Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Intel Chips Have Memory Access Design Flaw and Fix Could Lead to Performance Drop
- Thread starter MacRumors
- Start date
- Sort by reaction score
Sure sure.
- What problem does an ARM-based iMac Pro solve?
- How many years? Two? Five? Twenty?
Probably won‘t because they‘re fixing it at the OS level for everyone. I doubt you will be able to choose whether or not to use kernel page table isolation.
They would no longer be beholden to intel’s roadmap. They could achieve 20 percent increased performance for no power increase, or decrease power by 20 percent. They can put support circuits right on the cpu instead of external chips, allowing them to more easily do things like FaceID, high refresh rates, security, etc.
that was a layout bug, wasn't it? (FDIV bug, not sure if you are referring to the same thing or not). some kind of crosstalk problem leading to bit flips under certain data patterns. memory is hazy.
That has to be the worst analogy / excuse I've heard in MR. What........ mate, even if Apple created a gapping security the size of bridge you would deny it
Security flaws are bad......a bad analogy does not make it okay .
Actually, that does appear to be the argument of many people on these forums. Every little thing that Apple does gets spun around as a negative and yet another reason why Android/Samsung/Microsoft/Windows or whatever is superior to Apple.
If every single Apple product released in the last ten years had a critical flaw as serious as the one that Intel is dealing with now, you'd better believe many people would rightfully be done with Apple. Of course, some people think a single dropped frame in a user interface animation is the end of the world, so there's that.
I was in charge of the FPU on the x704. It could have been me
Yes.
Uhhhh. Based on what? This is a number you pulled out of a body part.
Also, is that once? So two years later, Intel will be better again? Or are you arguing Apple will somehow permanently be 20% ahead of Intel? Why? Where’s the evidence?
I designed CPUs for many years. Cisc and risc, including x86. I know what percentage of the pipeline and die area was dedicated to x86 eccentricities.
As usual there will be virtually no impact for the average Mac user, who predominantly runs Apple’s own apps. Pages, Numbers, Keynote, Photos, iTunes, iBooks, Safari, Preview, Maps, Notes, Messages, FaceTime. Before you can exploit any software or hardware vulnerability, you need to make me install your malware in the first place. Good luck with that! Most of my Macs are vintage Core2Duo machines. And most of the time I’m on my iPad anyway.
you are right. perhaps we were just speculating about the cause before intel revealed it. or there was another bug that was related to layout. 1994 was a long time ago.
X704 had a crosstalk problem, due to an automated power reduction tool that downsized current sources but forgot to take into account the effect on slew rates. Around that same time.
Ugh. I just finished installing new Intel processors in my Mac Pro, and they were so fast
Last edited:
https://www.fool.com/investing/2017/12/19/intels-ceo-just-sold-a-lot-of-stock.aspx
Intel's CEO sold all of his shares but the minimum amount required by Intel's corporate bylaws on December 19th.
That sounds fishy at best, maybe even fraudulent.
I was wondering about this. I doubt it unless they also exploit a browser vulnerability.
i was in grad school at that time so i wouldn't have known about what was going on in industry. i think someone i worked with might have been at exponential but it also could have been transmeta. bitrot, dontchaknow.
That’s why I am on my iPad most of the time, it’s a way better browsing device. There are less than a handful of websites, I still visit from my Macs. New Macs have been too expensive for me over the last decade in the Eurozone.
Name? I worked at exponential so maybe I knew him/her. Interviewed at transmeta when exponential failed, but wisely walked out after the first interviewer.