The CPU socket is the most important aspect of purchasing a motherboard. As a matter of fact, your motherboard choice really narrows down the choice of CPU that you can connect to it and vice versa. If you already own a CPU, you must also purchase a suitable motherboard.
To put it simply, CPU sockets on a motherboard are different for Intel and AMD CPUs. Furthermore, they are not even the same for all Intel processors or all AMD processors. For example, the latest AMD chips require AM4 sockets and the newest Intel processors use the LGA 1151 sockets.
As CPUs develop and become more advanced, so do the sockets that support them. For instance, Intel might release a new line of CPUs tomorrow that are not compatible with the latest LGA 1151 sockets and everyone who wanted to purchase the new Intel’s CPU would also have to buy a new motherboard.
What is LGA 1151?
LGA 1151, commonly known as Socket H4, is a CPU socket installed in motherboards and it’s compatible with Intel microprocessors. There are two revisions of the LGA 1151 socket: the first revision that was designed to support Intel’s Skylake and Kaby Lake CPUs, and the second revision that provides exclusive support for Intel’s Coffee Lake CPUs. The second revision LGA 1151 is often abbreviated as LGA 1151v2 or LGA 1151-2.
LGA 1151 is a successor to the LGA 1150 socket (commonly known as Socket H3). The number ‘1151’ is there for a reason: the socket has 1151 protruding pins that are used to connect to the pads on the processor. When compared to the LGA 1150 socket, it can be observed that the Fully Integrated Voltage Regulator has been returned to the motherboard (from the socket itself). The Fully Integrated Voltage Regulator is a module that converts the supply voltage of 5 V or 12 V to lower voltages required by the CPU. You might be thinking that 5 V is already a low voltage level but most modern CPUs require a voltage lower than 1.5 V to operate. The voltage level is low for a reason: it greatly reduces power dissipation and prevents the processor from overheating since it’s operating at high frequencies.
List of LGA 1151 CPUs
Making sense of it all
Mentioning all sorts of CPU characteristics such as its cores, frequencies, clocks speeds, and others might sound intimidating to users who don’t know just how to interpret these numbers. Sure, everyone knows that a higher frequency is probably better, but let’s check out the meaning of some of these phrases.
Number of CPU cores
The terms CPU and processor are often used interchangeably. However, a CPU and a processor don’t represent the same thing. A CPU core, on the other hand, is a processor. A processor can work on a single task at any given moment. More cores mean more tasks that can be run and computed. Simultaneously, making your computer faster in everyday use.
Not every process benefits from having too many cores. Video games usually don’t and they generally don’t use all of your cores. Video editing and animation, however, can run much faster on multi-core CPUs. Today, there are usually between 2 and 8 cores in a regular consumer CPU, with the number going as high as 18 core for workstation and CPUs for special purposes.
Besides how many cores you have, the second parameter that matters is just how fast they are. Clock speed represents cycles per second your CPU core can execute. It’s often measured in gigahertz (GHz) and it represents billions of cycles per second a CPU core can execute. However, there is another factor that plays a role: clock cycles per instruction. Executing a single instruction can take multiple clock cycles. Different processors require a different number of clock cycles to execute an instruction. To put things a perspective, a 2.4 GHz processor that takes 4 cycles to execute an instruction and a 1.2 GHz processor that takes 2 cycles per instruction are essentially working at the same speeds.
Still, comparing Intel processors can be done by comparing frequencies if they belong to the same architecture.
Max Turbo Frequency
Max Turbo Frequency is a term that’s used with Intel processors. By using the Intel Turbo Boost Technology, each CPU core can be overclocked and work at higher frequencies compared to its base frequency. This boost can be used without worries it might break as long as you prevent the CPU from overheating. Higher frequencies generally lead to higher power dissipation and that’s why the best CPUs can dissipate more than 100 W.
A CPU cache is a memory module located close to a processor. It makes copies of data from frequently visited memory locations and stores it for faster access. That’s right, your computer analyzes you and figures out which memory locations are needed the most. It stores it in its cache so your CPU doesn’t have to go all the way to your main memory to retrieve it.
Today, there are various levels of cache memories, usually denotes as L1, L2, L3, and so on. The first level (L1) is usually the fastest and the smallest, and these levels get slower and larger in memory as we go on. This is an important aspect of CPU performance as it can optimize your routine and help you in quickly responding to repetitive tasks.
TDP stands for Thermal Design Power and it represents how much heat your CPU can generate that your heat sink or cooling system will be able to dissipate. Intel defines this number as the average power dissipation of the CPU when operating at a high workload at the base frequency.
Some CPUs have a built-in cooling system that can take care of the TDP and some don’t. Whatever you do, you need to make sure that the amount of power dissipated is high enough to overheat and damage your CPU or any other component. Also, if you decide to overlock your CPU and work at higher frequencies, this number becomes irrelevant as the actual dissipation is much higher.
In any event you’ll be applying thermal paste to your CPU. If you’re planning on using some old thermal paste you have lying around make sure it hasn’t expired. Otherwise get a new syringe thermal paste, we personally recommend an ARCTIC MX-4, and you’re set.
IGP stands for Integrated Graphics Processor. This processor is a chip integrated into the motherboard and it uses some of the CPU’s processing power to process graphics, acting as a video card. However, their size is limited and they generally don’t have a dedicated cooling system like video cards. They are soldered to the motherboard and they can’t be replaced or upgraded.
These are some of the reasons why IGP can’t replace video cards as they are just not powerful enough. However, they have proven useful for laptops that were not meant for gaming in the first place.
Best LGA 1151 CPUs
Here are three recommendations of LGA 1151 compatible CPUs from Intel. Let’s start from the high-end option and work our way down towards our budget recommendation! In any event make sure you’re ordering an OEM model if you’re concerned about the warranty.
The letters don’t lie: K stands for unlocked, meaning you can overclock it and squeeze every bit of performance it has; and S stands for Special Edition! It’s difficult to imagine overclocking when you already have 8 cores with base frequencies for 4 GHz. However, with Turbo Boost, you can achieve a frequency of 5 GHz! Adding in a 16M cache and a bus speed of 8 GT/s, you’ve got yourself a beast as long as you’re ready to pay up.
At the price of around 20% of our high-end option, this CPU should definitely not be underrated. With a 9M cache and with six cored clocked at 3.7 GHz, this beast will probably satisfy your every need. Since the letter K stands for Unlocked, you can use Turbo Boost and clock in at a maximum of 4.6 GHz.
If you’re CPU shopping on a budget but you still want to experience the Coffee Lake technology in action, this 6M cache CPU with 4 x 3.6 GHz cores should surprise you positively. Around half the price of our mid-range option, it should treat your wallet well and still provide sufficient performance.
LGA 1151 revision 1 vs revision 2
LGA 1151 revision 1 supports Skylake and Kaby Lake CPUs. This means you’ll have to look towards revision 2 if you want to get your hands on some of the newest Intel CPUs, like some of the Intel Core i9 processors. Most motherboards with the LGA 1151 revision 1 socket only support DDR4 memory but there are a few with DDR3(L) support. This socket is compatible with Intel 100 and 200 chipset series, including H110, B150, Q150, B250, H270, and others.
LGA 1151 revision 2 supports Coffee Lake CPUs and it’s compatible with 8th and 9th generation CPUs that use the 300 series chipsets from Intel, including H310, B360, Z370, and Z390. Despite it being a revision of the original socket, it’s physically the same and compatible with the first revision. However, there are some pin reassignments to add more power lines that would support 6-core and 8-core CPUs.
Despite physical compatibility, there is no backward compatibility between the two revisions. The cause of this is a relocated processor detection pin. This means that you can’t use the revision 1 socket for 300 series chipsets and you can’t use revision 2 for 100 and 200 series Intel chipsets. This means you should be extra careful when buying a motherboard with an LGA 1151 socket. Intel doesn’t actually recognize the two revisions (formally) and they don’t refer to them with a distinct name. However, depending on your CPU chipset, you’ll need to use the proper revision! Each motherboard manufacturer keeps a list of supported chipsets so you should have no problems with finding that information.
LGA 1151 vs LGA 1150
The LGA 1150 socket, commonly known as the H3 Socket, came after the LGA 1155 H2 socket and it supports Intel’s CPUs that were built on the Haswell and the Broadwell microarchitectures. The support of Windows on LGA 1150 started on Windows 7 but certain CPUs worked with Windows XP as well.
Here are some core differences:
- LGA 1151 is a successor to LGA 1150
- LGA 1151 has 1151 contacts, one more than the LGA 1150
- LGA 1151 is oriented towards DDR4 memory whereas LGA 1150 exclusively supports DDR3 memory
- LGA 1151 is developed for Skylake, Kaby Lake, and Coffee Lake processors; whereas LGA 1150 supports Haswell, Haswell-WS, and Broadwell processors
- LGA 1151 is for 100, 200, and 300 chipset series and LGA 1151 is used with 8 and 9 series of Intel chipsets
LGA 1151 vs AM4
The same how LGA 1151 is the latest available socket for Intel CPUs, AM4 is the latest socket for AMD CPUs. Discussion about the sockets in their design, number of pins, or other characteristics, is pointless. You will settle for a CPU and later look for the motherboard with the socket you need. The real discussion is which CPU manufacturer is better.
However, there is an argument about which socket is more futureproof. Many gamers like to keep up-to-date when it comes to new processors from either Intel or AMD. Since the processor is your computer’s most important part, this is understandable and people are usually not afraid of paying up. However, the release of a new generation of CPUs from either manufacturer brings the fear of an entirely new socket being released, rendering your entire motherboard useless if you wish to upgrade to the newest generation. For example, the release of Coffee Lake Intel processors (9th generation of i3, i5, i7, and i9) resulted in a revised version of LGA 1151 that isn’t compatible with the first revision (in any direction). Intel is also quite notorious for recycling CPU sockets faster than users can buy new ones.
It’s difficult to answer the question of which socket is more future proof. Intel has a bad reputation and their sockets are not even backward compatible in most cases. Coffee Lake processors were launched in 2017 with the LGA 1151v2 socket and the socket named LGA 1200 has already been announced for the new generation of Intel chipsets. On the other hand, AMD has explicitly promised that the AM4 socket will be used until 2020 when it will be replaced with the AM5 socket.
To conclude, it’s probably the best wait out for the next wave of motherboards and CPUs if you really wish to own the newest possible setup. However, if you are a gamer or a regular user, it might be a bit of an overkill.