[0 GH/s 0% fee SMPPS] ArsBitcoin mining pool!

[theowalpott]

Good job so far BurningToad

Could I request a few stats related things?

- Show total generated shares for each worker.

- Time since last share for each worker was submitted

- Shares per worker per round.

These would be useful for monitoring! Adding things like this obviously increases load on the server and one thing I like about this pool is how responsive the site is.. so if this is gonna have a large impact then ignore my suggestions

[1713867002]

1713867002

[1713867002]

1713867002

[1713867002]

1713867002

[BurningToad]

OK everyone, here goes the update.  Stats/website might be down for about 10 minutes, but pool will keep accepting shares.

[BurningToad]

All done!  Things look ok so far.  Let me know if you see any issues.  The next test will be to make sure that existing blocks are correctly paid out the old way, and new blocks the new way. 

[BurningToad]

Good job so far BurningToad

Could I request a few stats related things?

- Show total generated shares for each worker.

- Time since last share for each worker was submitted

- Shares per worker per round.

These would be useful for monitoring! Adding things like this obviously increases load on the server and one thing I like about this pool is how responsive the site is.. so if this is gonna have a large impact then ignore my suggestions

Thanks for the suggestions!  I just spent a lot of time on the new payment system, but now it is done!  This means I can focus on other features.  I'll look into these, but I had a few other things at a higher priority first.

[BurningToad]

Have had our first complaint already

Someone is leaving because they are banking on luck.  It is true that we have been extremely lucky as a pool, and HIGHER than expected payouts have been sent.  However, we all know that can't last forever.  When the long rounds come around again, SMPPS ensures that you get paid for your work regardless of round length.  This, combined with protection from pool hopping (which can really reduce payouts for legitimate miners) and I think we have a good system.

[cuongnq]

I think you should open 2 mode to user switch like deepbit : Prop. and PPS.

[BurningToad]

I think you should open 2 mode to user switch like deepbit : Prop. and PPS.

Deepbit can do that because they just charge large fees for PPS and finance it out of their huge wallets.  However, if I implemented such a system, it would essentially just be creating two separate pools out of one.  We are not big enough to do that now, and there are plenty of proportional pools out there if people prefer that instead.

[Meatball]

+1 on the move to SMPPS vs. Proportional.  Was it nice getting lucky with 9 blocks in 4 days?  Yeah.  Would it suck when probability catches up to us and evens us back out to ~ 1 block a day, yeah even more...

Thanks for all the hard work BT.  Great job so far!

[GimEEE]

I was considering joining your pool. . .
I'm glad you showed a willingness to do a rapid change of the pool's ToS that significantly materially affects the mining agreement with little or no notice to your customers.
That alone is a deal-breaker for me.
The fact that the change was from 0% proportional to a system that credits users a fraction of that on short rounds (see my thread on reward comparison) is in itself a deal breaker even if you had given a bare minimum of 48 hrs or so notice by email or some other form like the header of your thread.
Finally, the fact your site doesn't have ssl as all the more popular ones is another.
It is unlikely these differences will ever be reconciled.

Best of luck though, I hope you make some money off the miners using this pps, and I hope they get what they deserve.

[BurningToad]

I was considering joining your pool. . .
I'm glad you showed a willingness to do a rapid change of the pool's ToS that significantly materially affects the mining agreement with little or no notice to your customers.
That alone is a deal-breaker for me.
The fact that the change was from 0% proportional to a system that credits users a fraction of that on short rounds (see my thread on reward comparison) is in itself a deal breaker even if you had given a bare minimum of 48 hrs or so notice by email or some other form like the header of your thread.
Finally, the fact your site doesn't have ssl as all the more popular ones is another.
It is unlikely these differences will ever be reconciled.

Best of luck though, I hope you make some money off the miners using this pps, and I hope they get what they deserve.

I have been mentioning the switch to SMPPS for quite a while now in this thread as well as on IRC.  I could have given better warning, but the expected reward for mining on the pool has not been reduced at all.

Your reward comparison thread shows some payment problems at Eligius, not payment problems here.  If the pool has BTC to pay, and someone is owed BTC, it WILL pay them.  It is very clear on our website how much the pool owes, how much it has, how much it has paid out, etc.  Miners will NOT get short changed on this system.  The pool will NOT "make money off the miners".  Over time, the SMPPS buffer will trend towards zero by definition.  I am not going to suddenly stop the pool with a large BTC buffer and not pay it out to miners.  

I do need to get SSL set up.

[gade202]

I'll try out the new SMPPS method, but, I'm not sure I'll be a fan.

I realize it gives steady payouts but at the same time, I very much liked the chance of getting 3 blocks in a day and didn't mind the risk of it taking a day and half to finish one block.  As of right now, I'm limited to an estimated 1.277BTC/day based on the current fixed share rate.  That would basically cut my earnings in half.

Sure, we're going to have a day where nothing is solved, but then the days where 3 blocks are solved offset this.  In my "lifetime" with the pool, I've been paid roughly 12BTC (over about 6 days).  Using my actual lifetime share count of 147,351, under the new SMPPS rate, I'd have only been paid 5.33BTC.  Granted, we've been lucky as of late but I don't think we're going to be so unlucky as to take a huge hit like that.  With the pool growing from ~50mhash to ~85mhash over my stay, the chances of turning blocks faster is better, albeit at a cut in proportional rates.

Like I said, I'm going to see how it goes over the next 2 days and re-evaluate from there.

Update: READ BELOW

Just to run some numbers by everyone:

Take a recent long block, completed on July 3rd at 12:00:03am.  Consisting of 1,839,526 shares, it took roughly 49.5 hours to complete.

The SMPPS on those shares, based on the current rate, would have been ~66.688525BTC paid out.  Hmm, that's a substantially lower figure for an unusually long block.

Let's average this out.  In the history of this pool 13,594,099 shares have been submitted, solving 18 blocks (900BTC).  That averages out to 755,277.72 shares per block.

Now, what if this policy been in place from the start of the pool?  At the current rate of 0.000036253103 per share, only 27.379BTC would have been paid out to miners PER BLOCK!  

Just a 54.75% payout.

Please help me wrap my head around this as I didn't expect the figures to look this bleak...

[Lightspeed]

Hi,

Tom Lightspeed here, did some stats work, if you like it please donate: 1LYjJCtP8RFJ6zz6orxf5ym9CrvxPS4pDr

I could do some of the things you guys want too from what I am reading

May the force be with you

[BurningToad]

Hi,

Tom Lightspeed here, did some stats work, if you like it please donate: 1LYjJCtP8RFJ6zz6orxf5ym9CrvxPS4pDr

I could do some of the things you guys want too from what I am reading

May the force be with you

I do appreciate it tom!  I sent you 0.5 BTC a while back.  If the pool becomes a net positive for me, or we request some work, I can send some more Others are of course welcome to donate too if you like the stats page!

[BurningToad]

Just to run some numbers by everyone:

Take a recent long block, completed on July 3rd at 12:00:03am.  Consisting of 1,839,526 shares, it took roughly 49.5 hours to complete.

The SMPPS on those shares, based on the current rate, would have been ~66.688525BTC paid out.  Hmm, that's a substantially lower figure for an unusually long block.

Let's average this out.  In the history of this pool 13,594,099 shares have been submitted, solving 18 blocks (900BTC).  That averages out to 755,277.72 shares per block.

Now, what if this policy been in place from the start of the pool?  At the current rate of 0.000036253103 per share, only 27.379BTC would have been paid out to miners PER BLOCK!  

Just a 54.75% payout.

Please help me wrap my head around this as I didn't expect the figures to look this bleak...

1)  A 1,839,526 share block is not "unusually" long.  BTC Guild has seen 10 million + share blocks.  66.68 BTC is the correct amount to pay for this block length.  If a block is double average length, then the pool pays 100 BTC, etc.
2) Our longest round was mostly completed at a difficulty of around ~800,000, which means that the PPS value would have been significantly higher.  That, PLUS the fact that we have been REALLY LUCKY mean that yes, the total payout so far would have been smaller.  However the pool would now have a large buffer (it has a 0 buffer now since we just converted) to pay out more for longer rounds.
3) Once again, the goal of SMPPS is to provide fair, constant payouts.  If you prefer to chase luck, then you are better off at some other small proportional pool, or solo mining.

I guess this won't make sense to everyone until we hit a 3+ million round block

[IceyEC]

If you want to see pool stats over time, Lego399 and myself have created some graphs with the API:

http://stats.nuradu.com/showstats.php

[goldcd]

SMPPS will go live in about 1 hour

Bah. I'm not going to quibble about the change as: a) it's not my pool b) you're not charging me to use it and c) I support anything that provides competition to the big pools.

but..

I liked the completely random payouts. The main reason I switched was (well OK, that was for the block bounties, but the other main reason was) just for a bit of a gamble. Days after day of nothing and then days where it came pouring in. Being a small pool the randomness was even more wonderful.

I can understand the change, and my love of variance isn't quite enough to drive me to mine solo..

How about bolting on something interesting to separate yourself from the other pools (I'm sure you're just twiddling your thumbs with nothing better to do). Opt in a lottery - say donate 10% of profits to a pool and then randomly give the pool (less admin costs naturally) to somebody in the round (each share gets one ticket).
 

[gade202]

10,000,000+ share blocks are possible but not probable.  BTCG has over 1,460 blocks solved and, as such, shows a wider range.  I'd be curious to see their stats averaged.

Basically, the break even point for the pool would be roughly 1,380,000 shares per block.

I understand spreading the risk out and the perks associated.  Boiled down, it's insurance.  I just don't like the idea of the pool taking on the risks and limiting the users possible rewards.  Personally, I don't mind the risks if everyone gets paid for their share of the work.  That's just makes the most sense to me as everyone equally shares the risks and rewards.  It just throws another cog into the chance wheel.  Eventually the pool will run a slight deficit or a surplus.  It's not going to be perfectly even.

[gentakin]

That's just makes the most sense to me as everyone equally shares the risks and rewards.

Actually, no: Pool hoppers have less risk and more reward than anybody else with proportional.

It just throws another cog into the chance wheel.  Eventually the pool will run a slight deficit or a surplus.

No: The pool won't run at a deficit. With MaxPPS, that's impossible (unless the implemenation wasn't done correctly).

[BurningToad]

It just throws another cog into the chance wheel.  Eventually the pool will run a slight deficit or a surplus.

No: The pool won't run at a deficit. With MaxPPS, that's impossible (unless the implemenation wasn't done correctly).

Right, it won't run at an actual deficit, but it could have a negative buffer, which means it can't afford to pay out the PPS value of all of the shares submitted yet.

[BurningToad]

Ok, I'm going to try to stop responding to every post.  This IS a new payment system, and we will have to see how it goes.  Almost every other pool out there is proportional, so there are plenty of options if you prefer that method.  For those that are interested in this method, you're welcome to stay.  I think it just comes down to preference.

Here are some more resources to compare payment methods, courtesy of luke-jr from Eligius.

Charts comparing real-world payouts:
http://eligius.st/~luke-jr/samples/800MH/

Notice the pros and cons of proportional:
    Easy math: (your shares / total shares) * reward
    Pays full reward for every block for work on that same block
    Pays in excess of work done for short blocks
    Pays less than work done for long blocks
    Easy payout improvement with pool hopping
    Risk from withholding is distributed among miners

And the pros and cons of SMPPS:
    Fixed rate of payout for work performed, limited overall by actual miner performance
    Never overpaid based on pool earnings; underpaid only when overall work performed has failed to deliver
    No benefit for pool hopping
    Risk from withholding is distributed among miners
    Non-reward credits carry over for future payment, regardless of future work

More discussion of various payment methods:
http://forum.bitcoin.org/index.php?topic=21999.0