Topic: [4+ EH] Slush Pool (slushpool.com); Overt AsicBoost; World First Mining Pool - page 846. (Read 4382675 times)

You might as well forget getwork. You were planning to close it down anyways.

Stratum and getwork servers crashed, looks like issue in datacenter, but I have no more information yet. Stratum looks recovered now, I'm working on Getwork.

i have problems too 

getting a lot of connect/disconnects from pool.  Is everything ok?

Is there a problem with the pool as I am unable to connect. I tried using both addresses on the site. Using GUIMiner, Win 7.

I'm pretty sure it has more to do with the time between shares, as opposed to where the share came in the round.  But it sounds like you have it figured out now...

Yap. You are right. Your screenshot shows the following:
The speed of your miner is not sufficient to difficulty 128, so you miss a lot of blocks providing 0 shares or providing 1 share(worth 128 shares of difficulty 1) at the very beginning of block calculation. Then this 1 share continuously goes to 0 as more and more new shares comes into round. It's the same as you hop to another pool after submitting first share.

You will be OK with difficulty 1. Set higher difficulties only if you experience problems with your connection to Internet.

Interesting, so shares actually drop to 0 value if the round goes long enough? I realized they were weighted toward later submissions but not to the point at which early shares would be worth 0. I was expecting a logarithmic curve.

Between this and what the other poster suggested (that my difficulty is set too high) I think I may have pieced together the cause (yours) and the fix (his). Testing now.

I don't think that could not be the issue here. If I were completing shares but not submiting them, why would the stats page be showing that I have 384 shares. Even if the page only displays shares given instead of shares received, it would make no sense that I would pick up multiple sets of work without submitting the first set.

I have dropped my suggested difficulty from 64 to 1 though to test this.

It's obvious that you put Difficulty setting on pool's web-site to 128. That means that your miners will not send ANYTHING unless they collect at least 128 shares. Only after that they will send them to the pool to request some reward.
To resolve your issue - decrease difficulty setting to 1.

P.S. Yes, I know, that actually difficulty works in some other way, but it's the most easy way to explain its meaning and why its better to put 1 on slow miners.

Why isnt cgminer working with:

"pools" : [
   {
      "url" : "stratum+tcp://ipv6.stratum.bitcoin.cz:3333",

A ping resolves fine, my ipv6 is fine.

I had that yesterday - a reward of none for a few hours but it git fixed after.

That's probably happening because the last share submitted was during the beginning of that round, and there weren't any submitted toward the end of the round.

For example.  If I am mining with a nice video card and cranking out 1 share per second for the first hour of a round, then the power goes out and the round lasts for 3 hours, my reward will drop to 0, or almost zero because the shares are weighted to discourage pool hopping.  The shares that come at the end of the round have more weight than the shares that come at the beginning of a round.

I'm currently looking into some unusual results from slush's pool payouts. Most blocks on which I complete work (shares) receive the appropriate payouts however some display shares completed but no payout amount.
A screenshot of this issue can be found at https://docs.google.com/file/d/0B_G2Isi7ITVMR25OcnRBYXkwams/edit?usp=sharing .
worker name: brinebold.worker1
uasername: brinebold

Can anyone explain this behavior. Please don't just blindly guess though. If you can't test your theory or provide a definitive answer, please move along.

I don't think these are stale shares because the odds of every single one of hundreds of shares being stale in the same block and this issue occurring on multiple blocks seem incredibly unlikely.
I don't believe this is a case where no work was successfully submitted because I was apparently awarded shares by the pool.
I don't think this is a case of simply being rounded down to 0 because I don't meet the minimum for a payout because the division works out to more than other blocks for which I have received a payout.
I don't think there is a problem with that blockchain, for example, it was rejected by the bitcoin swarm, because other people received a payout for one of these blocks.
These aren't simply unconfirmed blocks because there are blocks that are confirmed that also have shares but 0 payout.

However, if you can find some way to confirm any of those theories, I'd be glad to try it.

Currently the getwork interface has ~10x higher rejection rate than Stratum, because of scaling issues in getwork architecture; and I'm not going to waste a time to fix it.

There were many "stale" shares because of today's morning crash of the server.

aha. thanks for this info.
I am using Guiminer and I can see only total shares all. Only if i try to reset guiminer every time the block will be solved but..... waste of time.

You also get paid better with stratum.  Currently the getwork doesnt get any share of the fees, and pays a higher percentage overall.  That would make up for your stales.

You also might want to check the shares on your miner vs the shares actually getting counted by slush.  I've heard of miners counting shares as stale when the pool actually counted them as legit.

statrum - this means less stale share wright? But I am getting more stale share on statrum than over getwork.
in 24h I have together ~50 stale, before I had 1 maybe 2 stale

I am using a bamt machine in conjunction with a separate machine running the proxy.  I have the bamt machine running two video cards.  And a windows 7 machine with none, but with the proxy running.

The bamt machine has to be pointed at the ip address of the proxy machine instead of the ip address of the pool.  And make sure the port is correct too.  That's all that matters.  You can have multiple user names all going through the same proxy.  You can have multiple bamt machines all going through the same proxy.

HIt me back if you can't make it work.  I've worked with the proxy quite a bit...

+1