Author

Topic: Which pool participates in giving away miner's data? (Read 260 times)

member
Activity: 133
Merit: 33
I've brought this up before with the rather obvious fact that: it's completely wrong, since these 'studies' assume that all the other mining sources that they don't have data for, have a similar distribution to the pool data they use.
The other mining sources not part of the 'study' will be completely different.

And even if you calculate with the original assumptions from Marc Bevand, just replacing the gear from back in the times with modern gear, the figures are totally different to what they communicate on their homepage.

At the same time, I'm shocked how many newspapers just quote these figures "because CaMBriDgE..."
legendary
Activity: 4466
Merit: 1798
Linux since 1997 RedHat 4
I've brought this up before with the rather obvious fact that: it's completely wrong, since these 'studies' assume that all the other mining sources that they don't have data for, have a similar distribution to the pool data they use.
The other mining sources not part of the 'study' will be completely different.
member
Activity: 133
Merit: 33
With today's figures, and the current energy prices, most miners would not be profitable. Therefore, using best guess figures may lead to very pessimistic figures regarding power consumption, i.e. assuming that many miners just invest money without getting anything back.

There is still a lot of inexpensive power out there where you can host. But it's pure industrial mega space mega use locations.
Also, allowing for all costs and everything else in the end there is still and will probably continue to be a large market for used miners.

Even allowing for today's drop in price there is still the potential for profit, if you are running a lean operation.

As for the geolocation. Yeah well it will get you close, but sometimes it's not even within 50 miles close (80km)

https://www.iplocation.net/ip-lookup
And use as an example any of these IPs:
75.99.58.200
47.21.23.200
68.193.181.40
NONE of those listed are the correct city / town of where the equipment is.....

-Dave

Totally agree. My miners all run on excess solar power. The price is below 1 cent per kW/h. And I'm not the only one who's running a small operation. That's why I love used miners. Are they profitable? Yes, absolutely. Do they appear in the statistics of energy "waste"? No.

And geolocation is often a game of guesses. When I use industrial SIMs, equipment located in Spain appears as being in Latvia... Smiley
legendary
Activity: 3598
Merit: 2490
Evil beware: We have waffles!
Um, actually every block says who mined it. Case in point this one. Now *where* the actual miner that got it - the physical hardware - is located, thass another story. That said, AFAIK that finder info is optional so yes for some blocks the finder will be 'unknown'.
legendary
Activity: 3458
Merit: 6231
Crypto Swap Exchange
With today's figures, and the current energy prices, most miners would not be profitable. Therefore, using best guess figures may lead to very pessimistic figures regarding power consumption, i.e. assuming that many miners just invest money without getting anything back.

There is still a lot of inexpensive power out there where you can host. But it's pure industrial mega space mega use locations.
Also, allowing for all costs and everything else in the end there is still and will probably continue to be a large market for used miners.

Even allowing for today's drop in price there is still the potential for profit, if you are running a lean operation.

As for the geolocation. Yeah well it will get you close, but sometimes it's not even within 50 miles close (80km)

https://www.iplocation.net/ip-lookup
And use as an example any of these IPs:
75.99.58.200
47.21.23.200
68.193.181.40
NONE of those listed are the correct city / town of where the equipment is.....

-Dave
member
Activity: 133
Merit: 33
So what CCAF is furthermore writing:
Quote
Each participating mining pool aggregates IP addresses on their end to create an average geographic distribution of total pool hashpower by country and region. Pools then periodically push their individual distribution to our database via a dedicated API endpoint, connecting with a unique, pseudonymous access token that obfuscates the identity of the pool.
Their methodology is just crap.
And the more I dig into their methodologies for different data they produce, the more I am purely shocked that they aren't called out publicly.

I can see some flaw with their methodology, but it's not that bad and i won't call it "just crap". Identifying country based on IP address should be very accurate since each country has it's own IP range. Many miner probably don't use VPN/proxy to avoid risk of invalid/late submission share or unstable VPN connection. But personally i'd question accuracy on region level.

And it's not like they try to hide details of their methodology (such as requiring people to register). They emphasize their assumption under bolded orange box and limitation of their methodology on discussion section. Although i disagree with assumption 2 which state those data is good enough to be "representative sample".

Okay, I agree that "crap" is too strong a word. However, when you read Marc Bevand's paper, you see that he was calculating with some hardware that I look at with very nostalgic feelings. And the hashrate was 3250 PH/s.

With today's figures, and the current energy prices, most miners would not be profitable. Therefore, using best guess figures may lead to very pessimistic figures regarding power consumption, i.e. assuming that many miners just invest money without getting anything back.




legendary
Activity: 2170
Merit: 6279
be constructive or S.T.F.U
I totally agree that the data is meaningless.
Explicitly, they say they got the data from the following pools:
BTC.com, Poolin, ViaBTC, and Foundry.

And the more I dig into their methodologies for different data they produce, the more I am purely shocked that they aren't called out publicly.

When they first published the study (before China banned BTC mining), they didn't use Foundry in their data, and the 3 pools they used were making slightly above 30% of the total hashrate, by then, the study was somehow accurate and somehow reflected what the real world hashrate was, as of today, it's probably a lot less accurate but it remains one of the most accurate methods out there even if it was off by a margin of magnitude, there is no other way of getting these statistics with high accuracy.

member
Activity: 133
Merit: 33
Probably all the big ones. Don't see that it that bad as long as it's the pools giving the geo data and not the actual IP information.

It's also at this point just about meaningless. With the IPv4 exhaustion and the sale and leasing of IP space and the general fact that a lot of geo data is just flat out wrong. The best you are going to get is an approximation, and even then sometimes it's not even close. In the end most of the large mining operations are in known locations anyway. As for the smaller ones it should not be difficult to find a legitimate registered business. For the home / hobby miners it is what it is. If you want to talk to another service they are going to know your IP and I am sure their ToS says they can sell any of your info they want to.

-Dave






I totally agree that the data is meaningless.
Explicitly, they say they got the data from the following pools:
BTC.com, Poolin, ViaBTC, and Foundry.
Foundry has - as to what I understand - only US clients. May be wrong.
BTC.com represents 2.1% of the miners, Poolin and ViaBTC about 9% each.
So what CCAF is furthermore writing:
Quote
Each participating mining pool aggregates IP addresses on their end to create an average geographic distribution of total pool hashpower by country and region. Pools then periodically push their individual distribution to our database via a dedicated API endpoint, connecting with a unique, pseudonymous access token that obfuscates the identity of the pool.
Their methodology is just crap.
And the more I dig into their methodologies for different data they produce, the more I am purely shocked that they aren't called out publicly.
legendary
Activity: 3458
Merit: 6231
Crypto Swap Exchange
Probably all the big ones. Don't see that it that bad as long as it's the pools giving the geo data and not the actual IP information.

It's also at this point just about meaningless. With the IPv4 exhaustion and the sale and leasing of IP space and the general fact that a lot of geo data is just flat out wrong. The best you are going to get is an approximation, and even then sometimes it's not even close. In the end most of the large mining operations are in known locations anyway. As for the smaller ones it should not be difficult to find a legitimate registered business. For the home / hobby miners it is what it is. If you want to talk to another service they are going to know your IP and I am sure their ToS says they can sell any of your info they want to.

-Dave




member
Activity: 133
Merit: 33
The CBECI mining map claims to track the geographic distribution of Bitcoin’s total hashrate. For this, so their claim, they get geo-information from pools based on the IPs of their miners. They write:

Quote
"Data collection
 We have partnered with several Bitcoin mining pools to collect geolocational mining facility data in a non-obtrusive and privacy-preserving manner. This geolocational data is based on IP addresses of mining facility operators (‘hashers’) that connect to the servers of mining pools. "

Which pool is partnering with them? Are pools giving away the IP data? Really? Or is it just part of their overall FUD?
Jump to: