[~1000 GH/sec] BTC Guild - 0% Fee Pool, LP, SSL, Full Precision, and More

[kjj]

Heh, I wish you had given some notice that you were going to take down the miner1.btcguild.com URL.  All of my stuff was still pointed to it, and I didn't notice it was gone for about 10 hours.

I use the flex proxy, so I didn't lose any work, but I was pretty confused trying to figure out why my stuff wasn't working while the website kept counting up rounds.

Edit: fixed .org -> .com

[eleuthria]

Heh, I wish you had given some notice that you were going to take down the miner1.btcguild.com URL.  All of my stuff was still pointed to it, and I didn't notice it was gone for about 10 hours.

I use the flex proxy, so I didn't lose any work, but I was pretty confused trying to figure out why my stuff wasn't working while the website kept counting up rounds.

Edit: fixed .org -> .com

I am very sorry for this lack of notice on this.  I migrated our DNS servers to DNS Made Easy, and that entry was lost in the shuffle.

[kcobra]

Can anyone give me some advice on getting my stale rate down. Over a 24 hour period, running 13 gpu's on BTC Guild, I am getting a 5.2% stale rate according to the stats on the my account page. With deepbit I was getting around 3% and with slush I was getting around 4%. None of these numbers are good when I see people saying they get less that 1% stale rate.

Any ideas? I am running phoenix using the phatk kernel. Below is my command line:

phoenix.exe -u http://XXXXXXXX:YYYYYYYYY@us.btcguild.com:8332/ -k phatk DEVICE=1 VECTORS BFI_INT FASTLOOP=false WORKSIZE=256 AGGRESSION=13

Thanks.

[eleuthria]

Can anyone give me some advice on getting my stale rate down. Over a 24 hour period, running 13 gpu's on BTC Guild, I am getting a 5.2% stale rate according to the stats on the my account page. With deepbit I was getting around 3% and with slush I was getting around 4%. None of these numbers are good when I see people saying they get less that 1% stale rate.

Any ideas? I am running phoenix using the phatk kernel. Below is my command line:

phoenix.exe -u http://XXXXXXXX:YYYYYYYYY@us.btcguild.com:8332/ -k phatk DEVICE=1 VECTORS BFI_INT FASTLOOP=false WORKSIZE=256 AGGRESSION=13

Thanks.

Toning down the aggression can help.  Additionally, the IRC chat has universally agreed that phoenix's stale rate is completely terrible right now, since it seems to ignore clearing out its work queue after receiving a long poll.


Pool Status Update:
US West is getting bogged down compared to the other servers.  I'm working on shifting some of the frequent MySQL tasks to the slave pools to help it handle the load.  Eventually the frontend will get moved to a separate server so that the pool itself on US West is not significantly affected by website load.

[Clipse]

I would love to see a 24hr average BTC earned under the accounts page.

[eleuthria]

I would love to see a 24hr average BTC earned under the accounts page.

I'm still hoping to get that feature in this weekend.  I was sick most of the weekend, and my attention was focused on the server sync issues with the UK server.  Right now I'm working on getting a lot of the computational/SQL load for the frontend to be distributed to the slave servers, to speed up the My Account/API pages, as well as help the US West server from having idles.

[hchc]

this pool seems to have bad luck as of late If I tabulate the average round shares for the last 2 days. It is about 33% higher than the difficulty of 567358...while on deepbit it about 10%...meaning if i were to mine on deepbit i would have got 20% more...oh well lets lope for the tide to change

[eleuthria]

US Central (Chicago) will hopefully be online tomorrow or the day after.  Also obtaining a separate server to handle the website to get that load off of the US West pool server.

[BitMinerN8]

US Central (Chicago) will hopefully be online tomorrow or the day after.  Also obtaining a separate server to handle the website to get that load off of the US West pool server.

This is great news. Thanks for all your hard work.

[kokojie]

Can anyone give me some advice on getting my stale rate down. Over a 24 hour period, running 13 gpu's on BTC Guild, I am getting a 5.2% stale rate according to the stats on the my account page. With deepbit I was getting around 3% and with slush I was getting around 4%. None of these numbers are good when I see people saying they get less that 1% stale rate.

Any ideas? I am running phoenix using the phatk kernel. Below is my command line:

phoenix.exe -u http://XXXXXXXX:YYYYYYYYY@us.btcguild.com:8332/ -k phatk DEVICE=1 VECTORS BFI_INT FASTLOOP=false WORKSIZE=256 AGGRESSION=13

Thanks.

Toning down the aggression can help.  Additionally, the IRC chat has universally agreed that phoenix's stale rate is completely terrible right now, since it seems to ignore clearing out its work queue after receiving a long poll.


Pool Status Update:
US West is getting bogged down compared to the other servers.  I'm working on shifting some of the frequent MySQL tasks to the slave pools to help it handle the load.  Eventually the frontend will get moved to a separate server so that the pool itself on US West is not significantly affected by website load.

Which miner do you recommend to use?

[Spunky]

Is there currently a reporting problem?  I see this:

9,553 workers at 1,536.46 GH/s
Round Time   3:36:20      Round Shares   4577435

Statistically speaking this is extremely improbable.  Is all the server shuffling creating problems for the website such that it doesn't really know when a block was found?  Or are we really just winning the anti-lottery?

[IlbiStarz]

Is there currently a reporting problem?  I see this:

9,553 workers at 1,536.46 GH/s
Round Time   3:36:20      Round Shares   4577435

Statistically speaking this is extremely improbable.  Is all the server shuffling creating problems for the website such that it doesn't really know when a block was found?  Or are we really just winning the anti-lottery?

No, it's actually not that improbable. Just means the pool is unlucky, other pools are solving the block before we can.

[Fhtagn]

I appreciate all of your hard work, eleuthria. This has been a great pool, despite the growing pains; the growing hash rate reflects that.

I'm increasing my donations from 2.5% to 3%, to help pay for all of those new servers and maybe a few pots of coffee.

[Spunky]

Is there currently a reporting problem?  I see this:

9,553 workers at 1,536.46 GH/s
Round Time   3:36:20      Round Shares   4577435

Statistically speaking this is extremely improbable.  Is all the server shuffling creating problems for the website such that it doesn't really know when a block was found?  Or are we really just winning the anti-lottery?

No, it's actually not that improbable. Just means the pool is unlucky, other pools are solving the block before we can.

As I understand it other pools solving blocks doesn't affect our chances of solving a block at all, it is random chance when we hit one.  Having 1500GH/s just means we buy 1.5 trillion lottery tickets every second.  Sure we start using a new block header for hashes every time anyone solves a block, but that doesn't change the odds.  According to my limited understanding we have a 95% chance of finding a block within 1.5 hours at that hash rate.  A 2 hour round is an anomaly, a 3.5 hour round seems damn near impossible.  Not actually impossible, but so unlikely as to make me wonder if something else is going on.  It is certainly possible I'm misunderstanding things, I am no statistical genius.  Please educate me.

[kjj]

The thing is, and I've said this about 50 times in various threads around these boards, you can't make any claims about a single event.

[Serge]

I guess website is moving to a new home

[Spunky]

The thing is, and I've said this about 50 times in various threads around these boards, you can't make any claims about a single event.

If there is some other thread you can point me to where all of this is already covered that would be appreciated.

I originally just wanted to know if there was something going on with reporting, since there have been a lot of those kinds of problems in the last few days due to new servers coming online, and that round was very unusual.

I am now curious as to the statistics at work here.  Most of my analysis, such as it is, is based on the Bitcoin Generation Calculator.  I am assuming that the formula used there to calculate 50% / Average / 95% times to solve a block is valid, because the calculator does seem to predict fairly well how many BTC I will receive in a given time frame at a particular hash rate and difficulty.  I will admit to being a little confused here already, though, because it seems like the numbers for 50% and Average should be the same.  I clearly don't understand the finer points that lead to this distinction, or why my assertion that a 3.5 hour round @ 1.5TH/s is extremely unlikely using those numbers is wrong.  I am trying to learn the math though, so any links are definitely appreciated.  PM me if you don't want to clutter up this thread any further.  Thanks for your help.

[kjj]

Statistics has nasty dark corners.

If an event has probability X per attempt, your chances of getting the event after Y attempts is not X * Y.  It is 1 - ((1/X)^Y).

Because this is exponentially decreasing, events that happen before the 50% time will come quicker than the events after it are late.  As in, it takes more attempts to get from 50% to 51% than it does from 49% to 50%.

Consider rolling a die.  You have a 1 in 6 chance of getting a 6.  If you want at least a 30% chance of getting a single 6, you only need to roll it twice.  If you want a 50% chance, you need 4 rolls.  If you want at least a 70% chance, you have to roll it 7 times.

Moving the odds from 50% to 30% saves you 2 rolls, but moving it from 50% to 70% requires 3 more.  If you look at the 1 in 6 and 5 in 6 odds, they are even worse, 1 roll vs. 10 (savings of 3 vs. loss of 6).

Now instead of a die with 6 sides, imagine a bitcoin hash with 2.436 * 10^15 sides...

[Serge]

just the heads up
http://btcguild.com/  shows empty page, thought site wasn't working because moving to a new location until i decided to check www.btcguild.com few hours later

[fcmatt]

anyone having rpc connection problem right now? 2 AM eastern time as i post this.
using guiminer, newer version. hostname connected to is btcguild.com

edited to add:
well it might have just been me i guess. the problem is gone. false alarm.