All that the rules say (yes, I read them) is that a workstation with 32G RAM (formula is 10 + 0.03 * RAM_GiB) must not exceed ~ 11W in sleep mode and a desktop with 32G RAM must not exceed 6W in sleep mode (5 + 0.03 * RAM_GiB).
Basically, these desktops probably have very ancient/buggy motherboards with broken ACPI. Good riddance.
Fun fact: Desktops/workstations shipped without an operating system at purchasers request are exempt from the rule. So technically, if they don’t insist on shipping with OEM Windows, then the PC can be shipped just fine.
In any case, hopefully this rule will nudge them to use half-decent motherboards that maybe better supported under non-Windows operating systems.
>All that the rules say (yes, I read them) is that a workstation with 32G RAM (formula is 10 + 0.03 * RAM_GiB) must not exceed ~ 11W in sleep mode and a desktop with 32G RAM must not exceed 6W in sleep mode (5 + 0.03 * RAM_GiB).
That's it? The rules only mandate maximum power consumption for sleep mode? Some comments linked to the actual regulations[1], and they seem to contain tables with per-year maximums per device[2]. Can you comment on this?
It is based on the Energy Star energy consumption criteria. The formula is: 365*24/(1000)*weighted_average_power_consumption (the division appears to be due to using unitless formulas, and wanting a result value in kWH).
The weighted average power consumption is based on measured power draw in the following states: "off", "sleep", "Long_idle", and "short_idle". Most desktops under the California rules will use the "conventional" weighting factors of (45%, 5%, 15%, 35%) for the respective power modes in that average.
Short idle is basically what you expect. Screen is still on, drives are still spun up.
Long Idle allows basic power saving features to have kicked in (like spinning down hard drives), and assumes the screen has gone blank, but that the computer is still in ACPI G0/S0, not in some form of sleep mode.
Huh? That's not even remotely true, go read it again (and not just one of the table images this time).
All it requires of sleep mode is that it be configured to go to ACPI S3 within 30 minutes of inactivity. If it does not go to S3, then your quoted formulas are almost accurate, except all of the formulas subtract a base amount of memory; a workstation with 32G of RAM must not exceed 10W because the formula is (10 + 0.03 * (RAM - 32 > 0 ? RAM - 32 : 0)). Also, there's another formula for notebooks, and that one takes into account whether there's a dGPU.
But all of that is irrelevant because there's also a separate requirement, namely, all computers manufactured >=2019 shall "Comply with Table V-7" which has a yearly maximum power consumption (with base values and very complex "adders").
I once bought a workstation for a great price, wanted to make a good gaming PC. I discovered that the case didn't fit dedicated GPUs, except the short form factor ones. I ended up cutting some metal bays out with a metal saw to fit my GPU in.
Just want to remind everyone how far California has gone in energy efficiency:
> In 2018, California's energy consumption was second-highest among the states, but its per capita energy consumption was the fourth-lowest due in part to its mild climate and its energy efficiency programs.
I mean I'm all for reducing power consumption, but is having a mild climate an achievement to be proud of? I'd guess that California's main population centers have fewer degree days away from 72 F than any other state in the country. By default, you'd expect their power consumption to be the very lowest.
(The other states close to CA in consumption per capita have much worse climates, but their people live in denser urban centers and have smaller, likely more efficient dwellings.)
We put street signs on highways in Ohio. Quite handy for knowing whether you're in the correct lane, as it doesn't have the ambiguity that overhead signs can have.
Of course, we do also have overhead signs as a backup and for longer-range visibility. But it's more likely that the overhead signs are useless due to a setting sun than that the on-road ones are due to snow. The streets with on-road signs are very high priorities to plow, so the on-road signs are going to be visible unless it's a very bad storm where you probably shouldn't be on the road, and if you are you're probably going home and already know which lanes to be in.
I'm not saying this should be done for every road; it works best for roads known by a number (I-71; US-23; OH-315, etc.) where you don't have to read a long name. But it's not an daft as idea as it might sound.
I believe California does some of this in the really complex interchanges in the Bay Area, but I don't remember for certain and it's been years since I've been there.
High energy prices have probably done a lot to incentivize lower consumption & efficiency. But mild climate does still give it an unfair advantage if trying to compare it apples to apples with other states. South eastern states are always going to eat a huge chunk of Watt Hours in air conditioning, and in some states many people have little choice but to use electric heat or furnaces.
I'd be very interested in state-to-state comparisons that exclude internal climate control usage.
It might not be that California has an unfair advantage, but that southern states have an unfair disadvantage.
If southern states spike their watt hours due to aircon, we could say the cause is poor construction/architecture. The poorer the insulation (southern homes' insulation is crap compared to northern) the more you need to pump the aircon. And when homes are not designed for optimal environmental efficiency or passive temp regulation, you can't cool the home well except with more aircon.
(All this also assumes that aircon is some sort of human right. We could learn how to deal with heat without power, the way we had to for the past 300,000 years)
We dealt with disease without modern knowledge of medicine or hygiene for the past 300,000 years too. The fact that lacking something didn't make the human race go extinct doesn't mean it's something we should just give up.
In the long term, humans deal with everything by dying. It's rather inefficient and denotes a tendency to run away from any problem that can't be solved within the ~55 years of an educated human period of productivity. Maybe with a little more grit and resolve to avoid such cowardly tactics as death we would undertake longer term projects.
But as long as humans insist on dying, that is going to be a nearly impossible challenge. It is hard to get a person motivated to take on a high-stakes challenge when they plan on dying before the consequences of inaction impact them.
I propose that all humans make the sensible choice to not die, at least for a century or so longer than their current predilections in this area, to demonstrate their commitment to truly advancing the cause of human kind.
Yes, the argument from "But this is how natural humans used to live!" Has spawned any number of misaprehemsions, not to mention a few fad diets.
By all means, embrace food as eaten by out ancestors, if that's what someone really wants, but they should understand that such diets resulted in malnutrition and stunted growth, and we're "embraced" by our ancestors living on the brink of starvation & lack of choice.
And it doesn't mean we have to keep it, either. It helps to research the particular scenario and find out whether there's a better path than the one we default into.
Heating is much more energy intensive than cooling (unless you have a heat pump, aren't particularly common in the U.S). Air conditioning consumes a lot of electrical power in comparison to heating, but if we're counting overall energy efficiency it seems like darker, colder northern climates should have a disadvantage.
I completely, fully, and totally agree. Although that's also why, in terms of energy efficiency progress, I'd like to see those factors take out and compare where they stand on consumption that is less tied to climate and construction quality.
That's not so say they should be discounted, only that each segment of energy consumption should be compared like to like.
> (All this also assumes that aircon is some sort of human right. We could learn how to deal with heat without power, the way we had to for the past 300,000 years)
Well, sure, except that the way we dealt with the climate of the American South before air conditioning was by not living there. That is a major reason the South, and not the North, had so many slaves.
It's not an disadvantage, if you care about climate change you shouldn't flock to places where burnings tons of kwh's is the bare minimum to survive.
Of course it's easier to deflect the subject to plastic straws and other pseudo topics.
I meant that you can't get an apple-to-apples comparison on different states' commitment to energy efficiency if you don't break energy consumption up into different categories of usage. It's also important to consider sources of energy: All else being equal, I would rather a state shut down coal power plants in favor of building or importing renewable electricity, instead of keeping the coal while reducing consumption. (They can do both, but in terms of priorities I'd go with switching to renewables first) These are the important details of consumption patterns that we have to focus on to keep moving forward. Are they pseudo topics?
I also don't think the world's temperate zones are large enough to facilitate everyone living in climates that have minimal requirements for heating & cooling. Also many areas that require a lot of energy for heating also require very little for cooling, and vice versa: another reason why consumption details are important when evaluating these things.
Yep. According to Natural Resources Defense Council via NYT, "about a quarter of all residential energy consumption is used on devices in idle power mode" in Northern California [1]. Huge opportunity for efficiency improvement through regulatory action, given idle power consumption is otherwise an afterthought for both manufacturers and consumers.
It's not true, at least not in the sense that's implied. If you read the study you'll see they defined "idle load" not as electronics but as all devices which operate passively.
They did ten case studies, and in one they found the highest idle load device was a water recirculation pump and the second was a halogen light being left on 24/7 because no one cared.
If they studied my home they would consider my fish tanks as "idle load," because they are designed to run continuously.
This is not a study of devices in sleep mode. This is a study primarily measuring devices left on, or devices designed to always be on.
I suspect it’s not quite right either, but two of those appliances (dryer, heating) are often powered by natural gas in the US (and on the East Coast, heating is done with “heating oil”).
> Per research and analysis conducted by the Stanford Sustainable Systems Labs, idle load was 218 watts on average across the 70,000 homes in PG&E’s service territory. We estimated the always-on share of this idle load to be 164 watts.
Then they go on to say:
> ALWAYS-ON, 23% Continuous loads
> ACTIVE, 77% Variable loads, including heating and cooling, lighting, and actively used appliances
So their definition of idle load includes things like heating/cooling, wtf? Then they guesstimate that 23% of that is always-on load (continuous load) where do they get than number from?
Not really, I sat through the outages in Bay Area in the coastal area that basically got turned off by PG&E at the slightest hint of winds (yeah the coast is windy, that’s not new) and eventually broke down and bought a gasoline powered portable generator… not having hot water to shower with gets old very quickly, as does refilling ice chest with actual ice to have food.
Even then, it was at most a few weeks total. Annoying as hell, but probably not enough to really move the needle. Wish we could have had autonomous solar (but while renting, that wasn’t an option). Best I could do was a portable panel to recharge phones/power banks for the evening.
California's energy mix tends to be low due to a lack of both a heavy heating (northern climates) and cooling (southern / south-eastern) load. Housing stock has been notoriously inefficient in insulation, though newer construction gains on that.
Major usage has been in transportation. Sprawl'll do that to ya.
LLNL has historical energy usage charts, and began breaking that out by state sometime since the 1990s (possibly only in the 2000s / 2010s). History since 2010 is readily available here there's also an archive though you've got to hunt for it:
US EIA have 2019 data, but no trend-over-time. The map clearly shows that high per-capita energy usage is concentrated in states with either weather extremes, a large rural population, or both:
Intersting, New York is directly after California in per-capita usage. Rhode Island just edges out both.
A 2013 NRDC report shows that CA's per-capita usage has been flat since roughly 1980, whilst the rest of the US has increased. I suspect much of that growth is due to population increases in the South, with growing cooling load.
There may well be a case to be made that California's accomplished significant progress in energy efficiency. But I've a strong suspicion that what we're looking at instead is credit-claiming where not particularly due, based on dominant other factors. That's unfortunately rather common in this space, in both senses of the word.
Also, much of the industrial use seems likely to be energy-industry related. I'm presuming: Wyoming (coal), Louisiana (oil), North Dakota (fracked natural gas), Alaska (oil), Texas (oil and gas).
Or it may be electric power. These states' energy use is dominated by coal, and see Table 26 which bundles "industrial" and "power generation": https://www.eia.gov/coal/annual/pdf/table26.pdf
Texas is the only high-population state listed.
Note that small states have lower per-capita efficiencies across the board, though yes, those specific states have high industrial/transport usage. Note that per-capita variance in smaller states tends to be higher than large ones regardless (more of them, smaller populations, more opportunity for outliers).
This fact isn't in opposition to California's low energy consumption. It is one of the causes of California's low energy consumption. It is expensive to use power in Cali. This is a good thing, for the long term future of the environment. It would be better if using power stung a little everywhere, so people would make more considered choices about what to consume.
>just want to remind everyone how far California has gone in energy efficiency
How can we know how far California has come when the citation only has a single point in time referenced? What was the per capita consumption before 2018?
When I visited Hawaii long ago, typical homes were not even sealed. There'd be a large gap between the top of the walls and the roof. The idea was to let the breeze blow through. The air temperature never varied enough to make it worth installing an HVAC system.
In part, because the fixed infrastructure is built to serve peak demand, not average consumption. Bring down the peak (flatten the curve) and the rates will come down too.
Or, put differently, rates are high exactly because California per-capita usage is low -- fewer kWh across which to amortize costs.
Importantly, while California rates are relatively high, customer's bills are not -- because the customers use less electricity.
> In part, because the fixed infrastructure is built to serve peak demand, not average consumption.
Also, net metering and widespread consumer generation (largely solar) means that the fixed infrastructure costs are actually amortized not across the low total energy usage, but across a fraction of the total, already low, usage.
> Importantly, while California rates are relatively high, customer's bills are not
Clearly you have never dealt with SDG&E, home of the highest residential electricity rates in the country. And yes that translates to astronomical electricity bills for households
To be more fair, they also have to pay for fixing some of the infrastructure they let deteriorate with falsified maintenance reports over the last 15 plus years.
The issue seems to be that the devices don't have low-power modes that meet the criteria for annual idle power usage. However "If the model is shipped at the purchaser’s request with either a limited capability operating system or without an operating system, or if the model is not capable of having an operating system, the model is not required to comply with the power management requirements of Section 1605.3(v)(5)(B)." So it should be possible to buy the system without an OS from an appropriate provider.
I've never wanted an Alienware laptop in my life, but as a resident of Oregon there is absolutely nothing more that I want right now than an illegally powerful laptop.
These aren't that powerful from a computing standpoint; They use low-end i5 11400F along with a low-end nVidia 1650 Super and have 8GB single-channel memory. Also, I think it's more about inefficiency when sleeping and idle.
The relevant regulation here seems to be for laptops, that they must have net consumption of of 30 kWh/year or less. That comes to about 3W constant power, which is a lot.
Basically the law says Thou Shalt Not Ship a Computer Without Working Idle.
Why? Because it’s cool to break the rules? Does it matter to you that these rules are trying to protect our civilization from dissolving in a heap of wildfires, floods, power outages, droughts, storms, etc? This stuff is not just hypothetical any more.
You probably waste more electricity running your dishwasher instead of by hand than running a laptop. If battery usage is a problem because of climate change, then tax carbon and you can decide to cut back or pay more on dish washing, food deliveries, big houses with huge heating cost and the like, and everyone who wants a cool laptop can spend their money (or carbon credits) on that. If that's what's bothersome, then regulate the problem, not try to micromanage people's hobbies - which brings illegitimacy to people who advocate for actual change, instead of scoring petty political points making it look like you're doing something by adding another useless regulation with a short shelf-life of relevancy.
It’s funny how wrong people are about what consumes electricity and not. Washing dishes by hand is far less efficient than a dishwasher. It depends on exactly how you load it and how your water is heated, but running the dishwasher produces probably half the greenhouse gasses that washing by hand does.
As a general rule of thumb, the guide here is to track heat and where it goes. Hot water down the drain is a bad thing to be avoided as much as possible. Dishwashers reuse a small amount of hot water repeatedly in an insulated box until it’s fully spent (from a cleaning perspective). Hand washing sends a continuous stream of hot, soapy water down the drain, wasting both water, soap, and the energy required to bring that water up to temperature.
I wouldn't be wrong if you just charged me for carbon. I'm sure there's plenty of things people are wrong about carbon emissions and have no idea, because no one pays for the negative effects of pollution and therefor everyone does.
Carbon tax is a really nice and simple idea. But its key problem is that it costs the same whether or not there is a viable alternative for lowering carbon. So it punishes industries for which there is no viable alternative using today’s technologies. And it also provides relatively low pressure to adopt the technologies we know are on the cusp of working. In theory in the long term it encourages innovation in all the right ways, but in practice there’s no way to usefully spend billions of dollars on pure research projects to fix the problems we don’t know how to fix yet.
Targeted incentives and regulations, in contrast, can force industries to adapt quickly in ways that we know are actually not economically painful. (You just gotta enable low power mode on your laptops Dell, seriously it’s not hard just stop whining and do it.) Yeah it is a bit annoying, but it’s just annoying, no worse - not like putting a carbon tax on concrete which would be super expensive, basically impossible to avoid, and drag down economic growth everywhere.
In many cases this would be true. I think people would be a bit shocked if every pound of steel or concrete they bought came with a carbon tax attached, for example.
But in the case of dishwashing vs. hand washing you are already paying for the decreased efficiency in the form of your water and heating bill. The issue is that most of us have no way to measure the relative hot water consumption of either, so we’re left with the heuristic of “more convenient is less efficient”.
Dishwashers are, in fact, fabulously efficient. They use far less hot water than warning by hand, and the majority of the energy spent on dishwashing is spent on heating water.
> They use far less hot water than warning by hand
This must be based on a horribly inefficient method of hand washing. Google says that a modern, efficient dishwasher uses about 3 gallons of water per load. I just went and measured the volume of my kitchen sink to roughly where it gets filled to when I'm washing an equivalent of a full dishwasher's load of dishes, and it came out to a bit over 3 gallons.
If it's not summer, I then let my hot sink water sit until it's cooled down to room temperature to avoid wasting heat.
>Does it matter to you that these rules are trying to protect our civilization from dissolving in a heap of wildfires, floods, power outages, droughts, storms, etc? This stuff is not just hypothetical any more.
Half the issues you've listed are most certainly due to incompetence/mismanagement rather than PCs in CA/CO/HI/OR/VT using slightly more power. eg.
Not going to get into the laptop discussion but the wildfires are 100% due to heat and drought, and those are much worse recently because of global warning. PG&E could bury ever cable, they could increase the fire prevention budget by 10 times over, and it wouldn't fix it.
You're saying that the wildfires are 100% caused by heat and drought? I think you should have a look at the huge heap of evidence pointing to poor forest management being the true culprit. It's a fact of nature that forests WILL burn and the more humans interfere with nature, the more unintended consequences we'll cause. There are plenty of natural ignition sources and forests are equipped to deal with relatively low intensity fires. Here's just one example: https://www.forbes.com/sites/chuckdevore/2019/02/25/wildfire...
No, nothing is ever "100% caused by" any one thing, but it seems dishonest to act like there's no discernible link between record wildfires and record heat and drought.
It's equally dishonest to discount the positive mitigation of a reasonable fire prevention budget and burying/maintaining infrastructure. Especially when the presence of Fraud is taken into account.
The Climate may be famously closed to negotiation, but it doesn't mean human artifice need exacerbate the problem through avoiding doing necessary work.
I suppose if you could convert California into a rain forest, maybe they wouldn't have forest fires. Then they could run the power lines any which way, and everyone would be happy. Right?
You ask how to convert it into a rain forest? How about heating the ocean offshore, so more water evaporates?
Well duh, but since we're talking about climate change we should be comparing against a counterfactual where climate change hasn't happened. I'm sure heat/drought existed prior to global warming. Did wild fires not exist back then? Should Newsom be let off the hook for his budget cuts because the forest was going to catch fire anyways due to global warming?
Regulators have the power to actually do something but instead do the policy equivalent of a clickbait youtube thumbnail. This isn't behavior to celebrate or praise. Policy needs to actually do something.
Address the dirtiest sources and the largest sinks, not the symbolic tiny sinks. Your laptop using 6 watts of idle power is not going to make a meaningful difference.
Sure one laptop doesn’t matter. But if everybody in the country has one it does matter. 300 million laptops each drawing 6 watts 24/7 adds up to 15 terawatt*hrs per year, which is real power. That’s pushing 1% of all the coal-powered electricity in the country. A handful of policies like that, each of which just feels like a minor annoyance and thus pointless, but it adds up and starts to make a real difference.
That's an unrealistically pessimistic assumption though. Many laptops are off, not failing to sleep. The great majority of people aren't buying these niche devices.
And the source should be cleaned up since that is a vastly bigger lever impacting all loads. This is just limiting consumer choice for a largely symbolic gesture. I would also be really surprised if the rule is kept up to date and relevant, vs left in the codebase to rot.
I found this document published by Intel which has all sorts of technical information about "CEC Friendly Motherboards" that appears to describe the regulatory framework that rules out sale of certain computers to customers in those six states, https://processormatch.intel.com/Resources/Program%20for%20C....
“The National Commission for Computer Safety and Integrity requires that, as part of the 2032 Standards for Internet Safety, all devices must run manufacturer authorized code unless a permit has been granted, to supposedly prevent future pipeline and cyberwarfare attacks.”
“Home built PCs allow the use of crypto mining software and cryptocurrencies, which facilitate tax evasion and money laundering. We haven’t had this problem on computers which require signing… “
The freedoms of collaboration over the Internet can taken away with:
1. widespread trusted computing (we're close)
2. banning encrypted comms over unregulated spectrums (done)
3. political will to ban such freedoms (growing)
How can those that prefer such freedoms preserve them? Political activism can. Can censorship resistant tech help in that situation? Plausibly deniable mesh networks? Sneakernet?
This nightmare scenario was proposed in https://en.m.wikipedia.org/wiki/Consumer_Broadband_and_Digit... for combatting piracy, and instead we have settled on a lesser evil of limited DRM to access Netflix. This permits torrenting, but torrenting is less prevalent now.
A carbon tax is inevitable, and will alleviate the concerns such as OP's.
Control over E2E encrypted messaging is an ongoing policy dispute. I don't have a good feel for how the pieces will land. A lesser evil solution might be to require warranted escrow for popular messaging clients. This permits two malicious actors communicating over unpopular platforms, but catches cases like threats/harassment on popular platforms.
Another way this could play out is some gov official decides to create the National Commission on App Stores (to solve the Apple and Google abuses), and then a bureaucrat comes later and just says every device must have and use an App Store, and we have the NCAS to ensure it’s fair, so why would you want anything else?
You don't have to like it, but the right of the people to rise up against an oppressive government is a founding principle of the US.
That sentiment was all well and good back in the 18th and 19th century. But in the next civil war, the Mason-Dixon Line will run through your and my back yard.
I am going to be honest. I think the overlap between people for whom PC energy regulations are the last straw and people willing and able to go through the privations a civil war is sure to bring, particularly to the combatants, is pretty small.
You have no idea how powerful "Live Free or Die" is.
There is a reason the Declaration of Independence went to Great Lengths to articulate, and warn, future generations that it is their Right and Duty to rise against a tyrannical Government.
I'll be frank. The promulgation of "It's for your own good" types continue chipping away at liberties which should be some of the most prized things we cherish as a nation, has put an edge on life I've seen no precedent for in my life excepting firearm regulations, and plotted against a curve over time, it's only becoming faster to implement, more hospitable in the National Rhetoric, and less transparent as the Government goes the route of ignoring getting buy-in from the populace at large and contents itself with Industry collusions through influencing the means of production.
I'm actually kind of tired of people waving around Gadsden flags and getting to make life unlivable (sometimes literally) for everyone else. Freedom can be conceived of in more than one way and not all of them are compatible with each other.
Years ago I bought a tiny hockey puck sized computer to act as an internal web server. AFAIK there seemed to be no way for me to prevent it from going to sleep, I presume it was due to a similar regulation. This one from the EU. Had to throw the thing out.
I’m not looking forward to having to fight hardware / software locks on my computer that make it go to sleep all the time.
I've never heard of any power regulation like that in the EU. Any idea what directive that might even be from?
Edit: looks like 617/2013. Weird that I've never heard about this. This is about total power consumption per year though, so that thing would probably either be inefficient as hell or misclassified for some reason. None of the sleep mode regulations should apply to your case.
Sounds like a user misconfiguration issue to me. I've never had an issue keeping a device from going to sleep when it's... actively doing something... like running services.
It occurred in both Windows and Ubuntu. In Windows “Never” wasn’t an option for Sleep (edit; I forget if it wasn’t an option or selecting it didn’t work), in Ubuntu the usual disable sleep commands didn’t work. Someone in the Ubuntu forums suggested there was a recent power management change due to EU reg and the code around it was freshly buggy. I spent a day on it before cutting my losses.
Honestly, now that SSDs are near universal why are we bothering with sleep mode anymore anyways? Can't full hibernate be made fast enough for no-power standby?
Exactly! Sleep mode is known to have issues. On off over the years multiple different operating systems have had issues with sleep modes. Hibernate just always works and it's quick.
You have to copy things for hibernate and you don't for sleep. Also because you can go into/out of sleep quickly you can periodically wake back up for notifications. MacOS and some Linux distros do this.
Smuggling opportunity! Having legalized weed was a major blow to the unorganized carriage trade, I guess this counts as stimulus for that all important sector of the economy.
Would it really be smuggling (ie. something you could go to jail for if you did it)?
If someone took some new laptops with them in their carry-ons or checked luggage when they flew from one state to another, or even acted as an official courier, would that be illegal?
> If someone took some new laptops with them in their carry-ons or checked luggage when they flew from one state to another, or even acted as an official courier, would that be illegal?
If their plan is to sell them when they arrive then it probably is illegal.
These overzealous "green" regulations crack me up. Don't get me wrong, I'm very pro-green-energy: I think my own country should stop building coal power plants and start investing heavily in solar and wind. However, a lot of these consumer-targetted rules are just ludicrous bureaucratic overreach that cause more issues than they solve.
A quick example: My apartment block was just "upgraded" to meet some new energy-efficiency building codes, and the result is that the car park now looks like a cave. The new LED lights turn on and off automatically, but they turn on seconds after you drive past them. So you're basically driving along in the darkness, but followed in the distance by some lights. It's a lot like the glowing trails of bioluminescence you see behind dolphins sometimes. It's insane.
A more IT-focused example relevant to this article: I have made a name for myself running around to various customers doubling the performance of their core business applications by turning off dynamic power management on their database servers. There's a literal turbo button (in software) on every piece of server kit sold for about 15 years now because of some stupid regulation that was started by some hippie moron in California.
Think about how absolutely crazy it is that to be "energy efficient", companies all around the world are buying servers twice as powerful as they actually need, just to save $50 in electricity per annum. The amount of time, money, and resources wasted waiting around for CPUs running at 50% of their rated clock speed is just mindboggling.
This extra computer power they've been forced to buy isn't energy neutral! More chips have to be made, more assembly is required, etc... It burns up an unbelievable amount of resources in order to save a negligible amount of electricity, which is now becoming increasingly clean anyway...
I'm not making this up. A large telco here kept doubling down on the scale of their new LOB application server platform because the performance was bad. The irony is that the more hardware you throw at something like this, the slower it gets because the percentage load drops, which lets the power management kick in more often and more aggressively. Last I hard, they were planning on buying another 100 blade servers, despite 5% or lower loads on what they already had.
PS: Some cloud platforms like Azure leave this on the default settings at the hypervisor host level, so now there is nothing I can do to improve this situation. The control has been taken away from me, so I guess everyone just has to learn to live with 3 GHz CPUs running at 1.5 GHz forever...
by turning off dynamic power management on their database servers
That reminds me of how turning off "EuP mode" in the BIOS of an Atom nettop (it's an Atom, they're going to be using very little power even at full throttle) stopped complaints of stuttering and lag from its owner, several years ago.
A cynical take on these regulations is that they're a power grab by the CA and WA based big cloud vendors since users of powerful workstations will effectively be forced to offload their workloads to the cloud or buy a server if they want high performance with commercial business support. Well as it turns out, CA/WA is only a minority player in the desktop market, but they're significantly bigger in the cloud/server markets.
Netflix 4k isn't illegal despite being a completely frivolous use of electricity, but somebody professionals buying their own workstations is. Datacenters use 1% of power usage globally, the focus is on desktops.
I'm becoming increasingly extreme on actions were should take to fight climate change, but have to agree with you here. This type of weirdness happens because of super specific regulation instead of creating the right incentive structure. Just charge for negative externalities, in this case carbon emissions. That will incentivice consumers to seek out low-power consumption devices and energy providers to offer competitive pricing by avoiding carbon emissions. Set the right incentives and then the market/everyone together will solve the problem
Externalities is precisely correct, but even that can fall into traps.
For example, pure hosting providers care deeply about the electricity usage, because they have to pay for the air conditioning needed to cool the waste heat. From their perspective, that $50 extra in power is actually more like $100 in electricity + cooling costs for a server that might only bring in $500 in revenue to them in the same time frame.
Meanwhile that server could be running a database that costs $300,000 per year in licensing alone, hosting software that cost $100M to develop, and supports 10K staff managing $2B in turnover.
So of course, it's going to be slowed down to 50% by the guy in charge of the air conditioning to save him $50...
> I have made a name for myself running around to various customers doubling the performance of their core business applications by turning off dynamic power management on their database servers. There's a literal turbo button (in software) on every piece of server kit sold for about 15 years now because of some stupid regulation that was started by some hippie moron in California.
What regulation is this? AFAICT, this one doesn't regulate dynamic power management, it only requires a sleep mode for some computers (and the only mention of servers is "small-scale servers", which are exempted from all the sleep mode stuff)
Anyone know the actual rules? Is it a hard limit on the amount of power a home computer can consume, or is it something more subtle like a limit on the power when its idling?
From the fact sheet I found it’s mostly making sure the OS is configured for sleep by default and a minimum 80+ gold rated PSU if you have a 600W+ unit.
Knowing Dell, it’s probably the power supply not meeting 80+ gold efficiency.
EDIT: I guess there’s max allowed power during ACPI S3 state as well. 10.3W max for a system <32GB of RAM. That’s pretty rough, given even good quality DDR4 is going to need ~1.5W per DIMM during S3. But given base models of these systems will almost exclusively use 2 DIMM’s that still leaves 7W to idle with, tough but manageable.
Not that it matters - nobody should be buying any of Dell’s Alienware desktops. The chassis provides terrible airflow and the components inside will thermal throttle like crazy.
Is DDR4 refresh only usage power really that high?
Regardless, the standard is that the sleep mode must be ACPI S3 or if it is not S3 it must meet the stated power criteria criteria, (and for most desktops the limit is only 5W if under 32GB). I'm not sure why an S3 sleep is good enough to not need to meet the specific limits.
Searching the website for anything related to `computer` on the CA website doesn't give me the thing I'm looking for which is that table V-7 on a gov source (not a lawyer's website).
"This product cannot be shipped to the states of California, Colorado, Hawaii, Oregon, Vermont or Washington due to power consumption regulations adopted by those states. Any orders placed that are bound for those states will be canceled."
I haven't found a lot of information about this, but it appears that this is about idle power draw in electronics. I've been seeing a few things say that they they aren't California Title 20 compliant (like lightbulbs).
+. Transition connected displays into sleep mode within 15 minutes of user inactivity
+ Transition the computer into either the computer sleep mode or computer off mode within 30 minutes of user inactivity. If the transition is to a computer sleep mode, that sleep mode shall either as described in ACPI as S3
or
Consume power less than or equal to 10 + 0.03 * C, where C is the system memory capacity in gigabytes minus 32 gigabytes (see Table V-6 in Title 20 Section 1605.3(v))
i was gonna try to summarize in simpler words but im laughing too hard at the last bits there
> Consume power less than or equal to 10 + 0.03 * C, where C is the system memory capacity in gigabytes minus 32 gigabytes (see Table V-6 in Title 20 Section 1605.3(v))
Umm what? Who came up with this? Is there any scientific basis behind this?
It makes sense. Out of all components only RAM’s power usage scales linearly to it’s size. You need to periodically refresh each single memory cell to not lose it. Everything else you can either switch off or move to standby.
Is watching a video considered "input"? And are these regulations enforced as default settings and do users have the ability to customize the settings? Or do they want the manufacturers to disable customization too?
While I don't own any recent Dell hardware (maybe apart from two thin clients I got off eBay), I own a similarly-spec'd HP Omen 30L desktop (AMD Ryzen 5800X version) that was shipped to WA.
I did get the 30L from eBay without a GPU, but Best Buy was willing to ship the GPU-encumbered version to me (Full disclaimer: I don't game, I develop FreeBSD in my spare time and need a powerful PC to compile kernels/packages, I am nc@ in FreeBSD).
Maybe HP is already complying since they are based in California versus Dell being Texas-based, who knows?
This is legally codified horizontal scaling. Blue state legislators know that taping together a bunch of cheaper computers is better than a single expensive high powered device. It's very considerate of them to force this on those who might make the classic mistake.
Last time I checked during a previous discussion of this on HN, there were slightly more expensive CA (and presumably the other states) shippable versions of essentially the same models that had that warning available, that meet the power regs.
Is this basically just because of adopting newer "energy star certification" junk, and some of these devices just haven't been certified to comply to the new standards yet?
Obviously they're not singling out these specific laptops...
None of this is the real reason. The real reason is that computers are known to the State of California to cause cancer. The other states in the list are just hitch-hiking, I guess.
It makes a lot more sense to ban a car (regardless of price) for polluting the environment we all breathe and contributing to global warming, than it does to ban a computer that uses power the end user pays for by the kilowatt.
> It makes a lot more sense to ban a car (regardless of price) for polluting the environment we all breathe and contributing to global warming, than it does to ban a computer that uses power the end user pays for by the kilowatt.
What? That makes no sense. Unless you're off the grid and using solar - where do you think the electricity from your outlets comes from?
People still pay for the car and the gas - just like you do for the electricity that is result of work that pumps shit into the air too.
> where do you think the electricity from your outlets comes from
Power generation sources that can and should be made cleaner. And those sources can be made cleaner regardless of how much individuals use (and pay for).
> People still pay for the car and the gas - just like you do for the electricity that is result of work that pumps shit into the air too.
An individual's selection of vehicle directly affects carbon emissions in a way that can't be mitigated in a society-wide fashion. Electrical power generation, on the other hand, can get cleaner without the end user changing anything. So it makes more sense to have regulations about the carbon emissions on a car. A laptop doesn't have carbon emissions directly, it just uses electricity, so at a societal level that seems better managed by improving electricity generation.
I'm absolutely positive that the total carbon emissions from mid-grade retail gaming laptops outpaces the carbon emissions from super high-end luxury sports cars by a couple orders of magnitude at least. I'd be shocked if the ~25 Bugatti Veyrons in existence drive a collective 250 miles per years, compared to how many tens of thousands of wannabe streamers drop $1-1.5k on a shitty Dell every year?
I see your point but wonder if the legislation that bans these specifically mention computers or just any electronic device that draws more than XXXX watts per hour.
It does not matter. The correct way to reduce electricity consumption is to increase the price of electricity, aka a tax. No need for politicians to selectively pick and choose which products to gimp.
For those who will retort with “but those are regressive taxes and hurt the poor”, I will say that if you want to fix people being poor, then give them cash. It is unrelated to the goals of reducing electricity usage, and there is no need to mix the two.
For those who will say there is no money to give poor people cash, I will say that the difficult task of enacting even more wealth transfers is the only solution there.
Edit: It is the same exact thing with fuel efficiency in cars and gas taxes. There is no need for CAFE standards or mandated fuel efficiency if a simple gas tax was ratcheted up until usage dropped. Then we would not have pickup trucks twice the size they were 20 years ago consuming even more gas.
Of course this does not happen, because it is politically unpopular to enact legislation that broadly forces people to lower consumption. So we end up with these discriminatory band aids where you get screwed if you do not have sufficient political clout.
The objective is presumably to incentivize increased electrical efficiency in devices, by banning inefficient electrical devices. The goal wouldn't be to decrease consumption; it would be to increase the amount of useful work done while holding consumption constant.
That is semantics. The market of buyers will automatically reward sellers who develop more efficient devices once it starts hitting their wallet. If you want consumption to stay constant, then increase taxes a little. You can keep adjusting it up and down as one’s consumption goal changes.
> The market of buyers will automatically reward sellers who develop more efficient devices once it starts hitting their wallet.
The point is that it's not "hitting their wallet." Idle power draw is an externality — it's negligible per individual (i.e. something people neglect to care about), yet adds up to real money at the municipal level.
And yes, you can make electricity cost more until it's not nelgigible. But you (as a tax-code legislator) don't want to do that. With economic "force-multiplier" utilities like electricity or gasoline, you don't want consumption to stay constant — you actually want it to increase, because consumption of such utilities is in the direct causal chain for GDP growth. Spending energy is necessary-but-not-sufficient to drive your country's economy; disincentivizing your population as a whole from spending energy, is disincentivizing them from driving the economy!
An electricity tax is like a corporate income tax: it indirectly disincentivizes people from making money. Which, as a tax-code legislator, is the last thing you want.
The point of this weird and arcane approach to incentivization that you object to, is to simultaneously incentivize useful electricity consumption, while also disincentivizing useless electricity consumption. To ramp up productivity, by saying "use as much of this as you can — but only in such a way as to drive the economy!"
> the difficult task of enacting even more wealth transfers is the only solution there
What? Haven’t you heard of tax brackets? If you are so keen to tax electricity more than it already is , and are concerned it would affect the poor, then use tax bracketing just like income tax does. The poor pay less by being in a lower bracket.
Having said that, higher taxes on electricity may affect businesses in a way you don’t intend.
The same way that income taxes are calculated and paid. You pay tax once per year based on your income. But anyway, I don’t like the idea. It wasn’t mine.
I doubt the engineers can do a significantly better job, given the technology they have to work with; it's not their fault. Everyone could be tooling around in 60mpg Priuses, but they--the drivers--choose not to. (Disclaimer, I drive a Prius.) And if Ford, Chevy, Toyota, Hyundai, Kia and all but one of the other car companies announced that they'd only make Prius-equivalents starting next year, the one remaining ordinary car company would eat their lunches.
A policy designed to make more power available to a civilization is somehow seen as destroying it. It is honestly fascinating to watch the human brain at work.
They're not making more power available to citizens. They're literally trying to silence people they disagree with on Facebook because their ideas are more persuasive than their own.
Stop with the moral high horse, the left in this country fell off long ago. We're all on the ground now so it's time we look around and accept reality.
That is the least charitable interpretation of my comment. The sycophants and iconoclasts driving the modern left are a destructive force, it’s not about this particular “issue”.
You literally just went from being defensive about not being thought of charitably, to denouncing a gigantic group of people as destructive sycophantic iconoclasts, in the same breath.
There are two ways to solve climate change. Lower the population or lower consumption. Efficiency gains such as these are a way to lower consumption and we should be continue to focus on regulations that drive towards lower consumption. Population control is still largely taboo so lowering consumption is the only way out of climate change at this point.
Seems like it would make a lot more sense to focus on things that really use lots of energy. Focusing on sleeping computers is like a certain wise man said a couple thousand years ago: straining at a gnat and swallowing a camel.
The hand that prioritizes the execution order of that search seems to be taking the scenic route to effectual change to the point it makes one suspect whether they are optimizing to get anything done at all, or maximizing odds of holding on to the reins of power.
That is just shifting consumption but there are tradeoffs to that as well. Make sure you are mindful of the curves. Exponential year over year rise in greenhouse gas emissions in CO2 CH4 SF6 N20. Renewables help. But not enough. Lowering consumption is the main way to stop climate change.
This doesn't make any sense at all to me. Aren't renewable energy sources like solar and wind zero-emission? How am I contributing to climate change even slightly if I'm using power entirely from these sources?
&bhcp=1){kind=link}
Basically, these desktops probably have very ancient/buggy motherboards with broken ACPI. Good riddance.
Fun fact: Desktops/workstations shipped without an operating system at purchasers request are exempt from the rule. So technically, if they don’t insist on shipping with OEM Windows, then the PC can be shipped just fine.
In any case, hopefully this rule will nudge them to use half-decent motherboards that maybe better supported under non-Windows operating systems.