Deleting A Few Lines Of Code Can Yield "Significant" Power Savings On Some Linux Systems

[kcmichaelm]

Start simple- on AC, battery, or both? What does "top" say? What does "i7z" (or "turbostat") say about idle states? Are you disabling any power-management states via kernel command line (people do things like turn off ACPI, ASPM, etc. thinking it fixes problems)?

Thank you for the advice. Top will show essentially 0% activity, but the laptop fans will be louder than they should be, and the laptop will be generating heat. The fans do seem to work as they should, scaling up and down. At boot, the fans will be quiet for about 5-10 minutes, but once the laptop heats up it will just never return to a fully "idle" state. lm-sensors isn't able to pickup the laptop's sensors perfectly so I don't have fine-grain control over fan settings, though.

I did not intentionally disable anything, but it is a vanilla Arch install so it's definitely possible that I didn't include something which I was supposed to. I will look through your suggestions.

[kcrudup]

lm-sensors isn't able to pickup the laptop's sensors perfectly

... not even the core(s) or package(s) temps? Those are pretty much a gimme on any laptop.

What CPU are you running?
Which laptop?
What's "uname -a" say?
Could you PM me the output of "dmesg" after a fresh reboot?

[Djhg2000]

lol to the guy still compiling his kernel. Use threading you scrub! 🙂

Seriously, my 1950X compiles the kernel in like 30 seconds...you probably should have also ran lspci prior to going doen that road. 😉

Yeah, call out those still compiling their kernel. We obviously can't have that in a software-as-a-service society! /s

Not even sure which side you're on since you seem to be compiling kernels on your 1950X...

[Djhg2000]

If you're using solar power to do your bitcoin mining, no carbon emitted.

In the winter time, you'd be using energy to heat your house anyway, so mining bitcoin is really not a big deal.

But yeah, in the summer time, mining bitcoin isn't really worth the energy usage. I have solar, but it's much better for me to send that energy to other people's houses than to expend it here, heating up my house.

It always surprises me how much emphasis is put on "carbon emissions" when talking about being good for the environment.

First of all, plants need CO2 to survive and lots of it if you want nice forests and harvests. Nature already knows how to deal with CO2. CO is a slightly different story but it's not made into as much of an issue either.

Second, why is everyone ignoring the elephant in the room? Clearly dumping terawatts of heat into the atmosphere will have an effect on the global temperature. Sure, the sun is radiating about 50 terawatts onto the cross section of the Earth, but about a third of it is reflected by the stratosphere, clouds and ocean. More heat means more ocean evaporation and thus more clouds to cover non-reflective land masses and an increase in reflected radiation. It's a self regulating system as far as sun radiation is concerned. But massive heat generation from houses, vehicles and power plants happen below the natural protective layers. The US power plants alone produces about 0.1 terawatts on average over a year and have a capacity (peak output) of about 1 terawatt, most of which end up being converted into heat. Cars need somewhere in the magnitude of 10 kW to sustain normal traffic speeds and less than 1 kW when idle, again most of which end up converted into heat in one way or another (internal combustion, friction, etc.). Of course, it means we'd need more efficient thermal insulation in housing but it's a lot easier to just blame it on CO2.

The CO2 explanation holds up in theory (as in it increases radiation absorption in the stratosphere) but in practice the waste heat production on the surface clearly satisfies Occam's Razor. We keep acting like we can just dump heat into the atmosphere like it's an infinite mass able to absorb any amount of heat but it's not.

[kcmichaelm]

... not even the core(s) or package(s) temps? Those are pretty much a gimme on any laptop.

What CPU are you running?
Which laptop?
What's "uname -a" say?
Could you PM me the output of "dmesg" after a fresh reboot?

Sorry I left out a key point, I just meant to say that it doesn't see fan speeds, or the existence of fans within lm-sensors. It definitely gets me core and chipset temps.

It's a Sager NH58RC, so it's a Clevo OEM - but I was optimistic that the chipset should work because I believe it to be exactly what the System76 Gazelle is built from, and it's a supported config from System76 as well. I'll DM you the dmesg you asked, I'd be thrilled if you can find something I overlooked. Like I said I can easily believe I overlooked something installing Arch so hopefully there's just something missing.

uname -a outputs Linux truck 5.6.14-arch1-1 #1 SMP PREEMPT Wed, 20 May 2020 20:43:19 +0000 x86_64 GNU/Linux
The CPU is a i7-9750H.

[coder]

It always surprises me how much emphasis is put on "carbon emissions" when talking about being good for the environment.

First of all, plants need CO2 to survive and lots of it if you want nice forests and harvests. Nature already knows how to deal with CO2.

Oh dear. Looks like someone didn't make it much past grade school science.

Plants can't tolerate all CO2 concentrations equally well. As CO2 concentrations increase, they become less efficient carbon sinks. We're nearing (or even past) this point of peak efficiency.

Clearly dumping terawatts of heat into the atmosphere will have an effect on the global temperature.

Well, if you extrapolate current datacenter energy usage, I think they'll supposedly boil the oceans in something like 100 years. However, those trends will obviously break down, long before then.

More heat means more ocean evaporation and thus more clouds to cover non-reflective land masses and an increase in reflected radiation. It's a self regulating system as far as sun radiation is concerned.

The presence of a negative-feedback term doesn't make it self-regulating. The influence of the negative-feedbacks need to exceed the primary driver -- and all of the positive feedbacks that you conveniently ignore -- in order for self-regulation to occur. Positive feedbacks include things like melting tundra releasing methane, loss of snow and sea ice leading to more heat absorption. Also, consider that water vapor, itself, is a greenhouse gas. As you drive more evaporation, humidity rises.

Finally, do you know what happens when you drive more water into the atmosphere and heat the whole mess? That's exactly how you get bigger and more frequent superstorms.

it's a lot easier to just blame it on CO2.

Because that's what science says is the dominant factor.

But, from your perspective I don't get why you're even so concerned with the distinction. If energy is coming from renewables, then the net energy of the system isn't increased. Granted, some of the energy captured by solar power would've been reflected, but it's still introducing a lot less energy into the system than burning fossil fuels.

waste heat production on the surface clearly satisfies Occam's Razor.

If you keep misusing it like that, you're liable to cut yourself.

Occam's razor is a tool to help you figure out where to start investigating. It's not some kind of magic oracle or last word. In order to reach a robust conclusion, you still need theories which fit the data and withstand scrutiny.

We keep acting like we can just dump heat into the atmosphere like it's an infinite mass able to absorb any amount of heat but it's not.

Agreed. We shouldn't do that.

[Djhg2000]

Oh dear. Looks like someone didn't make it much past grade school science.

I actually did quite well in grade school science, but if you want to throw qualifications around then I take it you're the kind of person that listens more to qualifications rather than reason? If Einstein told you to jump off a cliff, what would you do?

I'd ask him why and pay close attention to my surroundings. Even Einstein can be wrong but he might see something running towards us which I'm not seeing.

Plants can't tolerate all CO2 concentrations equally well. As CO2 concentrations increase, they become less efficient carbon sinks. We're nearing (or even past) this point of peak efficiency.

Right, so you've established (1) input goes up and (2) efficiency goes down. What does that mean for the output?
A. It goes up
B. It goes down
C. It stays constant
D. Need more information

Well, if you extrapolate current datacenter energy usage, I think they'll supposedly boil the oceans in something like 100 years. However, those trends will obviously break down, long before then.

How about a more scientific approach and estimating what happens instead of guessing? We have a pretty good idea of how much energy is going into the system and roughly how much energy goes out, we know the mass and we know the temperature above absolute zero with historic data. Extrapolating that data isn't rocket science, it just takes a bit of thermodynamics.

The presence of a negative-feedback term doesn't make it self-regulating. The influence of the negative-feedbacks need to exceed the primary driver -- and all of the positive feedbacks that you conveniently ignore -- in order for self-regulation to occur. Positive feedbacks include things like melting tundra releasing methane, loss of snow and sea ice leading to more heat absorption. Also, consider that water vapor, itself, is a greenhouse gas. As you drive more evaporation, humidity rises.

Except I do not "conveniently ignore" the positive feedback, I'm ridiculing the idea that an arbitrarily selected single positive feedback contributor is causing the collapse of such a complex system. The Earth ecosystem isn't bulletproof but CO2 is one of the things it has a pretty solid negative feedback loop for handling. Deforestation isn't helping matters but by itself it's too small of a factor to cause literal doom.

Finally, do you know what happens when you drive more water into the atmosphere and heat the whole mess? That's exactly how you get bigger and more frequent superstorms.

Proving the ecosystem working as intended. You wanted negative feedback and you got it.

Because that's what science says is the dominant factor.

No, it's what some scientists are saying is the dominant factor. Science itself doesn't give you answers. It gives you tools to connect parameters and find answers, but tools are indifferent to what you think the answer should be.

But, from your perspective I don't get why you're even so concerned with the distinction. If energy is coming from renewables, then the net energy of the system isn't increased. Granted, some of the energy captured by solar power would've been reflected, but it's still introducing a lot less energy into the system than burning fossil fuels.

I'm concerned with the distinction because we're chasing a parameter that doesn't solve the equation. It's like trying to keep a house straight by piling massive amounts of books under the corners, only the house isn't standing straight because it's floating away on a tsunami.

If you keep misusing it like that, you're liable to cut yourself.

I know you're trying to discredit me and all, but that's some funny stuff.

Occam's razor is a tool to help you figure out where to start investigating. It's not some kind of magic oracle or last word. In order to reach a robust conclusion, you still need theories which fit the data and withstand scrutiny.

That's the point though, CO2 doesn't fit the data other than short term. It doesn't fit with the natural ice age cycles, it doesn't fit with the feedback mechanism scales, it doesn't fit with the achieved reduction of CO2, hell it doesn't even fit with the current disruption of the climate due to the novel corona virus!

Agreed. We shouldn't do that.

Well at least we agree on something.

[coder]

I actually did quite well in grade school science, but if you want to throw qualifications around

I admit that it was a cheap shot and not helpful.

What concerns me is not qualifications, but rather that you can't simply reason about something like the climate system. It's probably one of the most complicated, dynamic systems and modeling it depends on extensive data collection, modelling of individual contributions, rigorous simulation, and validation against historical data sets. Even then, it's still imprecise, which is why you see projections from bodies like the IPCC bracketed in terms of the best, worst, and most likely outcomes.

Right, so you've established (1) input goes up and (2) efficiency goes down. What does that mean for the output?
A. It goes up
B. It goes down
C. It stays constant
D. Need more information

I'd heard about this recently, but wasn't sure I could find the study. I'm not sure either is the one I heard about, but I did find two relevant studies, of note:

1. Don't look to mature forests to soak up carbon dioxide emissions
2. Under extreme heat and drought, trees hardly benefit from an increased carbon dioxide level

How about a more scientific approach and estimating what happens instead of guessing?

Because it's a trend that will break down, long before then. I mean, it might be fun for someone like Randall Munroe to explore, on his What If? blog, but it has about as much practical significance as this woman worrying about how she'll feed 4 dozen husbands.

I'm ridiculing the idea that an arbitrarily selected single positive feedback contributor is causing the collapse of such a complex system. The Earth ecosystem isn't bulletproof but CO2 is one of the things it has a pretty solid negative feedback loop for handling. Deforestation isn't helping matters but by itself it's too small of a factor to cause literal doom.

So, your argument boils down to this:

1. The climate system is complex and I don't believe scientists can understand it well enough to know what matters and what doesn't.
2. I, not being a climate scientist, believe I can sit back in my armchair and arrive at better conclusions just by thinking about it.

Good luck with that. It's hardly a recipe for success, in any other field of research - not sure why you'd expect it to work for climate science. Even when someone comes along and shakes up the dominant thinking in a field, It's someone who's at least well-versed in the subject.

I know you're trying to discredit me and all, but that's some funny stuff.

I didn't know you had any credit!
; )

I'm one who takes things on their merits. If you make a sound point, I'll agree with it, even if we've clashed before. If there are holes or faulty reasoning, I'll tend to point them out, even if I like or agree with the conclusion, and doing so potentially undermines that.

CO2 doesn't fit the data other than short term. It doesn't fit with the natural ice age cycles, it doesn't fit with the feedback mechanism scales, it doesn't fit with the achieved reduction of CO2, hell it doesn't even fit with the current disruption of the climate due to the novel corona virus!

Really? You mean according to some fake news being pushed by shills for the fossil fuel industry?

You really got me with that last point. The effect of the COVID-19 shutdowns on the cumulative amount of CO2 in the atmosphere is negligible. It's a significant reduction in the rate of contributions, but there's so much CO2 and the shutdowns have been so brief that you wouldn't expect it to meaningfully affect concentrations above the noise floor.

[Djhg2000]

What concerns me is not qualifications, but rather that you can't simply reason about something like the climate system. It's probably one of the most complicated, dynamic systems and modeling it depends on extensive data collection, modelling of individual contributions, rigorous simulation, and validation against historical data sets. Even then, it's still imprecise, which is why you see projections from bodies like the IPCC bracketed in terms of the best, worst, and most likely outcomes.

The IPCC has the sole purpose of investigating climate change. Right from the start they focused on greenhouse gasses (as cited from the UN in 1988 according to the article you linked). Their funding is very tightly connected to that one specific area of science and so of course they're going to claim it's greenhouse gasses.

They simply cannot reassess their foundational hypothesis and any scientifically inclined person should realize that's a fundamental flaw, because it's very rare for the initial hypothesis to last through decades of research. As a consequence they're left with the only option to instead prove the initial hypothesis by any means necessary to justify their existence. Any research commissioned by the IPCC is therefore effectively tainted by their core principles, and that's very unfortunate because we're dumping a lot of resources into the IPCC which could instead have been used for independent research.

I'd heard about this recently, but wasn't sure I could find the study. I'm not sure either is the one I heard about, but I did find two relevant studies, of note:

1. Don't look to mature forests to soak up carbon dioxide emissions
2. Under extreme heat and drought, trees hardly benefit from an increased carbon dioxide level

While scientifically significant as a study, it's not really applicable right off the bat with those extreme parameters. At least not yet.

We're not looking at leaving forests to grow for themselves until they collapse. We cut down and replant trees as well as reseed new trees. Just like the first study suggest, we need to have both new and old forests to get good coverage of CO2 processing.

In the second study they tried varying degrees of extreme environments for a strain of pine trees. Coniferous trees (I hope I got that translation right) are relatively slow in photosynthesis because of their different strategy for handling seasons. Trees with leafs shed the leafs as they get damaged to cold weather, conifers instead stay on because they can handle cold weather. The trade off is in their ability to effectively absorb photons and therefore the speed of photosynthesis.

Nevertheless, the study tried extreme heat and drought which made them retain water better and slowed down photosynthesis. It's an interesting survival mechanism for sure but humans will have other issues before it gets that extreme. Most significantly we won't be able to grow many kinds of crops before the trees kick the bucket.

Because it's a trend that will break down, long before then. I mean, it might be fun for someone like Randall Munroe to explore, on his What If? blog, but it has about as much practical significance as this woman worrying about how she'll feed 4 dozen husbands.

But how do you know our waste heat won't influence how we build datacenters in the future? Because the IPCC says "CO2 is definitely the culprit" with their highly questionable scientific methods? I'm a much stronger believer in the laws of thermodynamics. So far thermodynamics have served us much better than politics as far as finding answers to scientific issues. That's not to say we don't need initiatives for funding research, such as the emission regulations constantly pushing us to develop more efficient combustion engines, but the science itself should be left to researchers.

By the way, that example is inapplicable because it's assuming an unbounded function. Still funny for the same exact reason but pretty much any function we've talked about is bounded in one way or another.

So, your argument boils down to this:

1. The climate system is complex and I don't believe scientists can understand it well enough to know what matters and what doesn't.
2. I, not being a climate scientist, believe I can sit back in my armchair and arrive at better conclusions just by thinking about it.

Good luck with that. It's hardly a recipe for success, in any other field of research - not sure why you'd expect it to work for climate science. Even when someone comes along and shakes up the dominant thinking in a field, It's someone who's at least well-versed in the subject.

That's a straw man and you know it.

It is a complex system, but it's a self regulating system to a very large extent. Most of the scientists who claim CO2 are causing all of this are trying to prove the theory that CO2 is causing all of this. I guess you can say there are three kinds of climate researchers today; (1) those who want to prove CO2 leads to catastrophic breakdown, (2) those who are only given CO2 as a tool to explain climate change, and (3) those who try a wild variety of theories until they find which ones fit and which ones don't. 1 and 2 are who the IPCC are looking to fund. The 3rd kind is left to fight for themselves while being publicly shamed for being non-believers.

As for the second point, I think I have the capacity do decide for myself what makes sense and what doesn't. It's actually quite dangerous to not possess this skill because it makes you highly susceptible to manipulation. You can't just accept whatever answer comes from an authority for the sole reason that it comes from an authority. You need to think for yourself. I don't care if you're Stephen Hawking, if you can't explain why you're right in a convincing and provable way then I have no reason to take your statements as facts.

I could of course be wrong, not admitting otherwise would be foolish. But so far I find the CO2 explanation to be, at the very least, severely lacking in tangible proof. It's more like one of those stories you hear from those kind of conspiracy theorists where every statement is factually correct, but the conclusions are completely pulled out of their magic hat of imagination. They're entertaining but really nothing more.

As a side note my new armchair isn't happy about being used as a tool for your metaphors. It's not feeling anything because it's an inanimate object, but that just proves it isn't happy.

I'm one who takes things on their merits. If you make a sound point, I'll agree with it, even if we've clashed before. If there are holes or faulty reasoning, I'll tend to point them out, even if I like or agree with the conclusion, and doing so potentially undermines that.

Likewise. But then you'll have to admit there are some significant holes in the CO2 theory, right?

Really? You mean according to some fake news being pushed by shills for the fossil fuel industry?

You really got me with that last point. The effect of the COVID-19 shutdowns on the cumulative amount of CO2 in the atmosphere is negligible. It's a significant reduction in the rate of contributions, but there's so much CO2 and the shutdowns have been so brief that you wouldn't expect it to meaningfully affect concentrations above the noise floor.

While it is true that fake news are being pushed by shills, it goes both ways. It's not like one side consists of Devil Lawfirm Inc. and the other of literal saints. There are plenty of people pushing pseudoscience for either side for personal profit. Like BP and IPCC. They have somewhat different motivations for doing so but the end goal is always their extended survival, and it's a completely natural reaction when there's a lot at stake. You can't just reduce one side to being full of shills.

But there it is; you're finally starting to realize the CO2 theory doesn't make as much sense now that we've drastically reduced it. We've been told for decades it would take years if recovery was even possible. Yet, when factories shut down, it barely took a month for a downturn not even climate scientists could ignore. They went scrambling to come up with new models to prove CO2 would have short term effects and rolled with it like nothing happened.

However, heat output would have a very direct effect. Heat output is directly related to the total heat inside a system. If you consider the heat input from various sources as a scalar energy input and the amount of energy the Earth radiates out in space as work, then Earth becomes a closed system and you can just plug the numbers into the first law of thermodynamics.

It isn't bound by being a part of a very complex feedback system, it's straight up what the feedback system is trying to regulate (input energy vs output energy). If you take away the error then the error stops being compensated for. It really is that simple.

[coder]

The IPCC has the sole purpose of investigating climate change. Right from the start they focused on greenhouse gasses (as cited from the UN in 1988 according to the article you linked). Their funding is very tightly connected to that one specific area of science and so of course they're going to claim it's greenhouse gasses.

Chapter 1, paragraph 1 of the climate change deniers' handbook: scientists are somehow too invested in the CO2 hypothesis. I've been hearing this chestnut for more than 20 years.

Except, science doesn't really care what the culprit is. It's the fossil fuel lobby who are the ones invested in an outcome. The beauty of this tactic is that you neuter the damning indictment of the vested interest by claiming that it's somehow science that has a vested interest. No evidence, though, because it's pure nonsense.

Go on...

We're not looking at leaving forests to grow for themselves until they collapse. We cut down and replant trees as well as reseed new trees. Just like the first study suggest, we need to have both new and old forests to get good coverage of CO2 processing.

Who's "we" and why are you now suddenly an authority on global forestry practices?

When you clear a forest or convert a peat bog into farm land, you release vast amounts of CO2 and methane. They are not being replanted with trees - they're being used for farming and grazing, both of which tend to be net greenhouse emitters. Grazing is a greenhouse gas emitter for obvious reasons, but modern, non-organic farming is as well, especially if the crop stubble is burned.

the science itself should be left to researchers.

Agreed. If politicians stopped trying to meddle in science, we might actually have a hope of reigning in CO2 emissions before the climate reaches a tipping point.

It is a complex system, but it's a self regulating system to a very large extent.

The very definition of a tipping point is the point where the positive-feedbacks outweigh the negative ones. You take self-regulation for granted and downplay the sheer extent to which humans are bollocksing it up.

Most of the scientists who claim CO2 are causing all of this are trying to prove the theory that CO2 is causing all of this.

What's your real issue? Are you in the oil & gas industry? Or are you one of these conspiracy-theory types who just has a deep, contrarian urge to feel like you know something the "elites" don't?

Because the real issue isn't the science or the scientists. The real issue is you and some financial or emotional need that you have.

As for the second point, I think I have the capacity do decide for myself what makes sense and what doesn't.

That's a religious statement, not a scientific one. The way science works is to look at the data, find a theory, then try to disprove it. When enough people try hard enough to disprove it and they can't, then it becomes accepted. It doesn't work on the basis of someone just deciding they have the right answer and that's the end of the story.

It's actually quite dangerous to not possess this skill because it makes you highly susceptible to manipulation.

A favorite tool of demagogues is to appeal to rationality by peddling simplistic explanations. It's actually dangerous to believe that you have all the tools and knowledge you need in order to know what's true or not, simply on the basis of some narrative a person can weave. Most pseudo-science and the more successful conspiracy theories have a certain logic to them. By a somewhat Darwinian processes, most of those that are blatantly illogical don't survive long or spread far.

You can't tell whether they're true by simply looking at whether they seem to make sense. The real test is in the data, the testing methodology, whether others can replicate the tests, and whether anyone can disprove the theory. In science, this is a continual ongoing process that eventually weeds out bad theories and replaces them with better ones. We have that process to thank for virtually all the scientific advances of the last couple centuries.

Like BP and IPCC.

That's called a false equivalence. I'm sure most climate scientists would be happy if the CO2 hypothesis were wrong and just go study something else, whether climate-related or not. The fossil fuel industry has no plan B. They're too invested in what they're doing. They're fighting the truth with the ferocity and viciousness of a cornered animal.

But there it is; you're finally starting to realize the CO2 theory doesn't make as much sense now that we've drastically reduced it.

No, the only thing that doesn't make sense is you. The atmosphere is like a swimming pool and we've been filling it up with CO2. Just reducing the rate of output by some percentage, for a couple months, isn't going to suddenly empty it. Even if we completely stopped emitting greenhouse gasses, it'd probably take a couple centuries to return to historical norms, assuming we're not already at a tipping point.

it barely took a month for a downturn not even climate scientists could ignore. They went scrambling to come up with new models to prove CO2 would have short term effects and rolled with it like nothing happened.

Seriously, where do you even get this garbage? I doubt even Russian state media could spout such crap with a straight face.