Intel takes wraps off its 10nm SuperFin technology

[HEXUS]

Calls it the "largest single intranode enhancement," in its history.

Read more.

[Tabbykatze]

Calls it the "largest single intranode enhancement," in its history.

No, it's the single largest failure that was finally fixed 5 years later...in its history.

[Friesiansam]

No, it's the single largest failure that was finally fixed 5 years later...in its history.

Indeed, very late and, still lagging behind AMD.

[CAT-THE-FIFTH]

TBF,Intel were the first to mass produce Finfets,so it will be interested to see how this pans out!

[3dcandy]

I like the bit where it says waiting for 7nm processors...

[ik9000]

will total memory encryption bring any benefits? Won't this come at latency penalty? Is it their new way of dealing with side channel and bypass attacks - rather than solve the problem encrypt the information so memory leaks can't be easily understood?

[Tabbykatze]

will total memory encryption bring any benefits? Won't this come at latency penalty? Is it their new way of dealing with side channel and bypass attacks - rather than solve the problem encrypt the information so memory leaks can't be easily understood?

AMD incorporates this into EPYC and the impact is negligible

[zaph0d]

The "Biggest Performance Increase" is nothing to shout about, going from ~2GHz to ~3GHz might be netting a 50% boost, but it's still lagging some 40% behind the previous process

[LSG501]

will total memory encryption bring any benefits? Won't this come at latency penalty? Is it their new way of dealing with side channel and bypass attacks - rather than solve the problem encrypt the information so memory leaks can't be easily understood?

AMD incorporates this into EPYC and the impact is negligible

Got this vague recollection that upcoming desktop DDR5 has this built into the specs as well.

[edmundhonda]

Weren't they already calling their current process 10nm+ or ++?

[QuorTek]

so when they done with 10nm AMD will be advancing to 5 and 3nm?

[PMMEASURES]

so they moved to a 10nm node and seeing the benefits AMD/ARM/apple/samsung/qualcomm etc etc have seen experiencing dropping a node and upping the achievement\sales pitch\waffle as if it was a leap from an 200nm node to a 1nm node lol.

Hmmm nothing so see here it seems not already done by others. But as mentioned it is FinFet so be interesting but some of the waffle is just generation leaps like latest memory modules LP5-5400 type support which i will assume is DDR5 equivalents stuff hense the bump which is more on the memory supplier than there own CPU gains.

[[GSV]Trig]

I get that AMD being on 7nm+ and Intel being on a larger process seems like a fail, but its not like Intel are actually behind on performance is it, does them being on a larger process really matter in the grand scheme of things or is this node/process war just e-peen?

[ohmaheid]

"Rich I/O...and much, much more"

I was ready to give Intel the benefit of doubt, until i read that last line.
It all points to BS.

[Xlucine]

I get that AMD being on 7nm+ and Intel being on a larger process seems like a fail, but its not like Intel are actually behind on performance is it, does them being on a larger process really matter in the grand scheme of things or is this node/process war just e-peen?

The intel 10nm is widely considered comparable to TSMC 7nm, the branding numbers chosen are a bit out of step. The fail at intel is the delays to 10nm, as the 14++++++++nm node is definitely behind TSMC's latest greatest

[DanceswithUnix]

The fail at intel is the delays to 10nm,

Agreed, it sounds like with these updates we will get something like Intel wanted to release years ago.

Charlie has quite a positive overview on Intel's process tweaks https://semiaccurate.com/2020/08/13/...and-packaging/