Topic: Poolside user monitoring stats (Read 282 times)

Well, I guess its the end users job to worry about themselves first.

It's your job to worry about the rest of us.  I appreciate how seriously you take your responsibility.

Thanks!

Well if it was just me and you now have the short time to ramp up I COULD mine with you.

But we have four people in on the farm. we have a 1.2 ph a .5ph and .12ph

My share is 25%  of the 1.2 and the .12   so I could move back  .33ph.

The others want the certain payments.

Heh interesting comments by Phil ... one could easily, correctly, phrase it as:

If I get hit by another truck, people could lose a total of ... about 6.4BTC

However, if ViaBTC shuts down, and they only lose 24 hours worth of mining, that's ... 67.2 BTC a day lost ...
Current stats for them are: 9.3EH of 124EH or 7.3% of 144 blocks a day or about 10.5 blocks

I guess it depends on if you only think about yourself, or if you consider there are others involved also

Excellent. Now if we can just get some input from folks at poolin and Nova we'll be getting somewhere.

re:balls, thanks but not really. I play the long game and just have a lot of patience. Also helps that the majority of my farm is at work where I get the power as part of my compensation.  

Having been mining there since late 2014 I've been paid a fair bit over 100 BTC  - and as an aside plowed the vast majority of that back into the farm for upgrades/expansion over the years... All in all, excellent track record along with a pool op who has impressive IT chops with a passion to do things (programming/testing) right and who is willing to engage with the pools users is what keeps me there.

But kano.is is PPLNS and the N is 3 days

If you have 1.7PH at Via and you pointed it to kano.is it wouldn't matter if you mined for 38 days or 3 days you would get the percentage of the block that is equivalent to your percentage of the pool hashrate.

Which at todays rates adding your 1.7PH to the current pool average would give you 12.68% of the block or roughly 0.85 BTC

So:

If you mined for 35 days and then stopped for 3 days you would get 0 BTC

If you mined for 3 days and stopped for 3 days you would get 0 BTC

If you mined for 38 days you would get 0.85 BTC

If you waited 35 days, and then put your miner on for 3 days you would get 0.85 BTC

If you mined for 2 days you would get roughly 2/3 or 0.56 BTC

If you mined for 1 day you would get roughly 1/3 or 0.28 BTC

All numbers are "ish" I'm not going that deep into working out averages over time,  just basing it on here and now and the last block found.

Kano's pool has lots of good stats.

viabtc does have a lot of stats

here is a 10 s17pro account I manage.

90 days of stats for all 10 units.



last days in 10 min 1 hour 1 day



every block ever made by viabtc

https://www.viabtc.com/statistics


every time they have a block pay 6.75 btc.

it does this...  pay out. .96 x 6.25 = 6 and .98 x .50 =   .49

6.00 + 0.49 = 6.49 pay out per block. why is this figure big.

kano.is has averaged 1.0398 payout or 6.49875 per block. per his last 1000 blocks according to his stat page.

so of the blocks above  12 clearly are over kano.is 1000 block aver of 1.0398% and 6 are under it.

Still my account here earns 0.004 a day or 0.012 in three days.  that is my the pool vanished risk I lost all coins at the pool.

kano.is 38. block wait would be 38 x 0.004 = 0.152 that is a possible my the pool vanished risk I lost all the coins at the pool.

With kano.is it could go longer than 38 days making the 0.152 number worse. It could go lower than 38 days making that number lower.

So I get enough stats from kano.is and viabtc combined to show me I could mine some with kano.is and a lot with viabtc.

but to mine a lot with kano.is or all like notfuzzywarm does takes true balls of steel.

notfuzzy puts a lot at risk mining only with kano.is I give him credit for doing it.

Yes thanks Phil for that cheery thought  

He'd have to die very close to the time that a block was found. This is PPLNS remember, you only get paid if you are mining for the 3 days before the block is found.

The pool might go a year without finding a block!

Or I might stop mining for the three days before we find a block.

Either way I get nil-nada-zip-zilch till that block is found and I'm a lucky enough to have hit peak hash for the 3 days

Maybe the block find would bring on a cardiac arrest due to shock  

I don't mine 100% 24/7/365 at kano. I have bills to pay.

Your strategy is fine, its what you like, it means you are comfortable with the level of risk you have.

But you don't have an exclusive line on clever risk assessment, most of us are aware and act accordingly.

Getting back on topic, surely the best situation would be to have the best of both worlds, a massive pool with the level of back end checking and data availability of the smaller pool.

Or a medium sized pool, say 250-500PH with all the mod cons.

Solo pool mining is inherently a high risk .

It is why I have only https://solo.ckpool.org/users/146UJM5kgzLVUV23CXCf33KQKHckoX1gx3

3.21th mining at it

while I have 1.7ph on pps+

Would I be this generous if I was the guy that lost the block nope I would not.

As for stats I really don't care about any of them with pps+ mining.

This is the only number that matters

Last 24-Hour Earnings (BTC)
0.00745851 Avg. Daily Earnings 0.0000076 BTC/T  (≈PPS 106.14%)

which means I was paid at a rate of 6.25 x 1.0614 = 6.63375 coins a block

I am set to a 0.01 payout which takes me two days so I get about 0.014916 every 2 days. about 172 usd per payout

that is my risk factor. If they fail to pay me I leave losing two days of mining. about

Now if you mine at a small pool with  such as kano.is and you are biffa you have  about 1.2 ph and it earns about 1200 x 8.8 cents = 105.60 a day

what is your risk factor hmm 5 blocks last 40 weeks better 10 blocks last 52.9 weeks. that is 5.29 weeks a block

or 35 + 3 = 38 days x 105.60 = 4,012.80 usd at risk per payout.

Now I have a question Kano is human I actually remember him going out to dinner when he was in the states.

So being human he will die one day. Just like myself I will die one day.

So if kano dies does the pool pay?

I think if he dies there may be an issue.
Also how about right now the pool is 31 days since last block.  biffa is around 31 x 105 = 3255 unpaid.  So if kano dies today does the pool end?

Not a death threat or an ill wish for kano but something that no one mentions.

So I mine solo I am not risking much with only 3.2 th
I also mine at mmpool about 22th I am about .7btc at risk as we have not hit a block in years on that tiny pool.

I find these two risks acceptable.
Why because I have 1.7ph at viabtc.

Maybe biffa has more then 1.2 ph and the 3255 usd at risk is acceptable due to 10ph at viabtc.

All of what I wrote is pretty much correct.
including that I am willing to risk 3.2 th mining at ckpool
and 20th mining at mmpool.org

but I am not will to risk more then 1.7ph for more then two days at viabtc.

So while some may like to say I gamble I am playing a very safe risk factor.

Personally small pool mining has become very hard to justify.
I support solo.ckpool  with   3th
I support mmpool.org with 20th

I don't need a high end back end with viabtc.com

I only need to check on my payout every 2 days.

I don't worry what happens if kano gets sick and the pool shuts as I said biffa is on the hook for over 3200 based on 31 days at 1.2ph.

What happens if kano gets arrested and the pool shuts?
What happens if kano dies and the pool shuts?

So it is cool to support kano.is
or mmpool.org
or solo.ckpool.org

but all the stats in the world don't fix my questions of pool failures.
all those small pools a miner risks a lot of coin on a pool failure.

and on viabtc you risk 2 days if you set the payout to 0.01
only 1 day if you set the payout to 0.001

assuming you have decent hash rate over 1ph

Kano's individual Stats are over rated if you neglect to include pools average block time of 38 days for last 10 blocks.

What happens if his pool fails? Due to health accident death etc? These are real factors that don't get mentioned.

So basically I would be risking 38 days of payouts in a kano fail
vs risking 2 days of payout in a viabtc fail

19x the coins

I would say most of the large pools do have a back end database, you can see that just because it tracks history information for your account.

Pools that I have used before that have a plethora of historical data on my account are Slush and Via

Slush used to have more info, like you could see the blocks that you were responsible for finding, which was cool but they dropped that when they went hard core commercial and changed the web site, though I imagine the data is still in there somewhere.

ViaBTC probably also store a lot more information than they give out. Hell even kano probably has data that is only pertinent to running the pool opposed to what users want or need to see.

No one is perfect, I think that the reason why there is suspicion about what information is actually tracked is that there have been times in the past where things have gone wrong on some of these big pools and ironically enough its been kano and other smaller pool owners who have noticed it and held them to account or let them know there was a problem.

In 2016 Slush had a big player who was mining with broken software for a long time creating a witholding attack that impacted everyone mining there, it took them at least a month to admit there was a problem. See posts around here: https://bitcointalk.org/index.php?topic=1976.21580

In 2015 both f2pool and antpool forked off the main bitcoin network because of SPV mining, it was picked up on pretty much immediately by players on this forum, but took a while to sort out, this post in the thread sums it up nicely: https://bitcointalksearch.org/topic/m.11787367

So yes massive pools do screw up, do people care? I don't know, but it doesn't make it right, and my take is if you care about bitcoin then you should care. If you're just all about raking the cash, then you should still care because the stability and integrity of the pools and the network is paramount to its future.

Small pools screw up as well, I'm not sure what planet anyone else is on but if I lost out on the btc reward for finding a block on a pool because of a problem with the pool then I'd never mine there again, no matter how nice the guy appears who runs the pool. At a solo pool your talking about losing over $100K in bitcoin! If there is no database behind the pool to keep track of everything it would scare the shit out of me, I'd definitely run a country mile away from it as a basis for my own pool unless I built a db back end into it (why would you when one already exists)

It's great that you started a different topic so that we can stop pumping that one.

Although I feel that the main point of discussion has shifted a bit here, I should make clear that I don't think other pools provide more details to their clients than Kano, I simply imply that those pools PROBABLY do have and collect all the data Kano collects.


This is my point exactly, a pool has 15% or 20% of the total hashrate, pays in PPS and has survived for years does have all the reasons in the world to monitor every share submitted to them, in fact, they have more incentives to do that than Kano or any other PPLNS mining pools.

From a financial standpoint, if 20% of the hashrate is programmed to not find shares here is the outcome:

PPLNS pools:

Users lose 20% - 20% of mining fees, so if they pay 1% in fees they will lose 19.75% while the pool only loses 20% of its fees which is 0.25%

FPPS pools:

Users lose nothing, Pool loses everything ( expect if the block fees are paid in PPLNS then the users will encounter a small loss which is 20% block fees - 20% of the pool block fees)

I can't possibly imagine that a small FPPS pool like sigmapool with just about 850PH will survive all these years if they had just about 30PH of no-block-finding hashrate, they would survive for a week or two, maybe a month but paying 4% more than they could "potentially" make would have sent them to bankruptcy a long time ago, unless the remaining 820PH always have 4% more luck which is very unlikely.

Looking at large pools like F2pool, antpool, or all the other large PPS pools, they would have been losing MILLIONS of dollars on a daily bases if they allowed blockless hashrate to mine to their pool.

I agree that "Absence of evidence is not evidence of absence" but I can't believe that every other pool except Kano has been losing money to block-less hashrate knowing they don't even need a professional dev to figure that out after all these years, all they need is an accountant with mediocre skills.

Assuming F2pool paid 10,000 BTC throughout 1 year and found out they have only mined 9,000 BTC they would have closed down or hired professionals (assuming they don't have any which is also unlikely) and they would have figured out that say 10% of the hashrate they have been paying for is blockless.

But none of that happened, no pool blocks rentals or custom firmware expect for Kano, I need to see a real unicorn first for me to believe that Kano is smarter than the rest of the world combined and that every pool except for his runs at loss.

Now as far as stats that the user gets to see, and since you are " spoiled when it comes to stats" I suggest you stick to Kano   because I am sure most pools don't provide as many details, you usually get the average data of:


Online for/since
Offline for/since
Current hashrate
1H hashrate
24 hashrate
Accepted %
Rejected %
Hashrate Chart per worker and subaccount


I have yet to see a large pool that will show you the number of shares, the largest share, diff, and all the other that shows in the image above.

Honestly, I would love to get all these data, but it isn't a big deal, as long as I get to see my hashrate and the acceptance rate, that's good enough, I wouldn't mind the extra details but I certainly wouldn't leave a mining pool just because it doesn't provide that.

1st off, a disclaimer - I've exclusively mined @ kano.is since late 2014 so am rather spoiled when it comes to stats .

That said, in another discussion the topic of a pools back-end monitoring came up. If a pool has a comprehensive database running as a back-end tracking miner type, shares sent/received, worker diff, IP stats, share results eg stale, hi/lo/dupes etc. to make sure everything is running as expected then it should also provide their users with fairly detailed stats. After, all, data collection and generating tailored reports *is* what a DB is written for...

If other pools have a back-end DB, how do things happen like the Slush kerfuffle a few years back? Inattention to operations? Just not caring to track things?

Running a pool IS running a financial enterprise and must be treated as such.

My questions are: What kind of stats do the various pools  eg, Antpool, ViaBTC, F2pool, etc. provide? Can (and will) they pull up records of past performance when asked by their users?

Obviously, truly large farms with hundreds/thousands of miners will be connecting to a pool through a proxy(s) to minimize connections and their proxies will show up as a single (very large) miner or groups of miners but even they should be able to pull up stats regarding what the pool sees.

It'd be interesting to hear what users of the other pools have to say

edit: This is worker stats from part of my farm pointed at Kano.is: