Intel Debating Commercializing Broadwell-K, Might be Binned – Alleges Italian Report
The Italian publication Bitsandchips.it (which has been more or less authentic so far) is reporting that Intel is actually debating the very existence of its next iteration of Broadwell processors. That means that there is a very high likely hood that Intel cancels the desktop arrival of the Broadwell architecture altogether. Needless to say this will make for a very interesting development and depending on whether or not Intel decides to push back the release of Skylake-K, could seriously affect enthusiasts.
Broadwell-K might be binned, Intel could be planning to skip an architecture on the desktop front
Before I state the contents of this particular report, relevant background information would be prudent I think. Basically there have been numerous (confirmed) reports of Broadwell-K experiencing overheating and TDP problems - a fact that arises from its micro-architectural design (not exactly a design flaw per say, more like an expected iteration flaw). Broadwell-K is expected to fall around 65W TDP, however, at higher clocks the processor's unrestrained TDP rises to around 88W, infact a few SKUs of Broadwell-K are stated to be rated at max TDP of 95Ws. Although I would be surprised if any enthusiast thats going for a 'k' series CPU actually cares about power consumption, the fact remains that more power sippage = more electro migration = lowered life span. For those not familiar with the concept, electro migration is the erosion of the gates inside the processor due to excessive current.
Now the report actually stats that Intel top brass in Santa Clara is currently discussing completely canceling the Broadwell-K platform. Basically aimied at updating the Devil's Canyon platform, Broadwell-K was supposed to appease enthusiasts who go for an unlocked multiplier design till the time Skylake-K lands. While there are existing reports of Braodwell-K being delayed to 2H 2015, this is the first report we have received that states it might simply be binned. I would like to reflect a little on what would happen if it is canceled as well. Basically, Intel will either shift backwards to this year (most certainly 2H 2015) the launch of Skylake-K or enthusiasts who like an unlocked multiplier design will have to wait a long time. Skylake-S is already expected soon, so an unlocked multiplier design shouldn't really be an issue for Intel to pull off. I am also curious on how this affects the Broadwell-E platform, since HEDT is one of the better market skimming approaches for Intel.