"And while Haswell will undoubtedly be better than Ivy Bridge, which really couldn't hold its own with Trinity, there's a good chance it won't be good enough overall at a price point to be competitive with even Richland, forget trying to compete with Kabini. As I said in my article about it last week, Kabini and Temash will likely redefine the low end mobile x86 market with the complete failure of Clover Trail Atom. Given the hype surrounding Haswell, it didn't look possible, but the closer we get to the parts Intel is actually going to ship the more I'm convinced Intel doesn't know what it's doing.
Intel showing us a 10W Haswell part is one thing. Intel selling a Haswell part OEMs want to buy at a price they can sell the computer built around it is another. The battery performance on Haswell has to be lights out better than Richland to begin to justify Intel's pricing. I'm not talking an hour here; I'm talking 50% or more. It's possible but I'm not convinced. Apple (NASDAQ: AAPL) will pay for Intel's best parts because their margins are high enough to absorb their cost but everyone else is trying to figure out how to compete with tablets at a lower price point than an iPad.'
[from Tom Luongo's seeking alpha piece...seems to argue Haswell has missed the point altogether...won't compete at price point..and Ashraf is buckling at the knees saying if 2nd half bombs out he's an INTC seller...this one is NOT IN THE BAG apparently...could be Ottelini's major screw-up]
Wow, you're going to listen to Tom Luongo?
Tom Luongo is the rabid AMD fanboi who is as biased as the day is long. He admitted not buying an Intel processor in the last decade. Once I saw it was him you were getting all worked up about, I had to laugh. He does nothing but absurdly pro-AMD pieces.
According to Tom, Ivy Bridge couldn't hold its own with AMDs Trinity? In what alternate universe did that conclusion come from.
Here's something else you must have missed from Tom:
"Haswell won't have any issues continuing to win the CPU performance race, but that's a race that AMD and ARM (NASDAQ: ARMH) are not playing."
Hmmm. Wonder why they aren't playing? Because they can't. This fantasy that performance doesn't matter and that they don't need to compete is just that - a total fantasy spun by people who want to pretend that their fabrication crisis isn't going to change everything in the marketplace.
Tom's whole argument is that graphics will trump everything and make AMD a winner. Well, Tom, that's just not happening. Intel is steadily eating away at the graphics market and if that's all you have to run on then you and ARM and AMD are toast.
It isn't the criticism of Intel that is scary. But the fact you would take criticism so flimsy and from a person so completely biased at face value is...
Sentiment: Strong Buy
He gives examples of how if you're a serious gamer, you would buys the less performant CPU coupled with a better performing GPU to get better overall system performance for a limited budget. I think non-gamers face similar trade-offs. When Intel introduces 3 flavors of their CPU, most people go for the one in the middle not the top so there's obviously some truth to the fact that you cant arbitrarily keep selling up performance. At some point a consumer is going to ask you what difference it makes in real life scenarios. As Intel know really well, unless there's some killer apps that really demand the performance, its a tough sell. These days, the biggest volume of apps is being made for mobile phones and tablets. Desktops and laptops are being left behind -- look at the dismal app support for Windows 8, and that is after Microsoft has pumped billions of dollars of seed money.
Tom Luongo mentions something called Crystalwell.
"So, over the course of development in the past 16 months, initial features of Haswell got lopped off, most notably Crystalwell - the massive 8 GB integrated memory block to feed the new GPUs - which is where a great deal of the performance increase was going to come from. But, at this point, Crystalwell's cost is causing OEMs to scream. Even if Haswell with Crystalwell makes it into actual systems, at what price point will they be selling?"
Would 8GB (gigabytes, I am assuming) of integrated memory be reasonable on a CPU to "feed the new GPUs"? Or does he mean gigaBITS for 1gigabytes?
If Intel puts 8gigabytes in the package, that is too much just for "feeding the new GPUs". 8GB is enough to feed the GPU and to provide respectable system memory. If Intel is putting the system memory on the SoC Haswell, that could explain the high pricing and the reason the OEM's are screaming.
The Haswell definition of the Ultrabook specifies a "Haswell microarchitecture
SoC (10 or 15 W TDP)". The spec uses "SoC" which Intel uses to describe "anything in a single chip". The Haswell Ultrabook spec does not have dimensions or a RESUME spec defined.
This could be Otellini's "major screw up" or it could be a successful example of what Intel is really capable of doing.
search for :Haswell Will Have Powerful GPU Thanks to Crystalwell Technology
It looks like they are talking about a 64mb cache to keep 16 shaders happy.
I know when I write code I keep the inner loop data objects small enough to get mostly
good reads from the 8 mb cache on the i7's. For my code 64mb would give an amazing kick up in performance for the big calculations.
from semiacrurate last april...
We are told the GT3 variants of Haswell will have 64MB of on-package memory connected through an ultra-wide bus. While it won’t necessarily eliminate memory bandwidth problems, it will go a long long way toward minimizing the problem.
I'd take the amd fan boy tom luongo more seriously if he wrote tech articles, but he mostly write about gold, silver and bonds, not really a serious computer guy. seeking alpha tends to be
amateurish and random...