Yesterday, a report emerged claiming that Intel is planning to release its upcoming 14-nanometer Broadwell architecture processors as a ball grid array (BGA) rather than an land grid array (LGA) package.
This would have several widespread implications, including bringing to an end to processor (CPU) upgrades.
Traditionally, the processors in desktop systems are fitted into a socket on the motherboard that allows them to be removed and replaced, while systems such as notebooks and tablets have the CPU soldered onto the motherboard.
At present, Intel uses the LGA package design, which allows the processor to either be fitted into a socket or soldered directly to a motherboard. This gives the OEM down the line options as to how to mount the processor onto the motherboard.
A switch to BGA would mean that the processor could no longer be fitted into socket where it could be removed or replaced, and instead would be soldered to the motherboard much like processors for notebooks and tablets are nowadays.
I have now independent confirmation from a PC building OEM, who declined to be named, along with two motherboard makers, that Intel has briefed them of the switch from LGA to BGA for Broadwell architecture processors, which are expected to make an appearance next year.
Separately, tech site SemiAccurate has also received confirmation from two unnamed PC OEMs.
Why the switch?
First and foremost, at least from Intel's point of view, is that this move puts the chip giant in an even more commanding position, allowing it greater control over the motherboard market. More control means more money.
While it doesn't seen that Intel wants to cut existing motherboard makers out of the equation just yet, sources I have spoken to seem to be worried that this could happen in the mid-to-long-term.
The vast array of motherboard choices that both enthusiasts and OEMs currently enjoy could be a thing of the past in a couple of years.
It's a move that could make PC OEMs happy too. Soldering a component to a motherboard is cheaper than soldering a socket and then fitting that processor into the socket. The difference might only be pennies, but spread over millions of PCs, those pennies add up.
As far as the PC OEMs are concerned, killing off the PC upgrade market would be a good thing because it would push people to buy new PCs rather than upgrade their existing hardware. The PC industry is currently stagnant, partly because consumers and enterprise are making existing hardware last longer.
The casualties of this move will be upgraders and PC 'modders', the huge market that exists around them. While not many people bother to upgrade their PCs, instead choosing to buy a new one, the market is large enough to support countless manufacturers and vendors. This move by Intel would be the final nail in the coffin for this industry, taking down a number of players. This, unfortunately, would have a corresponding knock-on effect on jobs.
Intel wins. OEMS win. People wanting cheap PCs win. But there are a lot of losers.
According to SemiAccurate, the successor of the Broadwell architecture, called Skylake, will bring back a socketed CPU, "for a generation, possibly two," but I have not been able to confirm this independently.
This feels to me like the beginning of the end for the desktop PC. Modularity made the desktop PC, and removing this key feature will break it.