Topic: Why use a PPS pool when there are large PPLNS pools? (Read 686 times)

The large American invite only pools are pplns are they not? The ones currently ranking with the most hashrate...

I merited your post because of this


You make perfect sense, like what the hell am I doing here arguing about something knowing very well that nothing of value is going to come out of it, might as well just enjoy my time and not waste it in such a discussion, which is why I put him on ignore so that I don't have to end up having personal arguments with him that doesn't serve this forum in any way.

Happy new year.

For clarity, I don't want to take sides here and I don't care about merits in use such as this.

I just really hate, also in arguments related to my daily job, when people go personal with the arguing. Things too often go off rails to emotional level.

I don't have a bitcoin business career.

My bitcoin interest has been writing code, helping people for free, having fun and telling people about bitcoin.
While doing so, I'm very critical of scammers and stupidity.

Just so happens that I made 'me' some BTC along the way also, that now happens to be priced ridiculously high.
Until this happened, I had been broke for a long time - but that never bothered me.
Now I'm running a pool at a loss, not a problem, the pool made the BTC paying to run it anyway.

Not sure how you can say that after the last few posts of the guy who - lol - also 4 point meritted your post ...
I don't want merit points, but I love how he's apparently a merit source but uses them as an agenda against people
I've explained it to him even many times, it gets annoying ...

Kano you don't have to go personal, be the bigger person...

Merry xmas everyone and please set the pool arguments aside for a while, and enjoy your time with family etc.

You are a waste of time honestly, I would suggest that instead of wasting your time and everybody else's here showing off your understanding of statistics you should learn some business skills and try to revive your dead pool because apparently known of your alleged fields of expertise did you any good regarding your whole bitcoin business career, it also has to do with your attitude although I am not sure you can fix that just by reading some online books.

Meanwhile, I am putting you on ignore since I don't need the extra negativity coming from a random person online.

Try learning a little rather than showing how clearly you have no idea what you are talking about:
(If you can understand it ...)
https://kano.is/index.php?k=poisson

... and some magic words you clearly have no idea about: sample size

Edit: in fact I'll point out the obvious.
Since I've explained all this so many times ... (yes I am one of the few in bitcoin who seems to really have any idea about all this)
Go find a university somewhere.
Go on that massive trek across deserts and mountains, and find someone you think in knowledgeable about statistics.
Go ask them to explain it to you so that you are not posting silly comments: that show to the few who do know the topic, that you clearly do not.

I did not cherry pick the period,  I made the test because of this topic, I know the data could have given better results than PPS, but it also shows how great short term variance is, and it clearly shows the PPLNS sucks short term as shown in my data.

The numbers are what they are, your understanding of statistics has exactly zero value to me and my earnings, PPS+ has little variance and everyone should be using them if they want steady payouts which I believe the most folks are doing regardless of your useless statement.

After all, who would take a business mining advice from you given how terribly you failed managing a mining pool business despite having early mover advantage?  

Funny how people have no idea about statistics and report PPLNS as worse than PPS due toi selecting a time when the varinace was down

Yeah no chance PPLNS could be higher than PPS, not possible, they have ... ... lower fees ... ... oh wait, hang on somethings a miss in all this.

And the answer 'but this is real data' shows even a lesser understanding of statistics.

As I said, variance is simply up and down, not down, but of course you can pick a period when it's down and make up bullshit excuses.

@wndsnb I think the missing point here is the short-term variance, of course, if you measure PPLNS vs PPS against infinity then this whole discussion would be useless and PPLNS is the winner.

I went out of my way to test this in real-life and pointed about 800TH to viabtc PPLNS just for the sake of this topic, here is a week-long sample starting from 2021-12-12 to 2021-12-18, and here are my earnings:

Day                             Total
2021-12-18      0.00348205
2021-12-17      0.00421592
2021-12-16      0.0034405
2021-12-15      0.00372157
2021-12-14      0.00352051
2021-12-13      0.00221123
2021-12-12      0.0042225

on the 19th we had some cleaning and maintenance going on, and during that, I switched that 800th back to PPS+ and got the following:

Day                              Total                      
2021-12-21      0.00422157                
2021-12-20      0.00446968                


So it's an average 0.0035 PPLNS vs 0.0043 PPS+ , of course, 20 and 21 were likely lucky days for viabtc ("see the irony"?) and that's why the rewards were 0.00440, but still the base "promised," hashrate you get is almost always identical to what their PPS calculator which shows 0.00403BTC, so even if we were to work with that number I would still get at least 0.02821BTC in that week as opposed to 0.02481428 on PPLNS so that's a total "temporary loss" of 0.0033957 or 166$ based on current exchange rate.

** I know 7 days vs 2 days isn't exactly accurate which is why I used 0.00403 and not 0.0043.


Ya I know next month or next week I will likely have a week that nets me 0.0033957*2 - 2% in fees and that I will be earning more in the long run but is it really worth it when all I have to pay is 0.00008064   BTC a week ( at current diff for 800th )?  

Obviously, I can't answer for everyone, but to me, having a day like the 13th where I got only half the rewards I could get on PPS pisses me off slightly, a week like this would really piss me off as I would have been -$444 if I pointed all my miners to PPLNS.

It's not like next week is guaranteed to be a good week that would put me back at breakeven, my accumulative "temp loss" or "lost opportunity" could be in the thousands of dollars for God knows how long, I know I likely won't be mining for 10 years, maybe not even a whole more year, so all these theoretical calculations make no sense to me personally.

The other point is that I use almost MOST of the mining rewards as they come, I can not tolerate short-term variance, this doesn't have to make sense to everyone but I know enough people can relate.

Like I said many large pools have had long streaks of “bad” luck

There is no bad luck with pps+

If you are earning 100,000 a month and burning 95,000 during tight times your profit is 5 k

Now if you have variance you easily can drop under and into the red with bad luck.

Running at 95% luck will make you at a loss.

Although  if things are good like right now and you want to risk variance the profit margin will cover it.

This still just makes no sense to me. The "risk" we're talking about is getting >2% less reward in a year once every 10 years. So for the small guy, you'd pay an additional 2% all 10 years to avoid paying 2% 1 out of 10?

For the big guy, they'd pay maybe an extra 0.5% or 1% 10 out of 10.

0.5% for 500PH is ~$325K a year in today's conditions.

So the same guy that is putting up 10s of millions to buy miners, infrastructure, power contracts, etc..., in a brand new super volatile market, is worried enough about 2% less once in 10 years to pay $325K every year?

Mining is a risky business... it just doesn't make sense to me to give up so much profits for a "drop in the bucket" of additional risk.

A large miner may not want the variance risk of pplns.

remember that witholding / accidental bad software has hit pools for large amounts of coins.

A pps+ pool paying a big player daily means no variance risks.

the risk for that player is one days payment maybe two. as they would simply pick up and move to a different pool.

also when doing projections pps will have more stability.

and as I said if a big player is sending a lot of hash to viabtc and insists on pps+ vs pplns I am thinking viabtc may take on the big guy and give him what he wants.

Also, they lost a block the other day due to an orphan race ... so I guess they earn too much to care
(though that also reduces and FPPS/PPS+ they pay ... ...)

Reducing the fees on a larger *PPS* miner indeed increases their risk.
Being a large pool of course also means that risk is quicker to activate.
Well ... that does involve having someone who understand statistics ... so that also probably explains why ...

And why wouldn't Viabtc give an even lower rate for the same hashrate on PPLNS? If they give a spectacular rate for PPS, they are taking even more risk. If they encourage those big clients to go PPLNS, then it reduces their total risk.

And even large pools have had stupidly low months back to back .  Far beyond the actual odds allow.  Bitminter had back to back 99.9 cdf blocks in a 10 block run that had 60 or 7 90% cdf blocks. Very likely they were suffering from a withholding attack.  I left them.

Kano's first 1000 blocks were the opposite super good luck. Then in fall of 2017 that pools luck crumbled right when prices started to whale.

Neither pool was or is super big.  But BTC guild had huge issues and proven with holding attack as did Slush pool.  Finally ant pool with 20% of the entire mining gear had some terrible months.  Far worse then luck can explain. (so a possible attack)

Most people in this discussion are not considering that China could attack any pool with gear programmed to with hold blocks.

China to tell bitmain supply us with the gear and program it.

   The fact is badly programed gear has been proven to exist in the case of BTC Guild, and Slush Pool. Possible it was used against Kano pool back in 2017 maybe used against bitminter pool.  And I am sure that ant pool had a few super unlucky months back in 2016-2017 time period.

I will say this if you are a 10eh player
which is in s19's


10 x 100 = 1ph
1000 x 100 = 1eh
10000 x 100 = 10eh

you have say 1/17 of the network
you burn 10000 x 3.3 = 33000 kwatts or 33 megawatts
you could have your own pool which should hit 144/17 = 8.5 blocks in a day

I can see the argument for that player to self mine rather than pay 2.805 in btc every day to viabtc

but you have control it is all your gear so you know there are no bad actors.

current prices are 12k a unit cheapest price ever was 3k a unit so say 7.5 k x 10000 = 75,000,000 in gear add shipping and the buildings and this is about 25,000,000

So 100,000,000 gets you the 10eh

which pays 25.46 cents a th at viabtc or

25.46/.96 = 26.52 cents self mining

1 ph

$254.60 vs 265.20

1 eh

254600 vs 265200

10 eh

2,546,000 vs 2,652,000

Now if I was this kind of player

and I told viabtc I was pointing 10 eh at them

I would be certain I could get a better rate then 4% so the

  2,652,000
- 2,546,000
      106,000 daily cost  would be lowered to maybe 50,000

Most people don't realize the big players get better rates at pools

Some business people will go for the safer if less profitable route, others will go for the chance of more profit with the possibility of smaller profit or loss.
Depends on their plan and view. Variance is great to say if you can really ride out the highs and lows. But if you can't then it can kill a business.

It also looks at stuff in a bubble. If you have been selling BTC from mining to cover bills and such and the PPLNS pool has a run of bad luck when the BTC prices spikes up what did that really cost you? Yes, I know the opposite is also true the pool could have had a run of good luck during the spike. But in the end a lot of businesses want known numbers coming in.

Side note but relevant:
The known numbers think is actually hurting a lot of small businesses. You are actually seeing that now in a lot of retail rental spaces in hot markets. Landlords will ask for stupid amounts of money for small businesses but give large corporate entities a much lower rent sometimes 1/3 of asking *not 1/3 off, but 1/3 of asking*. People see it as gouging the little guy, the owners see it as no matter what Starbucks is going to pay the rent for the next 7 years. Even if they decide the store is not working and close it, they will buy out the lease. They never heard of "DaveF mikeywith kano coffee" so lets charge them full asking price If they make it great, if not and they only last a year we got the money. Because if they don't make it, we don't know what the market will look like in a year and that place might stay vacant for a while till we can rent it again. And if they don't make it collecting from a small closed business is just about impossible. But you have many many landlords willing to have vacant space (PPS pool) or rent a much lower price (PPS+) then take the risk of unknown (PPLNS)

-Dave

This is not how the merit system works, i merit posts that contribute to a good discussion even if i disgree with them or dislike the person posting them, i would still merit your posts but you do not participate in the merit system so those sMerit points will be wasted.


But you do not have to be rude, wndsnb is stating the same points as you are without being mean or rude to anyone, why is it so difficult for you to be nice?

You run a mining pool and being rude in a mining community is not helping your business at all, which is why i do not think anyone will take your advice related to business seriously.

Also judging by your attitude and how you managed to send away your clients to other pools and ended up being one of the smallest pools in the world despite having started long before most large pools makes you seem like a person who is not capable of giving business advice.

You really need to fix that attitude and stop calling people who disgree with you "ignorant". There is never a good excuse to call people such words.

... and yet the majority of businesses on the planet run on your definition of the word might.

The correct term of course is variance.

All I'm seeing here is comments about how miners have no business acumen and can't handle not getting the same income every day.
(as mentioned at the beginning of my first post) Interesting to see you confirm my post that was as much truth as comedy.

Wow, that's pretty shitty that you say miners are all like that, yet most if not all other businesses on the planet work on the fact that you will NOT get the same income every day.
Holidays, seasons, up turns and down turns in the market, even the day of the week affects businesses.

But miners can't handle that. Damn. Maybe all these miners that you and your biased pal are referring to need a single day in business school ... ?

I never really thought I'd have to explain English, but alas I'm not going to anyway.
It's there in back and white - I was quite clear about it.

Then this troll starts spreading lies again:

Of course he also goes and wastes those merit points to say "Hey someone agrees with me - heck I better throw them 4 merit points ... "

I'm pointing out your ignorance of the real world.
It's quite clear.
Since you don't want to live in the real world, that quite easily justifies the word 'ignorance'.

It's not even a question of explaining it to investors, it's a question of the opportunity cost of money.
If you are generating a fair about of BTC mining. You get a somewhat known amount with PPS you can then budget to get more miners & power knowing how much BTC you are going to take in.
You will probably over time get more BTC on a PPLNS pool, *but* if you are buying new miners out of the money coming from mining then unless you are in a huge pool, you would not want to take the risk since you want more miners coming in at a known rate. You might get more miners with PPLNS you might not. Can't run a business on might.

-Dave

lol ... let's keep it, civilized guys...

It doesn't make much of a difference now since profitability is so high, but when things get tighter that 0.5% or 2% fees can take a good chunk of your profits away depending on your power/operating costs. If you have nearly free power, it doesn't make much of a difference. If you're paying 10c/kwh and profitability drops out of the stratosphere then that 2% might be 10% of your profits...

So, obviously no universally correct answer for everyone. Everyone has their own risk tolerance, operating costs, business arrangements,... etc. People should make the decision with all the info though, and based on the popularity of PPS pools over PPLNS, I'm inclined to believe that a lot of people are not making an informed decision.

I'd guess large operations with multiple investors might have a problem with varying profitability when sending out financial reports. When you hit that 1 in 100 month and your income is 10% below expected it may be difficult to explain to investors.

Too bad ViaBTC doesn't publish the % of their hashrate that is PPLNS vs PPS.

Someone is jealous because nobody uses their PPLNS pool and wants to convince people that variance is a good thing.

Not sure why are you forcing your opinion upon us and calling those who disagree with it "ignorant" you need to get off your high horse, this is not good for your health.

Lulz, ignorance is bliss.

All the pools should be PPLNS, but alas people who think they are experts all give out the false information about why everyone should be mining PPS.
So everyone is happy paying more fees and the big pools are ripping you a new one every day making even more of a fortune out of you all wanting to pay high fees

Ya you are the only person in the world who understands what variance means.



Ya imagine having a good month when btc is down 40% in price and a bad month when its up 50% or simply a whole bad epoch when difficulty is down, just because you are narrow minded does not mean everyone has to think like you do.

No thanks i will leave both the ups and down for pool to handle and enjoy stable payouts.
 

Yup, see one con of using PPLNS is the chance of getting addicted to drugs.

Alas, ignorance of the meaning of the word 'variance'.

Variance means up and down.
Over time it averages out above due to lower fees (on large PPLNS pools).

I see comments all the time about how a 'bad' month can't be handled.
Yet they ignore the good months that make it average better ...

I guess the real problem is people having no impulse control:
 ... good month ... go blow all the extra on drugs ...
 ... bad month ... shit what did I do with all that extra BTC I had ... better move to a higher fee PPS pool.

I am aware of the chances and how low they are assuming your miners are up 24\7, when they are not -- variance can be largely affected by it, i had a few terrible months on slush pools years ago (which i know i probably wasnt going to face for years to come) but when I moved to PPS i can see how it is worth it, i don't care if it is one in a thousand or one in ten, i know any pool is subject to having a terrible day, week or month and I rather not be affected by it given that I only pay an extra of 2%.

But then speaking of fees alone, Binance pool has 2.5% PPS fees for that matter which is only 0.5% higher than viabtc PPLNS, there is also another small PPS pool that has 2% fees (i will try to remember the name later).

The point is some people rather use x pool for small features like a smooth app, smart mining or low payout threshold even at a higher fee, so slightly higher fees to avoid variance no matter big or small is indeed worth it to many people



Hashrate migration is a worse issue than difficulty increase, when people move to larger pools variance on small pools increases to a point that it becomes risky to use, the only reason viabtc is this big is because of PPS, eventually pools that offer only PPLNS will likely and sadly fade away.

The mining business is already full of variance and surprises in every aspect of it, mining to a PPS pool reduces that, you can even borrow money against your future payouts, but you can not do that with PPLNS because it does not matter how low the chances are even a large pool can have some bad months.

with the current viabtc hashrate, getting 20% less reward than pps in a month would be about one in a million...

10% less would be about 1 in 100.

Small pool issues are a whole different discussion, I'm only talking about large pools. Although, since difficulty has essentially been flat for the past year, mining on a small pool had less of a disadvantage than usual.

Taking FPPS out of the equation eliminates the need to discuss transaction fees as they are identical on both PPS+ and PPLNS payment methods, which brings us down to one key point and that is VARIANCE, paying 2% extra in fees while knowing in advance that for the next 2016 blocks I will be getting x amount of BTC is worth it for me, also the fact that both my internet connection and electricity aren't stable 24/7 makes variance even more noticeable and I simply hate variance.

Unlike what Kano thinks, this has nothing to do with business acumen, it all boils down to what the miner actually does with the payout as they get them, if someone hodl their payouts long enough to overcome any variance, then of course why pay more fees? but for someone that utilizes their payouts as they come, I am happy to hand out 2% more fees not to risk having a single month of -20% or anything like that.

And then imagine a pool that dies slowly like Kano.is, think of all the folks who mined for 10 months without hitting a block while paying the power bill, killing their miners and checking the status page every hour, and then they ended leaving to another pool, I am sure if you ask any of them, they would wish they paid 5 times the fees and got paid in PPS, of course, those who can afford to stay long enough will eventually end up earning higher than PPS pools, but that is a big if.

This PPS vs PPLNS is really similar to buying a house with cash upfront or installments, there is no right or wrong about it, it has everything to do with your risk tolerance, and wether you make use of less money today than more money the next month.

Heh I was going to do a thread a while back (but then forgot about it)
explaining how to work out the truth of the PPS+ fees any pool claims.

It all depends on the txn fees - and you can work that out simply by looking at the blocks they've found.
So e.g. if PoolX found X blocks in the previous diff change (14-Nov -> 28-Nov) you can determine the total fees in that range
and thus determine what the actual + they should be paying, vs the actual PPS value you get for your hash rate in that range.

i.e. it's all very straight forward to work out without having to believe what they claim.

I dug around at ViaBTC and F2pool for how they calculate payouts, the PPS+ at ViaBTC is the sum of PPS (based solely on difficulty) for the block reward and PPLNS for the transaction fees. 4% fee for the block reward, 2% fee on the transaction fees. PPS+ at F2pool is the same, except it's a 2.5% fee on both block reward and transaction fees.

For ViaBTC, if you compare PPLNS at 2% vs their PPS+, you'd expect to be ahead at the end of the year when using PPLNS about 9 out of 10 years. Maybe a bit less if transaction fees get really high. It's a lot closer with F2pool at 2.5%, but you should still be ahead 2 out of 3 years or so.

Heh - alas 80% doesn't necessarily mean 80 in 100.

80% means the block mean is 1.25 (i.e. 10/8)

Getting 8 blocks with a 1.25 mean
Erlang CDF -> 1 in 4.54 chance
If you are getting 8 blocks a month then that's once in 4.54 months - yep not unexpected

Getting 80 blocks with a 1.25 mean
Erlang CDF -> 1 in 57.3 chance
If you are getting 80 blocks a month then that's once in 57.3 months = 4.8 years - well not the end of the world I guess

However the specific case was 228 blocks with a 1.26 mean
Erlang CDF -> 1 in 7659.5 chance
If you are getting 228 blocks a month then that's once in 638.3 years - yeah ok that shouldn't happen

there were at least two proven witholding issues.

btcguild was attacked or accidentally with held
slush was attacked or accidentally with held

when bitminter got back to back 99.9% cdf which occurred in a string of 10 blocks that had 7 90%+ cdf blocks I quit them.

lastly kanos pool was god like first 1000 blocks ungodly good then the wheel fell off.

So I mine on viabtc and on mmpool .

If you check mmpool they had a four year drought and then popped 3 blocks in three months.
I picked up over 1.3 btc with just about a single s9 mining for the four years.

Ie under 1 trillion shares got me the 1.3 btc should have been 1/6 of 22 trillion which is 3.67 trillion shares

so i should have earned under .4 btc I also got the value of hodling without hodling as coins were 4k on 9/11/2017

and were an average of 50k for the 1.3 btc

may as well ask why dont more do what I did with mmpool.

ie act as a remora to a great white shark.

so the extra .9 btc makes up for the fees and certainty on viabtc.

btw you can do this with kanos pool and if you get hot you make out pretty well.

I would argue playing 20th to 40th on mmpool and 1000 th on viabtc is better than any other two pools in the time range of
9/11/2017 to 12/05/2021

That would just increase variance, not kill profits in the long run. Large enough pool and it is not an issue.


Again, large enough pool, and not really an issue.

Getting 80 instead of 100 in a month is a once in 4 years kind of occurrence.

1 Block a day is a lot different than 18 a day, but if you expected 20 blocks and only got 2 or fewer in a given time period, that is a 1 in 2,195,318 chance event. Probably more likely that the pool software was screwing up....

The issue is actually that most PPS miners think it's illegal, a hack, they can't trust the pool, it's gonna disappear tomorrow,

or have no business acumen.

PPLNS should only be higher variance, and can run at much lower fees.
But you will also find that most miners don't even know what variance means.

PPS fees are very high.
e.g. I wonder why people are happy to hand over more than $5.4mil a month in fees just to FUPool?

21.3EH/s = 119.9 BTC a day + tnx fees
= 119.9 * 30.5 * 3.0% = 109.7 BTC a month in pool fees

Of course some of these miners are also happy to pay another 2-3% in hack firmware fees and daily withdrawal fees.
Basically it's all a grab it and run mentality that been on going for a long time.

Money laundering has a high % loss - so I guess some follow that idea

As I've mentioned before on my pool, the top miner(s) left when we had a bad luck month.
Yet the full year including that month was above other pools ... so business acumen comes into it also.

The "I must get paid daily" mentality I guess is if you don't trust who you mine with.
Well, pick someone else I guess you do trust?

Have you ever had back to back 99.9% cdf blocks on a pplns pool?
Have you ever had 80% blocks for a month?

I have done both.

I have been on pools that should hit 100 blocks in a month and they hit 80

I have been on pools that should hit a block a day and they hit 2 blocks in 20 days.

All pplns pools are subject to witholding attacks.

and pps pays and pays and pays and pays.

One can argue viabtc grabs too much in fees for certainty  of payment.

But I also get a few referrals and that kind of fixes the issue.

Ideally I guess a USA based pplns pool would work, but I am simply not a big enough player.

Say 2eh of hash and I could try to just solo mine

Many people have different reasons.
Some want the consistent payouts.
Some small home miners don't have their miners on 24/7 and the ramp up and down can kill your profits.
Others like to hop between pools and possibly other SHA coins.
And so on.

As for are you getting all the TX fees, most pools do publish how they pay them, if you want to go back and do the math.

-Dave

Just wanted to get people's reasoning, since most people seem to choose PPS.

Take, for example, ViaBTC. Current pool hashrate is >20000PH, which gives about 18 blocks a day on average. So 126 blocks a week, and 540 blocks a month.

Using some Poisson magic...

The chances of finding 100 blocks or less when you'd expect 126 is 1 in 100. That would be about 20% (or more) less reward than expected once every couple of years.

The chances of finding 486 blocks or less when you'd expect 540 is 1 in 100. The chances of finding 486 or more is 99.1%.

Is it really worth paying more fees to avoid such a small chance of getting 10% less in a given month?

In the span of a year, it would be 1 in 1000 to find 4% fewer blocks, 1 in 100 to find 3% fewer, 1 in 10 to find 2% fewer.

Of course, it goes the other way too, so you can also get more blocks than expected.

Also, with a PPS pool, how do you know what fee you are actually paying? Just based on expected earnings? How would you know if you are getting the right amount of transaction fees (for pools that include it)? Do any publish the formula they use for calculating the transaction fee amount included in the payout?