Windows 10: Kernel memory leaking Intel processor design flaw

Oh? This is the first I've heard of it, but then again I live in a closet and rarely get out. *chuckle

Actually don't have the time to read much tech news other than what I see on the forum. I should peek in on the news thread more often.

I wonder if I can get a chip replacement for my Intel Pentium duo core D925 in my Acer. *chuckle

 

You obviously have evidence that shows Intel knew of the bug and continues to manufacture flawed CPUs anyway. Please provide a link.

 

That is one solution - but as everything is still being speculated upon (although the almost-emergency-like triage being performed on both the Linux kernel as well as the NT kernel says otherwise) we still don't know what is really going on.

Suffice it to say, though, that I'll be very interested to see which processors are affected and the level of severity by processor lines.

I'm already planning a new build for the summer, might have to move that timeline up a bit.

Then I have to call my parents - their 3 and 4 year old laptops will need replacing lol....

 

Lol

First of all, the bug has been in existence because of the way Intel CPUs perform tasks, particularly multi-tasking. More will be learned when the bug itself is un-embargoed soon. Until that happens, most everything is speculation.

Second of all, on the flip side of that argument, the fact that both the Linux kernel and the NT kernel were being revamped hardcore (without any real news about it) since at least November, and possibly even before then, since 17025 was released Oct 25, and Microsoft created this new screwy leap ahead program to go ahead and skip 16299 and start testing RS4 builds in the first place (which has always made me suspect something was up even as I also jumped on that bandwagon) means that it is not just a run of the mill bug.

One thing to note, we don't even know which processor lines are affected here - it may be only for the Core lines, leaving out the Pentiums, or it may be from later Pentiums all the way through modern processors, it may be for all processors with hyperthreading, who knows?

This may be the smoking gun AMD needs to pull ahead of Intel - or it may be nothing at all. We just don't know yet.

 

Thanks to the Admin that moved my post here. Guess I'm in the news thread now. LOL.

Depending on the first build the patches were included, I've noticed a significant slower change of speed on my Acer with 10 vs 8.1. It was when the weather started turning cold in late Fall that I got it out of moth balls to throw FCU and an IP on it. Couldn't be more than 2 months ago. I'm thinking 17046 was the first? If that be the case then yes, I think it has affected me, but we'll have to see. I'll have to look up when 17046 was released.

 

Seems to be very little difference.

 

17025 was released 25 October. It was then released to slow ring 1 Nov. So that is about the right time frame....

 

I didn’t experience a slow down, but I did experience a complete full stop from the very first RS-4 release


Flight Hub

 

Thanks, John. That's what I thought. I remember having either 040 or 046 updated on my AMD tower and making an ISO via UUPtoISO to install on the Acer. My 040 folder is dated 11/16 and the 046 is 11/22.

Anyone have Brandon LeBlanc's Insider build list page he made not too long ago? I thought I saved the link, but dang if I can find it or on Insider Corner.

Never mind. I see bordi just posted it. Thanks.

 

I will worry about it when I notice the difference in performance. Frankly, it was hard to tell the difference going from Devil's Canyon to Kaby Lake at about the same overclock. I also could not tell the difference when I dropped my overclock by 300MHz when I got rid of my noisy water cooler and went back to a quiet air cooler.
Now going from SATA SSD to NVME SSD WAS noticeable and there was the huge difference in going from a GTX 1080 to a GTX 1080 Ti in gameplay.
I wouldn't worry too much over it with both MS and Intel having a strong interest in minimizing the impact. it may take some time to tweak the bug fix tho so the first numbers out may be the worst case numbers which will gradually improve over time. If we fast ring insiders have had the fix applied for months already and no one noticed, I am not surprised *092

 

I wonder when Intel first knew of this bug?

https://www.fool.com/investing/2017/...-of-stock.aspx

 

Most noticeable impact is recorded by repeated writing to SSD (database). Also overclocking benchmarks will have to be revised *Think

 

Maybe he just needed some spending cash for New Yeas Eve (11$ mil)

 

Well, from a gaming perspective and benchmarking perspective, you aren't likely to see much difference whatsover. Neither gaming or basic rendering tasks are context-switching heavy payloads. It's your enterprise application, and notably cloud architectures (like Google Cloud, Azure and AWS) where a heavy impact could be noticed.

 

Source: https://newsroom.intel.com/news/inte...arch-findings/