Guide: How to check if your CPU is BAD!
Summary
TLDRThis video script addresses the ongoing Intel CPU degradation issue, offering viewers a comprehensive guide to assess their CPU's health. It discusses the natural degradation of all CPUs over time, but focuses on accelerated degradation affecting certain Intel models. The script provides a step-by-step approach using free software tools to test CPU stability and performance, helping users determine if their CPU is still reliable or needs an RMA. It also touches on the importance of updating BIOS and the potential impact of motherboard manufacturer practices on CPU longevity.
Takeaways
- 😀 The Intel CPU degradation issue is ongoing and has been a topic of discussion among PC enthusiasts.
- 🔍 The community has concerns about detecting CPU degradation, with many seeking ways to test their CPUs for stability.
- 🛠️ The video suggests using free software tools to test CPU stability and rule out the need for an RMA (Return Merchandise Authorization).
- 👕 A sponsorship mention for 'Jus sense melty connector shirt' is made, indicating a light-hearted moment in the script.
- ⚠️ It's emphasized that all CPUs degrade over time, but they are designed with headroom to maintain performance throughout their lifespan.
- 🔋 The script discusses various factors that could cause accelerated degradation in Intel's 13th and 14th generation CPUs, including microcode issues and motherboard manufacturer practices.
- 📈 The importance of updating the BIOS to the latest version, especially to include the most recent microcode, is highlighted to mitigate potential CPU issues.
- 🔬 The video recommends running specific tests like Intel Burn Test, Cinebench R15 and R23, and OCCT to check for CPU stability and degradation.
- 🎮 An unconventional test suggested is running a GPU stress test like Heaven Benchmark 4.0, which can also reveal CPU issues due to the CPU's role in shader compilation.
- 🛑 The script mentions that installing GPU drivers can be a stress test for CPUs, as it can cause crashes in systems with severely degraded CPUs.
- 🔄 The video concludes with the idea that not all CPUs are failing prematurely and that degradation is a natural process, but accelerated degradation is a concern.
Q & A
What is the main topic discussed in the video script?
-The main topic discussed in the video script is the issue of CPU degradation, specifically with Intel's 13th and 14th generation processors, and how users can test their CPUs to determine if they have been affected by this issue.
What is the significance of the term 'advanced degradation' in the context of the script?
-In the context of the script, 'advanced degradation' refers to an accelerated rate of CPU wear and tear that is not typical for the expected lifespan of a processor. It is a concern because it can lead to decreased performance or even failure of the CPU.
What is the role of microcode updates in addressing CPU degradation issues?
-Microcode updates play a crucial role in mitigating CPU degradation issues by providing fixes and improvements to the CPU's operation. They can help ensure that the CPU operates within safe voltage and temperature parameters to prevent accelerated degradation.
Why is it recommended to update the BIOS on a motherboard?
-Updating the BIOS on a motherboard is recommended because newer BIOS versions often include the latest microcode updates that can help prevent CPU degradation. Additionally, updates may include important stability improvements and compatibility fixes.
What is the purpose of running stress tests on a CPU?
-The purpose of running stress tests on a CPU is to simulate heavy workloads and determine if the processor can handle them without crashing or showing signs of degradation. This can help identify potential issues with the CPU's stability and performance.
What is the significance of running tests with XMP disabled?
-Running tests with XMP (Extreme Memory Profile) disabled is important because XMP may not always be stable, especially at high memory speeds. Disabling XMP ensures that any crashes or instability during testing are due to the CPU and not other system components like the RAM.
What is the 'Intel Burn Test' and how is it used in the script?
-The 'Intel Burn Test' is an older tool used to stress test the CPU by running a series of calculations. In the script, it is used to check for CPU stability and to see if the processor can maintain performance without crashing under heavy load.
What is Cinebench R15 and how does it relate to the script's discussion on CPU degradation?
-Cinebench R15 is a benchmarking tool that tests the CPU's performance in both single-core and multi-core workloads. In the script, it is used to test for stability and performance issues that might indicate CPU degradation.
What is the importance of monitoring voltages during CPU stress tests?
-Monitoring voltages during CPU stress tests is important because it allows users to ensure that the CPU is receiving the correct amount of voltage for its operation. Incorrect voltage levels can lead to instability, crashes, or accelerated degradation.
Why is the 'OCCT' software mentioned in the script, and what does it do?
-The 'OCCT' software is mentioned in the script as a free tool for stress testing the CPU. It is used to apply varying workloads to the processor and monitor its performance, voltages, and temperatures to identify any potential issues or signs of degradation.
What is the significance of the 'Heaven Benchmark' test in the context of the script?
-The 'Heaven Benchmark' is a GPU stress test that is mentioned in the script because it also puts significant load on the CPU, particularly during tasks like shader compilation. It can be used to identify CPU issues that might not be apparent during standard CPU-only stress tests.
What is the 'Driver Installer' test mentioned in the script, and why is it relevant to CPU degradation?
-The 'Driver Installer' test involves repeatedly installing graphics drivers, which can stress the CPU, particularly during the unpacking and installation process. It is relevant to CPU degradation because it can reveal issues with the CPU's stability and performance under heavy load.
What advice does the script give regarding the use of computers to prevent CPU degradation?
-The script advises against leaving computers on when not in use, especially when idle at the desktop or in the BIOS, as this can contribute to the CPU's lifespan being 'used up' more quickly. It suggests turning off the system when not in use to help prolong the CPU's life.
Outlines
🔍 Investigating Intel CPU Degradation
The speaker addresses concerns over Intel CPUs' longevity, particularly in the context of degradation. They discuss the difficulty in determining CPU health and introduce various tests to assess CPU stability. The video aims to provide viewers with tools to determine if their CPU has prematurely degraded and needs replacement or if it's still viable. The speaker also mentions the importance of BIOS updates and the potential for accelerated degradation due to factors like microcode issues, motherboard manufacturer practices, and material oxidation affecting CPU lifespan.
🛠️ BIOS Update and Testing Methodology
The speaker emphasizes the necessity of updating the BIOS for motherboards, especially for those purchased before a certain date, to ensure they have the latest microcode. They discuss the impact of motherboard manufacturer practices on CPU degradation and the need to check for proper voltage regulation and thermal throttling. The speaker then outlines a testing methodology that includes disabling XMP for stability, using monitoring software like Hardware Info 64, and running stress tests to observe CPU behavior under load.
💻 In-Depth CPU Testing with Benchmarks
The speaker delves into the specifics of CPU testing, discussing the importance of testing both multi-threaded and single-threaded workloads. They mention the use of Intel Burn Test and Cinebench R15 and R23 for evaluating CPU stability under stress. The speaker highlights the significance of observing CPU frequencies, voltages, and temperatures during these tests to identify potential issues. They also touch on the variability in CPU degradation and the difficulty in pinpointing exact causes due to the multitude of factors involved.
🎮 GPU Stress Test and Driver Installation
The speaker introduces GPU stress testing using Heaven Benchmark 4.0 to evaluate the system under load, which can also reveal CPU issues. They discuss the potential for crashes during GPU-intensive tasks and the importance of monitoring CPU performance during these tests. Additionally, they cover the peculiar finding that installing Nvidia drivers can be a consistent stress point for degraded CPUs, possibly due to the CPU's role in shader compilation. The speaker suggests running driver installation tests multiple times to observe any errors or crashes.
🏁 Final Thoughts on CPU Degradation and Stability
In the concluding part, the speaker reflects on the nature of CPU degradation, emphasizing that all CPUs degrade over time but at varying rates. They dispel the myth that all 13th and 14th gen Intel CPUs are faulty, stating that while every CPU degrades, only accelerated degradation is a concern. The speaker suggests practical steps for users to manage their CPU's health, such as avoiding leaving computers on idle and being mindful of the CPU's operational time. They invite viewers to share their experiences with CPU degradation and the tests that triggered failures in their systems.
Mindmap
Keywords
💡CPU Degradation
💡Microcode
💡BIOS Update
💡Electron Migration
💡VRM (Voltage Regulator Module)
💡Loadline Calibration
💡XMP (Extreme Memory Profile)
💡Stress Test
💡RMA (Return Merchandise Authorization)
💡Hardware Monitoring Software
💡E-Waste
Highlights
Intel CPUs are facing degradation issues that will persist, affecting performance over time.
PC enthusiasts are concerned about detecting CPU degradation through various tests and methods.
Degradation is a normal process for all CPUs, but the rate of degradation can vary significantly.
CPUs are built with headroom to accommodate expected degradation over their lifespan.
Microcode updates and motherboard manufacturer practices can influence CPU degradation rates.
Updating BIOS is crucial to mitigate potential CPU issues, but it's not a guaranteed fix.
The importance of running tests at stock settings to ensure accurate assessment of CPU stability.
Hardware monitoring tools like Hardware Info 64 can help track CPU performance and stability.
Intel Burn Test is an old but effective tool for stress-testing CPUs to identify stability issues.
Cinebench R15 and R23 are reliable benchmarks for testing both single-core and multi-core CPU stability.
OCCT is a versatile free software for stress-testing CPUs with variable workloads.
Gaming benchmarks like Heaven 4.0 can also reveal CPU issues under high stress conditions.
The peculiar effectiveness of the Nvidia driver installer as a test for CPU stability.
Different CPUs degrade in unique ways due to factors like voltage, temperature, and workload.
The video provides a comprehensive guide on how to test for CPU degradation using various software.
The video emphasizes that not all 13th and 14th gen Intel CPUs are problematic, contrary to some beliefs.
Best practices for CPU longevity include not leaving computers on when not in use to reduce unnecessary wear.
The video concludes by encouraging viewers to share their CPU degradation experiences and test methods.
Transcripts
so the Intel drama is here to stay and
it's going to be here for a while but
I've been getting a lot of messages and
reading a lot of forum posts and sort of
anonymously joined a lot of PC
Enthusiast groups just to kind of see
what people are saying and
overwhelmingly there's a there's this
notion of not being able to tell whether
or not your CPU has degraded or not um
we're going to talk about that today and
I'm going to show you some tests that
you can run to sort of rule out whether
or not your CPU has like Advanced
degraded and is ready for RMA or whether
or not your CPU is still stable with
lots and lots of life ahead of it so
that you can kind of figure out where
you might be in the whole drama scale of
things so anyway sit back we're I'll
talk about free software that doesn't
cost you any money to be able to do
these tests and hopefully this will give
you some peace of mind to determine
whether or not you should be ring right
now or hoping that a bios update with
the micro code fixes is enough to
hopefully stop the problem for
you today's video is brought to you by
the new really super cool Jus sense
melty connector shirt you know paying
tribute to one of the best designs
that's clearly ever existed and if you
don't like the vapor wave colors we got
our OG gray red colors available now so
get yours by following the link in the
description below nice ow I have a thigh
cramp I want to talk about degradation
real quick though because I did get on
the phone with Falcon Northwest they
Falcon Northwest is very um close with
Intel in that they have a lot of
discussions with them directly um
there's Kel the owner of Falcon
Northwest he's a nerd has been for over
30 years delving deep into this sort of
stuff and understands the AR the
architectures and and really has helped
get to the bottom of what's been going
on so he's a he's a source that I trust
when it comes to some of this info and
one thing that really needs to sort of
be driven home is the fact that every
single CPU degrades every single
processor yours mine everyone in here
amds intels arms qualcom they all
degrade the thing is they're built with
certain amount of Headroom into them so
that the degradation over time at least
is in theory supposed to La like the
performance that's advertised is
expected to last the life cycle of the
cpu's lifespan now the thing is nobody
ever advertises how long a CPU is
expected to last because there's a many
many factors that would determine how
fast or how slow a CPU degrades
unfortunately with all of the micr code
discussion and the multiple points of
accelerated degradation that's been
taking place whether it be micro or
whether it be damn it unfortunately
there's multiple ways that CPUs for 13th
and 14th gen have been sort of advanced
degrading whether it be Intel's own
fault with micro code stuff or Intel's
own fault with raining in their
motherboard manufacturers or partners
that are just going wild wild west with
voltages and just ignoring VRT all
together or just applying their own
ridiculous loadline calibrations Asus
I'm looking at you uh causing extreme
heat and extreme Advanced like electron
Migration by pumping so much voltage in
there um
unnecessarily or whether it be you got a
CPU from the 2023 era when there was
oxidation known to have apparently
affected x amount of CPUs that has yet
to be disclosed on how many have
actually been affected so there's
multiple things happening here so
whether or not you fall under one of
those categories the first thing you
should do and we've talked about this a
million times already is upgrade your
bios especially if you have never
touched the BIOS from when you took your
motherboard out of the box pretty much
any motherboard you get off the shelf
today is behind on the BIOS I've bet no
motherboards that are on the shelves
currently right now have the latest micr
code bios it's just been too soon
between when the BIOS launched for Asus
and MSI and the manufacturer delay
process for manufacturing boxing
shipping inventorying shipping to the
retailers eilers and getting them on
shelves so everybody who's built a
motherboard or built a system prior to
August pretty much even today if you
went out and bought the stuff today you
need to update your bios that's first
and foremost that doesn't guarantee
though that your CPU is not necessarily
going to still have some problems
because as I just mentioned it's being
discovered that even with the latest
micro code it's still on the motherboard
manufacturer's responsibility to make
sure that all the rtt or the voltage
regulation and thermal throttling tables
are taken into account and not pumping
too much voltage in um also too MSI had
a stin there where their load line
calibration was actually the opposite
and that it was causing severe
undervolting of the CPU so the crashing
and stuff that people were experiencing
was actually the opposite of degradation
it didn't have enough voltage suppli to
it to even Supply its stock boost clocks
and such it was going from something
like7 to8 volts under load which is not
enough voltage to even come close to
giving the advertised boost speeds and
stuff so the CPU would try and boost and
then the motherboard and vrms wouldn't
provide enough power and then it would
crash looking like a degradation problem
when it was actually the opposite um
anyway assuming now you've updated your
bios you've got the latest micro code
hopefully your motherboard manufacturers
those that haven't released theirs yet
will get on that soon that's currently
the 0x1 129 micro code is what we're
looking for we are now going to talk
about free software that you can
download and install and try and run to
see if your CPU crashes also too one
other thing I have to point out is it's
also not recommended it's kind of
required to do these tests at XMP off
you do not want XMP turned on you do not
want to add any sort of additional point
of potential crashing because XMP is not
always stable XMP especially if you're
running like 7400 MHz dims or 8,000 MHz
dims is not guaranteed to be stable on
all CPUs also if you're populating all
four Ram slots in your board it's it's
recommended to maybe take out two of
those sticks disable XMP to run the most
stable config we can think of we want to
make sure all the stress is on the CPU
nothing else could be causing potential
blue screens or crashing in our system
cuz how would you really know the
difference is your Ram's too fast or
heck even your Rams faulty versus your
CPU so I actually need to go and disable
XMP right now cuz I haven't done that
yet remember what I was just saying
about MSI boards sometimes undervolting
this is actually not bad but I'm in BIOS
right now which is actually a load a low
load state which would actually push the
clocks up and the voltage up it's
actually only at 0.890 volts but still
under load I know it passes cuz we ran
all of our CPU tests on this board with
this bios so we know we're good we know
the load line is probably going to
somewhere where it needs to be um but
anyway XMP profil is here oops I'm just
going to uncheck that boot and we're
basically just disabling XMP so we want
to make sure xmp's off that's first and
foremost okay so all of the software I'm
showing you is free this is actually
Hardware info 64 you guys have been at
me for years to use it I started using
it I actually really love it I might
actually do a video about it here but um
anyway this is a free piece of uh
monitoring software so that if you want
this is not prerequisite but if you want
to you can keep an eye on what your
voltages are bursting to what your
single core to multi-core workload burst
look looks like on voltage and stuff um
your temperatures and all that but
here's the thing if you're at stock
you're not doing any sort of
overclocking to your CPU you should not
be seeing temperatures go crazy
especially if you are running an I9
processor and you're at the 250 3 watt
or even an i7 with a 253 watt profile
for like 13 700 14 700 k um this just
allows you to keep an eye in some things
so you can see what our package power is
right here I've got a lot of it hidden
depending on what I'm doing I don't want
all the info so I I'll turn on and off
things just look up your CPU and see
what your TDP should be and then make
sure it's not exceeding that then you'll
know if it is your motherboard is to
blame for that and you need to make sure
that your Motherboard BIOS is updated to
stop it also we can see our frequencies
and stuff right here and then we can see
our vid our sa vid and cor vid right
there we're going to be using a few
tests here um Intel burn test it's
actually an old test but what what's
what's funny here is that most of the
degraded CPUs are not actually showing
their failures or their degradation in
multi-threaded workloads they're showing
it mostly in single threaded or or maybe
one or two thread workloads because
that's where the frequency is going to
its highest as you can see my frequency
is pegged right 5985 on these two cores
because those are the two best core so
they're the ones that are getting me up
to the 6 GHz boost technically it's 5.8
GHz but we can get a little spiky
workload same thing up here that's just
showing me the max core clock so 6 GHz
is where the CPU will go on single load
that also requires a significant amount
of voltage to make that happen so if
there's degradation it could happen in
those bursty workloads not just a
sustained hard workload so this is why
our tests going to show us a bunch of
different things here but the easier to
run test the more bursty the C the
processor and the core is can actually
go which is where the problem can start
to really show itself this is why I say
this is kind of an ongoing problem with
Intel because of the fact that they
we're still not 100% sure exactly what
is happening and the way that the CPUs
is degrade is not consistent some
degrade faster than others because what
motherboard combo was it in what
workloads were you doing what was your
ambient temperature how hot was a CPU
allowed to get those CPUs that are
running higher voltages with really big
coolers keeping the temps down will
degrade slower than those that have the
same voltage with lesser quality coolers
allowing the temperature to reach TJ
Maxx because we know as the temperature
increases higher temperature actually
with higher voltage leads to that um
electron migration faster than colder
CPUs with higher voltage so there's so
many different like Combos and
variations of the way these CPUs are
being loaded up it's very difficult to
kind of say hey this is exactly what
happened and this is what caused it so
first thing we're going to do right now
is I'm just going to start Intel burn
test how long you want to run these
tests is kind of up to you right now
it's set to run 10 times stress level
standard we're just going to leave it
out of the box settings for all threads
um and we're going to hit start the
reason why the fire is burning right
there is not well I mean it is kind of
putting your CPU under load I we spiked
to 1. 1469 right there 1.47 1. 1499 is
the highest but under load right now you
see we're we're jumping around 13s to
14s which is a lot but as you can see
for the core temperatures right 58c it's
not putting that much stress on the CPU
right now but we could set this thing to
run a thousand times or 100 times or
whatever we want and keep it going but
the fire is there just so that you can
see the fire will stop moving if your
CPU crashed so you may not get an air
that pops up on the screen that's why
it's called a freeze test um anyway this
is just getting us RS or G flops right
here and a very big result number and
the amount of I think that's math I'm
not entirely sure what it's doing and
it's saying how long it's taken to do it
but again you can let this thing run for
a long time if you want this used to be
like the preferred stability test for a
long time the thing is this is a very
outdated tool and it's something that
allows us to be able to just start light
on the CPU so my recommendation success
your system was able to maintain
stability so I would recommend actually
setting this to like
100 and then letting it complete all 100
tests and we're going to do that right
now all right so that's been running for
a while it hasn't done all 100 it's only
probably done like 20 of them and uh
that's good enough for right now because
we need to move on we have a video to
make I Can't Take 5 hours to make this
video uh we'll do the testing and stuff
later in fact I know it's it's a good
solid CPU but who knows maybe it was
solid and we're going to find out so the
other thing I would do here before we
move on is I would move this down to a
single thread the reason why I would go
down a single thread I've already
explained it changes the way that the
CPU reacts uh and the voltage is based a
single core workload versus a multi-core
workload I would run that test for maybe
not necessarily 100 times unless you
have hours to set aside and let it go
because it's on this CPU it's taking an
average of 16 half seconds to get that
test done so that's going to be,
1650 seconds okay so next up is
cinebench R15 I think buildzoid was even
showing R15 runs were instable or
unstable for some of his CPUs so that's
one of the reasons why we're using it
again an older Tesla will run super
super fast so we want to go ahead and do
let's see Advanced Benchmark so we can
do single core and multicore so that's
what we want to make sure we're running
so CPU is all cores CPU single core is
just that single cores so let's just go
ahead and loop a few times a
multicore I forgot what this scene
looked like it's been so long this is
going to destroy this
chart yeah so we can just sit here and
run it multiple times unfortunately R15
doesn't have a timer built in that's a
R23 r24 feature so we're just going to
do this manually a few times Is Here
Yeah so we're at 1 Point High 1 ons low
one 12s at 52 allore that's expected
behavior and we reached 70c so we're
definitely stressing our CPU harder with
R15 you could get soft crashes in in C
bench too so depending on how hard it
crashes you might just get an error
screen that pops up with a cin bench
window and it will have a bunch of
gibberish on it saying an error and a
bunch of stuff and if you start getting
that consistently but your system
doesn't hang and you can still close the
program and stuff that's going consider
a soft crash um which is where it's able
to recover but if you are at all stock
speeds and you're starting to get those
kinds of crashes that's a sign that
something is wrong with your your system
doesn't necessarily mean the CPU is
degraded if you can verify that your
voltages are operating as they're
supposed to that's why I have Hardware
info open because I saw that we're going
to one high 1 ons low one 12s on the
voltage I know that that voltage is
correct for stock speeds on a
multi-threaded workload based on the way
the load line is with this particular
motherboard um some of you might see
higher at like 13 or something like that
but if you saw like 1 or like n you know
900 molts then you would be able to say
maybe I'm having an under voltage
problem with that particular motherboard
I only say that because um again build
Zoid showed he had a motherboard that
was going extreme undervolting at stock
speeds so that was causing a problem and
with the latest bios and went the other
direction so he had an undervolting
board installed the latest bios for that
particular m sideboard I forget which
one it is and then started hitting six
on the volts so again latest micro code
but the motherboard was forcing itself
upon the CPU in a very unfriendly manner
causing an an offensive situation which
at this point would be the voltage so
that's why I say it's not just the CPU
that could be at fault here now we could
do the same thing with single core
single core will take longer cuz it's
only one core but we can then make sure
that our voltages and our single core
stability is there that's important
don't only check multi-threading single
core is actually a bigger piece of this
puzzle than you would think so single
cor is getting ready to finish and I can
see core six and seven on the PE cores
are my preferred cores they're the only
two hitting 6 GHz on the uh single core
workload and they're not doing it at the
same time obviously they're handing off
back and forth um anyway about to finish
I would probably run the single core at
least 10 times so moving on that's R15
you can save the score if you want sure
R23
is a little more useful in terms of
being able to walk away from it because
of the fact that we can come over here
and we can do a minimum test duration if
that's not showing up make sure that
advanced Benchmark is checked so go to
file Advanced check Benchmark and then
you get the minimum test duration 10
minutes should be plenty but if you have
more time on your hands you really want
to get a good lay of your systems
Cooling and the voltage and frequencies
and how they're they're working set it
to 30 minutes and then I'll leave it at
10 right now but you can set it to 30
minutes and then get your Hardware info
up clear your Min Maxes and then hit
start on multicore it'll take a little
longer to run this test cuz it's
designed to take longer to run but now
once again we can check our voltages in
this type of workload which should be
very similar to R15 you can see again
High 1. ons low 1.2s to mid 1.2s um
under load kind of a little throwback
here to when I was doing the uh AO
testing and I showed you guys just how
good the 420 is no pun
intended but yeah 420 62c right now
looping cin bench R23 63c excuse me
63 clearly it sucks all right so
presuming you passed uh your 10-minute
or your 30 minute or your multiple 30
minutes runs without any failures do the
same thing with single
core and I would assume we're going to
see the same type of behavior here so
core seven and six should be our Max
cores yep 5985 GHz on p7 then just hand
off to P6 for a second and hands off to
P6 for a second it's like it has to
catch it breath for just a quick moment
get a few Huffs and puffs in there and
then it's like okay give it back so it's
just handing off that workload between
those to and then if this passes we'll
move on to the next one all right so
occt another free software that as long
as you're not using it for commercial
purposes or you're not using it to say
stress test you're a system integrator
and you're using it to stress test
systems all day long that's a commercial
use which they want you to pay for it um
we're not and you're not using it in a
commercial um en right now so let's go
ahead and use this so we want to make
sure CPU is checked not CPU and RAM um
not linpack not memory not 3D not vram
you can test vram stuff if you want or
you can test memory and Ram or CPU and
RAM but just CPU I have the mode set to
variable now steady is what I use when
I'm doing thermal testing variable is
what we want because this will allow the
workloads to change they'll be a little
more bursty they'll be a little more
consistent they'll move up and down
they'll hand off certain cores it'll do
One Core four core 16 core whatever
it'll it'll change up the workload and
make the CPU have to change gears to
keep up with the instructions now we can
set it to Auto SS AVX avx2 or 512 I'm
just going to leave it on auto let's let
it change around a bunch of different
instructions um because we want to get
as much varying workload on our CPU as
we possibly can also too this test as
soon as you hit start because of the
free version it makes you wait 10
seconds before it'll start that's a fair
tradeoff for it being free uh it will
run until you stop it or until it hits 1
hour I'm pretty sure the commercial
version allows you to change how long it
runs for I think you can go longer than
an hour and I think you could set it for
less than an hour but right now
automatic up to 1 hour is all you get so
we can hit start and there it is right
there so it shows how long it's been
running we got this little worm that
keeps going around the logo to tell you
hey things aren't you know Frozen we can
see our CPU our Watts our volts um and
then our current frequencies so again we
would let this run for a while so I use
the OCC test for uh lunchtime so it's
now about 40 minutes later well 39
minutes and 7 seconds later it's gone
through the test eight times it will
continue to go until the 1 hour so far
zero errors which is kind of nice to see
um okay so from here we're going to move
on to the one like sort of gaming test
like I said and this is only because of
the fact that they had said that heaven
did show up some errors for the bad CPUs
so Heaven Benchmark 4.0 basic Edition
doesn't really matter what settings you
use to be honest we just want the
process running um your GPU is going to
make a bunch of noise during this
because it is a GPU stress test um but
I'm going to turn off the full screen
check because I still want to be able to
see what's happening in Hardware info um
tessellation extreme quality Ultra we're
going to use system resolution just to
put some load and then uh we can bring
Hardware info back up to the front so
look at our volts right now right our
volts are all the way up there goes the
GPU fan finally the volts are all the
way up at one .4 volts so there's an
example of the fact that um it does put
the see look our our CPU is also maxing
out the core speeds it's at 57 all core
essentially um and then our e cores are
all the way up at 44 so this is another
this is another reason why it's a good
test because it actually is using the
CPU in the state that's putting the
volts high it's putting the frequency
High which would be another opportunity
for any of the uh issues to sort of show
up so I would let this one run honestly
for about a solid
um let the system get nice and hot let
the system get saturated and you should
not see any crashing what would
typically happen if it crashes though
what that crash sort of looks like it's
going to be a black screen uh you might
get a crash to desktop you might get
some sort of a error popup you might get
a blue screen but one of the more common
types of crashes for Heaven is going to
be just a black screen so assuming you
don't have any crashes here we'll go
ahead and move on to the driver
installer so for the driver installer um
again unfortunately this only works for
people that have Nvidia based graphics
cards uh but just download the latest
driver put it on your desktop double
click it hit okay let it
unpack see the thing is we're not we
don't need to do necessarily a clean
install we can click the clean install
button if we want it might make the
system have to restart but we're going
to do this like five times because this
is the part right there when it was
unpackaging and then we're going to say
install the graphx driver agree and
continue we can just do Express that's
fine and then preparing for install
right now it's going to basically
extract the driver it's going to get it
ready to be installed um it's going to
go through the process and then when
it's done we'll be back at the desktop
it'll say driver install complete it
could come back and just show an error
it may just it could come back and just
say error installing driver within the
driver program it may throw the error or
you might get a popup on the screen you
might get a Windows eror you might get a
blue screen there's all kinds of
problems that show up or all kinds of
ways that the there's all kinds of ways
that the um unpacking or the unzipping
of the driver can show up when it comes
to this particular test test but it's
ironic that this is the test that seems
to show the most consistent on the more
severely degraded CPUs and it's kind of
ironic because I've had people emailing
me asking me why their system keeps
crashing when installing the Nvidia
driver now this was a while ago when the
Intel problems were just sort of
starting um and I just was assuming like
corrupt driver installed me bad download
um bad GPU potentially but obviously now
it's like looking more like the
the okay that's the driver that's normal
it will go off and come back on it's
installing the driver anyway see there
we go driver uh installer has or the
install has finished so I would just
literally do it again now and then for
good measure when I'm done here I would
probably ddu this driver and then
reinstall it fresh again but what I
started to say is there's a lot of
different ways that the error manifests
itself could be blue screen could be a
white pop-up screen um it could be a
shut down and restart does all sorts of
things but it's just ironic that
installing something for for the GPU was
really sort of exposing a problem with
the CPU just like the out of video
memory when launching certain games when
it was compiling shaders again the
compiling of shaders happens the CPU is
compiling the shaders for the GPU right
so that's a CPU task and it's just
ironic that it leads to thinking
something is wrong with a component that
is completely innocent in all of this so
I'm installing the driver one more time
and I know this is a good CPU um I'm
going to probably try all this again
with a bad CPU uh I'm just waiting for
one to arrive where we can then
potentially see what it actually looks
like and the thing is it's not
consistent it's not a guaranteed oh a
bad this bad CPU or a bad CPU will crash
on this test no matter what it's just
not that way because they don't degrade
the same way the electron migration it's
not a cut and dry black and white type
of thing it happens weirdly or different
spots of the do it the wafer and the way
that the and the way the migration
happens is just not consistent so
because of that the way that the cores
are hit based on the instructions that
are being you know done or performed by
the CPU based on what background tasks
were run there's a million different
combinations of things that could be
happening to cause the errors and it's
just extremely difficult to actually
figure out what those are but I thought
it was important to do this video today
to give you guys some empowerment to say
hey you know what I want to make sure my
CPU is working um you see a lot of
comments online saying oh every single
13th and 14th gen is broken that's not
true every single CPU in the market
right now is degrading but it's
degrading at a at a expected rate great
it's like an engine in a car when your
engine is brand new and it's got 5 miles
on it it's certainly going to operate a
little differently than that same engine
200,000 mi from now in 200,000 Mi it's
going to have a little bit more rattly
noise it's going to have a little bit
more play the cylinder walls are a
little more worn in your piston rings
are a little bit worn you're going to
have a little bit of blowby maybe some
smoke you know but it's still running
right it's not failed on you yet until
something breaks so this is not
defensive Intel it's just I want to put
this out there that when people thrown
around the word degradation every CPU is
degrading but Advanced degradation or an
accelerated
deg an accelerated degradation
rate is not okay cuz that's what we're
dealing with here so anyway there we go
best practices right now might honestly
be don't leave your computers on when
you're not around don't leave them on
sitting at your desktop don't leave them
on sitting in your bios
just now know that there's a timer on
many of these CPU use and any time that
your system's on when you're not there
is just clicking away life it's like
that movie uh with Justin Timberlake
where you had a time on your arm and
that was how long you had left to live
imagine there's that on your CPU right
now and if you could turn off this the
know the system to save some of that
time you should probably do it anyway
hope this video has helped if you guys
have a bad CPU comment down below what
you were doing that you could like
recreate your scenario for your CPU to
fail everyone's scenarios are different
everyone's CPUs and and systems have
different programs and probably
different versions of Windows and all
different bios and stuff say down below
what your motherboard is which CPU you
have and what guaranteed test for you
can make your CPU give up the ghost all
right guys thanks for watching share
this video with someone that it will
help and as always we'll see you in the
next one
تصفح المزيد من مقاطع الفيديو ذات الصلة
Intel's new Microcode patch is HERE! Impact Testing Performance...
Motherboard Default settings could be COOKING your CPU!
This trend NEEDS to stop with motherboards!
TCHAU PLACA DE VÍDEO? AMD Ryzen 8600G e 8700G em MUITOS TESTES
Intel's Biggest Failure in Years: Confirmed Oxidation & Excessive Voltage
No One Is Buying AMD Zen 5, Post Launch Update
5.0 / 5 (0 votes)