Topic: [1050 TH] BitMinter.com [1% PPLNS,Pays TxFees +MergedMining,Stratum,GBT,vardiff] - page 217. (Read 837101 times)

I will try a simple analogy:
If you have 500 coins worth $0.10 in your wallet, you have $50.
If you have one $50 note in your wallet, you have $50 as well.

Now if you go to a shop, and try to pay a $50 item with 500 $0.10 coins, the cashier may look at you in a weird way and you will get many haters if there is a queue behind you. That does not happen with the $50 note.

This is because altough theoretically with the same value, in practice, dealing with 500 small coins for a $50 payment bears a much higher side-cost (read money handling) than a $50 note. The same applies to Bitcoins, and it translates as much higher transaction fees.

Being new I assumed it was safer to take coins at 0.001 to my wallet, to protect against the pool being hacked or going away ...
Why does it matter what my wallet was built with?
If the wallet now holds BTC50 and I pay out BTC10 surely it matters not that the BTC50 was saved up pennies ?

It's good, actually.

The higher the pool's hashrate, the lower the miners' daily variance. The expected rewards remain the same, but the actual rewards fluctuate less.

Now, a single fast miner will mine at a high worker difficulty, which means boosting the pool's hashrate without putting a strain on the pool's network and CPU usage.

I always thought BitMinter already had a 0.01 threshold for payouts, every other pool I know of are already at 0.01. And I totally support your analysis on protecting the newbies from shooting themselves in the foot and avoid them piling up tons of very small coins without knowing what they're doing.


No, it doesn't really change anything for others. To get the idea, if the pool hashrate goes x10 while your hashrate stays the same, you will be paid 10x less for each block found, but it will happen 10x more often.

Can I ask is it bad to have a miner as powerful as realasicminer in the pool ?

BitMinter is changing the minimum auto cash out threshold from 0.001 to 0.01.

The problem with 0.001 cashouts is that it gives you a "wallet full of pennies" and you will be forced to pay fees to send all those tiny amounts elsewhere later. With over 5000 active miners it is also starting to cost the pool a bit to do free cash outs, especially of such tiny amounts that demand a fee. So essentially we were spending money to help newbies shoot themselves in the foot.

Starting now, the website won't allow setting thresholds below 0.01. If you already have your threshold set below that, then it is recommended to change it to 0.01 or higher. If you don't, then that change will be forced on saturday june first (2013.06.01). Contact me ([email protected]) if you have any problems.

I'm working on stability issues right now. Haven't even had a chance to look at Litecoin at all yet. Can't make any promises.

Yes, I counted only the orphans and not the stale ones. But I think you misread my post* because I was saying indeed that the orphans ratio was less than 2%, or even much less than 1.5%, so basically I agree with you.

My point was that the pool was worth a 1.5% donation anyway (full disclosure I am only donating at 0.5% right now but I was at 2.5% with the perks enabled, prior to the 1% fee).

*edit: either misread my post or my grammar was too broken, sorry about that

Are you sure you counted "orphan" blocks? Remember, stale blocks do not count and even with the perk, you do not get BTC for stales. I believe actual orphan blocks are much, much less than 1.5%.

Haha, pathetic luck today. After 24 hrs at 9.5 TH/s, we only found 8 blocks!

I've always donated to the site, 'cos it's a very easy to use site and full of info. And "when" my BFL 5GH/s and 25GH/s ASIC's come along I know it will work right out of the box 

I browsed through 20-something pages of block history to estimate that there are less that 2% (it was 2% at the time, before 1% fee) orphan blocks, so I would mainly see it as as the "interests" on the pre-pay advance + "insurance" for orphaned blocks. It depends on your expectation.

At any rate I think the excellent work of DrHaribo deserves at least a small donation, so if not for the arguments above, you can do it just to support him ;-)

so is the 1.50% perk + 1% pool fee worth it, to avoid the dry spell? just curious

yeah I just did the math, I think this more than alleviates the risk of bad luck, and I still get namecoins, compared to other pools

Looks like Dr. H beat me to the punch!  Pretty much what he said.

And now OzCoin is down.  I mined there for 5 months, but it's not very reliable for me right now.  I would happily pay Graeme and his team 6% fee for stability.


Dr. H - any timeline for getting idle miner notification working?

You can mention other pools.  It's not like it's hidden data.  I assume you're talking about 50BTC.  Well I was mining on there yesterday and thanks to the downtown I lost about 3 hours of revenue.  The OP claims the shares were counted but on a pure PPS system it's very easy to see that my earnings are exactly 7/8th of what was expected yesterday, so 3 hour downtime.  Which happens to coincide with the 3 hours missing on the graphs.

This wouldn't be a big deal if the pool OP shut down the pool or redirected traffic.  Any sensible miner not new to the game would have backups.  The problem occurs when they OP says there's nothing to worry about and basically miners are spinning their wheels.  I even found a block for them yesterday.  But since CGMiner didn't detect a disconnect it didn't switch over to backups.

And this is where you can tell the good pools from the bad.  BTCGuild has proven itself time and time again.  Sure it gets hit with DDoS but E is very much on top of it.  That is probably the most stable pool.  So what does it matter if you pay a 5% higher fee if the pool allows 5% more income due to more reliability.

As this thread is about Bitminter we all know Dr. H is very much on top of his creation.  I'm sure for most pool ops it has become a full time job.  I used to mine here and will happily switch back - the only thing holding me off is the lack of miner offline notification.  It's pretty much summer here in Southern California and some of my older cards are not tolerant of this heat.  2 cards are already running 1/2 speed and goes Sick/Dead quite often - can't wait for ASIC.

So again choice of a pool is determined by the end user.  Those who can micromanage a 200 GPU farm as their daily job can afford to sit at a pool with no notifications.  I'll happily pay a higher fee for a pool that has better reliability as long as it doesn't concentrate the network.

So 3% fee, and 1.5% loss of transaction fee income and 6.5% loss of namecoin. Total = 3 + 1.5 + 6.5 = 11%

But I agree that 7.5 + 1.5 + 6.5= 15.5% is more expensive.

Anyway, the difference in average income between BitMinter at 1% fee and a "3% PPS" pool is 10%. It looks like a 2% difference. But it is actually a 10% difference. An important word there is "average". You may want to pay 10% to take luck out of the equation. It looks rather expensive to me, though.

I remember the uproar when OzCoin introduced a 3% fee (later reduced). And yet there are people mining with a big smile and getting 14.5% less than the average payout at BitMinter. I guess those are two different groups of miners. What I could do is offer miners a choice between 1% and 15% fee, to cater for both groups.

I don't want to promote any competing pool here so I won't mention its name but there is another major pool running pure PPS at 3% fees. I never really understood why BTC Guild was so dominant (not implying they are bad at anything here, just others aren't bad either). And another one on CPPSRB (quite as linear and variance-proof as PPS, just without the bankruptcy risk for the pool operator) that only keeps tx fees profits for the pool (so the actual payout rate is ALWAYS 100% PPS in the long run, except if they cease operations and end up with unrecoverable shelved shares).

Thank you for mentioning this famous unlucky round, I feel like a noob now :-)

Yes I get at what you're trying to say.  The chance of something rare happening twice is 1/rare as likely.  Akin to having lightning strike a human being twice.  But if you consider a human being who has already been struck by lightning, his chance of being hit again should be the same as you and me (assuming he doesn't have a metal scalp or something silly).

So if the first run of bad luck happened, the chance for it to happen again is now the same as it was before it happened the first time.  According to OoC the pool didn't exhibit anything that deviated significantly from the expected distribution.

All you new miners missed BTCGuild's horrible 3 month run in middle/late 2011.  So many threads were made accusing E of scamming/stealing.  And 18 months later he has to increase fees to keep people away from his pool 

People either need to spread their mining power, accept a 8-10% fee at a PPS pool, or just live with PPLNS/Slush/DGM based scoring.

You know it probably better than me, and you are right about the 50/50% chance of outcome when tossing a coin no matter what the past results were. But it's very subtle as you know, and if you consider a series of tosses, the odds of an "all heads" or "all tails" series is 1/n 1/(2^(n-1)) where n = number of tosses, or am I wrong? So the chance of having a series of n bad rounds is (twice) lower than n-1 bad rounds, or am I wrong again?

Edit: Ahem I always sucked at math. But you get the idea, I hope

What?  How can you say such a thing?  Past "luck" has no bearing on future "luck".  If you tossed heads 25 times in a row on a nickel you may feel it to be a special nickel but it will still have a 50/50% chance of landing on tails or heads on the next toss (assuming it's not loaded or a 2 headed coin ie cheating).

In the long run (long as in 1 year of mining) the rewards so approach close to 100% of expected, but that's not to say bad or good luck won't continue.

Having said that I'm still taking offers to move away from certain pools as I can offer stats showing I am the most cursed miner