Topic: [0Th]Ozcoin Pooled Mining |DGM 1%|Stratum+VarDiff port 80|NEW CN mining| - page 130. (Read 398185 times)

the first error was fixed, not sure what time you edited your post.

automatic payments were fixed yesterday.

sorry if my life has interupted your bitcoining - off for surgery tomorrow - guess i been busy preparing for that

second issue fixed - thanks for letting us know

Notice: Undefined variable: bitcoinController in /var/www/accountdetails.php on line 35 Call Stack: 0.0000 640440 1. {main}() /var/www/accountdetails.php:0 Fatal error: Call to a member function validateaddress() on a non-object in /var/www/accountdetails.php on line 35 Call Stack: 0.0000 640440 1. {main}() /var/www/accountdetails.php:0

I think you need to clean out your cookies  

Update: Fatal error: Call to undefined function islocked() in /var/www/accountdetails.php on line 38 Call Stack: 0.0000 640440 1. {main}() /var/www/accountdetails.php:0


Dude, what's up?  Was there some thunder down under?  

Are you on a walkabout to the local billabong in search of a tasty blooming onion?

Why is it that you see pools as a bet and not what they are - collective mining cooperatives?  While you can "bet" on a pool before a round start, it's ultimately silly since you are fighting against pure randomness.  In the end, you can't but hope to come out even at best so it's really a lot of time and energy spent on nothing.


Hopping pools does absolutely nothing to "secure the network" or provide "hashing where it's needed."  Whether you mine solo or on a pool is completely immaterial to the security of the network, unless that pool has >50% of the total network hashing.  Other than that one factor, there is no network advantage to you hashing for different pools or a single pool.

Hashing is never "needed" anywhere - each share submitted on the network as a whole has an equal chance against any other share submitted. Whether you're providing your hashing at pool A or pool X, your hashes are equivalent. Spreading them around provides absolutely no benefit to the network.


What is it you have questions about?  DGM, in a brief nutshell, will ramp up when you first start on a pool - you'll receive less payout for the first few blocks than you normally would.  Then while you are hashing, after the initial ramp up phase, you'll receive the same amount as you would on a proportional pool (plus any gains from pool hoppers).  Then when you quit, your payout will slowly ramp down to cover what you did not receive during the ramp up phase - you'll be paid for blocks you didn't even participate in. 


DGM is going to be one of your best bets to combat the pain of intermittent connections.  GM, PPLNS, etc... can be really painful if your connection drops at an inopportune time.

Payouts broken, no auto either...

both pools are, we are running DGM alongside the LTC pool to get some stats atm

when we decide to change payout system we wll post in forums, update 1st post in this thread and email members

so, as it stands now, btc pool is still prop?

I think the main issue that worries paraipan and other miners is that intuitively you look at prop or pps or smpps and you see every share paid. A lot of people don't see a difference between that and 'the expected value of every share is 1.0' and so don't realise that they'd get the same payout on pps and dgm - only the variance is different. Instead, many people just see a bunch of shares unpaid and don't realise that this will be (or was) offset by previous times that shares were paid more than 1.0.

Inaba, pool op for EclipseMC, makes this explicit - his stats inform miners what they earned compared to what they would have at a proportional pool - sometimes more, sometimes less. Spelling it out like that for miners increases their confidence in the payout system. If you also include what they would have earned if the pool was proportional and being hopped, then they'll see they are much more often ahead than not.

Feeless PPS is unsustainable long term, so if you can put up with some variance and you don't want to pay the 7-10% fee, DGM or PPLNS is your best bet. Since the parameters of DGM are variable and can be adjusted, I personally thing DGM is the better bet. You'll be able to ask your miners what they prefer, eg less variance vs longer time to payout. If the majority of miners are under 300Mhps, then you can reduce variance.

As the pool hashrate increases (as I expect it will) you can take advantage of shorter round lengths that reduce variance and payout time for you and tweak the DGM parameters further. PPLNS can do this too by changing the N value to something smaller, but there is not nearly as much control as DGM has.

Finally, as well as informing miners what their earnings under the proportional system would have been (hopped or unhopped), a pool luck indicator is another important stats tool that lets miners see at a glance how lucky the pool has been both short and long term. The developer of the DGM payout system has suggested a good pool luck indicator, and a much better one than I was using previously: https://bitcointalksearch.org/topic/m.625307

yep, you got me Graet, i don't really know what i want...
the only thing i know bitcoin is like a roulette and we're all in for the winnings. I see pools more like numbers you bet on before a round starts, so the measures ops are taking to "force" miners bet only on their pool are a bit strange.
The logic always tells me to have my "eggs" in more than one basket after seeing what happens when pools get attacked by malicious people.

So, yes, i defend free roam between pools. You have to spend time and resources to make it right but you increase your income and help secure the network by hashing where it's needed. Few times "hoppers" decide to voluntarily help a pool in need if i'm not mistaken. They don't have any obligation but they do it anyways to support a good service.

Sorry for spreading FUD about DGM, i don't know how it really works and can't find good explanations either. I suppose Meni have done a good job with the math and it's doing it for a living too so, like many others, i'm lost there.

Organofcorti has done a good job with the graphs btw. I managed to understand why i had so much payout variance on score pools compared to prop, because my internet connection isn't perfect and i have to take that into account too.

Please ignore my rambling if it doesn't make too much sense, it's more like how i feel right now about the issue

thank your for your opinion.

From your  active participation in this thread 'How to hop' blog
5th post
"we" being hoppers

rather active from here
https://bitcointalksearch.org/topic/m.614646
just after some graphs showing the impact of hoppers on regular miners.
So in a thread that talks about how hoppers know prop is a problem to showing graphs of rewards for hoppers (going up) and regular miners (going down) to telling me the pool will die if we change payout system - excuse my confusion at your motives...

regular miners are working hard for the shares to 100% of round - not just difficulty. this round is over 7M shares so far....
they are sustaining this pool while you "help" another pool to 43% then leave thier miners to complete the round and "help" some other pool, or park on a PPS pool in times of slim pickings on the prop pools.

Show me evidence the DGM punishes small regular miners and i will reconsider, I have not found anything to show this (maybe its one if the reasons for running some tests)... but until you do that is just FUD. This pool was built by miners for miners, all are welcome.

The bitcoin protocol is awesome, but to be vilified and threatened for exploring a potentially fairer payout system... huh?
I haven't stated we are changing yet but simply testing to see if it will be fair... now IF we had implemented with no testing you might have a point in there ...somewhere...

I think I already covered your last point.Be aware your 200Mhash/s is 10x that of some of Ozcoins miners...

Oh and why do you hop rather than help a pool to 100% - be honest now...

indeed, interesting times... we're complicating the payouts with more and more stranger formulas thinking that we will solve some nonexistent "pool cheating" by miners. You're in for a surprise indeed...

Those miners are actually working hard for the shares you get until 43% but they choose to help some other pool after that percentage. Paying them just like the others and helping all miners understand the service they do for the pool is an operator moral obligation. Lately i see almost all pool ops choosing what they think is the best path for their business and that is implementing a new payout algorithm that "punishes", yes that is what is does, miners that are too "small" to fill their charts.

I'm mining at only 200 Mh/s and prefer small daily payouts. I will not even consider a pplns or geometric pool in the future just because it will be like signing a very-long-time-contract with the pool owner. If prop. will be the black sheep in the future i will definitely prefer any form of PPS, even with a greater fee to maintain the service.

Bitcoin protocol is an awesome invention indeed and not even one person truly understands how it works as a whole, only Satoshi did. We are exploring it slowly with our bare hands, trying not to get burned on the way and the fact is that one of our "senses" tells us that miner nodes, or pools, need a lot of entropy (randomness) in this process called mining.

Contrary what some "experts" could tell you the popular experience taught us that the most successful pools are not the ones with 2-3 big miners but the ones that accept any kind of miners out there independent of their hashing power (botnets not included). So yes you're in for a not very pleasant surprise if you manage to scare the greater mass of small occasional miners from your pools people. I will trust my senses and small experience on that and hope you reconsider when you will start reviewing your charts sometime in the future...

Good to hear! Your miners will thank you for it.

Thanks to everyone that took the time to fill out our survey

Some good suggestions,some interesting comments and some that left me a little confused

After much discussion, research and though we have decided to look closely at Double Geometric Method. This would currently appear to be the fairest payout system for both miners and poolops.

We are currently setting up Double Geometric Method on the Litecoin pool to monitor how our chosen variables will look. We are running it alongside the Litecoin pool to gather stats as Litecoin blocks come a lot quicker
Full info on Double Geometric Method
https://bitcointalksearch.org/topic/double-geometric-method-hopping-proof-low-variance-reward-system-39497

I will post info on our results asap - interesting times

Our survey at
http://kwiksurveys.com?u=ozcoinpayment
is bringing some good info
If you havent checked it out do so now.

55BTC/round continues until we move to a fairer payout system
enjoy

US server has been up for a week now, PSJ been running smooth as - thanks Shadders
mining now at
us.ozco.in:8999
and
us.ozco.in:80

the litecoin pool crashed and has been rebuilt ovenight

this thread I though was for BTC pools - litecoin pool has its own thread in the "Altenate Currencies" forum https://bitcointalksearch.org/topic/ozcoin-litecoin-dgmpps-48653

the Pool page has a link beside last 10 blocks found to the "all blocks" page a pretty logical and easy to find link I had always thought, but i guess we can add another to the header, it shows not only unconfirmed blocks and the number of confirmations but unconfirmed earnings by block and value of shares by block.

I have found most "fresh" miners to be experienced people joining a new coin, the ones that arent have been able to click the "webchat" link and get help from our irc chanel or by posting in forums

when i can find a coder with time we will add a "estimated total unconfirmed earnings" for all unconfirmed blocks. the "Est earings" comes from the bitcoin pool and I am aware with the quick rate of ltc blocks this can be confusing, I had thought of removing that all together.

Thanks for thye feedback,
 
also i have just found that by adding a -r -1 flag minerd will not auto quit

Some problems with http://lc.ozco.in or its just me?


it worked well till yesterday - going to alt pool for now...

I would really appreciate more information about your pool in this thread.

ALSO, you need to add "Pending Confirms" field in stats if you want more people on your pool (maybe direct link to blocks.php in header too)
Rename "Est. Earnings" to "Est. Earnings this round" because it is for the current round only.

Those things needed in order to not confuse fresh miners and scare them away (ie. no coins after 3hrs of mining)

++edit: Now it works. Thanks for quick response

55BTC/block bonus is ongoing - 10% more at ozcoin

ahh sorry late reply
i hope we got it sorted at https://bitcointalksearch.org/topic/m.577765

I pointed one of my testminers your way, is it normal that the stats stay at:

Current Hashrate 0 KH/s
Lifetime Shares 0
Lifetime Stales 0
[...]
Round Shares 11,947 shares

Minerd has stated 'PROOF OF WORK RESULT: true (yay!!!)' several times now but the stats never seem to update... Any insight ?

https://bitcointalk.org/index.php?topic=47417.700
Ozcoin has opened a Litecoin pool for those interested in cpu mining
lc.ozco.in

Migration succesful
1st block on new server found 12mins after we started accepting shares on new server :D