PC Building - general topic

[RefluxSemantic]

I prefer to trust science on topics that I dont have a lot of expertise in.

https://repository.tudelft.nl/islandora ... 249c0c3868

This article from 2019 specifically states that "as far as known to the authors, there does not exist any estimation of power consumption of Proof of Stake networks". The article that estimates >75% energy reduction is from 2020, but I cant find anything else. Its interesting that a group of scientists claim this estimation does not exist if its that easy to calculate.

Again I like basing my opinion on trustworthy sources. With a subject where I dont fully know all the details, I find it hard to form an opinion based on reddit comments and such.

[RefluxSemantic]

https://www.statista.com/statistics/881 ... ison-visa/

BTC apparently uses about 600000 times more energy per transaction than VISA. The ADA network currently uses 30000 times less energy than BTC, so it's safe to say ADA would use significantly more energy than a credit card company (between 10-100x the amount), but definitely not even close to 1 MW more. One thing to note as well is that as the network grows in size, it gets more energy-efficient.

It is true that any decentralized platform will always use more energy than a centralized one. After all, transactions need to be validated by the network and that necessitates trust between the network nodes. In a centralized platform like VISA, any VISA transaction is inherently trustworthy because we trust the company.
The creation and maintenance of trust between network nodes will always cost some energy. However it can be an effectively negligible amount, and there are significant benefits to decentralization as well.

Btw a mW is a milliwatt, 1/1000th of a watt amd 1/360000th of a Wh, gotta run the numbers myself but Id estimate ADA is wasting multiple watts compared to credit card transactions

Edit: removed some stuff because I misread. Do you have a source for the 30000 number?

[Goodspeed]

https://www.statista.com/statistics/881 ... ison-visa/

BTC apparently uses about 600000 times more energy per transaction than VISA. The ADA network currently uses 30000 times less energy than BTC, so it's safe to say ADA would use significantly more energy than a credit card company (between 10-100x the amount), but definitely not even close to 1 MW more. One thing to note as well is that as the network grows in size, it gets more energy-efficient.

It is true that any decentralized platform will always use more energy than a centralized one. After all, transactions need to be validated by the network and that necessitates trust between the network nodes. In a centralized platform like VISA, any VISA transaction is inherently trustworthy because we trust the company.
The creation and maintenance of trust between network nodes will always cost some energy. However it can be an effectively negligible amount, and there are significant benefits to decentralization as well.

So ADA is 95% more energy efficient than bitcoin? Big oof.

Where'd you get that number? It uses 30k times less energy atm, whatever percentage that is.

(Btw a mW is a milliwatt, 1/1000th of a watt amd 1/360000th of a Wh, gotta run the numbers myself but Id estimate ADA is wasting multiple watts compared to credit card transactions)

Ah I interpreted it as Mega but it was a small m.
1 mW more than a CC company would be negligible. I wouldn't consider that an issue, and again there are benefits to decentralization as well.

For example, overseas transactions are still significantly more expensive and inefficient than CC transactions, and that problem could be solved by a crypto bridge currency. Decentralization makes it more trustworthy in such scenarios.

[RefluxSemantic]

See my edit. Curious about the 30,000 source.

[Goodspeed]

There are currently 1672 stake pools running on the ADA network. Going by this guide we can assume you need about 3 servers per pool, and per server you'd probably be using about 100W. 1672 * 3 * 100 = 501600W which is about 4.4 million KWh per year. Side note: A household in the US uses about 4500 KWh per year so the entire ADA network is using the equivalent of 1000 US households of energy.

BTC on the other hand uses 130 TWh per year. 4 400 000 / 130 000 000 000 =~ 0.000034.

So theory aside, real-world data tells us the ADA network uses around 0.0034% of the energy used by the BTC network.

[RefluxSemantic]

Is that the reddit thread again? Im sorry but its just not good enough for me, I have no way to verify the credibility of that.

[Goodspeed]

No that's my post

If you're doubting the 1672 number: https://adapools.org/
The source of the amount of computing power needed per pool is that guide I linked to.

BTC consumption I just plucked from the internet, 130 TWh per year seems like a generally accepted estimate.

[RefluxSemantic]

Yeah, you dont have any credibility in my book. Sorry.

[harcha]

Btw a mW is a milliwatt, 1/1000th of a watt amd 1/360000th of a Wh

I think this is the main take-away from this discussion.

[Goodspeed]

Yeah, you dont have any credibility in my book. Sorry.

Do the maths yourself then it's really not hard? If you don't understand this enough to even do that, we should probably just leave it here.

[Goodspeed]

Btw a mW is a milliwatt, 1/1000th of a watt amd 1/360000th of a Wh

I think this is the main take-away from this discussion.

The takeaway is BTC bad, crypto not necessarily bad

[RefluxSemantic]

Btw a mW is a milliwatt, 1/1000th of a watt amd 1/360000th of a Wh

I think this is the main take-away from this discussion.

Its also wrong. I missed a zero :)

[RefluxSemantic]

There are currently 1672 stake pools running on the ADA network. Going by this guide we can assume you need about 3 servers per pool, and per server you'd probably be using about 100W. 1672 * 3 * 100 = 501600W which is about 4.4 million KWh per year. Side note: A household in the US uses about 4500 KWh per year so the entire ADA network is using the equivalent of 1000 US households of energy.

BTC on the other hand uses 130 TWh per year. 4 400 000 / 130 000 000 000 =~ 0.000034.

So theory aside, real-world data tells us the ADA network uses around 0.0034% of the energy used by the BTC network.

Isnt your calculation of the energy consumption completely wrong? 501 kW comes down to 1.57*10^13 kWh, or ~16 TWh. ADA has about a tenth of the number of daily transactions, so by your estimates it should be less energy efficient?

[RefluxSemantic]

There are currently 1672 stake pools running on the ADA network. Going by this guide we can assume you need about 3 servers per pool, and per server you'd probably be using about 100W. 1672 * 3 * 100 = 501600W which is about 4.4 million KWh per year. Side note: A household in the US uses about 4500 KWh per year so the entire ADA network is using the equivalent of 1000 US households of energy.

BTC on the other hand uses 130 TWh per year. 4 400 000 / 130 000 000 000 =~ 0.000034.

So theory aside, real-world data tells us the ADA network uses around 0.0034% of the energy used by the BTC network.

Isnt your calculation of the energy consumption completely wrong? 501 kW comes down to 1.57*10^13 kWh, or ~16 TWh. ADA has about a tenth of the number of daily transactions, so by your estimates it should be less energy efficient?

Looks like you think watt is energy/hour. Its energy/second buddy. This is why I try not to trust random people posting crap on forums.

[Goodspeed]

This is really embarrassing for you tbh

[harcha]

Looks like you think watt is energy/hour. Its energy/second buddy. This is why I try not to trust random people posting crap on forums.

Oh buddy!

[RefluxSemantic]

This is really embarrassing for you tbh

? Watt is joule/s??

[Goodspeed]

If my PC is using 100W then after 1 hour it has used 100 Wh, after 2 hours 200 Wh etc. Apply that to above calculation

[RefluxSemantic]

If my PC is usuing 100W then after 1 hour it has used 100 Wh, after 2 hours 200 Wh etc. Apply that to above calculation

My bad, you are right. This is why we dont trust random strangers on the internet :)

[harcha]

Basically you can't equate watts to watt hours (or joules) because they are measures of two different things.

[RefluxSemantic]

Basically you can't equate watts to watt hours (or joules) because they are measures of two different things.

Yes stupid mistake by me.

[RefluxSemantic]

The problem I have with your post is that I have no way of verifying what a 'stake' is in relation to the entire network. It seems like you host a stake and others participate? What is their energy consumption then? I dont really know and there's no material provided to verify the reasoning.

[RefluxSemantic]

https://www.researchgate.net/publicatio ... onsumption

Finally found another source. The final graph is nice but not a lot of explanation

[princeofcarthage]

You are trying so hard. It almost seems as if you want convince yourself.

[RefluxSemantic]

You are trying so hard. It almost seems as if you want convince yourself.

Yes, I tried really hard to find another article about PoS energy consumption. I found one and it shows significant reduction in energy consumption so I shared it. Looks like PoS is pretty nice possibly (still one article against the other, but its somewhat more reasonable to believe now).

Contrary to what people on this forum might think, Im actually pretty open minded. But the arguments need to be of a certain quality and on esoc you generally dont see that happen a lot.