Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
The takeaway message from our discussions with Intel is that there are some hardware mitigations in the new Whiskey Lake processors. In fact, there are almost as many as the upcoming Cascade Lake enterprise parts. Intel told us that while the goal was to be transparent in general with how these mitigations were being fixed, they misread the level of interest in the specifics in advance of the Whiskey Lake launch, especially when the situation is not a simple yes/no.
@jigger I'm sure if Intel did the same it would also fly under the radar... or not, you know there would be a lot more outcry. GF blatantly lied and there was no optical shrink, 12LP only had some very minor changes. At least AMD is moving away from GF.
And you obviously can mitigate those vulnerabilities, maybe apart from Spectre Var 1 which seems to be tackled at a software level.
GlobalFoundries 12LP which ramped earlier this year was designed to make it as seamless as possible for their 14nm customers to migrate over to 12nm. GF robust 14nm platform and IPs are largely compatible with 12nm. Unlike other foundries, GF made no optical shrinks or other major design rule changes that would force customers to make more significant design reworks. Instead, they opted for improving the device characteristics as well as introducing a new higher-density library that brings a modest density and performance improvement over the prior 14LPP high-performance library. Customers can also continue to use their existing 9T and 10.5T libraries while taking advantage of the higher performance from the raw device improvements that were made to 12LP.
Overall, the low-risk, high reward strategy for 12LP has worked well and GlobalFoundries is reporting yield comparable to 14LPP with a number of customers, including AMD, already leveraging the process for mass production products. GlobalFoundries has also extended the platform to include new RF and analog applications as well as automotive-grade applications.
@jigger I'm sure if Intel did the same it would also fly under the radar... or not, you know there would be a lot more outcry. GF blatantly lied and there was no optical shrink, 12LP only had some very minor changes. At least AMD is moving away from GF.
And you obviously can mitigate those vulnerabilities, maybe apart from Spectre Var 1 which seems to be tackled at a software level.
Tell me, how long have these vulnerabilities been about on Intel CPUs? It must take a very long time to get into trouble i suppose.If only this situation was so trivial. L1 cache wipe is 4 clock cycles?
Anyway if you are an Intel guy, because Intel must be promoted, I wouldn't even mention security and how critical it is or isn't. Keep that hush because Intel are in deep trouble.
Tell me, how long have these vulnerabilities been about on Intel CPUs? It must take a very long time to get into trouble i suppose.
I guess things that are realistically easy to perform or obtain are just decades in the finding seeing as nobody notably has performed or obtained anything.
Another Intel related thread derailed by the usual bunch, of course.
Also did anyone notice that Whiskey Lake doesn't have any fixes for Spectre V2 which is one of the bigger performance impacts? Apparently Amber Lake doesn't have any fixes because it's just a higher Kaby Lake Refresh bin, but Whiskey Lake is fabbed using 14nm+++ so they could probably implement a few silicon fixes. If Coffee Lake Refresh is using the improved 14nm node, wouldn't that mean they'd implement some spectre fixes? Going to be interesting to see if they do.
Also the mitigation they implemented in Cascade Lake silicon should put them roughly on par with AMD again.
Another Intel related thread derailed by the usual bunch, of course.
Also did anyone notice that Whiskey Lake doesn't have any fixes for Spectre V2 which is one of the bigger performance impacts? Apparently Amber Lake doesn't have any fixes because it's just a higher Kaby Lake Refresh bin, but Whiskey Lake is fabbed using 14nm+++ so they could probably implement a few silicon fixes. If Coffee Lake Refresh is using the improved 14nm node, wouldn't that mean they'd implement some spectre fixes? Going to be interesting to see if they do.
Also the mitigation they implemented in Cascade Lake silicon should put them roughly on par with AMD again.
Tell me, how long have these vulnerabilities been about on Intel CPUs? It must take a very long time to get into trouble i suppose.
I guess things that are realistically easy to perform or obtain are just decades in the finding seeing as nobody notably has performed or obtained anything.
Which was one of the reasons people got angry with Intel since they were publicly selling Coffee Lake as a secure CPU at launch even though Intel already knew it had serious security issues.It seems Intel got 6-12 months heads up before the flaws went public.
Intel is caught between a rock and a hard place, if they implement hardware fixes, which im not sure they can without completely redesigning the branch prediction in their CPUs, they will suffer performance penalties for doing so, which may well bring their current IPC lead to being only parallel with AMD's IPC, as lets face it, CFL is not massively better IPC now than Zen.
They could keep their advantage by refining their designs and adding hardware level fixes and at best this may keep their IPC lead mainly due to the refinements as much would probably be negated by the branch prediction fixes and other bits n bobs, then AMD would move to 7nm and refinements and could potentially bypass Intels IPC lead.
AMD may well have played the long game here by luck or judgement, if they had known of these types of exploits and vulnerabilities, which in all honesty they should be, as anyone designing this type of architecture should be well versed in what is and is not possible, and could potentially have known that Intels IPC lead was due to the way they favoured speed over these types of security exploits, and well.. chickens have come home to roost.
Its very interesting, id love to know if AMD merely lucked into this current scenario, and if it was a conscious decision by Intel to forgo this type of security in favour of performance.
Looking ahead you can see Intel has deep issues, if they cannot patch the immediate future chips fully from these exploits, and its going to be a couple more generations before they can, that could be a good few years, with the rate cyber attacks increase and become increasingly more clever, and the rate people are playing cloud based games or gobbling up cloud based services, its inevitable that Intel is going to be on the sharp end of their chips being at the heart of a severe security breach somewhere and fingers pointing in their direction, especially if home users start getting attacked.
Which was one of the reasons people got angry with Intel since they were publicly selling Coffee Lake as a secure CPU at launch even though Intel already knew it had serious security issues.
For your average gamer these flaws are mostly irrelevant, unless any fixes substantially impact performance.
At what point did Intel know about the flaws is the big question.
For your average gamer these flaws are mostly irrelevant, unless any fixes substantially impact performance.