Topic: [1050 TH] BitMinter.com [1% PPLNS,Pays TxFees +MergedMining,Stratum,GBT,vardiff] - page 21. (Read 837101 times)

8 months near enough.  I started when BTC prices were low, saw the £1000 spike, saw people poop a brick, and then saw the '£300 period' where people started looking again.

So far it's as follows:

Bitminter.  Small & reliable.  Income fluctuates, and this scares some people.  1450 TH.

Eligius.  6000 TH and reliable. Income fluctuates less, but any good luck is spread out over a lot of people so will impact on how big your 'good luck' payment is.

GHash.  40000 TH.  So reliable in it's return that people are now buying life-time mining contracts on it ( getting a host to mine for them ).  The big data-centres come here too.  Income really does not fluctuate ( so no panics happen really ), but it's boring.  If you get any good luck you just won't notice, where it is spread across so many different people.


Bitminter has the best USB client ( it just works ), and also a VERY nice GUI.  The Perk System is also a real selling point.  And yet we're small.

If it's not the product ( and obviously it's not ) it must be the marketing.  So you need to check stuff out to see what you can say truthfully to show you are the best.

The only fly in the ointment is people panicking over luck ( there have been plenty of posts on this ), and that is why I am looking for ways to stabilise payments ( so that we appear as reliable as Eligius or GHash to someone who is new to mining ).

If we can look nice and reliable to a newb they may well get that Credit Card out and give Doc some more hashes.


Anyway, I'll let this chat cool down for now.  I know some are interested but others aren't.

How long have you been mining?

I think you are mis-judging the intelligence of some of your miners.

Considering how nice a Pool Bitminter is ( the nice GUI and decent USB client etc ) have you not wondered by you are so small compared to others?

Do you want to grow or not?  Keep talking Casino and you won't grow by much.

And before you say anything I'm an engineer with a corporate background ( including Legal, Finance, & Management ) and 20 years of reading under my belt.  Investors do not want Casino.  They want stability, so that they can stick their feet up and let someone else do the running around.  Casino is not an attractive investment; stability is.  Hence why the big pools do so well.   They may be boring ( no real cut & thrust ), but they do have a reliable ( enough ) return that means you can budget.


If I can find a pattern that The Doc can use to improve stability then a lot more people will reach for a Credit Card and add an s3 to this pool.  Which is what we want.

It's simple business.


Spreading your bets over 2 pools is the first stabiliser.  I'm simply testing whether pool-jumping is still possible if you crunch the figures once a week ( which isn't much effort I assure you ).

If you could average out a month's payments based on predictions etc you could create a perk that 'stabilises' people's payments.  So if you know someone at their present hash will earn 1 BTC over  a month based on your graphs then you can work out what an average payment would be per week, and then offer that.  Nice and stable, rather than Casino.

Charge for it to create a reserve to pay for any amounts up front that will be recouped later.  You might even squeeze a profit if your luck holds.

That's another way to stop people from jumping ship after a bad week or 2.  And may well attract investors.


Investors do not want 'Casino'.  They want risks kept to a minimum.  Bitminter is a small pool, and if you want to grow it you have to play the game.  If you give them Casino they'll go to a big pool that is good at hunting for blocks, and also more stable in it's payments ( if less lucky on a personal level due to how your luck is being shared with so many others ).  That's why GHash is the man now.  Profits are fairly dull ( not very exciting ), but they are reliable.  There's no Casino.  So investors will sink cash into a datacenter loyal to GHash.  You need to find a way to make Bitminter look as reliable on return to get these people to come to you.

So talking Casino is a real bad plan here ( even if it is true ).  You need to think of a way to show stability both on the graphs and in the payments sent to people.

Your perk system is good.  But you need a 'payment averaging' one for those sitting there looking at an s3 and wondering whether to get the Credit Card out.  As I've said before I know VERY few people who can afford to take £400 and just set it on fire ( for example ), and in a Casino you can lose everything.


I'm happy to help if I see anything

I cannot even begin to understand what you are thinking but please stop trying to explain to the rest of us.
And you are splitting up your 1.5TH/s of miners to chase this luck?  Even if your theory was possible, the amount of hashing power you have does not really make it worth the effort you will put into this.

It takes about 4-5 days for Bitminter to get up to speed when I add some hashes. The other pool was the same.

Bitminter's peaks and troughs ( the graph) last for 2-3 weeks from what I can see.


That's the test. Read the graph, make your choice, and then go from there. It's only 1/3 of my hash, so shouldn't be too big a gamble if I get it wrong.

However...


If you want people to hang around then you need to give them options should they panic over a week of bad luck. Or they'll see you as a bad pool. Mining techniques to smooth things out should help here. Using 2 pools is a basic ( it's nice & simple to understand) , but there will be profit-chasers as well.

Do you realize that chasing luck is pointless? You'll be leaving a pool after there was bad luck and you already got a low payout, and you'll be joining a pool after there was good luck and everyone else got high payouts.

How do you plan to predict the future? Do you believe old luck causes the opposite luck in the future? Or that it causes more of the same luck? Both are popular superstitions.

In the past you could sometimes tell if future payouts would be low on some pools, pools that used the proportional reward system. And also Slush's score-based reward system, which can probably be abused still if they didn't switch reward system. This resulted in something called pool hopping. People would switch pools at the right time and get more than their fair share of the mining rewards, while 24/7 miners got less than fair pay. PPLNS, which Bitminter uses, is designed specifically to make this impossible.


It looks good on first glance, but this would just increase variance. Usually you get low payouts because someone else found the block, but sometimes you may find the block and get a high payout. Over a very long time it evens out, but in a shorter timespan it just creates variance. "Shorter" here means forever for small miners.

That's not to say it wouldn't work. Many miners don't understand variance, and there's also the psychological aspect of saying "you get paid a bonus if you find a block!" but not mentioning that you get a lower payout the rest of the time.

Worse is paying for stale work. It's a horrible idea. I think there are still some pools doing it, though. It just means that people with proper miners have to give some of their income to people with broken miners. If you have 1 TH/s and get 0.1% stales, and another user has 1 TH/s with 5% stales, you get the same payouts. Even though you contributed more to the mining effort. The other person was doing a lot of useless work, working on old data long after it was too late. So the effect is that some of your income is given to someone else. That's why you should not mine in such pools if your equipment is working properly. They are great if there is something wrong with your miners though, and they are producing a lot of stale work. Just like pool-hopping vulnerable reward systems, paying for stale work creates a pool where some users get more than their fair share, while others get less.

On the other hand, if the pool says "we pay for stale work, so you don't lose anything - it can be an extra 1% income!" then it sounds like a great thing. What you don't realize is that in reality you may be making 1% less than your fair share because your hardware, software and internet connection are actually working.

That's a lot of rambling. Anyway, that's why I don't add these two "features" that some pools have. I couldn't stomach marketing it as a good deal either.

I'll be running an experiment:


I'll be splitting my hash ( currently 900 GH/s) 50:50 between Doc & another pool I won't name ( out of respect to the good Dr ).

Gonna see if that helps to stabilise earnings/ luck a bit. Bitminter still averages over a month, but these peaks & troughs obviously spook some people.


I'll also be adding an S3 to my mining kit later, which will be moved between these 2 pools to chase the luck ( it's the only machine that will pool-jump ).

The idea is to go for max profit, without damaging Bitminter's block-creation abilities ( by moving too much hash away & thus destabilising the pool if enough people start pool-jumping ). Only 1/3 of my hash will be mobile.

It may well make no difference, but the next month ( before the S3 shows ) will be very interesting. I will keep you informed.

The better way to improve and fasten payouts is to add hashing power.

And maybe I have an idea that you should consider to attract bigger hashers...

It could be wise to give a reward to the block finder in addition to his payout.

The amount is to be consider, but if the blockfinder could received, maybe 0,25 btc + his payout, this would probably attract more user to the pool. Bringing more hashpower, bringing more regularity, but in that case, losing 1/10 of revenue for each user....

on the s1 under miner configuration tab use:
Pool 1             stratum+tcp://mint.bitminter.com:3333
Pool 1 worker  yourusernameonthatsite.workername

on the site just workername as the above.

stratum+tcp://mint.bitminter.com:3333 works just fine with s1

There doesn't appear to be any user named Biosman on Bitminter. I guess this was just an example you were using, but if not, make sure you use the correct user name - it's displayed in the upper right corner of the website when logged in.

There is Stratum protocol on port 443. This is the most efficient option.
There are getwork and getblocktemplate protocols on port 80.
Eventually I want to get completely rid of getwork, so please don't use it.


Maybe you need to connect through a proxy at work?



Some mining client interfaces don't like the stratum+tcp prefix. I can't remember if Antminer S1 is one of those. You could try with just: mint.bitminter.com:443



That's up to you. You can run all your hardware on one worker. New users get a worker named "1" which is configured as a default worker. That means you can mine with just your user name and it will pick the default worker. On the other hand if you want separate stats and slow worker emails then it can be handy to create multiple workers, especially if you have machines on separate physical locations and one of them stop working. You can do this under "my account" -> "workers" in the website menu.


There's a short tooltip text that appears if you hold the mouse pointer over the underlined words, that can help a bit. Ask if there's something specific you want to know about. Most of those settings will do fine with the defaults. The most useful thing is probably low hashrate warnings. You can choose a minimum MH/s there and if your worker goes below that you will receive an email alert.

Hmm, yeah ok, I was a bit quick there. You're hashing an 80 byte bitcoin header. Of that you'll just want to change the 4 byte nonce and the 32 byte merkle root (determined by the transactions you include). So that's 36 bytes, or 288 bits, you change while hashing. That's 2 to the power of 288 possible permutations, which is 4.9732323640978664E86. If we round it up a bit the number of possible headers you can hash is 5 followed by 86 zeroes. As a second passes you can update the time field and you get that many new headers to try.

If you had a lot of hashing power I guess you could compute all those before everyone else, basically buying all the lottery tickets.

It's also possible that at some point, with a certain previous block hash, difficulty, and time value, that there is no valid solution until the time field has advanced months or years. It's just not very likely. It's actually more likely that Slender Man lives under your bed. So check on that first.

50% CDF means 50% of blocks are found with this amount of work or less at the difficulty in question. At 99% CDF only 1% of blocks take more work, so that's pretty unlucky. On average it happens with 1 out of 100 blocks though, so it's not like some people think; "bad luck like this is impossible, something must be wrong". Mining is like a lottery and on average 1 out of 100 blocks will take that much work. If you expect that to never happen, it's your thinking that is wrong.

Merged mining NMC is free for you at no computational, power or financial cost. If you do not want your NMC, consider to move the slider for NMC donations to the pool to 100%.

Behind a Firewall: How do i get pool connection to Bitminter under Port 80?

Hello Community!

I am working at home with Antminer S1, every thing is fine there.

I want to run 2 Antminer in my office, but we have there a firewall, only following ports are open:

SMTP: Port 25
HTTP: Port 80
POP3: Port 110
NNTP: Port 119
EPMAP: Port 135
Shared Network Folder: 139
IMAP: 143
HTTPS: 443

I found now your Bitminter pool, which is connectable with Port 80:

http://mint.bitminter.com:80

At home this adress works fine but in office it does not work
Does my Antminer need another port to run or where ist the problem?
I use PortScan 1.46 to watch the ports.

----------------------------------
Another questions and problems:

The adress: stratum+tcp://mint.bitminter.com:443 does not work at home nether in office

Does every Antminer working unit need an extra name e.g. Biosman_Worker01, Biosman_Worker02 etc or is one name for all Antminer enough e.g. Biosman_Worker?

Can i turn off under Bitminter mining for Namecoins, i think i just want all my power for BTC Mining

What are these settings and do i have to change something?

   

----------------------------------

Thx for your help
Gr€€tz Biosman

Makes sense. But what about the case in which I'm brute forcing a private key. And my machine is the only one working on the problem. At some point in time a million bajillion years from now I would expect to have been able to try all possible combinations. Now assuming there is only one matching private key, it would be possible to find that key on the very last sequence tried. What is the CDF? Would 50% CDF be half that time, in this case?

Thanks for your explanations. Very lucid.

Where someone can learn/read about all these?

0 CDF would mean "it's impossible to find a block with this amount of work". That would be true only with 0 work done. Even with just 1 proof of work accepted by the pool there is a chance it will create a block, so that's not 0% CDF. If you found a block, then the round is not 0 CDF, per definition.

100% CDF would mean "with this amount of work you are absolutely certain to find a block". But that is never true. Unlike a lottery where you would perhaps be able to buy all the tickets, you can't do that in bitcoin.

As time approaches infinity, CDF approaches 100%. But just like time never reaches infinity, CDF can never be 100%.