50 comments

[ 5.5 ms ] story [ 114 ms ] thread
Here's a 2016 Nokia blog article about it: https://www.nokia.com/blog/water-cool-new-way-take-heat-base...

Besides the video being totally cute I remember Ben Krasnow (Applied Science) mentioning in his video about making liquid air that this was frequently used in mobile base stations. I thought it would perhaps be used for super conductivity, but pretty definitely for cooling (dude in the video says other stations are air cooled). Am I thinking of something else?

How much power does a 5G base station consume compared to a 4G? From what I understand, more stations will be needed for this technology.

The European Union has been talking about digitisation and green new deal for economic recovery, but investing in 5G means spending a lot of materials on both stations and mobile phones. It can also mean an increase in energy consumption for both stations and mobile phones.

It seems to me that the two objectives may conflict with each other.

from what i've heard, 5g base stations definitely consume significantly more power, on average. but on a "per bit moved" basis, they are more power efficient.
The question is, do we really need more bandwidth? What I see is that we're already swamped with the amount of information reaching our cell phones. Our attention is also a limited resource.
I think people often misunderstand how bandwidth (and generally any utilization) works. It's not about using 10% of the speed for some time, but using 100% of it for a fraction. You're practically always bottlenecked by transfer.

Major improvement is latency, which not only enhances existing application experience significantly, makes development easier by removing lots of concern, but it also enables completely new tech that has been stopped because of the latency profile of earlier tech.

What you're propably thinking of is the smallest use case like a webpage (which can still significantly be sped up) or a video. But have you though about loading several GiB sized app for just product demonstration purposes on the fly? The things fast connections enable can be hard to even imagine. You could have an entire VR/AR experience where-ever you are

Bandwidth or transfer the idea remains. We already have a very short attention span and we have many companies fighting for this finite resource.For example, we have companies that are using casino strategies in games for children.

Increasing the speed of the hardware can bring advances but it can also exacerbate the problems that we are already feeling with this hyperconnectivity.

Increasing the responsiveness of everything should have the exact opposite effect, because attention is most affected by latency. If you're always immediately getting what you're looking for immediately you can't be distracted either.
That's magical thinking. The gas will expand to fill the container. People will use it as much but download more.
Increasing attention available should be good regardless of how it would be used, even the worst-case scenario would be where we are now, only we get more done
Isn't the additional energy consumed also increasing the EM radiation levels?
It's not harmful to life, so where's the problem in that?
Is there already a concensus that there's no biological response?
Behavior of microwaves is well-understood yes. Proving a negative would be rather difficult however, much like I find it hard to prove that no magical gnome exists at the bottom of my well, or that there is no renaissance-era teapot floating at the other side of our solar system perpendicular to our planet.
What we all want to hear is that they blasted some poor animals with radiation their whole lives and nothing happened. And that they exposed people for a few years to no ill effects.

You know kinda like they do for new medicines.

Observing complex reactions with chemicals requires much longer study than simply beaming things with microwaves. Results are nearly immediate, and any long term effect would have to be shared with everything else that warms things.
Using more energy isn’t bad. It will drive efficiency improvements such as achieving a higher PUE via liquid cooling
Well, inefficiency in a way does promote efficiency, since it's inherently bad, but that doesn't make it... good.
That's not a reasoned argument.

More energy... greater carbon footprint... faster climate change.

"the industry consensus is that 5G will double to triple energy consumption for mobile operators" - https://www.fiercewireless.com/tech/5g-base-stations-use-a-l...

Needless to say that ecologists see the 5G deployment as utter madness, especially since, in their view, 4G is good enough.

I've yet to read a compelling argument in favor of large-scale 5G deployment, and I'm far from being anti-tech. Except for self-driving cars, I suppose.
More efficient use of spectrum. LTE spectrum is already overloaded in many places.
Latency benefits would enable many new realtime applications both in business and consumer end and speed up many existing ones so they'd be much less janky to use, and thus more accessible to people.

It's impossible to see all the future exploits the new technology allows. All the things we do are built upon other existing ideas, moving to higher speed, lower latency connection is another step in progress.

Think about being able to download several GiB application on the fly just to view a demo for a house, or get the schematics for what you're viewing in realtime with AR

There's a reason China is so keen on it (let's not get on the state surveillence bandwagon yet). Their population and population density is consistently high in so many cities, that they face contention problems more than anyone else. If we're talking relatively sparse cities found in many places around the world, then no, the extra local capacity is probably not in high demand.

I would bet that 5G has cut down base stations which operate similarly to 4G, but the option to not use a new generation only works for maybe 1 generation before it gets too far behind. Data consumption is still going up after all, and some countries (such as Australia) have already mandated the shutdown of older generations, with 2G switched off to refarm spectrum, and 3G scheduled for switch off.

That's my opinion also. 4G is good enough, it consumes not only less material resources but also our atention that is a scarse resource also.
Attention is highly latency sensitive, and bringing reaction for your action a second sooner has major implications, and significantly increases your ability to interact with and understand content
"Good enough" does not mean anything in itself. Good enough for what?

More and more of the use cases move away from providing higher data rates to smartphones, so yes, oone could argue that 4G is currently good enough for that specific use case, but that's just one among many.

5G is about supporting greater capacity in general (so maybe the same rate as 4G but to more users at the same time, which is important in dense areas), massive IOT, extremely short latency, high reliability. These are important use cases but they are mostly off the radar of the average person.

It looks like a site may consume about 20kW. This sounds like a good bit, and is more than an LTE mast, but you have to keep in mind that it’s much less than a car and about equivalent to two electric showers. From a societal energy usage point of view it’s irrelevant.
20 kW in what period of time? 20 kW in a year is nothing. 20 kW in a second is quite a lot.

Edit: Facepalm!

A watt is a measure of power, not energy.
I think you're thinking of kWh, which are kilowatt hours. In which case, the answer is 20 kWh per hour.
Depending on how many people a single point serves, it can imply a pretty sizeable chunk of energy. Smaller things add up when you're talking of hundreds of millions of people
Thanks for the information.

I know there are other things that consume more energy, but this is indicative of Jevons' paradox. When we have the choice between efficiency and productivity we choose the second. We could be replacing 4G technology with one with the same speed but less energy consumption, but that would never make economic sense.

In my opinion resource efficiency is not an irrelevant goal but a goal contrary to this society. We must always consume more because otherwise there is no growth. Without that there is no creation of jobs and profits.

The goals of the economy are opposite to those of ecology. The first is to produce as much as possible. It is no wonder that carbon emissions have only actually been reduced when there has been an economic decline.

Now we are going to revitalise the economy with the green new deal, but if it is really efficient it will not generate profits or jobs. So it will only make the situation worse. I think 5G is another desperate measure to maintain this infinite growth in a world of finite resources.

Efficiency isn’t irrelevant, but if you’re worried about efficiency, this isn’t worth caring about. A mast that can covers tens of thousands of people uses as much energy per day as driving an SUV for a couple of hours. If that mast lets one person work from home a few days a week, say, it has paid for itself energy-wise.

If you’re worried about energy inefficiency, the best things to focus on are probably domestic insulation and reducing car usage. People tend to get hung up on the small stuff.

It’s really bizarre to me that THIS is the aspect of consumerism that people go after for energy use. Improved access to the internet has substantial benefits to society; a little energy use is acceptable. SUVs and 5000sqft houses have no obvious benefit to society at all, and use vastly more energy than the telecoms industry could dream of using.

We shouldn't counter-argument against something by saying there are worse things. Just because there are worse things does not mean that less bad things do not matter, especially since we have to reduce our carbon emissions substantially.

Personally I don't have a car and I don't fly. I live in an area that has a climate that allows me not to spend too much on air conditioning (I don't have air conditioning), but again that's not the point [1].

We currently spend 5.6 billion tons of materials on communication equipment 5.6% of all our consumption [2]. These are not figures that we can ignore, especially since in my opinion it is one of the areas that we least need to spend.

[1] https://en.wikipedia.org/wiki/Whataboutism [2] https://www.theguardian.com/environment/2020/jan/22/worlds-c...

The thing is, I don't think it's crazy to imagine that this could be a net benefit. Better comms, especially in rural areas where current comms can be extremely limited, enable people to travel less.

Of course, if we had been optimising society purely for energy usage from the start, no-one would live in rural areas anyway. But that's not the world we have.

I think there are benefits but they don't justify the costs and the associated disadvantages. Ivan Illich said that the faster we can go, the further away things get.For example if we all have cars we will prepare the environment for this reality and instead of having stores close to home we will have a big shopping center miles away. I think the same can happen with information technology.

I'm not really familiar with 5G technology, but I have an idea that the stations had less range than 4G although they are faster. If that were the case maybe there would be other better ways to give good network to the rural environment

> I'm not really familiar with 5G technology, but I have an idea that the stations had less range than 4G although they are faster. If that were the case maybe there would be other better ways to give good network to the rural environment

No, that's not correct.

Most countries are looking at 5G on three separate bands. Usually something in the 700MHz area (for use in rural areas, similar range to the longest-range LTE), something in the 2.5GHz area (for use in urban areas, similar range to current shorter range LTE), and ~30GHz (very short range, dense urban areas only).

For the 700MHz area what are the diferences in speed compared to 4G LTE?
Where are these 10kw electric showers? Even 240v at 20amp is only 4.8kw.
BEHOLD, terrifying 40 amp showers: https://www.screwfix.ie/c/bathrooms-kitchens/showers/cat8202...

Getting less popular now that tankless gas and electric heating is a thing, but they're historically pretty common in some countries. They're more efficient than electric tank water heating approaches, and arguably more efficient than gas tanked systems, sometimes (usual caveats for comparing gas and electric heating apply).

A major downside is that you typically can't have more than one, or at _most_ two, in one home. Some installations have a scary setup where there are two or more of them, but only one will operate at a time.

(comment deleted)
It seems it's only worth it if you can plumb the waste heat into a nearby building.
Because of the short range 5G has, it seems like they would only put towers on or very near to buildings.
This is misunderstanding.

5G spectrum cover spectrum including low band and mid band spectrum of the 1G through 4G LTE frequencies. Then there is the new high-band. First 5G installations are deployed into new high-band.

5G installations in countryside will be low band that is more efficient than 4G/LTE.

Which is exactly what the article says...
It's sad to see initiatives like this come and go but never filed ready or viable to be deployed realtime
Sorry to hijack, but is anyone else getting a Netscape Navigator favicon when visiting the site? Nice throwback.