[4+ EH] Slush Pool (slushpool.com); Overt AsicBoost; World First Mining Pool - page 1088.

dacoinminster

legendary

Activity: 1260

Merit: 1031

Rational Exuberance

Quote from: dishwara on March 17, 2011, 02:07:29 PM

I see a bug or problem using ur script on google chrome.
It doesn't save the check boxes. save given by slush has no use.
Every time i closes google chrome completely & open chrome, every thing is back to NO SCRIPT page, then have to change Time offset, & check the check boxes needed again.
Please make it save the changes. I used google chrome 10.0.648.134 version.

Hi dishwara,

I was able to reproduce this problem, and I believe I fixed it. Please install my latest version of the script, then refresh the mining page and then change a setting. From then on, your settings should be saved between browser sessions.

If this script is making your life easier, please consider a donation: 19hMEAaRMbEhfSkeU4GT8mgSuyR4t4M6TH

Thanks!

dishwara

legendary

Activity: 1855

Merit: 1016

Quote from: dacoinminster on March 17, 2011, 10:44:45 AM

Hey Slush,

Have you tried my UI improvements greasemonkey script? (http://userscripts.org/scripts/show/98553)

Would you be willing to link to it from mining.bitcoin.cz and/or the top post of this thread? Obviously you would want to say that it is a 3rd-party script which could be changed by me to something evil at any time (not that I would do that).

Another possibility is you could host an approved version of the script, or link to an approved version in the version history on userscripts.org.

I assume that some people on this thread have reviewed the code and can vouch that it is not evil (can anyone back me up on this?).

My script also puts your bitcoin donation address (and mine) at the bottom of the workers page.

I see a bug or problem using ur script on google chrome.
It doesn't save the check boxes. save given by slush has no use.
Every time i closes google chrome completely & open chrome, every thing is back to NO SCRIPT page, then have to change Time offset, & check the check boxes needed again.
Please make it save the changes. I used google chrome 10.0.648.134 version.

slush

legendary

Activity: 1386

Merit: 1097

Quote from: dacoinminster on March 17, 2011, 10:44:45 AM

Have you tried my UI improvements greasemonkey script? (http://userscripts.org/scripts/show/98553)

Not yet, but I read the code right now. Looks like a lot of work

. I'll add link to the first post and I'll consider to adopt some info you've tweaked into main GUI. Mainly switch the "last share at" to "last share before", which I consider as really useful. But there are lot of other feature requests, so I cannot promise it to some date

.

demonofelru

full member

Activity: 238

Merit: 100

Quote from: slush on March 17, 2011, 01:26:58 PM

Quote from: demonofelru on March 17, 2011, 10:49:44 AM

Hey Slush sent you a PM about a possible bug in the pool.

I know, no need to poke me here

.

From my investigation the sending part of the payment is definitely not this pool. You probably used the same wallet somewhere else Wink

. I checked my bitcoind logs, database logs and blockexplorer details and it is absolutely clear

.

Yeah it's a bug in my fat fingers it seems I pasted wrong address sorry for the bother

slush

legendary

Activity: 1386

Merit: 1097

Quote from: demonofelru on March 17, 2011, 10:49:44 AM

Hey Slush sent you a PM about a possible bug in the pool.

I know, no need to poke me here

.

From my investigation the sending part of the payment is definitely not this pool. You probably used the same wallet somewhere else Wink

. I checked my bitcoind logs, database logs and blockexplorer details and it is absolutely clear

.

bitjet

hero member

Activity: 696

Merit: 500

Quote from: demonofelru on March 17, 2011, 10:49:44 AM

Hey Slush sent you a PM about a possible bug in the pool.

He should really replace his filter and use the skimmer daily.

demonofelru

full member

Activity: 238

Merit: 100

Hey Slush sent you a PM about a possible bug in the pool.

dacoinminster

legendary

Activity: 1260

Merit: 1031

Rational Exuberance

Hey Slush,

Have you tried my UI improvements greasemonkey script? (http://userscripts.org/scripts/show/98553)

Would you be willing to link to it from mining.bitcoin.cz and/or the top post of this thread? Obviously you would want to say that it is a 3rd-party script which could be changed by me to something evil at any time (not that I would do that).

Another possibility is you could host an approved version of the script, or link to an approved version in the version history on userscripts.org.

I assume that some people on this thread have reviewed the code and can vouch that it is not evil (can anyone back me up on this?).

My script also puts your bitcoin donation address (and mine) at the bottom of the workers page.

Grinder

legendary

Activity: 1284

Merit: 1001

Quote from: marcus_of_augustus on March 17, 2011, 05:55:52 AM

It appears to me to be already happening, have you not noticed a surge in hash power in your pool after the mean time has been exceeded for a block?

If it's happening, it's the participants in the other pools that are being cheated. They will end up sharing some of the profits for the long blocks, and all of the profits for the short ones. For those who use Slush's pool all the time it will mean that they will get a lower share for the long blocks, but they will also take shorter time to solve, so it averages out.

Grinder

legendary

Activity: 1284

Merit: 1001

Quote from: marcus_of_augustus on March 17, 2011, 05:48:44 AM

Indulge me. How is it that "most" of the rounds will last shorter than the mean?
Aren't half the rounds shorter than the mean and half the rounds longer than the mean? Isn't that what the mean is in a normal distribution?

If you want to see the exact statistics you can try putting any number into the mining calculator. http://www.alloscomp.com/bitcoin/calculator.php. It will show you that the average time to solve blocks is higher than it takes to solve the quickest 50% of the blocks. The problem is that you don't know in advance which blocks will end up within those 50%.

The distribution isn't normal, there is a graph at the pool which shows it: http://mining.bitcoin.cz/stats/graphs/

As you can se, it's kind of open ended. Your suggestion for fooling the pool requires there to be an upper limit to how low it can take to solve a block, and a normal distribution within this range. That's not how it works. In theory there can be a block that will never be solved.

marcus_of_augustus

legendary

Activity: 3920

Merit: 2349

Eadem mutata resurgo

Quote

The point is that nobody knows when the round finish. You'll probably miss many rounds by connecting after the half of expected round duration. You can do this, you'll have +/- same payouts for those blocks, but much less often, so it is not a vulnerability.

Yes, but now there are other pools the miner can work on equal shares payment (or solo) when not working on late round shares on your pool, so in aggregate you'll be ahead. It appears to me to be already happening, have you not noticed a surge in hash power in your pool after the mean time has been exceeded for a block?

marcus_of_augustus

legendary

Activity: 3920

Merit: 2349

Eadem mutata resurgo

Quote

No, it won't. Your mistake is to not understand how the randomness of mining works. If you do it your way, your average earnings for the time spent mining will be pretty much exactly the same as if you joined every round from the start. Hint: most rounds will last shorter than average.

Indulge me. How is it that "most" of the rounds will last shorter than the mean?

Aren't half the rounds shorter than the mean and half the rounds longer than the mean? Isn't that what the mean is in a normal distribution?

slush

legendary

Activity: 1386

Merit: 1097

Quote from: marcus_of_augustus on March 17, 2011, 03:50:51 AM

Which actually is part of my point. You get paid more for the second half of a round shares

The point is that nobody knows when the round finish. You'll probably miss many rounds by connecting after the half of expected round duration. You can do this, you'll have +/- same payouts for those blocks, but much less often, so it is not a vulnerability. There are some graphs and analysis done in this thread before. It shows that connecting after some time is not worth of the effort.

slush

legendary

Activity: 1386

Merit: 1097

Quote from: kindle on March 17, 2011, 02:11:53 AM

Slush is some thing wrong with the server?
Active workers (at least one share in last hour) = 0

It's only minor bug on stats page. Number goes to zero when round is longer than one hour, but there are constantly over 1000 active workers now.

Quote

Anyway I noticed that 113846 seems to be the last block reflected in blockexplorer.

No, there is just some delay before block is shown in blockexplorer. Now you can see all the blocks in BE too.

Grinder

legendary

Activity: 1284

Merit: 1001

Quote from: marcus_of_augustus on March 17, 2011, 03:50:51 AM

How do you know it when the second half of the round starts you ask? Well you just join rounds that have gone longer than the average length for slush's pool ... so at present difficulty joining rounds at 50 mins in is going to earn more than mining the whole block, no?

No, it won't. Your mistake is to not understand how the randomness of mining works. If you do it your way, your average earnings for the time spent mining will be pretty much exactly the same as if you joined every round from the start. Hint: most rounds will last shorter than average.

marcus_of_augustus

legendary

Activity: 3920

Merit: 2349

Eadem mutata resurgo

Quote from: Dude65535 on March 17, 2011, 01:40:16 AM

A round that has been running for a long time has no better odds of finding a block in the given time period than a new round would.

Which actually is part of my point. You get paid more for the second half of a round shares even though they have exactly the same chance of finding the block. So on law of averages if you only mine the second half of the rounds you will get paid more.

How do you know it when the second half of the round starts you ask? Well you just join rounds that have gone longer than the average length for slush's pool ... so at present difficulty joining rounds at 50 mins in is going to earn more than mining the whole block, no?

kindle

member

Activity: 84

Merit: 10

Slush the server seems fine now. And yay us for solving block 113846.
Anyway I noticed that 113846 seems to be the last block reflected in blockexplorer. Does that mean that the blocks generation has stopped? I read some where that the network tries to generate 6 blocks/hr. So have the blocks for the hour maxed out?

Sorry for the rookie question, I am just puzzled and have yet to stumbled across the answer to that question after digging through the web.

Cheers!

kindle

member

Activity: 84

Merit: 10

Slush is some thing wrong with the server?

Active workers (at least one share in last hour) = 0

Dude65535

full member

Activity: 126

Merit: 101

A round that has been running for a long time has no better odds of finding a block in the given time period than a new round would.

marcus_of_augustus

legendary

Activity: 3920

Merit: 2349

Eadem mutata resurgo

Slush, now that your pool shares are heavily weighted to those found late in the round what is to stop miners cheating by joining late in the round to cash in on all the easy money then?

By this, I mean that they only join your pool when a round has already gone on for longer than the average time for your pool to find a block. So at present they join your pool after ~50 mins into the round, and at other times go solo or with one of the "equal" share pool.

Topic: [4+ EH] Slush Pool (slushpool.com); Overt AsicBoost; World First Mining Pool - page 1088. (Read 4382648 times)