Reduce EM emissions from Cpu and Gpu

[debucoslid]

Hi everyone,

this is my first post on this forum, I see is very active and interesting. Thanks to everyone to keep the forum active.

I'm moving from Mac to Linux. I bought a Fujitsu laptop with an Intel® Core™ Ultra 7 155H processor, running Lubuntu.

I am extremely sensitive to electromagnetic emissions, and I've noticed high EM activity from the laptop, coming specially from CPU and GPU.

So, my goal is to reduce electromagnetic emissions as much as possible by further limiting/slowing down the processor beyond what’s possible via standard tools and power profiles.

So far, here’s what I’ve already done to reduce EM emissions, based on principle to slow down CPU use.

• Deactivated HT (Hyper-Threading) in the BIOS
• Installed power-profiles-daemon, using powerprofilesctl set power-saver
• Installed cpupower-gui and set the CPU maximum frequency to 1.2 GHz
• Manually disabled logical CPU threads (through /sys/devices/system/cpu/cpu*/online)
• Disabled Intel Turbo Boost (echo 1 | sudo tee /sys/devices/system/cpu/intel_pstate/no_turbo)
• Set p_state to passive (added intel_pstate=passive to /etc/default/grub, then updated grub)
• Lowered integrated GPU frequency to 800 MHz (echo 800 > /sys/class/drm/card1/gt_max_freq_mhz etc.)

After all those steps, I reduced activity of CPU and GPU but EM fields are still strong and I cannot use laptop for more than 15-20 minutes.

Is there anything else I can do to “slow down” or further limit the Core Ultra 7 processor (or the whole SoC) on Linux to minimize EM emissions, beyond what I’ve already done?

Are there kernel parameters, patches, or methods (maybe undervolting, more aggressive frequency limiting, etc.) that you would recommend?

Has anyone successfully achieved even lower frequencies/states with this CPU on Linux for minimum possible EM output?

About undervolting, I read on a forum that Intel disabled the undervolting on the Core Ultras, so is not possible to underclock and undervolt processor, but I'm not sure on this.

Any help would be greatly appreciated.

Thank you in advance.

 

[Trml]

Hi,
I have no specific insight into EM fields, etc, and just want to drop-in two general suggestions as thought bites:

1. There was a time when using power-saving profiles and related was pretty effective. However, this ended about 10 years ago. There are various technical design reasons for it (e.g. different E/P cpu chiplets just to name one), but most importantly the modules for Linux changed. Nowadays, any measure you list to save power/emission basically results in the CPU being active for a longer time. A "power saving mode" works different today, it will easily clock to turbo speeds and the power saving is result of other design features. Measures like undervolting can have a good effect on thermals, but that's other physics.
2. I suggest with a keyboard/mouse and monitor you can place the laptop at a distance, even shield it off, and achieve better results. I understand that's not doable with mobile use, but doubt a sensitivity like you describe it can be mitigated with kernel parameters and related configuration.

Good luck.

 

[Brickwizard]

one of your main problems is the amount of plastic in the computer casing, in order to pack more in and reduce weight most are now mainly plastic, how to control spurious emissions, stick the drive unit into an earthed Faraday cage, but that will not help unless you also have/use a second wired screen and a USB keyboard+mouse [Bluetooth is no good as a good cage will stop the signal and a Ethernet connection [or a short USB extender that terminates outside the cage with a USB-wifi dongle.

 

[VP9KS]

Hi,
I have no specific insight into EM fields, etc, and just want to drop-in two general suggestions as thought bites:

1. There was a time when using power-saving profiles and related was pretty effective. However, this ended about 10 years ago. There are various technical design reasons for it (e.g. different E/P cpu chiplets just to name one), but most importantly the modules for Linux changed. Nowadays, any measure you list to save power/emission basically results in the CPU being active for a longer time. A "power saving mode" works different today, it will easily clock to turbo speeds and the power saving is result of other design features. Measures like undervolting can have a good effect on thermals, but that's other physics.
2. I suggest with a keyboard/mouse and monitor you can place the laptop at a distance, even shield it off, and achieve better results. I understand that's not doable with mobile use, but doubt a sensitivity like you describe it can be mitigated with kernel parameters and related configuration.

Good luck.

@Trml Your item 2 took the words right out of my mouth.

Have you considered using the mouse/keyboard/monitor option, and placing the unit inside a Faraday cage?

My question is: What effect do you get from it, and how did you come to the conclusion that it was emf causing it? I wonder if it is radiation from the display causing your symptoms. That would be interesting to experiment with.

Paul

 

[VP9KS]

one of your main problems is the amount of plastic in the computer casing, in order to pack more in and reduce weight most are now mainly plastic, how to control spurious emissions, stick the drive unit into an earthed Faraday cage, but that will not help unless you also have/use a second wired screen and a USB keyboard+mouse [Bluetooth is no good as a good cage will stop the signal and a Ethernet connection [or a short USB extender that terminates outside the cage with a USB-wifi dongle.

Spot on mate!

 

[CaffeineAddict]

In general to deal with dust, moisture or EM emissions I would consider a specialized computer case and would place it into a separate room that would need a special treatment such as having moisture regulator, be regularly cleaned and ventilated.

 

[VP9KS]

one of your main problems is the amount of plastic in the computer casing, in order to pack more in and reduce weight most are now mainly plastic, how to control spurious emissions, stick the drive unit into an earthed Faraday cage, but that will not help unless you also have/use a second wired screen and a USB keyboard+mouse [Bluetooth is no good as a good cage will stop the signal and a Ethernet connection [or a short USB extender that terminates outside the cage with a USB-wifi dongle.

That is why I would not have a case with a clear side, not to mention adding RGB leds to the inside of the case is a Serious waste of money. Now if I could get my hands on some "clear aluminum", I would be ok with the clear side, but the RGB would still be a waste. NYUK NYUK

 

[bob466]

If you're that sensitive to EM emissions...then you don't own or use any of these...
https://emfcaution.com/emf-sources-homes/

 

[Deleted member 210773]

Hi everyone,

this is my first post on this forum, I see is very active and interesting. Thanks to everyone to keep the forum active.

I'm moving from Mac to Linux. I bought a Fujitsu laptop with an Intel® Core™ Ultra 7 155H processor, running Lubuntu.

I am extremely sensitive to electromagnetic emissions, and I've noticed high EM activity from the laptop, coming specially from CPU and GPU.

So, my goal is to reduce electromagnetic emissions as much as possible by further limiting/slowing down the processor beyond what’s possible via standard tools and power profiles.

So far, here’s what I’ve already done to reduce EM emissions, based on principle to slow down CPU use.

• Deactivated HT (Hyper-Threading) in the BIOS
• Installed power-profiles-daemon, using powerprofilesctl set power-saver
• Installed cpupower-gui and set the CPU maximum frequency to 1.2 GHz
• Manually disabled logical CPU threads (through /sys/devices/system/cpu/cpu*/online)
• Disabled Intel Turbo Boost (echo 1 | sudo tee /sys/devices/system/cpu/intel_pstate/no_turbo)
• Set p_state to passive (added intel_pstate=passive to /etc/default/grub, then updated grub)
• Lowered integrated GPU frequency to 800 MHz (echo 800 > /sys/class/drm/card1/gt_max_freq_mhz etc.)

After all those steps, I reduced activity of CPU and GPU but EM fields are still strong and I cannot use laptop for more than 15-20 minutes.

Is there anything else I can do to “slow down” or further limit the Core Ultra 7 processor (or the whole SoC) on Linux to minimize EM emissions, beyond what I’ve already done?

Are there kernel parameters, patches, or methods (maybe undervolting, more aggressive frequency limiting, etc.) that you would recommend?

Has anyone successfully achieved even lower frequencies/states with this CPU on Linux for minimum possible EM output?

About undervolting, I read on a forum that Intel disabled the undervolting on the Core Ultras, so is not possible to underclock and undervolt processor, but I'm not sure on this.

Any help would be greatly appreciated.

Thank you in advance.

I think your laptop has very high specifications, with a powerful CPU and GPU.
There are some systems that can mitigate this problem,
this link below provides several tips for mitigating the problem:

7 Simple Ways To Reduce EMF Radiation From Your Laptop - EMF Insiders

This article presents 7 simple steps to reduce EMF radiation from your laptop. Learn how to invest in an EMF radiation protection case, use a wired mouse, keep your laptop away from your body, switch to Ethernet cables instead of Wi-Fi, turn off your Wi-Fi when not in use, reduce exposure to...

emfinsiders.com

But if the problem persists, the best thing to do is to buy a used laptop in good condition: I read that you use Lubuntu Linux, generally, in most cases, Linux distributions do not require high hardware specifications and can work very fine because they do not require the use of many resources.

 

[tinfoil-hat]

@debucoslid Hi, sorry for not adding any value to your problem, but may I ask what symptoms you have? What happens when you are around a 5G towers or similar circumstances? Just asking for my own interest

 

[VP9KS]

If you're that sensitive to EM emissions...then you don't own or use any of these...
https://emfcaution.com/emf-sources-homes/

View attachment 27997

Geese Louise, you need to be a hermit living off grid in the mountains to be clear of rf fields, mate!!

 

[bob466]

Geese Louise, you need to be a hermit living off grid in the mountains to be clear of rf fields, mate!!

Exactly or live in a cave...maybe wear a tin-foil hat.

 

[KGIII]

I'll leave this here:

Electromagnetic hypersensitivity - Wikipedia

en.wikipedia.org

There's a giant star nearby that emits more EMR than any device we humans have made. The universe is pretty much full of the stuff.

 

[VP9KS]

I'll leave this here:

Electromagnetic hypersensitivity - Wikipedia

en.wikipedia.org

There's a giant star nearby that emits more EMR than any device we humans have made. The universe is pretty much full of the stuff.

The author of that article seems to believe that it is all in our heads. In my opinion, considering that our brains control our muscles using electrical impulses, it is not that far fetched that we could be injuring ourselves with the saturation of emf present in modern society. I know that my Software Defined Radio has detected some really strong fields, and having a device which emits a strong microwave field (cell Phone) glued to the side of your face all the time can't be good for your brain. Of course the cell phone companies will tell you "that's a load of tosh".

 

[VP9KS]

Exactly or live in a cave...maybe wear a tin-foil hat. View attachment 28005

Just build a Big Faraday cage inside the cave. You may find it difficult to earth it though, unless there is water in the cave.

 

[tinfoil-hat]

maybe wear a tin-foil hat. View attachment 28005

A gold hat would do a better job

 

[KGIII]

it is not that far fetched that we could be injuring ourselves with the saturation of emf present in modern society.

It's not 'far-fetched' to believe all sorts of things. Me? I prefer the scientific method.

Sure, this is an FDA site, but it has some helpful links.

Do Cell Phones Pose a Health Hazard?

The weight of scientific evidence has not linked exposure to radio frequency energy from cell phone use with any health problems.

www.fda.gov

Y'all can believe what you want, however. At the same time, the current understanding via the scientific method should be represented in the thread.

 

[Trml]

I prefer the scientific method.

Me too. However, there is still a lot un-researched and research particularly needs to catch up when there are leaps in technology.

There are historic examples that show science does not always catch up with research very quick, e.g. non-stick cooking pans. Teflon variants found toxic were banned in 2014, after it being invented in 1938. Science takes time.

For example, to stick to the link about cellphones, there are precautionary power limits for cellphones. I'm surprised these are not mentioned on the FDA site right away. They are regulated in the EU (2 W/kg max) at least. There are also major smartphone models which failed to comply limits and got reworked.

We can apply above mentioning of faraday cages to vehicles. Cars have external antennas not only for better reception, they also help to keep transmit power down. If you hold the phone to the ear in the stationary car, you sit in a faraday cage. I.e. the phone will max transmit power much quicker than outside. To the average user it shows in the phone getting hot faster. A sensitive user might get migraine or whatever.

 

[KGIII]

However, there is still a lot un-researched and research particularly needs to catch up when there are leaps in technology.

The subject of this thread is actually well researched. Not one study has shown that anyone can reliably tell when they're being hit with non-ionizing radiation

Electromagnetic hypersensitivity: a critical review of explanatory hypotheses - PMC

Electromagnetic hypersensitivity (EHS) is a condition defined by the attribution of non-specific symptoms to electromagnetic fields (EMF) of anthropogenic origin. Despite its repercussions on the lives of its sufferers, and its potential to become a ...

pmc.ncbi.nlm.nih.gov

As for cell phones, the EU's precautionary measures haven't resulted in any measurable differences in outcome. But, as they do release microwave radiation, it is possible that they could warm the area (a trivial amount) with extreme long-term use. It's also not really backed by science. (Legislation often isn't.) They've not done so because they can prove it causes harm.

There have been a number of debunked studies, such as the NTP study in 2018, where they just bombarded rats witih extreme levels of radiation and then tried to claim that cell phones could do the same. That was quickly debunked when folks learned the methodology but, by then, the papers were all trotting it out as 'science'.

But, the evidence just isn't there.

For example, cancer rates have remained stable since 2000. If cell phones had increased cancer rates, we'd see them in the rate of change. There are more than 8 billion cell phone subscriptions on the planet.

The evidence just doesn't show any harm (physically). Science is about evidence. You're bombarded by EMR all the time. Radiation can indeed be harmful, but the radiation we're talking about here is non-ionizing radiation.

If you want to worry about the harm cell phones do, worry about something we can actually prove like the negative results it can have for people with addictive personalities and the negative results it has on our youth.

Like I said in my previous post, y'all can believe what you want. As for me, I'm going to stick with science. If the evidence ever changes, I'll change my views accordingly.

 

[Trml]

As for cell phones, the EU's precautionary measures haven't resulted in any measurable differences in outcome.

... yet. - I miss that word at the end of your sentence. And if I used the word precautionary, the reason was exactly what you describe in the rest of the paragraph. 2014-1938=76

I can attribute migraine mostly to a specific pollen allergy, but even that took some years to figure out.