Guide: How to check if your CPU is BAD!

Summary

This 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

00:00

🔍 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.

05:00

🛠️ 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.

10:01

💻 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.

15:01

🎮 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.

20:02

🏁 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.

Keywords

💡CPU Degradation

CPU Degradation refers to the decline in performance of a Central Processing Unit over time due to various factors such as heat, voltage, and usage patterns. In the video's context, it is a central theme where the script discusses the accelerated degradation of certain Intel CPUs and how it differs from the normal aging process of a CPU. The script mentions that every CPU degrades, but the concern is about the rate at which this degradation occurs.

💡Microcode

Microcode is a layer of code that is embedded in a CPU and is used to help the CPU execute machine code instructions more efficiently. In the video, the script discusses how updates to microcode can affect CPU performance and stability. It points out that even with the latest microcode, issues with voltage regulation and thermal throttling can still cause problems with CPU stability.

💡BIOS Update

A BIOS Update refers to the process of updating the Basic Input/Output System firmware of a computer to incorporate support for new hardware, fix bugs, or improve performance. The script emphasizes the importance of updating the BIOS, especially for motherboards that are behind on the latest microcode, as it can help mitigate issues related to CPU degradation.

💡Electron Migration

Electron Migration is a phenomenon that occurs in semiconductors where the flow of electric current causes atoms to move, potentially leading to the failure of components over time. In the video, it is mentioned as one of the factors that could contribute to accelerated CPU degradation, particularly when high voltage is applied over time.

💡VRM (Voltage Regulator Module)

VRM stands for Voltage Regulator Module, which is responsible for converting DC input voltage to the precise voltage levels required by the CPU. The script discusses how certain motherboard manufacturers might be 'going wild west with voltages,' potentially contributing to accelerated CPU degradation by not properly managing VRM settings.

💡Loadline Calibration

Loadline Calibration is the process of adjusting the voltage levels supplied to the CPU under different load conditions to ensure stability and prevent overheating. The script points out that improper loadline calibration by motherboard manufacturers could lead to excessive heat and advanced degradation of the CPU.

💡XMP (Extreme Memory Profile)

XMP is a configuration profile for RAM that allows for automatic overclocking of memory modules to their rated speed and timings. The script advises turning off XMP during CPU stability testing to ensure that any crashes or instability are not due to memory issues, but rather the CPU itself.

💡Stress Test

A stress test is a type of software test that is designed to evaluate the stability and reliability of a system under heavy load conditions. The video script mentions several stress tests, such as Intel Burn Test and Cinebench, which are used to determine if a CPU is stable and not showing signs of degradation.

💡RMA (Return Merchandise Authorization)

RMA stands for Return Merchandise Authorization, a part of the return process where a customer is given permission to return a product for repair or replacement. The script discusses the possibility of a CPU being ready for RMA if it has significantly degraded and is no longer stable.

💡Hardware Monitoring Software

Hardware monitoring software is a type of program that tracks the performance and health of a computer's components, such as voltages, temperatures, and frequencies. The script mentions using such software, like Hardware Info 64, to keep an eye on CPU parameters during testing to ensure they are within safe limits.

💡E-Waste

While not explicitly mentioned in the script, the concept of e-waste is relevant to the discussion of CPU degradation and the lifecycle of electronic components. E-waste refers to electronic products that are discarded after their useful life has ended, often due to performance issues or obsolescence. The video's theme of CPU degradation ties into the broader issue of e-waste and the need for sustainable practices in technology.

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