Intel has problem with their high end chips. 13900k's and 14900k's have stability issues after some time(even few months of use) likely due to degradation caused by running at a too high of a voltage. It's similar to what happened with 7800x3d.
It's just funny how many intel shills were saying that chip degradation was only an amd issue caused by their inferior quality.
https://www.theverge.com/2024/4/9/24125036/intel-game-crash-13900k-14900k-fortnite-unreal-engine-investigation
 It's All Fucked Shirt $22.14 |
 |
It's All Fucked Shirt $22.14 |
c2d chads can't stop winning.
2500k is fine too and it can take shitload of voltage
the smaller the node, the more prone to these issues chips become, factory overclocking just doesnt make sense on modern stuff
>the smaller the node, the more prone to these issues chips become
This is something I was wondering about the other day, but as much as I searched I could just not find information about it.
So which is it IQfy? Do they really become more failure prone at smaller nodes? And what was the sweetspot node? 14nm?
well its very node dependent, some node are just more fragile than others, there is a downward trend with voltage tolerance on newer nodes, but its not a guarantee, for instance, intel's 45nm more than 1.45v is bad, where as Glofo 32nm soi, is good up to 1.55v maybe even 1.6v, just depends on the design goals for the node, durability is something they can tweak, which costs density, and power efficiency.
2500k will start to degrade too if you push it beyond 1.4v (including transient overshoots). same applies to 14900K, which goes beyond 1.4v out of the box, kek.
>t.poorgays who can't run shit
I had a C2Q that got progressively less stable when overclocking, but that took many years.
in reality chip degradation is caused by morons overclocking, everyone intelligent underclocks regardless of where they got their cpu from
nobody does underclocks, intelaviv shill
do you also run your monitor at 100% brightness at all times then complain about it dying in 3 years?
I would complain.
I used my old monitor like that for close to 10 years and it's still working
your monitor is less than half as bright as it was new, you just haven't noticed.
this is normal (backlight losing brightness) and happens to all screens no matter how well kept (low brightness) they are. It's all a matter of (voltage) brightness and hours used.
My monitor is still bright enough that my eyes hurt from 70%+ brightness despite being on for hours a day since 2009, I think you're just making up excuses and bad arguments for shitty products sold by terrible companies. Which is weird, because you're arguing against your own interests. (assuming you arent some corporate shill wiener sucker)
There wouldn't be a 100% setting if the monitor wasn't meant to be using it.
yeah and manual car wouldn't allow you to reverse while going at 100kmh either if it wasn't meant to be done, dimwit
>y-your a dickrider or something
K chips simply mean that they are simply unlocked, no guarantees and no refunds
Are you really comparing manual machinery to electronics with limits set by the manufacturer?
Automatic cars don't allow you to go into reverse at those speeds
but you're complaining about chips that by design don't have limits?
Go overclock it to 9Ghz
Nah, i was talking about monitors.
And stock chips do have limits, they are just set a bit too high because intel is just being intel
>Intel has problem with their high end chips. 13900k's and 14900k's have stability issues
this is what OP posted on my monitor
the point is those chips degrade at stock limits and start crashing for many users.
>buy overclockable chip (aka too defective to sell normally)
>totally didn't overclock it
>crash at stock limits
I'm done responding to this thread
Are you moronic? overclockable intel chips run faster than their non overclockable ones. Non overclockable are the defective ones.
Too many are dying to blame the users, intel is currently investigating the issue.
>overclockable chip
if you have had any intel chip post 10series, they aren't worth overclocking. businesses like siliconlottery shutdown overnight because intel FINALLY started to send their chips out with settings that make sense. You're not squeezing noticeable performance out of chips by overclocking anymore, they're built to run strong out of the box.
Brainlet take, the real reason is that AMD/Intel has been factory overclocking their gayming SKUs to the bleeding edge since Kaby Lake/Zen+. Overclocking died that day outside of suicide runs.
kaby lake was 7000 series, you could overclock the frick out 7000, 8000, 9000, and even 10000.
poorgay larping like he owned any of these, probably pajeet still using 2 core amd from 2011
>you could overclock the frick out 7000, 8000, 9000, and even 10000.
Nope, all of those SKUs had almost no overclocking headroom outside of striking gold on the silicon lottery and/or exotic cooling.
6/10 got me to reply twice
>Poor-gay hard-stuck on decade+ old platforms detected
not without buying a delid kit and liquid metal
None of that stuff did jack with overclocking ceiling outside of striking gold on the silicon lottery.
It only made the thermals more tolerable at maximum load.
the more you reply the more moronic everyone sees you are. having lower temps = less voltage required = more stable at higher frequency.
Wrong, you need more voltage for stability at higher clockspeeds and this generates more heat. The vast majority of chips of that era at stock were practically near their ceiling and require a insane of votls to go beyond this.
Delidding on modern SKUs is practically ricing for a little extra thermal headroom. Outside of striking gold in the silicon lottery, you are lucky to get 5% more then maximum boost speed.
The days of chips being binned ultra-conservatively and having ~20-50% headroom with some basic armchair overclocking are long over.
Again, CPU overclocking is dead outside of suicide runs for epenis benchmarks.
>The days of chips being binned ultra-conservatively and having ~20-50% headroom with some basic armchair overclocking are long over.
>CPU overclocking is dead
We're not disagreeing here, I'm fully refuting the starting point is all. Rocket lake (11th gen) is the death of overclock. Prior to that you could still squeeze an extra 15%+ performance out of k chips.
>. Prior to that you could still squeeze an extra 15%+ performance out of k chips.
Only with golden sample yields. The vast majority of SKUs of that era you were lucky to get ~5-10% (that's 100-300Mhz) extra beyond maximum boost speed before resorting to massive overvolting.
Overclocking died with Kaby Lake/Coffee Lake on Intel front. Users were in denial then until it became painfully obvious with Comet Lake-Rocket Lake era. It is similar story with AMD front with Zen+/Zen2 when both companies decide to frick it with ultra-conservative binning to win epenis benchmarks and impress the "min-max" types chasing those extra 2-5% gains.
> Users were in denial then until it became painfully obvious with
People are still in denial
The whole industry is in denial
PC performance tuning is effectively dead but both MB vendors and users utterly refuse to acknowledge that.
MB vendors especially made their bed by basically cutting out actual PC users and instead segmenting motherboards by VRMs and memory speeds.
They would be losing a cash cow if CPUs are essentially untouchable and it made zero difference if you had a $200 board versus a $600 one.
Users are also in denial because it seems every year the standard for stability gets lower and lower as headroom disappears.
Going to be interesting the next few years. I sincerely hope rationality wins out and Intel/AMD really put their foot down for the sake of actual users and not moronic tuners.
>users want cpu manufacturers to use smaller and smaller transisters
>nooo, why is technology that requires more precision failing now?
we need to go back to 50nm
>transisters
heh
>PC performance tuning is effectively dead but both MB vendors and users utterly refuse to acknowledge that.
Only for laptops/desktop platforms as silicon has been plateaued. All of the excitement is elsewhere.
Yeah where?
Everything else is lock down.
Servers, HPC and workstations. Performance and effiecny has been jumps. It is the reason why whole ML meme is even a thing. Such commercial applications were not economically feasible due to be limited to expensive and PITA to operate supercomputers.
Again, all of those are locked down
For those who actually care about real work and compute are not concerned.
Tweaking and min-maxing limited hardware platform is a silly exercise.
I don't want a $200 motherboard that doesn't have good routing and signal stability for RAM tuning, and was designed to have a 12,000 hour lifespan at 55C temps with 70 amps on a low end 4 core 60 watt CPU.
and the UEFI on cheaper boards is usually SHIT
>and was designed to have a 12,000 hour lifespan at 55C temps with 70 amps on a low end 4 core 60 watt CPU.
That's just your paranoia
MB vendors predict no difference in durability between motherboards. It's why they all have the same warranty.
you didn't have to buy a delid kit for 7000 or 8000 series, they didn't solder the shit til 9000, you could just throw the chip in the oven and melt the epoxy and easily pull it apart.
>comparing an LED backlight to a CPU with billions of transistors
dumb Black person
same thing idiot, if you ramp up power high enough, your CPU will glow even brighter than the monitor
LED screens are literally meant to be ran at 100% brightness. Not everyone is a night owl or lives in a dark basement.
I've never had a monitor die before the ten year mark, running them always at 100% brightness, /b/ro. I gave away my old monitors to Goodwill, no idea what happened to them later. They might still be in use in new homes.
Yes because I didn't buy OLED meme monitors so I can actually use my monitor without a billion caveats
mkg and crt gays winning. wasnt everything better.
Yes, people that care sane thermal output and power consumption have been undervolting and underclocking. They are essentially reversing the silly trend of factory-sanctioned overclocking that AMD/Intel have been doing Ryzen+/Kaby Lake.
I undervolt my 13700k.
2 cores boost to 5.4Ghz at stock while the rest run at 5.3Ghz.
So I just run all core 5.3Ghz and undervolted. Dropped an entire 0.1mv, 25C and 50W.
i did the same thing, it's absurd that that isn't the default, I hate intel
That is the motherboards doing it, not Intel specifically.
However I wouldn't be surprised if Intel was telling them to.
Motherboards are allowed to set higher power limits. Intel allows it which means their boost algorithm should take care of keeping the system stable. AMD doesn't allow mobo makers to mess with power settings so boards never ship with PBO enabled.
what's the name of the turbo/overvolting that takes place on a stock ryzen (motherboard)?
Precision Boost. AFAIK it's locked down under AMD's control and motherboard makers can't modify the algorithm. You can use Precision Boost Overdrive to set power limits, voltage offset, boost target, etc but it voids the warranty and is never enabled by default.
I love when things have the word boost in it. You know it's about to be stronger
so there's no way to switch off or reduce Ryzen boost from the BIOS?
based autist
my friends zen 2 ryzen motherboard do 1.4v voltage spikes stock, is that really within amd spec?
anecdotal evidence but buildzoid and many others zen 2+ have died andor degraded within stock or light overclocks
worrying about cpu defects and the like we've had to do with ryzen is unprecedented, and now it seems like the hybrid slop intel cpus are pulling the same shit. like others said i firmly believe monolithic intel 10th gen were the last good cpus. now it's just flimsy slop
>my friends zen 2 ryzen motherboard do 1.4v voltage spikes stock, is that really within amd spec?
Yes, my 5800X sits at 1.45 V most of the time because I'm running the most aggressive performance settings.
>anecdotal evidence but buildzoid and many others zen 2+ have died andor degraded within stock or light overclocks
Zen 2 could only overclock with a static voltage which is really dangerous. On stock it goes down to like 1.2 V in benchmarks.
>i firmly believe monolithic intel 10th gen were the last good cpus
I can agree with that, the thing I hate the most about Ryzen is how it goes behind your back and starts silently error correcting. For example, I can set Vsoc to 1.0 V or 1.15 V and the system seems fine on both but memory performance is much better on 1.15 V. CPUs should be straightforward and predictable.
anything above 1.3v vcore can kill cpu (except sandybridge toward comet lake)
yestech city post this so keep it low temp no overclocking and vcore at stock
I haven't tweaked anything from the stock settings since re-installing Windows a few months back so I've been letting it hit up to 1.450V by virtue of the clock speeds being unlocked but other than that I've always kept the fans running on the Hyper 2012 attached to it along with keeping the case open all the time so iirc the temperatures never exceed anything beyond 70C. Voltage usually seems to spike when in desktop on this chip (bought it in 2019). Might go back into the BIOS to take a look at it however or just lock it down again.
Also, as far as I remember it can also be dependent on the motherboard as far as how aggressively the voltage is used in addition to the chipset drivers. Though 1.488V seems consistent for most users of the 3700X along with idle temps of 40-50C.
Try updating the bios. My asus b450 was running 5700x at up to 1.45v. After update it's under around 1.22v and the performance didn't change.
3600s pull 30watt on average what the frick is this moron talking about
10th gen has a chronically unstable ring bus that rears it's head in RT games and other specific scenarios
That only applies to the i9
Nope. i5 through i9 share the same ring bus
>RT games and other specific scenarios
Examples?
Ryzen CPU's have been risky to run higher voltages. We saw the end result of that with the 7800X3D chips blowing up. They have returned to safer voltages entirely.
OCCT stable, good enough for me.
>We saw the end result of that with the 7800X3D chips blowing up
That was the IF/memory controller voltage which CAN be modified by motherboard makers (and they were pushing it way too hard to get DDR5 stable). AMD CPUs have high core voltages at stock but this has never been a problem, lots of people still have Zen 2 and they work fine.
You can disable Precision Boost, makes the CPU run at base clocks.
>You can disable Precision Boost, makes the CPU run at base clocks.
I can only see something called Core Boost, which expands into its own menu. I'm on a B450 motherboard
That's probably it but there's no reason to disable it. All it does is make your computer slower for no reason. If you want to save power, enable Eco Mode.
>Eco Mode
i could be wrong but i don't think i saw anything like that. i run a passive cooling setup so i just don't want any unnecessary voltage/heat spikes if i can help it.
I thought my -0.03v undervolted 13900ks was stable until I ran Tekken 8 demo which crashed on the first time I started a new game
Raptor lake instability is a real thing in UE5 games tbh
>a whole 0.0001 volts
I put my 5950X in a slow mode and it still runs perfectly fine. This means it stays in the 30-40C range - never going above 40C and never going higher than 3.6Ghz which is more than enough. It even plays X4 Foundations perfectly to my surprise.
Do you put plastic wrap over your toilet paper too? <10 nm CPUs are consumables like SSDs. Use em up and throw em out.
SSDs last longer now, CPU transisters, we are dying too fast, our response?
It's not about that. It's the typical annoying ayyymd random boosts that annoy me which flare up the fans every 2-5 minutes. Who fricking cares tho? The pc runs exactly the same for what I use it for. If I do need the 5ghz boost, I wil turn it off.
This.
I swap out my CPU daily. Microcenter sells 14 packs
If by morons you mean motherboard manufacturers, yes.
Intel just didn't stop them from doing it while with amd few manufacturers went over the spec
this.
My MSI motherboard was shoving in way more power than it needed. Because motherboard manufacterers don't want their motherboard returned on a bad cpu that doesn't run within spec.
I would.
If I ran my motherboard at 110% brightness I wouldn't.
my phone warns me every time I set brightness to max, and it is really bright
Thank you for correcting the record, rabbi!
>buy chip with K unlocked mulitplier
>literally meant to be overclocked
>d-don't overclock everyone intelligent wastes their money buying a high-perf CPU and running it like a laptop chip
you are worse than Nvidia fanboys at this point, at least they get decent products and not this stupid hybrid architecture that Intel cheaply copied from their mobile chips.
By the way 10nm was already around for smartphones in 2017, but yes goy keep buying the next 10nm+++ 450W (undervolted) space heater.
you can't underclock a non k cpu.
i run mine at 2.5Ghz for daily use and only at 4 Ghz when gayming
Yes, you can. Non-K SKUs are half-locked you can only decrease the multiplier as this is done for EIST/Speedstepping. Only Uncore, reference clock speeds are fully locked.
Imagine being this much of a dick rider. Not only are they K chips explicitly unlocked for overclocking, a feature that intel ~~*innovated*~~ selling, but even if you cuck to not overclocking several of the big motherboard manufacturers were overclocking the chips by default out of the box without Intel stepping in to tell them no while riding the performance numbers this bullshit yielded. Frick off you absolute homosexual.
Nah, K series main thing is that Intel factory overclock thems with their boosting which is enabled by default.
Overclocking has been dead for years outside of suicide run. The overclocking crowd has shifted their gears towards undervolting. Trying to get the silicon to operate at stock speed with as little power and voltage as possible.
Overclocked for 12 years with one of their older chips with 0 issues
>underclocks
You mean underVOLT, right?
13900k and 14900k are basically factory-overclocked to the max already. you can get like +200 MHz at best without LN2
>t. used mining gpu seller
Intel motherboard manufacturers defaults to unlimited OC to push their motherboard sales and for Intel to get good CPU reviews. Intel loves it
Shalom!
>at least they get decent products
lol novideo shill really wrote this
Their chips at least don't degrade so rapidly
It is because both companies are trying keep "min-max" types excited over 2% gains and are resorting to what formerly as "hardcore overclocking" on cherry-pick SKUs at the factory. This is all to mask the cold reality that silicon has been plateauing on miniaturization.
this is why 10th gen 10700k is endgame
while in reality all you need is i3 (and AMD equivalent of that)
>and AMD equivalent of that
no
all ryzens are minmaxxed with buggy dogshit firmware
nice
https://cpu.userbenchmark.com/Compare/Intel-Core-i7-12700-vs-Intel-Core-i7-10700K/m1750830vs4070
k
>3.5 gazillion homiehurtz
>7 trillion homiewatts
>somehow not stable, melting and catching fire
how could this be???
why cant they just put less voltage innit?
nah g you gotta crank it senpai 350W gang melt it
>tell me you know nothing about electricity without telling me
you can send 350w at 1.1v or you can send 350w at 1.7v, these are very different things
>you can send 350w at 1.1v or you can send 350w at 1.7v, these are very different things
i was taking the piss homeboy, but ok, since you brought it up, why don't you tell me the difference between the two and i might let you off the hook
the difference is i can't jelly my 350w dick in your mom with 1.1v but i can jam it in with 1.7
man i was hoping for something a bit more uh, ejoo-kayshonal from a big man like you
but i guess you was bluffin and you cant into science fo shieeeet
Assuming resistance is the same, to send the same amount of wattage at a lower voltage you need to increase the amperage. That means the conductor needs to be physically larger since more heat will be produced. That's why sockets and cables have amp ratings, not volt ratings.
So undervolting my GPU will lead to higher temperatures?
lower temperatures, more crashes and unstability.
No, because you cannot "overamp" a CPU while undervolting it in order to maintain a constant wattage.
consumerism created the most moronic peoples in history
Karma for lying about Ryzen chips degrading
chip by default requests insane voltage, if you have a x900k you should set voltage manually
Aren't both of these made on TSMC 5mn?
Could be a TSMC issue
I thought they were going to use tsmc starting from 3nm
no all the "nm process" shit isn't a measurement, it's marketing jargon.
>he buyed intel
have any of you read the article? its complete nonsense there is no indication of degrading chips. why would a cpu issue only be happening in some ue5 games and be giving “out of memory” errors? clickbait nonsense
You can find more information on reddit or youtube but i didn't dare to link that in op.
The issue happens only on intel's 13 and 14 gen and nvidia even said to contact intel support and not them
https://www.tomshardware.com/pc-components/cpus/nvidia-blames-intel-for-gpu-vram-errors-tells-geforce-gamers-experiencing-13th-or-14th-gen-cpu-instability-to-contact-intel-support
buttholes just want more money
why are intcelaviv boys so sensitive to literally any intcelaviv criticism? imagine obsessing this much over israeli corporation
>intcelaviv
remove the t, homosexual
Why are AMD users so asshurt about intel all the time?
Does AMD not work unless you shit on intel everyday?
Criticizing a company doesn't mean you're vouching for the competition, especially not when there's multiple companies on the same market
are companies handing little IQfy trolling playbooks as standard guerrilla marketing tactics nowdays?
It's exactly the same as snoy threads on vee
It's just turd worlder kids getting their first CPU for scraps.
Chips don't degrade, users are just moronic or it was a poor chip/bin from the start.
They do degrade from overvolting and overclocking from abuse. AMD and Intel have been low-key doing this on their gayming SKUs since Kaby Lake/Zen+. High-end 14th series and X3D SKUs have been dancing at the literal bleeding edge. It is the 1.13Ghz Pentium III and FX-9590 over all again.
>FX-9590
Did it degrade? Those CPUs could take a lot of abuse and survive for years. Same for the 28nm APUs (excluding Excavator). It was fine as long as you kept it cool
yes, all electronics degrade.
>Chips don't degrade
ALL THINGS DEGRADE FROM USE PERIOD THIS IS A LAW OF THE UNIVERSE.
The question is how rapid.
>The question is how rapid.
depends on clock rate and voltage
>Chips don't degrade
Are you moronic? Haven't you hear about post mining gpus unable to work at stock clocks?
i love virtua fighter
Ensure your shit is running BELOW 1.3v if you want it to last. Applies to both amd and intel, these israelites tell you that >1.45v and 95c during loads is normal.
>Have 14900k
>Running at stock speeds.
meh
How long ago did you get it?
We might soon get some tools to test if the cpu is working properly, in case of any issues you could use warranty
>by running at a too high of a voltage.
Skill issue
I wonder how many of those are caused by running them under air cooling
>intel in current year
Lol
People still buy them because they trust them. with all the people saying amd is unreliable compared to intel i'm not surprised
>Meltdown, Spectre
Those alone have destroyed Intel's prestige that it build over the decades.
Name one (1) actual case of that being exploited in the wild.
Probably was used but likely nobody had noticed. The problem is that was problem was revealed to the public and enterprise/SMB market can't ignore it or be liable for nasty lawsuits in event of a reported attack. This forced downtime on updates and you are stuck with hardware that will never operate and perform as advertised increasing TCO and reducing life cycle.
Sysadmins and PHB-types will remember this for a long time. It shattered the multi-decade notion that "nobody got fired for buying Intel"
How much is high of a voltage? My R3600 runs at 1.4v when full load
don't care, still sticking to my 13900k
Until it randomly dies in a year
>degrading
AMD shill alert. Not even the source you link suggested degrading
>The only workarounds that seem to improve stability involve manually downclocking or undervolting Intel’s processors.
>changing the SVID behavior to Intel Fail Safe in the BIOS settings of Asus, Gigabyte, or MSI motherboards
So I gather they are being pushed too far and also overheating. But who knows, normalgays aren't known for having quality setups.
Works on my machine.
It's worse than degradation
A portion of 14900k/KS chips are outright unstable at stock clocks in cinebench if you lift the current and power limits to unlimited
You're telling me running silicon at 5-6 GHz pulling close to 500 watts at the wall isn't sustainable?!
Intel bros?! what are we going to do?!
>500W
>gaming
Here me out for a second......
HAMSTER
WHEEL
Think about it
>motherboard pushes CPU up to 1.4V
>running XMP on DDR5 pushes the SA up to 1.4V
Intel chips have always been monsters in terms of stability and overclocking but since they aren't withholding 25% clockspeeds anymore, the cracks are showing.
Nah, they haven't been monsters since Haswell. The cracks have been there now it is massive gaps building up.
>xhe needs to do all sorts of tweaks to keep xir cpu from blowing up
KWAB
A rare look into one of the JIDF headquarters. Imagine the smell
i have a 13900k, no issues so far
because it's nothingburger, there's always CPU's that are broken but somehow get past the testing, some people get unlucky
>CPUs degrading like milk when mobile archs are pushed to limits
>DDR5 barely surviving with on-chip VRM & on-chip ECC
>NAND quality nosediving with QLC
Enjoy the era of fragile hardware, IQfy.
Smaller things require higher precision equipment that's more expensive, and lithography is already a method where 30%, sometimes 40% and even 50% of chips are dead before they ever get out of the factory. This is a problem at all sizes.
>NAND quality nosediving with QLC
the point of QLC is being cheap, and it unavoidably comes at the cost of quality, but again, that's precisely the point of QLC
Intel is the blizzard of cpu manufacturers, all the talent has left. Pushing an extra 100w for a few more fps and in return forever damaging your image as reliable. moronic is an understatement here.
Powerlimiting my 12900k to 125w, and have yet to notice anything but lower temps. It's sad that 90% of users will never realize this.
Most "gaymers" are moronic goyslop addicted honkey donkey brain damaged serotine rich soft skin baby man childs.
too many buzzwords, go and fast from online posting
the issue is motherboard manufacturers by the sounds of it, using reasonable settings doesn't seem to cause any issues. the default settings on most mobos try to heavily overclock cpus that are already heavily overclocked out of the box.
i was also able to undervolt my 14900k to around 90w and still maintain most of the performance, i wouldn't call them bad cpus just that theyre poorly configured to achieve higher benchmarks for their marketing
Raptor Lake and Zen4 are both extremely efficient if tweak the chips to follow their efficiency curve. This is what CPU vendors used to do on their desktop chip before Kaby Lake/Zen+. The factory-sanctioned overclocking for a 5-10% increase in silly benchmarks to the wow the min-maxers for nearly double the power consumption is insane. Leave that to the hardcore overclockers who know the risks.
Marketing people are really cancerous
just
turn off
turbo boost
>buy car
>take it to auto shop that does modifications
>blows up
yes, and? your post isn't related to what's happening
Well it is. Intel has their specs but motherboards push them to their limit to look "better" for investors / consumers.
yet somehow amd doesn't blow up with boards made by those same manufacturers, very antisemitic
It literally did. amd had this problem first and it's fixed now.
when it was happening with amd they called out mb manufacturers fast while intel is holding on. they might've set their own spec too high
>they called out mb manufacturers fast
With AMD it was pretty obvious what was happening, in hindsight anyway.
With Intel, it's more difficult because this mainly seems to affect the highest end parts and you have to investigate how much of this is tweaker morons or motherboard vendors.
Intel attracts some of the biggest morons that'll blame windows for kernel panicking instead of their "stable" 7200mhz ram OC
It's also known that motherboard just do whatever for power limits.
post examples of intel chips being shit
13900K almost 2 yrs old on a 240AIO running PL1/125 PL2/253 stock settings and everything is fine.
Wasn't this said to be a problem with the 3700X and Ryzen 3000 in general too when running unlocked? I recently started running it without having it locked to 3.6 GHz again.
>anything above 1.3v vcore can kill cpu
why are browns typing about technology
I'd sooner believe AMD marketing was behind the entire wave of identical copypasted GPT articles claiming intel chips were suddenly "being returned by the millions" for issues they don't have.
You can believe in whatever you want. some people believe in chemtrails, some in flat earth
So I'm right, and you're coping by citing flat earth and other boomer conspiracy theories.
>AMD having good marketing
we both know that aint it
Running at 1.5V to push that 6 Ghz while consuming 400W for few months at a time can degrade the CPU? WHO KNEW?
saved
10900f master race wins yet again!
werks on my machine
>It's just funny how many intel shills were saying that chip degradation was only an amd issue caused by their inferior quality.
It was an AMD issue. I guess now all chips have this issue.
12900bros, we won
It has nothing to do with electromigration and everything to do with Black personlicious motherboard settings.
So how many people in this thread have a 13th or 14th gen i9? And how many have had issues?
ran Pentium 4 for about 15 years on 100c
ring-bus anus architecture and estrogen cores are not it
sorry