This site may earn chapter commissions from the links on this folio. Terms of use.

Intel-Chipset

Intel has taken the unusual pace of releasing (or leaking) data on the company's next-adjacent-generation processor. The new, 10nm chip will be codenamed Ice Lake and isn't expected to debut until after Coffee Lake, which is expected to launch afterward this twelvemonth on the 14nm++ process. We already knew Intel was going to reveal new information on its upcoming CPU family on the 21st (the same mean solar day every bit the eclipse), merely it'southward rare for the company to let much skid about its hardware plans more than ane generation at a time.

10nm-1

Click to enlarge.

Right now, it looks as if Intel volition innovate 10nm parts (codenamed Cannon Lake) merely in 2022. Just as Intel brought 14nm up with low-power Broadwell chips before deploying it for other cores, Cannon Lake is likely to focus on lower-ability fries first. This was actually implied by Intel's 10nm documents, released earlier this year. We've put the relevant slideshow below; all the slides are relevant to Intel's 10nm plans, but the terminal slide is particularly useful.

Intel told us back on March 31 that it would introduce a new set of products congenital on 14nm++, and that 14nm++ would really offer college operation than its showtime-generation 10nm. 14nm++ isn't expected to be as strong a leap over 14nm+ as 14nm+ was over 14nm, but information technology should nevertheless deliver some significant improvements. Assuming Intel also plans to move its Skylake-Ten / Skylake-SP cache configurations to its Coffee Lake CPUs, we should see some intrinsic uplift in that location likewise. Retrieve, the gains shown here but reverberate transistor improvements, non architectural changes.

Anandtech thinks nosotros'll see Coffee Lake and 14nm++ as the only selection on desktops, laptop parts are probable to be a mix of 14nm++ (at or above 35W), and 10nm will exist reserved for 15W and lower-TDP parts. This would mimic how Intel debuted its 14nm technology and it makes sense — early in a process cycle, when yields are the worst, the best way to make certain you can minimize wafer cost is to build as many CPUs per wafer as possible. The more good chips you lot go per wafer, the more CPUs y'all have to blot the cost of low-yield production.

As for why Intel'due south 10nm timeline has slipped so much as to necessitate pushing 10nm back, Ars Technica reports hearing that the company had to reassign engineers from its 10nm procedure to fix issues it was having with 14nm. This took long plenty that the visitor couldn't make up the divergence and hit its original launch targets with 10nm. This isn't particularly surprising, either, though it wouldn't surprise us if Intel adopted a different tactic for its own process nodes going forward.

Correct now, the foundries and Intel accept pursued very dissimilar strategies for targeting new nodes. Samsung and GlobalFoundries are both taking an incrementalist approach. Their 10nm technology is similar in key metrics to Intel'south 14nm tech and their 7nm applied science, when information technology arrives, will be like to Intel'south 10nm tech. GlobalFoundries hopes to have its first-generation 10nm applied science in loftier book manufacturing by the second half of 2022, though no firm introduction date has been given. It'south non unusual for customer shipments to lag 4 to vi months behind the introduction of loftier volume manufacturing, depending on which chips are existence built and how the process node launch aligns with the client's product launch schedule. TSMC, for example, started its high volume 10nm manufacturing at the beginning of the yr only those chips didn't hit the market until June.

It looks correct at present as if the foundries may take fabricated up some ground on Intel, just nosotros'll wait and meet how 7nm shapes up before drawing whatsoever house conclusions there. If they have, Intel may well motility to a new model for its own parts, 1 that simplifies the update procedure past taking smaller steps at once.

It's also all besides likely that delayed EUV rollouts or trouble with the 7nm node could delay Samsung, TSMC, or GlobalFoundries. Information technology's getting harder and harder to hit new nodes, and every foundry has bug from time to time. TSMC had a terrible 40nm deployment, GlobalFoundries had to cancel its ain 14nm technology and prefer Samsung's, and Samsung…well, Samsung hasn't really had any cataclysms, only its first implementation of big.Piffling was a hot mess and it's only been fairly recently that the company started pushing to expand its foundry business. Even when information technology built chips for Apple tree, Samsung'southward foundry business more often than not served its own needs get-go. Give it enough time, and Samsung will trip as well.