Topic: [300 PH][BTC PPS 100%][EN/CN/RU] Sigmapool.com - page 5. (Read 1879 times)

Yet the halving will simply halve the calculated value as is obvious for anyone with basic math skills
... and I have already said that.

too bad you don't understand me

you could read this parable

https://en.wikipedia.org/wiki/Blind_men_and_an_elephant

and we can leave it at that.

At least we agree he will go broke  so we are better off then the blind men

@ Meni Rosenfeld

https://bitcoil.co.il/pool_analysis.pdf

I would love to discuss

Appendix C
Safety nets for PPS pools

Since I fail to agree with using  B  in the formula  and disregarding ½  ing effect

To leave out  the 50% drop in B  means  formula is most accurate day 1 of ½ ing

least accurate day before new ½ ing why did you leave it out  or not mention it .

As If I do the calculation  and ignore it I over estimate reserve the closer I get to  the  next ½ ing

Yep it sure is again.


You are using specific examples that the pool must be limited to and adhere to, before your example can work.
That's the point I've already made that you are assuming certain things, that you can in no way assume to be correct.
... and everything after the above quote, is doing exactly that, assuming that the pool size must be exactly what you want it to be.

Oddly enough the pool size is probably 0PH and the rest of the pool (240PH) is their own hardware, so that even says they can have no BTC in their wallet and be fine since they are currently effectively solo mining.

You don't know what their public hash rate will be (nor do I) so there's no point using specific examples to say, yeah it's OK.
You also keep choosing small pool sizes to say that you can ignore the calculation.
As I already stated, that's pointless, since it doesn't cover what's necessary, any pool size.
If the pool advertises hard about zero fee PPS they may well be able to get a lot of idiots to join who don't understand the statistics ... just like you ... and then the public pool size will no longer be some tiny value that works for your specific example.

I've already suggested you read Meni's analysis and posted the link to it - but you (obviously) haven't and wont.
Thus you keep trying to make the argument that your specific case of a specific sized pool, with a specific level of luck wont fail.
Which is, to be blunt, pointless.
I'll not bother including any more of the above post cos it's a waste of time.

... and here I am repeating myself, pointing out why your calculations don't matter, coz you, for some unknown reason, seem to want to keep repeating them over and over.

Oh and most countries consider 4 years to be ... 1461 days.

I have some more garbage to post here.

This pool makes about 5 blocks in six days  25 blocks in 30 days.  Lets say a 50 to one shot occurs  and the pdf for the next 25 blocks is about 0.98

the luck would be about 66%

the shares would be about 150%

I took all this info from here: https://www.kano.is/index.php?k=pblocks

Last 25 Blocks   14.9wks   150.17%   101.19%   0.9876   66.59%   66.78%

funny how karma is that the above pool has a nice example of why a 140 coin reserve for sigma pool.com would work.

so if  karma pool suffers the same bad luck as kano.is did for 25 blocks  will 140 coin reserver work as I said.

well  150% shares bumps the time out to 45 days not 34 days.

So I was wrong as it is better then 34 days.

the pool needs to pay out about 450 coins for the 45 days  

I will leave the fees out since we work against the 25 blocks earned

I will then add tx fees back

so 12.5 x 25 = 312.5 coins + 140 reserve = 452.5  so owner of pool would have 2.5 coins as a reserve and if you add in tx fees he has  3.125  add and get net 5.625  so pool owner would be able to pay for at least 45 days.

now the miner that kano said is at risk due to a lack of 4300+ reserve  lets see were he stands

45 x 97 (viaBTC pps)    4365
45 x 100 (sigmapool pps) 4500

you are now ahead 135 in earnings

after 45 days if the owner had a reserve of 140 coins

since you the miner risk 1 day of mining earnings  the next day you could lose 97

so  135/97   is your payout   on a 1/50 chance of going broke.

assuming the owner has the 140 coin reserve.

so if the owner sent a real escrow 140 coins  mining here is smart not stupid

the problem is the owner almost certainly does not have a 140 coin reserve
he certainly won't put it in escrow and risk it.
and if he did most likely the pool grows to 500 ph  which makes these calculations no good.

Once again I have shown that 140 coin reserve works for a miner.
with 240 ph and the certainty the owner pays it.

you risk a 97 dollar loss if you earn 97 daily at viabtc if you switch here.

once again  that while a reserve of 4317 coins would be the answer to the formula as applied by kano it is simply not applicable.  to my facts above

I simply ask you in a true game  were you put up 97 dollars for a 135 profit   and I have to hit a 50 to 1 shot for you to lose  you should do it.

The problem does not lie in the reserve need for the owner to stay afloat.
The problem lies in the willingness for the op to really lose 140 coins in 45 days.  Ie a true 140 coin reserve.

Back to the formula.

B=12.5
1n .1000 =6.9079
2  = a constant
0.01 = true fees  based on about .126 in tx fee

so 12.5 x 6.9079/ (2x0.01) = 86.348/0.02 comes to 4317.4  kanos number  a true number  not false just applied wrongly in this case.

kano said I did not understand the math and was not entitled to an opinion in this thread.

well I had forgotten how to get 1n.1000  to become 6.9079   and asked him to figure it out in the other examples.

he insulted me but he did do it thinking it proved me wrong about him misapplying the reserve.

well I had an error right here

I decided to show you were wrong to apply the formula

and here is why you ignored that B is linear
you ignore or don't understand that if tx fees are .125  now and stay at .125

the op fee  2x  as .125/12.5 = .01
and .125/6.125 = .02

so the 4300/4 = 1075 at the ½ ing

and 4300 was only true on day one of the ½

put another way so that you will give  up and simply agree that the number 4317 is wrong

if the ½ were in 2 days  the 1075 number would be correct.
and today the formula shows  4300

so zero blocks means  op pays 20 coins at 10 coins a block  and we are at 1075    so reserve is 1095 if ½ is in 2 days.

I normally don't push here but you said pool size does not matter.

I have shown  1ph  a small reserve lasts for more then 100 years even if he never hits a block

I have shown that at 240 ph   he only need sto have a 140 coins  for me to have a solid shot at a profit.

bet 97  to gain 135 and a 50 to 1 shot is needed to beat you.   that is a good move.

but we both know he  simply is not going to stay here and drop 140 coins he will bail.

which is why I won't mine at him.

I also have shown that meni's formula needs to be applied on day 1 of a ½ ing to be most accurate you applied it on day  600  948  it is due in 466 days

so at best

july 9 2016 to feb 13 2016 is 948  

feb 13 to say may 25 2020 is 467 days

this means B is not really 12.5 for practical purposes

I agree with you he will go broke

you erroneously used 4317 as the proper reserve

Peace kano
 
if the ½ ing was in 2 days   we would go

It's not an argument ... ... ...

The problem is the garbage you keep spouting ...

Firstly, the parts out of the first mod merited 2 point post of yours:
This is garbage.
It's where I guess you get your first 140 BTC number from ... ... ...
and then throw in the comment about withholding being the point ... ... ...

More completely incorrect garbage - they don't give a big difference.

False. More completely incorrect garbage. It's not magically suddenly 1/10 ... as I said you don't understand.

False. Read my reply quoted below.


Lets continue with (all) the parts out of the next mod merited 2 point post of yours:
And here you are proving what I said ... you don't understand that 199/200 you said would be OK is 1/200 chance of failing is ... from further up
1/200 3311 BTC

1/200, 1/1000 both give very large BTC numbers.

You said that they only needed 140 BTC


No, it's not an argument, it's me explaining with various data, that you keep posting rubbish based on what is probably misunderstanding.

No, it proves how to run a scam so that 240 PH of NEW people wont notice it for a month ...
The pool itself is probably already 240PH of their own hardware.

No, the formula has nothing to do with halvings.
When the next halving occurs, then indeed the 12.5 BTC becomes 6.25 BTC
But lets give a simple (mathematical) example of that effect ...

1% fee pool
Phil's dodgy 0.5% chance of failure pool (that's the number he said was ok 199/200 ...)
12.5 BTC reward
Reserve = B x (ln (1/δ)) / (2 x f) = 12.5 x ln(1/0.005) / (2 x 0.01) = 12.5 x 5.2983 / (2 x 0.01) = 3311 BTC

So if we hit the halving, the B value halves, so you simply halve the answer = 1656 BTC

So does it really make thatr much difference?
Well it's half the BTC, but still a lot of BTC ... ... ... ...


The only loser in this is someone who spouts so much incorrect information and then says they won an argument.
Again, it's not an argument, I'm pointing out where you've gone wrong in MANY places throughout this discussion.

Seriously?

You're gonna go down a conspiracy theory path for a reason to not mine at a pool who looks like they own 240 PH ?

What about a 10PH pool you like to send mining to ... that's OK?
They don't even check any of the statistics to determine what's going on, so that would be a simple case of just point withholding miners there ... ... ... ...
But that's OK? Or are they just too small it really doesn't matter at all?

...

That is the point of the math I've been providing from the start before you began spouting rubbish.
To give actual grounds to not risk mining here.

Not some random 'gut feeling' or random 'conspiracy theory' you seem to enjoy using.

...

If you want to turn this into an argument, then first correct ALL that garbage you've posted so far.

Edit: Phil, you haven't been around bitcoin for very long, go find and read the threads for the 2 best known pools that have paid PPS and gone broke and the end game for the two of them. Not very nice (BitAffNet and MtRed)

I never said to mine here I simply said you using a reserve  number of 4300 is wrong
you then went all out to win your arguement


I completely proof this part is wrong by showing  multiple linear mathematical example of why it was wrong

linear math proves you to be wrong in the application of the formula

It also proves  that the formula has better accuracy on day one of a ½ ing vs day 600 or 700 and you think I am out to defend pool op.

I am not all I did was want to show that he did not need a reserve that high.

Just because you lost the argument does not mean any one should mine here.

Peace out I am done here.

To all that think I want you to ignore kano and mine here:

I don't and  I don't mine here.

My reasons for not mining here  are practical and don't involve difficult to understand math.

1) bitmain has admitted to mining with 10 percent of the world's hash

so we are at about 43,390 PH/s worldwide
bitmain by there own admission is 4,339 ph

this pool is at about 240 PH

So if bitmain points 60 ph  to do a with holding attack  it could cost the pool  around 75 btc a month

https://bitcoinwisdom.com/bitcoin/difficulty  just put in 60000 th and up pops 74.67 coins a month

now if bitmain can write  3 speed s9 software (they can)
they can write 4 speed s9 software   with
low
medium
high
with hold
they can rotate  the 60ph around their 4,339 ph  with proxy and vpn
this pool would not be sure  if they were being with held or not.

what is the cost to this pool  about 74.67 coins a month

lets see what does bitmain lose the tx fees  which are around 1 %  or .75btc

if they pointed it to solo antpool and used it to pay off  cloud contracts  they could suffer variance
if they pointed it to antpool and used it to pay of cloud contracts they could suffer variance.

their profit margin on a cloud contract  is well over 4 %

so they pay .75 btc in tx fees collect well over 3 btc  in cloud mining fees

and cost this pool  75 btc.   this reason may never be done ,but it can be done

and while as Kano says I don't know stats I do know that the above is an issue that stops me from mining here.

This issue  causes me to feel the pool op will fold up shop.
This issue  is far more real world then kano's use of meni's formula for reserve needed.

I don't think it can be stopped if bitmain did this to a pool. Other then banning all bitmain gear from mining on the pool

Well yeah size matters if you are talking about pissy little PPLNS pools that take 2 months to find a block.
But in that case you are already losing BTC BADLY due to diff changes making your shares worth less across the time frame of the N in PPLNS.

If you want to talk about such pools (PPLNS or PPS), then go find a sandpit to sit in, and get the other kids around, to sit and listen.

But what is going on here, so far, is a pool saying how great they are, and no fees, paying PPS, and Phil backing them up saying (incorrectly) how little BTC they need so everyone will be happy and make PPS similar to a 1% fee PPLNS pool but with no PPLNS variance at all.

Wow sounds great, lets get lotsa people here mining and making great rewards ... ... ... hmm so that means it does matter if they grow to even a 1EH pool ...

So either people will see that they don't have the BTC if the pool grows and wont be able to pay a down turn in luck, or people will see Phil posting about how this is all OK, they don't need a lot of BTC to be able to afford to run the pool so everyone should join in ... ... ...

This thread is about this pool and I'm trying to say that there's a problem with the PPS payout, and I see you saying: no my statistics are wrong or deceiving people ... ... meaning yeah it's OK go for it

Ignoring the fact that it is off topic, altcoins in general are worthless (otherwise why would you get them for free ... )

If the pool is merge mining some tiny altcoin and keeping it, they need to be making a % of the pool's expected income to thus increase the % fee for the calculation.
i.e. if the fee is 1% and they make another 0.01% by keeping merged mined coins and not giving them to the miners, then they could say the pool fee is 1.01% in all the calculations (which is basically a waste of time)

Which I didn't and I clearly proved that by doing a very simple bit of math that you wouldn't do, and as usual in this discussion, you're just ignoring it.
0.1% or even 1% chance of failure - both require a large amount of BTC

Your argument is saying I'm deceiving people by using 0.1%, but who cares, if we switch it to a be a "Phil" pool where 1% chance of failure is OK, it still requires 2878 BTC instead of 4317.375 BTC

Yeah you're pretending that's not the case ... dood you're wrong, OK, just live with it, and stop giving people REALLY bad advice.

... and I've no idea why an idiot mod would give merit to that post that gives people REALLY BAD advice ...

I think, statistics matters here and even at 1% OP may go a long way and get away with it, but he'll have been lucky. Probably even more so if pool doesn't get bigger quickly (again, statistics but wider chance for variance with more participants?).

Bitcoin aside, is it possible somehow merge mining could offset some of that risk? I never really wondered until now if a new pool might actually gear to optimise that.

pool size matters as a practical matter.

If 1 ph is mining at 100% pps  payout is 1.245 btc

a 4300 coin reserve  would last for 3453 months  if a block was never hit.  many ½ ings would happen so it would last much longer.

if 240 ph is mining at 100% pps payout is  300 btc a month  

a 4300 coin reserve would last for 14 months with no blocks hit.  about may next year.

block would change to 6.25  and even if tx fees stayed same size it would be 2x the %

so 4300 would drop to 1075.

B 1n 1000

2 x .02

or 6.25 x 6.9../  2 x .02 =  about 1078 coin reserve

tired but above is accurate.

below I am not as sure.

Since pool should do  at least 90% luck  practical number is not 4300 more like 300 -270 = 30 x 12.5 = 375 coins if he stays at

240ph    which gets him to ½ ing

so what is the odds if 14 months at 90%  ?

or 270 block when you should get 300

edit

1)The reserve formula  from meni  is accurate at day 1 or 2 of ½ ing start

the closer you get to the next ½ ing   it is less accurate  2x on last day Due to B changing

2) when fee is the tax fee  the  change is 2x the fee or ½ the  old reserve due to f changing in ratio to B

3) so   2 x 2 as applied to this pool  so reserve is 1075 vs 4300  next ½ ing

4) pool size is a factor as the smaller  the pool the more likely you can reach the next ½ ing  with a smaller reserve since you pay out much less

5) close to ½ ing time wise the pool is more likely to reach a new lower reserve/

@ op and kano   thanks for this it was fun

@ meni  if you still read this forum  tell me I am right or wrong. about your reserve  formula as I applied it.

nite all  zzzz

I never said the statistics you gave were wrong I said you used them deceptively.

try another way.

practical is what matters.

240 ph =  10 coins a day  or 3650 coins a year

so if the op's pool  hits zero blocks for a year vs the almost 300 it should hit  the op would  pay out 3650 coins of his 4300 coin reserve

fuck no he would shut the pool down thinking he is being robbed some how.

to be a little more realistic   lets say he hits  290 blocks  down 10 =  125 coins

if I mine there for 365 days  earning 100 a day vs 97 a day  I earn 36500 vs  35405

and his reserve of under 140 coins   worked for me for an entire year.

your use of meni's numbers does not apply  for the miner.

Now if you want to argue you don't care about the miner in your example and you are concerned for the pool op  then you are correct you are warning him he should go broke and truth is he should go broke without  a huge backup  of coins.  

why risk the coin if you have it?  I guess you and i look at this differently My angle is that of being a miner and your is that of being a pool operator.

So pool op charge 2% plus tx fees  not 0%  plus tx fees  you will still beat viabtc rates  and stand a better chance of not going broke.

3311/3 = 1104 coins (reserve) gives you a 199/200 rate of success.

good luck to the op  since I am betting he won't last past summer at just the tx fees

So you've given up and are ignoring the statistics now ... (since it doesn't match what you want)

Pulling the number '34 days' out of your ass is pointless.

I'll link it yet again https://bitcoil.co.il/pool_analysis.pdf

... and it has nothing to do with pool size.

it is late I am tired.

I can show that miners only need for him to have  a reserve to pay 34 days in a row.

via btc pays me 97  op pays me 100 each day in mining

34 days in a row = 3298 from via btc

34 days in a row = 3400 from op

op folds   runs skips  etc  I loose on 35th day

so  via btc would have paid me 3395
op would have paid me            3400     so if he had enough coin to pay for 34 days  I am better off with op.

since pool should make   24 blocks in a month   it would need to do 11 blcoks  and the reserve of 140 would  be enough

making 11 blocks vs expected 24 is more then 100 to one.

my point was he may need a much smaller reserve if he is legit for a miner to profit. and that asking for him to have a 4300 coin is too much

But honestly  everyone that runs a high paying pps  seems to go broke

why is that?

they are simply hoping to have an early hot streak  and if they don't they bail.

I doubt he has 2 blocks in reserve no less 13  or 350

thanks for the break down on bankruptcy for op.

I would guess it is more like a 95% chance for him to go broke.

Where does this bullshit 140 come from?
Seriously can't you just put the simple numbers into the equation on the other page?


SO ... did you try the: VERY SIMPLE to calculate, and very bad suggestion of, a 1% chance of failure?

OBVIOUSLY not.

See, I do realise why this doesn't matter, but to make you look even more stupid I'll do the calculations

We already know 1/1000 that I used the same number that Meni used - that produces that number in the middle ln(1/δ) = ln (1/0.001) = "6.9078"
Then the equation is (as already stated) 12.5 x 6.9078 / (2 x 0.01) = 4317.375 BTC

But what the heck, lets jump to the bottom of your list and go for a ridiculously risky pool with a 1% chance of going bankrupt.
ln(1/δ) = ln(1/0.01) = 4.6052 HOLD YOUR HORSES! that's not much less than "6.9078" (yes I did realise this)
So what do we get for the BTC required?
12.5 x 4.6052 / (2 x 0.01) = 2878 BTC

No doubt the pool doesn't have 2878 BTC in it's wallet either.

But what the hey, I'll do a quick table of all 4 numbers:
1/1000 4317.375 BTC
1/500 3884 BTC
1/200 3311 BTC
1/100 2878 BTC

Yeah I don't see anything as tiny as 140 BTC ANYWHERE in that
I do see a bit more than 20 times that if the pool is going down the path of a high 1% chance of going bankrupt ...

Don't you hate non-linear maths?

@ hagssfin. I understand it.

Problem is math Kano gave was deceptive. And he implied I did not understand it.

At 240 ph  he needs a practical reserve of more then 140 coins as I explained.

If he is with held attacked he will need more.

And even a reserve of 4300 coins he will bust out if he suffers from a large withholding attack.

So in a magical world of zero withholding attacks 4300 coins. Would give hive a 999 out of 1000 chance

edit Of not going  bust.

I don’t really care about the op having a reserve that big. As a 999/1000 safety factor is more then enough.

I would argue I would mine with this pool if he had enough coins to make the safety factor 199/200

As he pays 100% of 12.5
And via btc pays 97% of 12.5

I resent Kano tossing out the number he did and then acting that he is smart and I don’t understand the math. And that I am not entitled to have an opinion on the thread.

Here is my opinion.

I think a reserve of enough coins to make the risk  of going broke 1/200 would make the pool worth playing.

I think the op does not have that many coins.

I think that if the pool catches on it will be with holding attacked.

I won’t mine at the pool since I think the pool will fail.

To further explain that if I had hundreds of coins for reserve I would not open the pool and do just tx fees.

I would do 98% which is still better then  anybody else.
My guess is op is hoping the pool has a hot streak early.
My guess is if pool has a cold streak early op,will fold it.
My guess is op has very little coins in reserve and is following the plan I just outlined.

I also think I would like to see Kano calculate

For 1/1000
For 1/500
For 1/200
For 1/100.

And say phil you are right I used the 1/1000 figure and did not explain it.

But Kano has never admitted being wrong for anything I ever called him on.

Also I should  shut the fuck up as this post will piss more people off.

Kano crossed posted and stayed true to form.

did a bit of edits

I used Meni's choice 0.1% - phil can go argue with Meni if he doesn't think that's realistic - though phil seems to know nothing about statistics ...

I explained the equation quite clearly on the other thread - so anyone can simply just plug numbers into it.

But assuming it's ok for a pool to have a 1% (or even 10%) chance of going broke is ludicrous.

Phil, 1 in a 100 or 1 in a 50 factor would mean that you agree him having a higher chance to go into the bankrupcy.

I think Kano's numbers are good,
because a pool with those numbers is not a risk for the miners.

So you said simply put  and showed that number  I ask again did you use a 1 in 1000 factor?

Did you use a 1 in 100 factor?

as I see it

 1 n 1000
 1 n 100
 1 n 50

all give big difference in the reserve needed.

So what was the 1 n factor.

1000 , 100 , 50

if I mine at viabtc.com   and get 97% pps with a very small chance of bankruptcy
it could be worth losing 3%

if I mine at op's pool and get 100% pps  with a larger chance of bankruptcy  it may be worth it to me.

so if you are using 1 in 1000 to get your reserve number

you have attempted to deceive all the mathematically challenged miners reading this.

so what is the answer how did you get that 6.9078 number?

I can tell you if you used 1000 and I don't mind 100 then he needs 1/10 the reserve

BTW   I am pretty sure you used 1 in 1000

now  using 1 in 100  he needs a reserve of much less  then what you put in correct?

and if he stays at 240ph which is what he was at  he would make a block  about every 1.2 days which means he should make 100 blocks in 4 months  now if he has a reserve of 50 blocks or 600 coins. he would not go bankrupt in the first 4 months.

Honestly speaking I think he is crazy trying to do a 100% pps and I won't mine with him.

but if you toss up a number that says he needs a reserve of 4317 BTC  you should say  that is for a 999/1000 chance of not going bankrupt
you did not.  you mislead people .  I normally do not go off on people but to top it off you tell me if I don't understand don't voice an opinion.

Dude I do understand.  so my opinion is  you left out that 1/1000 or 999/1000  number out to mislead people that struggle with the math.
I can't code but I know math.

 So to the op
if you want a 99.9% chance of not going bankrupt you need  4317 coins

If you want a 99.0% chance of not going bankrupt you need less then 4317 coins.

Maybe as a peace leaf kano could calculate that for you.

Haw about that k man

do 99.9%  you did  4317 coins
do 99.8%
do 99.7%
do 99.6%
do 99.5%
do 99.4%
do 99.3%
do 99.2%
do 99.1%
do 99.0%

or leave the guy alone as you were pretty much out of line here.

At the op  my guess is you don't even have a 50 coin reserve and will go broke but that is a completely different story.

MY alternate account as I want to space this. Not double post.