WEBVTT

00:00:00.120 --> 00:00:07.600
modern CPUs are usually pretty reliable so it came as a shock to people when

00:00:04.960 --> 00:00:12.040
their shiny new Intel Core I9 processor started crashing for reasons unknown and

00:00:10.440 --> 00:00:16.240
was even more shocking when they went online and found that there's been a

00:00:14.080 --> 00:00:21.720
great deal of frustration about instability issues with Intel's 13th and

00:00:18.920 --> 00:00:26.640
14th gen core I 9s along with some core i7s and even lower tier chips that has

00:00:24.680 --> 00:00:31.439
gotten so bad that motherboard manufacturers were compelled to push out

00:00:28.840 --> 00:00:35.879
BIOS updates to to help fix the problem but who's to blame for these issues is

00:00:33.440 --> 00:00:39.200
Intel pushing out shoddy chips is another component manufacturer at fault

00:00:38.120 --> 00:00:43.320
was it Colonel Mustard in the conservatory with the Candlestick well

00:00:41.440 --> 00:00:48.239
the story is actually pretty complicated and has its roots in competition between

00:00:45.719 --> 00:00:51.879
motherboard manufacturers however that doesn't mean that Intel shouldn't be

00:00:49.960 --> 00:00:55.600
shouldering some of the blame as well let's go back though compared to a

00:00:53.480 --> 00:01:00.000
couple of decades ago these days there isn't much to separate one motherboard

00:00:57.600 --> 00:01:04.519
from another in terms of raw performance with manufacturers instead focusing more

00:01:02.199 --> 00:01:08.320
on the feature set that a motherboard offers but that doesn't mean that the

00:01:06.680 --> 00:01:12.119
motherboard companies aren't trying to convince people that their products are

00:01:10.040 --> 00:01:17.000
faster than the other guys even if it's only by a few percentage points and this

00:01:14.880 --> 00:01:21.840
is where the trouble begins you see Intel tells the motherboard companies

00:01:19.159 --> 00:01:27.079
not to enable CPU overclocking out of the box users can of course overclock

00:01:24.880 --> 00:01:30.520
their chips later if they want to but when you turn your PC on for the first

00:01:28.759 --> 00:01:36.399
time you expect the things to be running at stock right well the motherboard

00:01:33.520 --> 00:01:40.159
companies took no overclocking mostly to mean don't mess with the base clock and

00:01:38.439 --> 00:01:44.159
the multiplier which are the main settings used to derive the clock speed

00:01:42.479 --> 00:01:48.719
and they didn't mess with those but unfortunately they took liberties with

00:01:46.200 --> 00:01:52.640
other settings specifically motherboard manufacturers were setting the load

00:01:50.600 --> 00:01:56.960
lines incorrectly and if you don't know what load lines do they basically help

00:01:55.000 --> 00:02:02.000
keep the CPU stable when it's dealing with a significant load doing this had

00:02:00.000 --> 00:02:07.399
the intended effect of the CPU running cooler than it otherwise would have been

00:02:04.280 --> 00:02:09.679
and when the system saw the CPU was cool

00:02:07.399 --> 00:02:13.680
it would boost the CPU's speed which the mother manufacturers could possibly use

00:02:11.760 --> 00:02:18.120
to say that their board was delivering better performance but the load line

00:02:16.200 --> 00:02:23.680
configuration meant that they couldn't ensure stability at those boosted speeds

00:02:21.000 --> 00:02:27.840
meanwhile the CPU cores would boost too much resulting in instability and the

00:02:26.040 --> 00:02:33.000
load lines being out of whack meant that the CPU voltages would then over boost

00:02:30.599 --> 00:02:37.319
which could damage the CPU permanently this was all compounded by some bioses

00:02:35.319 --> 00:02:43.519
not lowering the clock speeds until the CPU temperature hit around 100° C which

00:02:40.360 --> 00:02:45.239
is much warmer than the usual threshold

00:02:43.519 --> 00:02:49.239
of course the motherboard companies were hoping the chips would be fine but

00:02:47.040 --> 00:02:53.400
unfortunately reports of instability popped up from users running cinebench

00:02:51.200 --> 00:02:58.200
Unreal Engine based and similar games and decompression software higher end

00:02:56.239 --> 00:03:02.640
chips were hit harder as their high voltages speeds and temperatures caused

00:03:00.400 --> 00:03:07.200
greater instability and even chip degradation in some cases resulting in

00:03:04.799 --> 00:03:10.480
the need for an RMA to Intel investigations like the one undertaken

00:03:08.920 --> 00:03:17.519
by the developers of the game Warframe showed that the problems were overwhelmingly on core i9s and some core

00:03:14.920 --> 00:03:22.560
i7s now we did say that motherboard manufacturers tried to address these

00:03:19.480 --> 00:03:24.599
issues unfortunately their initial fixes

00:03:22.560 --> 00:03:29.560
didn't do much good or even made the situation worse and we'll tell you why

00:03:27.280 --> 00:03:33.920
and how Intel themselves share the blame here right after we thank mend mend is

00:03:32.360 --> 00:03:37.760
an email marketing platform that provides business owners with a suite of

00:03:35.760 --> 00:03:41.439
tools to improve customer relationships through personalized Outreach you can

00:03:39.799 --> 00:03:45.439
send professionall looking emails and craft visually stunning landing pages

00:03:43.159 --> 00:03:49.319
with moen's campaign editor they also integrate a whole SLE of popular

00:03:47.280 --> 00:03:53.079
e-commerce platforms allowing businesses to streamline all their marketing

00:03:50.920 --> 00:03:58.000
efforts so what are you waiting for get a moo on it and follow the link in the

00:03:55.959 --> 00:04:03.640
video description to try mend for free for 30 days and save $10 % after the

00:04:00.920 --> 00:04:08.200
news and people's CPUs broke BIOS updates started popping up that were

00:04:05.239 --> 00:04:13.680
described as Intel baseline or Intel failsafe except this didn't actually

00:04:10.920 --> 00:04:17.840
mean Intel's intended default settings for example Asus came out with a BIOS

00:04:15.920 --> 00:04:22.520
that just jammed More Voltage through the chip to try to stabilize it even

00:04:20.000 --> 00:04:27.880
though Intel Baseline would imply that it uses more conservative settings this

00:04:25.320 --> 00:04:33.039
did help with stability for many users but it also had the unint side effect of

00:04:30.680 --> 00:04:36.320
making their chips degrade faster and then once that happens you need even

00:04:34.560 --> 00:04:40.560
more voltage to keep them stable and so on and so forth until the point of no

00:04:38.240 --> 00:04:43.720
return but while motherboard companies stepped Over The Line in trying to

00:04:42.120 --> 00:04:49.039
achieve performance advantages and in trying to fix this some of the blame

00:04:46.000 --> 00:04:51.039
does lie with Intel for starters one

00:04:49.039 --> 00:04:54.960
reason motherboard manufacturers felt that they were able to mess with the

00:04:52.440 --> 00:04:59.320
overclock settings the way they did and why later BIOS fixes didn't completely

00:04:56.919 --> 00:05:05.039
work is because Intel's guidance about prop voltages was actually quite vague

00:05:02.600 --> 00:05:12.000
including for how long the load lines could be set Intel also identified a bug

00:05:08.639 --> 00:05:13.759
in an Intel feature called ETB and

00:05:12.000 --> 00:05:18.039
admitted that there was an issue with the CPUs themselves that resulted in

00:05:16.039 --> 00:05:23.680
incorrect voltage requests to the processor that affected all 13th and

00:05:21.199 --> 00:05:27.720
14th gen chips that were rated at 65 watts and higher and although they

00:05:25.759 --> 00:05:30.919
didn't give any additional details it sounds like this may have been a

00:05:29.080 --> 00:05:35.759
contributing factor to motherboards delivering too many volts to the CPU

00:05:32.880 --> 00:05:39.560
course whoever owns Which percentage of the blame though the truth is Intel's

00:05:37.800 --> 00:05:42.800
reputation has been tarnished to the point where one game developer has

00:05:41.039 --> 00:05:49.720
announced they're migrating all of their servers to AMD CPUs and while it isn't

00:05:45.960 --> 00:05:52.479
clear exactly why the ETB bug is present

00:05:49.720 --> 00:05:56.759
it appears that toward the end of 2023 Intel was already trying to get

00:05:54.479 --> 00:06:00.759
motherboard companies to rush out BIOS updates right around the time this

00:05:58.800 --> 00:06:05.880
problem started to rear its ugly head which may have been the origin of the

00:06:02.960 --> 00:06:10.479
ETB problem but take that with a grain of salt the good news at least is that

00:06:08.280 --> 00:06:14.440
Intel has now released recommendations for the correct settings to be applied

00:06:12.080 --> 00:06:17.919
to fix these instability problems and as of the time we wrote this the major

00:06:16.080 --> 00:06:23.280
motherboard manufacturers have put these defaults into updated bioses as well as

00:06:20.840 --> 00:06:29.000
a micro code patch for the ETV B bug for boards with 600 and 700 series chipsets

00:06:26.240 --> 00:06:34.199
so please go download the latest BIOS for your motherboard as soon as possible

00:06:31.319 --> 00:06:37.800
like right now even though Intel has said the fix for the root issue

00:06:36.000 --> 00:06:41.639
contributing to excessive voltage isn't coming until the middle of August the

00:06:39.560 --> 00:06:47.400
current BIOS updates should make your system more stable and at least prevent

00:06:44.280 --> 00:06:48.680
damage to your CPU in the meantime if

00:06:47.400 --> 00:06:53.240
for some reason you're still having stability issues and you don't have time

00:06:50.880 --> 00:06:58.240
for system downtime to RMA your CPU to Intel you can try turning your CPU's

00:06:55.360 --> 00:07:01.400
clock speed down by about 300 MHz by adjusting the multiplier in your BIOS

00:06:59.960 --> 00:07:04.879
hopefully that'll at least keep you limping along in the meantime the one

00:07:03.400 --> 00:07:09.400
positive that we hope comes out of all of this is Intel hopefully keeping a

00:07:06.759 --> 00:07:12.759
closer eye on exactly how motherboard manufacturers build their products and

00:07:11.080 --> 00:07:17.680
configure them before they leave the factory because it appears there's been

00:07:14.800 --> 00:07:21.800
not enough oversight I understand it I mean with so many Boards out there it's

00:07:19.280 --> 00:07:25.759
harder for Intel than say NVIDIA who has a more limited number of addin board

00:07:23.680 --> 00:07:29.639
partner graphics cards to examine but team blue obviously is going to just

00:07:28.039 --> 00:07:32.840
have to bite the bullet here no matter how many bunder boards are out there

00:07:31.240 --> 00:07:38.000
because another hit to their reputation like this is not going to feel good by

00:07:36.280 --> 00:07:42.639
the way there is another Wrinkle in all of this Intel has said that some reports

00:07:40.599 --> 00:07:47.120
of instability can be traced back to Intel themselves to a manufacturing

00:07:45.039 --> 00:07:52.639
error that involved oxidation of some early 13th gen chips so if you bought

00:07:49.639 --> 00:07:54.479
your chip in early 2023 or prior that

00:07:52.639 --> 00:07:58.360
could be part of your problem and seeing as it's a physical issue with the chip

00:07:56.440 --> 00:08:03.479
that wouldn't be fixed with a micro code update so you will need to R Intel have

00:08:01.879 --> 00:08:07.280
you had any problems with Intel's latest chips comment down below and if you want

00:08:05.560 --> 00:08:11.759
to find out why electronics die in general go watch this video next