News Ryzen Burnout? AMD Board Power Cheats May Shorten CPU Lifespan

[Admin]

HWinfo releases a new feature that detects which motherboard vendors are cheating AMD Ryzen power limits, which can result in decreased CPU lifespan.

Ryzen Burnout? AMD Board Power Cheats May Shorten CPU Lifespan : Read more

 

[flyrobot27]

Bruh Isn't that the same thing as Intel's MCE? This is such a shill post

 

[Deicidium369]

So admitting that the Ryzen CPU WILL burn out - even without the motherboard shenanigans. Is this called "Heritage Mode" by chance - because the Athlon XPs loved to burn up. Some things never change.

This is why AMD keeps AM4 around - since it's CPUs WILL burn out at some point - and need to be replaced. All makes sense now.

 

[JamesSneed]

Bruh Isn't that the same thing as Intel's MCE? This is such a shill post

He wast calling out AMD but instead the motherboard manufactures. Did you even read this article?

 

[Alvar "Miles" Udell]

What happens when you take a 3700X, manually set it to 4300mhz all cores, and use 1.2v? Efficiency!

 

[escksu]

"however, it also makes the CPU run hotter and potentially negatively affects its life-span, the same ways as overclocking ".

"Intel expressly approves and even encourages motherboard vendors to adjust power limits to differentiate their products, and those adjustments don't impact chip longevity within the warranty period. "

"In contrast, this practice reportedly isn't sanctioned by AMD and could kill your chip sooner than expected. "

PBO itself is overclocking, period.......

And then, I am not sure what the article is implying..... Is it saying that Intel chips are good for oc and AMD sucks?? I don't know....

 

[ezst036]

So admitting that the Ryzen CPU WILL burn out - even without the motherboard shenanigans. Is this called "Heritage Mode" by chance - because the Athlon XPs loved to burn up. Some things never change.

This is why AMD keeps AM4 around - since it's CPUs WILL burn out at some point - and need to be replaced. All makes sense now.

If you read the link that goes to hwinfo .com you will see that this is the motherboard vendors doing it, not AMD. Specifically named was MSI. It looks like another user reported their Gigabyte board doing it.

I doubt AMD thought to themselves in early/mid 2016 "You know what? MSI is going to juice our processors in 2020, so we better build AM4 to last!"

Separately I had a thought about running processors on the hot side, albeit at stock ratings.

Personally, I keep a spare processor on hand. I run my processors fanless, with massive heatsinks in order to get 100% silence. No fans, fanless power supply, everything. Even at full load I rarely go above the mid-hi 40C range. As I type this it's just under 20C.

Processors get cheap fairly quickly on Ebay, after a few years. So that its said, I hope the motherboard vendors issue a BIOS update to correct this, now that they have been caught and called out.

 

[escksu]

What happens when you take a 3700X, manually set it to 4300mhz all cores, and use 1.2v? Efficiency!

LOL, I think some of us have read about excessive voltage used by PBO....So, how ironic.....

 

[Dijky]

So admitting that the Ryzen CPU WILL burn out - even without the motherboard shenanigans.

Where did you get that from?

Anyway, it's true that silicon ICs will degrade over time. That applies to varying extents at all voltage/power/temperature levels across all manufacturers.
The typical effect is that an IC will gradually requiring higher voltage to achieve a certain frequency, and ultimately fail.
The specification targets a certain rather long lifetime, but increasing the operating parameters beyond specification can significantly reduce it.
This is not news and is widely known to overclockers.

This is why AMD keeps AM4 around - since it's CPUs WILL burn out at some point - and need to be replaced. All makes sense now.

Be advised that excessive weight of tinfoil on your head can cause neck pain.

And then, I am not sure what the article is implying..... Is it saying that Intel chips are good for oc and AMD sucks?? I don't know....

I'm reading the article and also The Stilt's original post as putting the blame on mainboard makers here, because they are violating the specification:

"I'd like to stress that despite this exploit is essentially made possible by something AMD has included in the specification, the use of this exploit is not something AMD condones with, let alone promotes. Instead they have rather actively put pressure on the motherboard manufacturers, who have been caught using this exploit," The Stilt added.

 

[escksu]

If you read the link that goes to hwinfo .com you will see that this is the motherboard vendors doing it, not AMD. Specifically named was MSI. It looks like another user reported their Gigabyte board doing it.

I doubt AMD thought to themselves in early/mid 2016 "You know what? MSI is going to juice our processors in 2020, so we better build AM4 to last!"

Separately I had a thought about running processors on the hot side, albeit at stock ratings.

Personally, I keep a spare processor on hand. I run my processors fanless, with massive heatsinks in order to get 100% silence. No fans, fanless power supply, everything. Even at full load I rarely go above the mid-hi 40C range. As I type this it's just under 20C.

Processors get cheap fairly quickly on Ebay, after a few years.

Sounds like you are living near north pole......

 

[Alvar "Miles" Udell]

LOL, I think some of us have read about excessive voltage used by PBO....So, how ironic.....

If a value of 50% means it is consuming twice as much power as it reports, then a value of 150% means it is consuming half as much power as it reports, then again it expects to be running at an all core speed of something like 4100mhz at 1.4v, not 4300mhz at 1.2v.

 

[Math Geek]

i'm not sure what the net negative effect is for this??

by reporting lower power usage then PBO would bump up performance accordingly. however, there are still thermal limits that keep it in check. seems to me like this is a way to kind of fake PBO functionality if the feature is turned off.

sneeky and all that but i'm not sure there should be any negative effect from this that enabling PBO would not accomplish.

i doubt that the mobo maker would allow this to crank up the power and fry the board itself since that would be counterproductive for them. so considering the thermal limits are still there, there would not be any runaway power usage that would go unchecked.

or am i missing something here?

 

[PaulAlcorn]

"however, it also makes the CPU run hotter and potentially negatively affects its life-span, the same ways as overclocking ".

"Intel expressly approves and even encourages motherboard vendors to adjust power limits to differentiate their products, and those adjustments don't impact chip longevity within the warranty period. "

"In contrast, this practice reportedly isn't sanctioned by AMD and could kill your chip sooner than expected. "

PBO itself is overclocking, period.......

And then, I am not sure what the article is implying..... Is it saying that Intel chips are good for oc and AMD sucks?? I don't know....

This isn't with PBO though. Straight stock, no OC of any variety.

 

[PaulAlcorn]

i'm not sure what the net negative effect is for this??

by reporting lower power usage then PBO would bump up performance accordingly. however, there are still thermal limits that keep it in check. seems to me like this is a way to kind of fake PBO functionality if the feature is turned off.

sneeky and all that but i'm not sure there should be any negative effect from this that enabling PBO would not accomplish.

i doubt that the mobo maker would allow this to crank up the power and fry the board itself since that would be counterproductive for them. so considering the thermal limits are still there, there would not be any runaway power usage that would go unchecked.

or am i missing something here?

yeah, but PBO voids your warranty. This 'hack' isn't exposed to the user as happening, so they have no idea. This occurs at stock settings.

 

[Math Geek]

ok so that's the part i was missing. did not realize PBO voided warranty since it is an amd feature.

i'm interested in what the testing shows. i still don't see how this would be a long term problem other than the owner not knowing it was happening. not a fan of sneeky like this for sure.

 

[escksu]

ok so that's the part i was missing. did not realize PBO voided warranty since it is an amd feature.

i'm interested in what the testing shows. i still don't see how this would be a long term problem other than the owner not knowing it was happening. not a fan of sneeky like this for sure.

Ya, I didn't know that too....

 

[Math Geek]

i expect Paul knows what he is talking about, and i have never read up on it myself. not that important to me really.

it does seem un-related to the topic either way. my thought is faking voltage numbers to trick the cpu to speed up to make the mob look better, though sneeky and i don't like sneeky, would have little effect on how long the cpu lasts considering this is basically PBO without the owner knowing it was happening. again i don't like them doing this either, but not sure the effect would be any different than actually turning on PBO as far as longevity is concerned. of course if you turned off PBO and NEVER intended to use it, then you could argue the cpu might not last as long as it would if the mobo maker was not forcing un-stock settings through this exploit. i agree there.

i'm waiting for the testing to see what conclusion if any can be reached. but that's just my guess based on my understanding on how things work. Paul gets paid to do this for a reason as he knows more than i do for sure

 

[agello24]

you guys along with Hardware unboxed are always going after AMD. have you guys changed out your old overpriced xeons for epyc chips yet?

 

[Dijky]

this is basically PBO without the owner knowing it was happening.

Another key aspect is that not only the consumer doesn't know, but the CPU's power management doesn't know either.
I don't know what the processor may be able to do to protect itself when exceeding stock configuration, but it definitely can't do that if it is not aware of it. For example (entirely hypothetical) the CPU might still cap the power or voltage after a certain duration of beyond-stock boosting, or apply a non-configurable upper limit to certain parameters.

 

[Math Geek]

Another key aspect is that not only the consumer doesn't know, but the CPU's power management doesn't know either.
I don't know what the processor may be able to do to protect itself when exceeding stock configuration, but it definitely can't do that if it is not aware of it. For example (entirely hypothetical) the CPU might still cap the power or voltage after a certain duration of beyond-stock boosting, or apply a non-configurable upper limit to certain parameters.

that is def an unkown about this whole business. that's why i was thinking thermal limits would come into play. if false readings are given, i'm not sure how any power limits would be effective like they should be. i have to agree with you 100% there. hopefully that is what further testing will investigate as it does seem like any power limits to prevent damage would be compromised.

 

[helper800]

With R20 I just tested my setup which is a 3900x on all stock settings in an Asus x570-f motherboard with BIOS version 1407 AGESA version 1.0.0.4. Hwinfo reported within The Stilts margin of error (+-5%) at between 95.5-98%. Interestingly enough CPU-Z benchmark made my motherboard dip to 90-95%. So there may be a minor amount of dubious editing by ASUS but seems to be within the margin of error to me.

On another note, some user's here seem to be confusing this practice with PBO and it is not the same thing. This is the motherboard manufacturers' maligned effort without consumers' knowledge to make their products seem higher performance than others even when there is little to differentiate themselves from each other. The motherboard manufacturers are modifying the underlying AGESA code in their firmware to trick the processor into believing there is more room wattage wise to boost and or maintain higher clocks. AMD's limits power in wattage for each SKU so at least this practice is not frying CPU's, however, what The Stilt is trying to do is inform users that this practice may impact the longevity of each processor.

 

[escksu]

The article needs to explain it properly.

1. This is pretty much same as increasing the power limits on GPU. The article should explain properly what happens and how it can affect CPU, not just 1 sentence "CPU would allow itself to consume twice the power of its set power limits, even when at stock. ". It only confuse readers, making pple think that CPU power consumption at stock is doubled.....Should explain properly that this affect the TDP limit set by AMD. It does NOT mean its consuming 2x the power all the time.

2. It may affect CPU lifespan, but under certain condition. Eg. 3600 TDP is 65W and 3600x is 95W. Did AMD did some magic to allow 3600X to have 30W higher TDP?? Yeah, bigger heatsink.....Something all aftermarket coolers can do. Its not like 3600 cannot do 95W, the crappy stock cooler is to blame.

3. "same ways as overclocking does. " This is the kind of statement that further confuse people. Because we all now know that both Intel and AMD does dynamic overclocking based on factors like power, thermal and load. Thats why we have base clock and boost clock. And today, factory boost clocks are very close to limits.

 

[helper800]

you guys along with Hardware unboxed are always going after AMD. have you guys changed out your old overpriced xeons for epyc chips yet?

You either didn't actually read the article or, you didn't understand it. This is motherboard manufacturers doing stuff behind consumer's backs to the condemnation of AMD. AMD is not being attacked in any way. Motherboard manufacturers are going against AMD's strict guidelines for how power management of their CPU's should be managed.

 

[helper800]

The article needs to explain it properly.

1. This is pretty much same as increasing the power limits on GPU. The article should explain properly what happens and how it can affect CPU, not just 1 sentence "CPU would allow itself to consume twice the power of its set power limits, even when at stock. ". It only confuse readers, making pple think that CPU power consumption at stock is doubled.....Should explain properly that this affect the TDP limit set by AMD. It does NOT mean its consuming 2x the power all the time.

2. It may affect CPU lifespan, but under certain condition. Eg. 3600 TDP is 65W and 3600x is 95W. Did AMD did some magic to allow 3600X to have 30W higher TDP?? Yeah, bigger heatsink.....Something all aftermarket coolers can do. Its not like 3600 cannot do 95W, the crappy stock cooler is to blame.

3. "same ways as overclocking does. " This is the kind of statement that further confuse people. Because we all now know that both Intel and AMD does dynamic overclocking based on factors like power, thermal and load. Thats why we have base clock and boost clock. And today, factory boost clocks are very close to limits.

1. I agree, more could have been said to inform the less advanced users.

2. No. The TDP of a processor is not determined by the stock cooler included for the CPU's. The difference in TDP between the 3600 X and non X is determined by AMD's intended purpose of the CPU. AMD wants the 3600X to have a higher TDP because it is a better-binned silicone and its purpose is to be a higher performance part than the 3600. You need more power to get more performance, within reason, for CPU's.

3. No, you did not understand that statement with the proper context.

 

[NightHawkRMX]

In fact, nearly every motherboard vendor makes adjustments with Intel's chips, but there's a big difference: Intel expressly approves and even encourages motherboard vendors to adjust power limits to differentiate their products, and those adjustments don't impact chip longevity within the warranty period.

Look at what you say here: "Within the warranty period." The warranty period on most Intel CPUs is 3 years, or a pathetic 1-year warranty for a CPU like a 9900ks. So Intel isn't saying this won't effect longevity at all, and it certainly will. It is a well-known FACT that If a CPU runs very hot and the motherboard is allowing a lot of extra current through the CPU, it would have a negative effect on any processor, regardless of brand.

How is it that when Intel motherboards violate power limits by default and allow the CPU to draw far more current than is within Intel's spec which causes the CPUs to run hotter, you say doesn't have any effect of CPU longevity and don't bother to make any article. However, when AMD motherboards violate power by default limits causing the cpus to run hotter, you make an article titled "Ryzen burnout" and say that this may have damage to the processor.

Explain this?