Intel Haswell Info: Single Chip for Ultrabooks, GT3 GPU for Mobile, LGA-1150 for Desktopby Anand Lal Shimpi on November 9, 2011 5:37 PM EST
VR-Zone spotted a bunch of very interesting slides about Haswell over at Chiphell. The slides and information both look fairly believable so let's get with the analyzing shall we?
Haswell is Intel's next tock, meaning it's a brand new architecture. Haswell will debut sometime in 2013 on Intel's 22nm process, first introduced with Ivy Bridge at the beginning of 2012. The information on Haswell spans three major platforms: notebooks, desktops and Ultrabooks.
Haswell for Ultrabooks will be available in a 15W TDP, similar to where SNB based Ultrabooks are today. The big news here is Intel will move the PCH (Platform Controller Hub) onto the same package as the CPU, making the Ultrabook version of Haswell a single chip solution. With Sandy Bridge you needed two parts from Intel, the CPU and the PCH, with Haswell you only need the Haswell MCP (multi-chip package). That's two individual die on a single package, often the precursor to outright die integration (perhaps at 14nm?). The combined MCP should require a smaller footprint than the CPU + PCH arrangement we have today, allowing for less cramped (or smaller) motherboards and potentially even larger batteries in Ultrabooks. This is a huge move as it really starts to blur the line between Ultrabook and tablet hardware.
While Haswell for Ultrabooks tops out at two cores, you can get 2 or 4 core versions in notebooks and desktops.
Both the Ultrabook and notebook Haswell platforms will feature one of three different on-die GPU configurations: GT1, GT2 or GT3. Desktop Haswell will only be offered (as of now) in GT1 or GT2 configurations. No word on the differences between each configuration, but the fact that there are three in Haswell (vs 2 in SNB/IVB) indicates Intel may be exploring an ultra high performance GPU option to further encroach on discrete mobile GPU territory. An even higher performance GPU option for Haswell is something we hinted at in our Ivy Bridge architecture discussion.
Lower Power Memory & A New Socket
The list of memory support is also fairly power optimized. All three Haswell targets will support DDR3L, while the desktop version can use regular DDR3 and the Ultrabook version can use LPDDR3. All three implementations feature two memory channels.
It's important to note that despite Haswell's retarget to focus on 10 - 20W TDPs, the architecture appears to be capable of scaling nearly as high as Sandy Bridge (95W desktop parts will be available, although TDPs may not be directly comparable). This makes sense given that a single architecture can usually span an order of magnitude of TDPs without losing its edge.
Other Haswell features include integrated voltage regulators (should simplify things on the motherboard side), AVX 2.0 instruction support and of course things like AES-NI and Hyper Threading. Haswell will require a new socket: LGA-1150 for desktops.
Nothing here is really all that surprising. We knew that faster integrated graphics was coming, I am curious to see just how powerful this GT3 option will be. The true test is whether or not it will be enough to steer customers like Apple away from including a discrete GPU in their 15-inch Macbook Pro for example. From what I've heard, much of Intel's integrated graphics roadmap has been strongly "encouraged" by Apple.
The move to a single-chip solution for a high-end x86 CPU is a pretty significant step. That line between tablets and notebooks is going to become mighty blurry come 2013.
Post Your CommentPlease log in or sign up to comment.
View All Comments
phatboye - Wednesday, November 9, 2011 - linkWhy am I not surprised to see Intel push out another socket which I am going to bet is incompatible with the previous socket. This is exactly why I love AMD. To bad they don't have a desktop CPU that can match Intel's performance.
GreenEnergy - Wednesday, November 9, 2011 - linkI guess you missed the VRM goes ondie with Haswell. Meaning all the VRM part on motherboards around CPUs dissapear. Tho Anandtech is to blame for some since they missed quite a few slides.
And with boards hitting 30-40$ with Haswell. I think we survieve.
GreenEnergy - Wednesday, November 9, 2011 - linkForgot to add the slide:
phatboye - Wednesday, November 9, 2011 - linkI doubt high end board will be anywhere near $30-$40. Maybe cheap-o netbook boards though.
Yeah I see the ondie VRM but I wasn't aware of that till this article.
DanNeely - Wednesday, November 9, 2011 - linkUrgh!
Intel might claim "overclocking improvements", but I fear what we're going to see is VRMs with virtually no overhead to increase power except on the most expensive binnings.
GreenEnergy - Wednesday, November 9, 2011 - linkWe heard that story before. Yet it turned out for Intels advantage.
wifiwolf - Thursday, November 10, 2011 - linkyup. Thought the move from 3 chips to 2 would make it cheaper but the opposite happened.
SleepyFE - Wednesday, November 9, 2011 - linkFinaly someone else who thinks that way. I too am a fan of using a bit of an older board, so i can keep the cost down. I found that Gigabyte has enough firmware updates to use newer CPU-s. I can upgrade to a 6 core and it wasn't even out when i bought the board. Unfortunetly i am stuck with DDR2 RAM (fortunetly i bought RAM before the price went up).
Starfireaw11 - Wednesday, November 9, 2011 - linkIn real terms, very few people actually upgrade their processors over the lifetime of a PC - it's just us techie types and enthusiats that do, or would like that option. The vast majority of users either don't upgrade at all and simply replace the entire system when it's past its prime or upgrade the "holy trinity" of motherboard, processor and RAM.
By changing to a new socket, intel is not limited to an established and ageing baseline and is free to make changes to the supporting technology as they see fit, at the cost of ruffling a few feathers in the "upper middle class" of their user-base, which is neither particularly large nor profitable. The mainstream and low-end customers will replace their entire system at the end of its service life and their enthusiast, professional and server customers always want bleeding-edge and will replace components or systems regularly in order to stay there.
By staying with rigidly defined socket specifications, AMD is forced to make compromises in their CPU designs or to limit features that can be enabled. They are also forced to rely on a northbridge chipset for much of the system functionality that intel has been slowly pulling on-die. Another downside of the fixed socket specification is that the user experience varies greatly depending on their choice of motherboard - some pretty good AMD processors have been hobbled by being packaged with an outdated chipset or motherboard. The AMD solution does allow for cheap, incremental upgrades though.
There are arguments for and against each setup and as always, it's up to the customer to decide.
retrospooty - Thursday, November 10, 2011 - link"Why am I not surprised to see Intel push out another socket which I am going to bet is incompatible with the previous socket. This is exactly why I love AMD."
Alot of the time it is necessary to push performance upward.
Intel changes sockets more, sure... But the alternative side, AMD is sort of stagnating on the performance side. Their new chips are hardly any faster than the ones they had 3 years ago. In fact, Intel's chips from 3 years ago (45nm Core2 quad) still beat todays AMD chips in most cases.