:
1. True, but they can base TSMC's numbers on what they publicly disclosed. (Even then, the graph could still give information on their own scaling plans.)
2. True, but they can base Intel's numbers on what they publicly disclosed, in this case, they just copied the graph with Intel's numbers.
3. I don't think that's true, they're just speaking about regular density improvements from new nodes.
As I said I don't trust TSMC. So it's likely their slide doesn't mean much. My point is it is likely Intel slide doesn't mean much either. You can't on one hand claim TSMC is lying while claiming Intel is right, both are into a marketing war and recently Intel claims have been proven wrong too many times to give them the benefit of the doubt (though I still tend to trust them much more than any other tech company, except in some domains where I happen to know much more than what is publicly known).So TSMC copied Intel's graph and made some changes:
1. There will be 16nm FinFET Plus version of their 20nm node, which will give a 15% area improvement. I don't expect that this will be used immediately, probably 1 year later, but they used that node to make 16nm look better.
2. According to TSMC's own words, they also plan to scale aggressively at 10nm, so to make the graph look nice, they just connected the line to be on par with Intel's data, without real numbers that are able to support that.
Density numbers in regards to this topic are invariably their highest density 6T-SRAM.Both Intel and TSMC graphs are wrong and misleading, these are just marketing slides.
1. Intel can't know for sure where TSMC stands
2. TSMC can't know for sure where Intel stands.
3. each of them might be measuring density of different kind of cells.
Pseudo-technical marketing BS at his best.
OTOH TSMC have a very poor track record in the comm department (though recently Intel is trying really hard to catch up...).
To sum up, I don't believe either of them.
Density numbers in regards to this topic are invariably their highest density 6T-SRAM.
And I actually believe TSMC, partially. Intel clumsily ignored the small but still existent density improvements coming with 16FF/14XM. 10nm is quite a way off, and I think TSMC's claim is a load of garbage there, however.
Do you mean both Intel and TSMC?Density numbers in regards to this topic are invariably their highest density 6T-SRAM.
And I think Intel gave correct numbers for their 10nm, though I guess it might change in the future depending on how hard it is to get high yields.oAnd I actually believe TSMC, partially. Intel clumsily ignored the small but still existent density improvements coming with 16FF/14XM. 10nm is quite a way off, and I think TSMC's claim is a load of garbage there, however.