Status as of Tuesday, March 5, 2019

Discussion of development releases of Prohashing.
User avatar
Steve Sokolowski
Posts: 3612
Joined: Wed Aug 27, 2014 3:27 pm
Location: State College, PA
Contact:

Re: Status as of Tuesday, March 5, 2019

Postby Steve Sokolowski » Thu Mar 07, 2019 4:32 pm

CSZiggy wrote:Is there any way to tell if the decrease in SHA-256 was from people who own their own miners or from a decrease in hashrates sent from nicehash?

Maybe just 1 or 2 people that used to mine here stopped or moved to another pool and took the nicehash hashrate with them?


A query could be run on that. It's complicated to write and would take a long time to execute, so I think I'll do it over the weekend. But even if we find that Nicehash hashrate is what left, I don't think we should be focusing on trying to obtain hashrate from Nicehash. That company's irresponsibility on a number of fronts affects both us and the community in general. For example, they did not reply to our July support ticket asking them what protections they took against low luck miners on their service, and they have allowed several 51% attacks to be conducted against coins using their services. Both of these issues can be detected with high probability, and yet Nicehash has not implemented limits to how much hashrate can be used to mine easy coins.

We'll be starting some advertising next month, consisting of videos aimed at convincing people to connect directly to pools rather than to them.

What's unusual this time around is that the number of customers is remaining the same, but the number of connected workers has been declining dramatically recently. I'm not aware of any pools that have recently launched that offer a compelling feature, so we are missing something.
User avatar
CSZiggy
Posts: 692
Joined: Wed Jan 31, 2018 2:44 pm

Re: Status as of Tuesday, March 5, 2019

Postby CSZiggy » Thu Mar 07, 2019 6:39 pm

Personally, in the past year I went from running 30 machines down to just 2, so you lost 28 of my miners. Electricity costs more than they make and I don't mine at a negative HOPING the prices come back up before tax season. The crash in November pretty much killed my business last year.

Its less of a business and more of a way to heat a few rooms in my house now. On the plus side I made a ton of money 2 years ago, and the latest round of machines I bought with that all got ROI, but I didn't make over ROI, just broke even. With no foreseeable increase or profit for what I am guessing to be the next 2 years, I am less enthusiastic about mining or cryptocurrency. IF prices bounce back up in another 2 years, if the machines aren't all obsolete or difficulty not crept up so high as to become unprofitable I look forward to switching them all back on and making some $.

Until then....not worth the hassle, taxes, tracking coins, paying short term-long term, the extra costs, invested money that may never be seen again.
Other issue, as Mr. Trump pointed out, the USD is just too damned strong. If I lived in Denmark and was mining coins and getting paid out in Krones I could be paying 4X more in electricity and still be making profits.

Another criteria to look at, not sure if possible, check IPs of incoming miners for country of origin codes? Proxy and VPN not withstanding, maybe there is a trend globally....less chinese miners mining abroad in some type of hinted-forced country wide policy to not profit foreign pools/exchanges for instance. Or if electric costs have doubled in England maybe half the lost miners came from there? Anything regional that might explain the decreased hashs?
ajs
Posts: 44
Joined: Sat Jun 16, 2018 2:43 pm

Re: Status as of Tuesday, March 5, 2019

Postby ajs » Thu Mar 07, 2019 8:14 pm

CSZiggy wrote:With no foreseeable increase or profit for what I am guessing to be the next 2 years, I am less enthusiastic about mining or cryptocurrency. IF prices bounce back up in another 2 years, if the machines aren't all obsolete or difficulty not crept up so high as to become unprofitable I look forward to switching them all back on and making some $.


Exactly why I suggested a small portion of the business on PoS. While, as Steve correctly points out, it will never make the big money that mining could make, it will stay a modest fixed percentage of the amount people delegate to the pool, it would not take a hit due to a flooded market of new ASICs, and will make money from HODLer's (investment mechanism not mining folks as strugmo correctly points out) broadening (diversifying) the client base and providing an income that moves differently from the mining pool. Diversity is very important in all settings.

The sad truth though, is I can't think of a way to distinguish yourself in that market, as all the delegation pools will likely run that exact same percentages, and thus what would draw a client to this pool vs another pool (that might develop on that coin ecosystem) other than the solid reputation and transparency of the owners. I agree with Steve that that is not enough to draw sufficient delegation unless you are a very early PoS pool on a successful coin. And even then, if Binance starts staking (rumor is they intend to) they could easily dominate.

As an investor and very minimal miner, I am very intrigued with the potential ecosystem around PoS, and am not sure how it will pan out and work. They need to make it profitable or people won't do it - functionally killing those projects - but the current examples I am watching (EOS, TRON, QTUM, and soon Cardano) seem very early in the process and the pools that exist are all somewhat related to the dev team. But I expect that will change as the eco system matures (or dies), and there might be opportunity for the early adopters to set up differentiators. That said, I would not make a business move based on that hope.

https://stakingrewards.com

Oh well, it is an interesting thought experiment.
User avatar
Steve Sokolowski
Posts: 3612
Joined: Wed Aug 27, 2014 3:27 pm
Location: State College, PA
Contact:

Re: Status as of Tuesday, March 5, 2019

Postby Steve Sokolowski » Fri Mar 08, 2019 9:20 am

CSZiggy wrote:Personally, in the past year I went from running 30 machines down to just 2, so you lost 28 of my miners. Electricity costs more than they make and I don't mine at a negative HOPING the prices come back up before tax season. The crash in November pretty much killed my business last year.

Its less of a business and more of a way to heat a few rooms in my house now. On the plus side I made a ton of money 2 years ago, and the latest round of machines I bought with that all got ROI, but I didn't make over ROI, just broke even. With no foreseeable increase or profit for what I am guessing to be the next 2 years, I am less enthusiastic about mining or cryptocurrency. IF prices bounce back up in another 2 years, if the machines aren't all obsolete or difficulty not crept up so high as to become unprofitable I look forward to switching them all back on and making some $.

Until then....not worth the hassle, taxes, tracking coins, paying short term-long term, the extra costs, invested money that may never be seen again.
Other issue, as Mr. Trump pointed out, the USD is just too damned strong. If I lived in Denmark and was mining coins and getting paid out in Krones I could be paying 4X more in electricity and still be making profits.

Another criteria to look at, not sure if possible, check IPs of incoming miners for country of origin codes? Proxy and VPN not withstanding, maybe there is a trend globally....less chinese miners mining abroad in some type of hinted-forced country wide policy to not profit foreign pools/exchanges for instance. Or if electric costs have doubled in England maybe half the lost miners came from there? Anything regional that might explain the decreased hashs?


I think that you're misunderstanding what the issue is.

It does make sense that you personally, as well as many people in general, decided to stop mining because of lower prices. That said, the bitcoin network hashrate has risen during that same period of time. That means that even though you stopped mining, someone else started mining in your place.

We would expect the pool's hashrate to fall as the entire market contracts, and it did. But now that that contraction has ended, and bitcoin network hashrate has started to rise again, SHA-256 hashrate here has declined dramatically, especially within the past week. That's despite our increasing relative profitability and fixing bugs.

For scrypt, our hashrate has declined at about the same rate as that of litecoinpool.org's, which means that we have maintained about the same share of customers. When prices of litecoins rose again, our hashrate actually increased more than litecoinpool.org's. But that isn't the case for SHA-256. That could imply there is a bug in our system we don't know about, that there is some change in mining hardware we don't know about, or that some new pool that I haven't been able to find launched with a compelling feature. In all of these cases, once the problem is identified, we can work hard and address the issue.

While I agree with the other comments that we can add new coins, and we plan to do so, shouldn't the loss of payout coins affect all algorithms equally? That's why I'm still of the opinion that something specific occurred, within the past 10 days, that changed conditions for SHA-256.
ajs
Posts: 44
Joined: Sat Jun 16, 2018 2:43 pm

Re: Status as of Tuesday, March 5, 2019

Postby ajs » Fri Mar 08, 2019 9:34 am

Steve Sokolowski wrote:That something specific occurred, within the past 10 days, that changed conditions for SHA-256.


My guess is that Bitmain (AntPool) turned on a bunch of these:

https://www.prnewswire.com/news-releases/bitmain-announces-next-generation-7nm-asic-chip-for-sha256-mining-delivering-breakthrough-energy-efficiency-300797186.html
User avatar
Steve Sokolowski
Posts: 3612
Joined: Wed Aug 27, 2014 3:27 pm
Location: State College, PA
Contact:

Re: Status as of Tuesday, March 5, 2019

Postby Steve Sokolowski » Fri Mar 08, 2019 9:58 am

ajs wrote:
Steve Sokolowski wrote:That something specific occurred, within the past 10 days, that changed conditions for SHA-256.


My guess is that Bitmain (AntPool) turned on a bunch of these:

https://www.prnewswire.com/news-releases/bitmain-announces-next-generation-7nm-asic-chip-for-sha256-mining-delivering-breakthrough-energy-efficiency-300797186.html


This is interesting. I didn't even know that it was possible to make any chip at the 7nm process size yet. I thought that Intel had been having trouble for years with that.
User avatar
CSZiggy
Posts: 692
Joined: Wed Jan 31, 2018 2:44 pm

Re: Status as of Tuesday, March 5, 2019

Postby CSZiggy » Fri Mar 08, 2019 2:53 pm

Intel has had issues, but AMD and its fab just rolled out video cards with 7nm chips on them for public sale. So there are 7nm chips being made.

https://www.theverge.com/2019/1/9/18175 ... aphics-gpu
User avatar
CSZiggy
Posts: 692
Joined: Wed Jan 31, 2018 2:44 pm

Re: Status as of Tuesday, March 5, 2019

Postby CSZiggy » Mon Mar 11, 2019 2:27 am

I noticed on the main page of the website, it lists NUMBER OF MINERS(2161, 2319, 2366), but at the top of the page and on the forums it lists CONNECTED WORKERS(12144, 11766, 11834).

What is the correlation between MINERS and WORKERS?
When you split out the NUMBER OF MINERS tracking on the main page by algo, can you also add a NUMBER OF WORKERS by algo also for comparisons and keep a running tracking on that number.
User avatar
Steve Sokolowski
Posts: 3612
Joined: Wed Aug 27, 2014 3:27 pm
Location: State College, PA
Contact:

Re: Status as of Tuesday, March 5, 2019

Postby Steve Sokolowski » Mon Mar 11, 2019 12:12 pm

CSZiggy wrote:I noticed on the main page of the website, it lists NUMBER OF MINERS(2161, 2319, 2366), but at the top of the page and on the forums it lists CONNECTED WORKERS(12144, 11766, 11834).

What is the correlation between MINERS and WORKERS?
When you split out the NUMBER OF MINERS tracking on the main page by algo, can you also add a NUMBER OF WORKERS by algo also for comparisons and keep a running tracking on that number.


The number of connected workers is the number of rigs that are connected right now. The number of miners per day is the count of unique users. Since many people own multiple miners, the number of workers is always going to be higher than the number of users.
User avatar
CSZiggy
Posts: 692
Joined: Wed Jan 31, 2018 2:44 pm

Re: Status as of Tuesday, March 5, 2019

Postby CSZiggy » Mon Mar 11, 2019 8:57 pm

The pool has a graph for pool hashrate over time, can we get a graph for number of pool workers over time/by algo also?

The number of miners per day is count of unique users. So if a user turns off half of their machines, or all but 1, then the data recorded doesnt change. Maybe tracking how many machines are connected to each algo may give you better data you can find a use for to track trends.

Who is online

Users browsing this forum: No registered users and 2 guests