Topic: [OLD] Eligius: ASIC, no registration, no fee CPPSRB BTC + 105% PPS NMC, 877 # - page 74. (Read 458255 times)

I'm not sure what wiki you mean, but a nice thing about wikis is that anyone can add useful information to them

Things should be pretty much back to normal.  If you're still stuck on work that is too high of a difficulty, please reconnect, although I'm pretty sure I reset everyone correctly.

-wk

Seeing the same as someone above.  Until the 4th Nov my hashrate graph was smooth, between 25GH and 30GH.  Never above, never below.  The past few days it's fluctuating between 20GH and 40GH.  Same hashrate (28GH) at this end.

Thanks! Works great. Maybe something for the wiki?

this pool are crazy

The hash rate on the site shows the effective hash rate in two minute time intervals.  So if you have lucky/unlucky streaks, you will see fluctuation.  The 3 hour and 12 hour has rates reported should match much more closely what your miners are reporting (assuming continuous mining for the prior 3 and 12 hour time frames.

Someone feel free to correct me if I'm wrong, but that's what it looks like to me.

Wow.    I think you haven't got a clue what you're talking about.

Hey, its happening again?

My hashrate is fluctuating ..... see some dips down in hashrate. Check miners and they're working ....

You can make workers by using a user name of _ for each worker instead of just for all of them.

Are you running a server at your house or something? Seeing your own name "~wizkid057.....:" in some of the URL's makes me think something isn't setup right somewhere. Hopefullly you are on a dedicated or a good VPS. Doesn't seem setup right though, no?

Anyone notice the little easter egg in the block statistics?  I got a nice chuckle out of it


"Vegeta! What does the scouter say about his power level?!"

"It's over NINE THOUSAAAAAAND!!!"

hi guys, how do i track miners separately as i see some addresses on the top list doing it? i have tried googling and your docs, with no info!
Would be great to be able to track each worker separately..

when fix all problems?

Thanks for the update.

But, please don't close the getwork service.  I sort of reply on it as backup to my Blades - normally they run on Eligius through BFG's proxy, with backup to Eligius' getwork server.  I'd be a sad panda if the getwork interface closed, as has happened on most other pools.  Nor am I going to run another PC just to have a proxy running, I've got my miners running on OpenWRT, so I can turn off all my PCs, I don't want to have to run a PC 24/7 again.

I might even have to use Eclipse again, it's the only other pool that has a decent getwork interface....

Whow something new is going on....

First my pool dif was "Always" 127 it just jumped to 255 (ok that can happen) then 511 now 1023 or like cgminer likes to report 1.02k

Isn't this to much for my 550Ghash ?

(Don't get me wrong, i like to show more muscles then i have   )

WK "This is what is causing the stats to act completely crazy with hash rates jumping all over the place.  Balances and related stats are correct, however."

My hash rate was varying to the 50% side of the equation, but this sentence comforts me!

Hi everyone.

Looks like some pool server side load issues have caused some oddities with the shares database being delayed a bit.  The load balancing I had in place for stratum was not properly distributing the load between servers and eloipool instances and instead was maxing out just a handful of them on one server.  This is due to the way I originally designed the load balancing to be compatible with getwork mining, but, is not really imperative anymore. I'm working on this now.

This is what is causing the stats to act completely crazy with hash rates jumping all over the place.  Balances and related stats are correct, however.

I temporarily halved the target shares per minute for variable difficulty to reduce database load on this server until the load balances out.  Right now it is targeting 16 shares per minute instead of the normal 32 shares per minute like it normally does.  This is why some people are seeing higher work diff than normal.  I will set this back to normal as soon as I get the load balancing stuff straightened out, which shouldn't be long.

The stats are another issue, since this delay actually delays the timestamping of the shares, so, I won't be able to go back and correct the errors in the hash rate graphs and such, unfortunately.

--------

Now, as for people complaining that their balances are incorrect or less than the should be, I advise you to read up on CPPSRB.  CPPSRB is NOT proportional or PPLNS.  The basics are, you don't just get a set % of every block the pool finds.  That is a broken reward system open to exploits.  
The funds under CPPSRB are paid to the share log.  If all of yours are paid (ie: you have been paid 100% PPS for all of your work) then the pool pays other miners who have not yet been paid.  You can not get paid more than the work you put in.  Every share is worth, currently, 25/network_difficulty.  No more, no less.  When all of yours are paid, you are paid.  Usually, you will have some shelved shares in the share log.  When that happens, at the beginning of each round your estimate will "jump" up to what appears to be a proportional amount.  But it can never give you more than you have in the share log.  If you're paid up completely (shares rewarded=100%) then you are paid.  You only get what you put in.

Capped Pay-per-share with recent backpay (CPPSRB)

-wk

this is weird. stats showing wrong I guess. getting like 120GH which is way more than I'm capable of.

something is wrong with this pool
many things change

Yeah something is funky -- right now vardiff seems to think diff 512 makes sense for a miner running 200gh...