[4+ EH] Slush Pool (slushpool.com); Overt AsicBoost; World First Mining Pool

[DigitalDoom]

So....what's going on? I started up my miners after noticing they had shut down for some unknown reason.  

At first, they weren't registering any work at all. Now the site says down for maintenance and to come here for info.

Well....I'm here! Where's the info?  


EDIT:

NM....it's all up and running now. No idea what happened there, but all is well now 

[johnyj]

In long run this does not evens out, my ISP's network is not that stable, I often get that kind of network blackout for a while. So, for most of the blocks that took a long time to solve, I will get poor payment. Since this pool is not big enough to shorten the block generation time, they often have blocks that took several hours to solve

And I never really understand this pool hopping thing.

Suppose the pool has generated several blocks in a very short time frame, is it the time to switch to another pool due to I expect the pool will become unlucky in the next a few of hours?

Suppose the current block has not been found for 2 hours, that means all the other pools also have not found a solution for 2 hours, how could I make my chance of finding a block bigger by just switch to another pool?

Isn't this the same as forecasting which side of the coin is up after tossing a coin?

[organofcorti]

[.....] And I never really understand this pool hopping thing.

Suppose the pool has generated several blocks in a very short time frame, is it the time to switch to another pool due to I expect the pool will become unlucky in the next a few of hours?

No.

Suppose the current block has not been found for 2 hours, that means all the other pools also have not found a solution for 2 hours, how could I make my chance of finding a block bigger by just switch to another pool?

Still no.

Isn't this the same as forecasting which side of the coin is up after tossing a coin?

Yes. Pretty much, anyway. If you want to understand how pool hopping affects your earnings, try:

http://organofcorti.blogspot.com/2012/04/41-slushs-pool.html
http://organofcorti.blogspot.com/2012/08/42-slushs-score-method-and-miner.html
http://organofcorti.blogspot.com/2012/09/43-slushs-score-method-and-miner.html

[solitude]

I'm still getting the "No connections could be made because the target machine actively refused it" error in guiminer

It's only when I use it through my vpn, which I consider essential for using the internet.  Why is my VPN fucking this up?

I downloaded the latest guiminer before I made this post and am still getting the error.

Slush? You around bro?  Any idea how I can fix this?

I'm pretty sure it starts hashing automatically after a few minutes of getting the error.  I'll report back.

Add this in the server.ini of guiminer.

restart guiminer, Try the alt server. All your setting should be there still.

Code:

        {
            "balance_host": "mining.bitcoin.cz", 
            "balance_token_url": "http://mining.bitcoin.cz/accounts/token-manage/", 
            "balance_url": "/accounts/profile/json/%s", 
            "host": "stratum.bitcoin.cz", 
            "name": "slush's pool(IPv6/alt)", 
            "port": 3333, 
            "url": "http://mining.bitcoin.cz"
        }, 

Thanks so much!  Worked first try, hopefully no errors in the future.

The problem is back.  Same vpn server, really pissed.

[johnyj]

[.....] And I never really understand this pool hopping thing.

Suppose the pool has generated several blocks in a very short time frame, is it the time to switch to another pool due to I expect the pool will become unlucky in the next a few of hours?

No.

Suppose the current block has not been found for 2 hours, that means all the other pools also have not found a solution for 2 hours, how could I make my chance of finding a block bigger by just switch to another pool?

Still no.

Isn't this the same as forecasting which side of the coin is up after tossing a coin?

Yes. Pretty much, anyway. If you want to understand how pool hopping affects your earnings, try:

http://organofcorti.blogspot.com/2012/04/41-slushs-pool.html
http://organofcorti.blogspot.com/2012/08/42-slushs-score-method-and-miner.html
http://organofcorti.blogspot.com/2012/09/43-slushs-score-method-and-miner.html

I understand this pool setup something to prevent pool hopping, but if pool hopping itself does not provide any benefit for miners, why should miners do pool hopping at the first place? And if they seldom do this, why trying to prevent this?

[organofcorti]

[.....] And I never really understand this pool hopping thing.

Suppose the pool has generated several blocks in a very short time frame, is it the time to switch to another pool due to I expect the pool will become unlucky in the next a few of hours?

No.

Suppose the current block has not been found for 2 hours, that means all the other pools also have not found a solution for 2 hours, how could I make my chance of finding a block bigger by just switch to another pool?

Still no.

Isn't this the same as forecasting which side of the coin is up after tossing a coin?

Yes. Pretty much, anyway. If you want to understand how pool hopping affects your earnings, try:

http://organofcorti.blogspot.com/2012/04/41-slushs-pool.html
http://organofcorti.blogspot.com/2012/08/42-slushs-score-method-and-miner.html
http://organofcorti.blogspot.com/2012/09/43-slushs-score-method-and-miner.html

I understand this pool setup something to prevent pool hopping, ...

Again, nope. The score method reduces the effect of pool hopping significantly, but does not prevent it.

..... but if pool hopping itself does not provide any benefit for miners, why should miners do pool hopping at the first place?

If you're engaged in pool hopping, you'll earn more per share than full time miners - last I checked this figure was around 20% more, but it's probably a bit less now.

And if they seldom do this, why trying to prevent this?

If pool hoppers earn more than fulltime miners, then fulltime miners are earning less than expected - by about two percent when I checked last year. If you prevent pool hopping then everyone can earn the same average payment per share. PPLNS, DGM, PPS and PPS-like methods are fair in that over time you earn the expected amount per share submitted. Proportional methods will on average pay more to strategic pool hoppers than to fulltime miners.

If you want a more in depth explanation I encourage you to read the posts to which I directed you.

[The Mafia]

FINALLY I can reply

I have a question and a suggestion.

1. Is it better to have say, 5 separate workers, 500mh/s each, on difficulty 1, or make them all one worker at 2500mh/s and make them difficulty 2?

2. Also, Slush, this one is for you. Would it be hard to implement a feature where it emails you when a worker is idle for more than xxx mins?

Thanks, and thanks for the great pool

[slush]

I have been on some conferences and I also took a short break in recent days; that's one reason why I didn't respond to many questions here, on FB page and from email.

The second reason is that it began impossible to respond to everybody, because in recent days I had a storm of emails of any kind. I even have many phone calls a day "where can I buy ASIC and is it worth the money?". Once I realized that I spend over three hours a day just by responding to common questions, I decided to stop answering such questions at all. Maybe it sounds impolite, but pool now has over 160.000 registered people and answering their questions distracts me from any real work. In contrary to situation a year or two ago, now there're many manuals, howtos and FAQs about mining, ASIC etc so I don't think answering everything personally is needed anymore.

I'm writing regards to recent issues on pool. You may notice that there're some consistency issues on Stats page, e.g. rounds 17370/17371, website also sometimes hangs with "in maintenance" message. The reason is overloaded database; just in last month, the traffic on website quadrupled, much more people started using some monitoring tools polling the API from their widgets, others are refreshing profile page again and again; the traffic on website is just crazy. This *doesn't* affect mining, because mining backends are completely asynchronous - shares are queued and registered to the database lazily, which also explains these inconsistences in statistics page for very short rounds.

On friday I ordered new database machine which will take such load, which was prepared yesterday. I'm finishing the configuration and at today's night I'll migrate database there. There'll be just a ~15 minute downtime of website and few-second downtime of mining (backend restarts). I believe that it fixes all website issue.

[Newar]

On a side note, how come Namecoin rewards are no longer being awarded on a regular basis.  I am lucky if I get namecoins 1 out of every 10 blocks

https://bitcointalk.org/index.php?topic=1976.msg1764838;topicseen#msg1764838

[slush]

Also thanks to everybody who helps me answering these newbie questions here on forum, on IRC and FB!

[qbits]

Hi Slush,

I've been mining on your pool for about 6 months now and it seems it is time to quit. Not that there is anything wrong with the pool, on the contrary it is great and I still have one worker (about 1gh/s) left on the pool, but the pool hash rate and difficulty are slowly eroding the motivation to do so. the only way that keeps motivation up the the BTC price

Most notably for some rounds like:

17371   2013-04-09 07:57:02   0:00:02   2486   none   none   none   230426   25.32750321    87 confirmations left
and
17365   2013-04-09 04:45:16   0:00:10   3559   none   none   none   230396   25.14230000    57 confirmations left

rounds were so short that my gh/s worker running on difficulty 1, did not manage to submit a single share, therefore my earning is '0'. I know statistically in the long term I'm still  earning based on what my hash rate is, however emotionally I'm getting the feeling all those earnings will eventually drop to '0'.

In your recent post you stated that "LTC brings nothing new to the table", well, I'm not so sure about that. In the world of ASICs, GPU miner has no place. Opening LTC pool would allow us to continue mining with bitcon.cz...

[JimiQ84]

Hi Slush,

I've been mining on your pool for about 6 months now and it seems it is time to quit. Not that there is anything wrong with the pool, on the contrary it is great and I still have one worker (about 1gh/s) left on the pool, but the pool hash rate and difficulty are slowly eroding the motivation to do so. the only way that keeps motivation up the the BTC price

Most notably for some rounds like:

17371   2013-04-09 07:57:02   0:00:02   2486   none   none   none   230426   25.32750321    87 confirmations left
and
17365   2013-04-09 04:45:16   0:00:10   3559   none   none   none   230396   25.14230000    57 confirmations left

rounds were so short that my gh/s worker running on difficulty 1, did not manage to submit a single share, therefore my earning is '0'. I know statistically in the long term I'm still  earning based on what my hash rate is, however emotionally I'm getting the feeling all those earnings will eventually drop to '0'.

In your recent post you stated that "LTC brings nothing new to the table", well, I'm not so sure about that. In the world of ASICs, GPU miner has no place. Opening LTC pool would allow us to continue mining with bitcon.cz...

this is problem with Slush's anti-hopping feature. PPLNS is much better in this case.

[scouzi]

Hi Slush,

I've been mining on your pool for about 6 months now and it seems it is time to quit. Not that there is anything wrong with the pool, on the contrary it is great and I still have one worker (about 1gh/s) left on the pool, but the pool hash rate and difficulty are slowly eroding the motivation to do so. the only way that keeps motivation up the the BTC price

Most notably for some rounds like:

17371   2013-04-09 07:57:02   0:00:02   2486   none   none   none   230426   25.32750321    87 confirmations left
and
17365   2013-04-09 04:45:16   0:00:10   3559   none   none   none   230396   25.14230000    57 confirmations left

rounds were so short that my gh/s worker running on difficulty 1, did not manage to submit a single share, therefore my earning is '0'. I know statistically in the long term I'm still  earning based on what my hash rate is, however emotionally I'm getting the feeling all those earnings will eventually drop to '0'.

In your recent post you stated that "LTC brings nothing new to the table", well, I'm not so sure about that. In the world of ASICs, GPU miner has no place. Opening LTC pool would allow us to continue mining with bitcon.cz...

this is problem with Slush's anti-hopping feature. PPLNS is much better in this case.

Its not a function of anything. The round lasted a whole 2 secs. I have 2 GPUS (7970 and 6670) and none found anything in that interval. Meanwhile, we just got a block after a 4 hour dry spell.

[johnyj]

If pool hoppers earn more than fulltime miners, then fulltime miners are earning less than expected - by about two percent when I checked last year. If you prevent pool hopping then everyone can earn the same average payment per share. PPLNS, DGM, PPS and PPS-like methods are fair in that over time you earn the expected amount per share submitted. Proportional methods will on average pay more to strategic pool hoppers than to fulltime miners.

If you want a more in depth explanation I encourage you to read the posts to which I directed you.

You said there is no gain from thereotically to do pool hopping, and you said again there is a 20% gain, which one is true?

I don't want to dig into this kind of complexity, I never had this kind of problem with other pools, thought this pool is PPLNS based, it isn't?

[scouzi]

If pool hoppers earn more than fulltime miners, then fulltime miners are earning less than expected - by about two percent when I checked last year. If you prevent pool hopping then everyone can earn the same average payment per share. PPLNS, DGM, PPS and PPS-like methods are fair in that over time you earn the expected amount per share submitted. Proportional methods will on average pay more to strategic pool hoppers than to fulltime miners.

If you want a more in depth explanation I encourage you to read the posts to which I directed you.

You said there is no gain from thereotically to do pool hopping, and you said again there is a 20% gain, which one is true?

I don't want to dig into this kind of complexity, I never had this kind of problem with other pools, thought this pool is PPLNS based, it isn't?

To me it seems that Slush's method penalizes pool hopper leaving it but incentivizes late round pool hopper from other pools joining. Am I missing something here?

[organofcorti]

If pool hoppers earn more than fulltime miners, then fulltime miners are earning less than expected - by about two percent when I checked last year. If you prevent pool hopping then everyone can earn the same average payment per share. PPLNS, DGM, PPS and PPS-like methods are fair in that over time you earn the expected amount per share submitted. Proportional methods will on average pay more to strategic pool hoppers than to fulltime miners.

If you want a more in depth explanation I encourage you to read the posts to which I directed you.

You said there is no gain from thereotically to do pool hopping, and you said again there is a 20% gain, which one is true?

No, I didn't - perhaps you misunderstood me. Again, read these posts:  http://organofcorti.blogspot.com.au/search?q=MAREK

I don't want to dig into this kind of complexity,

This is fun and very interesting. Slush's score method gave me months of interesting research last year. Try digging into the complexity - you might enjoy it

I never had this kind of problem with other pools, thought this pool is PPLNS based, it isn't?

No, the pool has it's own special scoring method. It's not yet PPLNS, although Slush has been thinking about changing it for a while. The method is outlined here: https://bitcointalk.org/index.php?topic=1976.msg50002#msg50002


I think you're over-thinking all this. Do you want to stay at slush's pool? Are you ok with losing a tiny percentage of your earnings to pool hoppers in exchange for a stable pool and a responsible pool op with a good reputation who has the longest running top 5 pool? You don't need to be concerned with the technical aspects of pool hopping if you don't want to.

[Quantus]

I'm back with more dumb questions. Is it better to have 5 shitty GPU's under one Worker name and pass? Or to give each Worker its own gpu.

my workers scores are

132.1497
186.6425
258.3901
66.6967

Where dose this number get plunged in? What do these numbers stand for and how do they get used?

[slush]

Judging because of extremely rare 2/10 second rounds? People are upset when the luck is bad,  but people are also upset when the luck is good... :-) I don't remember how such "good luck" happen last time, it is really rare.

[Lucko]

Well none would ba upset if he/she got a share in. I didn't but one's who did are smiling...

EDIT:

Did you consider adding NMC to Stratum? Or is this imposible?

[Richy_T]

I'm writing regards to recent issues on pool. You may notice that there're some consistency issues on Stats page, e.g. rounds 17370/17371, website also sometimes hangs with "in maintenance" message. The reason is overloaded database; just in last month, the traffic on website quadrupled, much more people started using some monitoring tools polling the API from their widgets, others are refreshing profile page again and again; the traffic on website is just crazy. This *doesn't* affect mining, because mining backends are completely asynchronous - shares are queued and registered to the database lazily, which also explains these inconsistences in statistics page for very short rounds.

I'd suggest you may want a separate machine for querying statistics which caches them locally and doesn't query the database each time.