Back in August 2010, AnandTech published its Sandy Bridge preview—an in-depth examination designed to tantalize consumers and industry alike as to what Intel’s latest production has to offer. I would like to review some of the major points as an introduction to the platform.

As you would expect, the new socket 1155 processors are incompatible with socket 1156 motherboards. The new motherboards will come in H and P varieties, with the H series taking advantage of the graphics on the processor die, whereas the P series will utilize discrete graphics only. At launch, both P67 and H67 chipsets will be available, with the H61 chipset released during Q1 2011.

Despite losing the on-chip graphics with the P series, these boards will support dual PCIe lanes running at x8 speed. The PCIe lane bandwidth of the new chipset is double that of previous Intel chipsets, firstly to increase correlation with chipsets, but also to help support SATA 6 Gb/s which runs over PCIe 1x, and future movement into USB 3.0.

The P/H67 chipsets will natively support two SATA 6Gb/s ports, with the possibility of some manufacturers adding an NEC/Marvell/Etron chip to increase this to four. Four SATA 3Gb/s will be included as standard. No USB 3.0 native support is included, much to the disappointment of some consumers, but again manufacturers at their own discretion can add an chip to give a couple of ports in the back panel, or a few more through onboard headers. USB 2.0 is provided copiously, with at least 10 ports available across the range, through either the back panel or onboard headers.

One major benefit, which I wholeheartedly approve of, is that the holes for the new coolers are identical to the socket 1156 coolers, and various board manufacturers may include socket 775 holes as well, allowing customers to keep their old air or water coolers.

You may remember the following comparison table:

Chipset Comparison
  P67 H67 H61 P55 H57 H55
CPU Support Sandy Bridge
LGA-1155
Sandy Bridge
LGA-1155
Sandy Bridge
LGA-1155
Lynnfield / Clarkdale LGA-1156 Lynnfield / Clarkdale LGA-1156 Lynnfield / Clarkdale LGA-1156
CPU PCIe Config 1 x 16 or 2 x 8 PCIe 2.0 1 x 16
PCIe 2.0
1 x 16
PCIe 2.0
1 x 16 or 2 x 8 PCIe 2.0 1 x 16
PCIe 2.0
1 x 16
PCIe 2.0
RAID Support Yes Yes No Yes Yes No
USB 2.0 Ports 14 14 10 14 14 12
SATA Total (Max Number of 6Gbps Ports) 6 (2) 6 (2) 4 (0) 6 (0) 6 (0) 6 (0)
PCIe Lanes 8 (5GT/s) 8 (5GT/s) 6 (5GT/s) 8 (2.5GT/s) 8 (2.5GT/s) 6 (2.5GT/s)

Another major step over previous sockets and chipsets is the introduction of UEFI (Unified Extensible Firmware Interface). UEFI is a replacement for the older BIOS firmware present in most motherboards, designed to enhance the usability of the motherboard firmware over previous BIOS implementations. Major benefits include a graphical interface, the ability to use a mouse, improved possibility for overclocking on the fly, and the capability of booting from GPT hard drives (those with capacities over 2.2TB where the MBR partition table is unsuitable). Each implementation of the UEFI will be different from manufacturer to manufacturer, and we will look at ASRock’s implementation here today.

One other aspect of note with the new chipsets is the overclockability prospects on Sandy Bridge. Intel’s decision to integrate the clock generator onto the chipset die means that every BUS speed is a derivative of the clock speed. Various buses are highly sensitive to the clock speed, and will allow very little overclocking—maybe, at most, 1 or 2MHz above 100MHz. Thus, for the most part, people will leave the clock speed alone and end up adjusting the multiplier to increase the CPU speed (with the appropriate multiplier unlocked chip), and the divider to increase the RAM speed. For complete scalability, both will have to be adjusted in order for the CPU to reach optimum efficiency. On the P67 chipset, memory speeds up to 2133MHz are selectable, as well as adjustment of the memory sub-timings. We will examine the UEFI options ASRock provides, but please check Anand’s CPU overview for the full Sandy Bridge scaling picture.

On that note, let us have a look at one of the first Sandy Bridge motherboards to market, the ASRock P67 Extreme4.

Visual Inspection
Comments Locked

55 Comments

View All Comments

  • Wander7 - Monday, January 3, 2011 - link

    Still using "ATI" on the motherboard.. I guess they don't want to use "AMD" or didn't get the memo.
  • anandskent - Monday, January 3, 2011 - link

    Why are there no results for the overclocked system in the benchmarks for either the SNB CPU article or this mobo article?

    Also, isnt that a fan header by the 8pin cpu power? it would make the board have 6 total then
  • IanCutress - Wednesday, January 5, 2011 - link

    The 3D Movement multithreaded benchmark was tested at the 4.6 GHz overclock. This is mentioned in the text, and in the graph for that benchmark.

    And yes indeed, that is a 6th fan header. It has no fan control in the UEFI or OS.

    Ian
  • justaviking - Monday, January 3, 2011 - link

    "the H series taking advantage of the graphics on the processor die"

    Sorry if I missed this, but I wanted to be sure on something obvious.

    P-series = No onboard graphics, discrete only. That's clear.
    H-series = Onboard graphics... but can it ALSO USE DISCRETE GRAPHICS???

    Maybe it's so boneheaded it wasn't worth pointing out. Maybe I should have a second cup of coffee.
  • sviola - Monday, January 3, 2011 - link

    From what I understood, none have on-board graphics, and the H series is designed to use the on-chip (CPU) graphics, while the P series is designed for discrete graphics (altough you can use the on-chip graphics as well).
  • justaviking - Monday, January 3, 2011 - link

    Thanks. That is what I assumed, but I didn't see it explicitly spelled out.

    P.S.
    Sorry about the sloppy terminology. I mean "on-chip" when I wrote "on-board". I understand the difference. Thanks being gentle with me.
  • evilspoons - Monday, January 3, 2011 - link

    Not quite. If you Anand's article from today about Sandy Bridge, I got the opposite impression.

    P can not use the on-CPU graphics at all but supports full overclocking of CPU and memory. H can use on-CPU or discrete graphics (or both simultaneously, it would seem), but only supports memory overclocking. Z is meant to fix that by allowing full overclocking and including the "FDI" (flexible display interface) needed to access the on-CPU graphics processor.
  • bah12 - Monday, January 3, 2011 - link

    What I'm confused about is this. The K series gives us the faster GPU, and the ability to fully overclock the cpu. However you cant use the GPU since the whole reason for getting the K sku is the overclock, thus the P series board and no GPU.

    Where does this leave QuickSync (arguably the top selling point)? From what I understand we cannot have a fully overclocked QuickSync CPU right?
  • ericore - Tuesday, January 4, 2011 - link

    Overclock option is only for CPU + Discreet GPU (limitation of a current motherboards; a stupid one at that; cheap Intel bast****)

    QuickSync comes with i5 and greater regardless of the motherboard (it is an instruction set; all of which are built-in to CPU)

    I want want AMD Bulldozer to kill your royal heighness, and bring hardware market back to democracy.
  • vol7ron - Monday, January 3, 2011 - link

    You forgot to mention that P allows for SLI/x-Fire, whereas H does not (since it only supports one discrete gpu).

Log in

Don't have an account? Sign up now