Introducing CherryPicking - new Windows & Linux Pool Hopper

[Bloodred]

Type info and press Enter.

[1714933733]

1714933733

[1714933733]

1714933733

[1714933733]

1714933733

[Bloody Angel]

Cool,

it works!

Last question, how can I set the maximum amount of round shares for a pool?

I want to set, e.g. mining at slush pool while round shares are under 400000, then mining at BTCPool24 while round Shares under 1000000 and else mining at arsbitcoin


Is this possible?


Thnaks a lot!


Bloody

[Bloodred]

No, that's not possible currently. The optimal switch point is at 0.43 * diff, if you changed that it would hurt your efficiency anyway.

[Bloody Angel]

So,

there are some more questions I have after testing and playing with Cherry:

1. When the "0.43 * diff" Level is reached at all pools, is cherry switching to the Backup Pool?

2. How is cherry switching when the Pools have different payment models!? Such as PROP, PPLNS, SCORE

3. When is Cherry Switching to other Pools!? When they have a lower Submitted Shares Rate, or wenn the reach "0.43 * diff" or when the pool have SCORE when it reaches 0.43 * diff, else when the pool have PROP, when the submitted shares are lower?

3. What is with delayed stats (such as bitcoins.lc!?)



Thank for you fast help and congratulations to your Program.


Bye Bloody from Germany

[muyoso]

So,

there are some more questions I have after testing and playing with Cherry:

1. When the "0.43 * diff" Level is reached at all pools, is cherry switching to the Backup Pool?

2. How is cherry switching when the Pools have different payment models!? Such as PROP, PPLNS, SCORE

3. When is Cherry Switching to other Pools!? When they have a lower Submitted Shares Rate, or wenn the reach "0.43 * diff" or when the pool have SCORE when it reaches 0.43 * diff, else when the pool have PROP, when the submitted shares are lower?

3. What is with delayed stats (such as bitcoins.lc!?)



Thank for you fast help and congratulations to your Program.


Bye Bloody from Germany

1.  Yes
2.  PROP and SCORE hop onto a pool when they are at 0*difficulty, ie just found a block.  Prop hops OFF of a pool at .43*difficulty and SCORE hops off a pool around .1*difficulty.
3.  CP switches to the pool with the fewest shares until it reaches the hop off point, and then it will either go to a backup pool if no other pools are available under .43*difficulty shares. 
4.  They delay the stats so that we don't know exactly when a round starts.  This reduced our payout greatly as we can't hit the fastest blocks and we miss out on the most profitable shares, which are the first ones of a block.

[ampirebus]

please fix the polmine update I was making alot of revenue with it, the bot only sees the polmine stats like 4% of the time if that. considers it "lagging or down" when not really lagging or down

[muyoso]

please fix the polmine update I was making alot of revenue with it, the bot only sees the polmine stats like 4% of the time if that. considers it "lagging or down" when not really lagging or down

Yea, Polmine and bitcoinpool are both huge losses.  They both implemented anti-hopper measures, but both should be able to be circumvented.

Polmine doesn't advertise stats any longer, but they do advertise the current round duration, which can be used to know when to hop to the pool, and they do still advertise their hashrate, which can be used to calculate the estimated number of shares to know when to hop off of a pool.

Bitcoinpool advertises all of its users shares for the current round on a single page, and all we would need to do is add them up to know the current share count of the round.   That page is here: http://bitcoinpool.com/index.php?page=1&ipp=All&do=currentround .

[scatterbrain]

please fix the polmine update I was making alot of revenue with it, the bot only sees the polmine stats like 4% of the time if that. considers it "lagging or down" when not really lagging or down

same request but for bitcoinpool. their json data keeps resetting several times within a round so i had to stop hopping there. is it possible to make a cfg file that pulls the data from their main page? i believe all the info you need is at the top of the page

[Bloodred]

I am working on on time-based hopping and it will definitely be available in v0.6.5. By time-based hopping I mean using the round time instead of round shares to hop. This will enable both bitcoinpool and polmine, as they both provide the round duration on their web pages.

There's also a new hopping algorithm that takes pool hash rate into account coming, it should be better than the current simplistic STATIC_FAST.

@scatterbrain I can only see the round duration, are they advertising the current number of shares somewhere and I've missed it?

Thank for you fast help and congratulations to your Program.

Thanks, doing my best to provide a good product! muyoso already beat me to answering your questions.

[ampirebus]

please fix the polmine update I was making alot of revenue with it, the bot only sees the polmine stats like 4% of the time if that. considers it "lagging or down" when not really lagging or down

Yea, Polmine and bitcoinpool are both huge losses.  They both implemented anti-hopper measures, but both should be able to be circumvented.

Polmine doesn't advertise stats any longer, but they do advertise the current round duration, which can be used to know when to hop to the pool, and they do still advertise their hashrate, which can be used to calculate the estimated number of shares to know when to hop off of a pool.

Bitcoinpool advertises all of its users shares for the current round on a single page, and all we would need to do is add them up to know the current share count of the round.   That page is here: http://bitcoinpool.com/index.php?page=1&ipp=All&do=currentround .

thats weird because I got .839 BTC from bitcoinpool.com in the past 3 days and from yesterday i have another .19 unconfirmed waiting

8/10/2011   1.6755
8/11/2011   2.2012
8/12/2011   2.6328
13-Aug   3.63296747
14-Aug   3.95496747 (aggregate) 3.95 BTC/6 days = .65 BTC for ~600 mhash.   Been having problems with my other 2 cards overheating and shutting down but once i iron that out they will be cherrypicking too

I've just donated .1 BTc to your sig address blood red


Uploaded with ImageShack.us
heres a spreadsheet in excel for anyone sitting on the fence

[Folax]

Hi Guys, just joined the CP customer base some days ago and got some great hand-holding from one of you (thank you). Great product although I got something I would like to see come available in a future release:

Continuous hash rate counter (yes INFO gives a snapshot but when tweaking your GPU it's not really sufficient)
I have only read a couple pages in this thread so could be I missed the request already done and the answer to it, in that case I apologize for requesting this again.

I got 2 questions:

Previously I used Phoenix miner with Phatk kernel, my 5870 GPU's were hashing at 420ish mhash/s, when I type INFO now in the CP box I see a hashrate between 750 and 800 (2GPU's) so poclbm does not seems to hash as fast as Phoenix did.
With Phoenix I used: -k phatk VECTORS BFI_INT AGGRESSION=7 WORKSIZE=128 FASTLOOP=false
What would the poclbm/CP equivalent be for these settings?
I am using this now: Args=-v -w128 -f70

My 2nd question: is the Linux version of CP already available? Would like to get my Linux box cherrypicking too :-)

Thanks for this great product and for any answers given.

-F

[muyoso]

thats weird because I got .839 BTC from bitcoinpool.com in the past 3 days and from yesterday i have another .19 unconfirmed waiting

I know, thats why I want so bad to find a solution to their new pool hopping protection.  They just started with it yesterday.  By sheer coincidence no one minded at all because yesterday they were extremely lucky and found like 6 blocks so constantly mining on them was actually a benefit and cherrypicker probably would have done it anyway.  

[Bloodred]

@Folax The continuous hash rate display has already been suggested and I'll try to implement something like that, but now I'm working on time-based hopping which is more important. I'll try to add it, but don't expect it in the next version, 0.6.5.
Your questions:
1. Try something like -f5 (or lower), the higher the number after -f is the less aggressive poclbm is. Also keep in mind that the hash rate displayed is poclbm's own average which takes into account actual share submissions, so depending on your luck it may differ from what Phoenix would display.
2. No, not yet. You can try the same .jar if you want, the feedback could actually be useful since I don't have any Linux distro installed right now (this will change after I implement some of the more immediate requirements). Last time I tried poclbm wouldn't mine, but that was a very early version and may have been doing something else wrong, so early that it wasn't even CherryPicking yet. For all I know it may work, but I don't advertise this because it's not tested and I wouldn't want somebody paying for it and then seeing it doesn't work.

I've just donated .1 BTc to your sig address blood red

Thank you!

[Milkshanks]

@Folax

-f70 will actually give you less performance, try using -f10 or something like that. "Agression" is kinda backwards on poclbm


@Ampirebus

I got roughly the same hashrate as you ( 670ish) and I wish I could be getting as much are you are, guess I'm just unlucky

[ampirebus]

@Folax

-f70 will actually give you less performance, try using -f10 or something like that. "Agression" is kinda backwards on poclbm


@Ampirebus

I got roughly the same hashrate as you ( 670ish) and I wish I could be getting as much are you are, guess I'm just unlucky

are you running completely 24/7 or are you turning your rig off for a few hours? i leave mine running completely 24/7 unless i have to do video rendering projects or am playing a video game, for some reason i can still mine while i play SC2 though. the short fast rounds that last under an hour or last 2 hours are the make or breaks for whether or not your payouts are above normal or not.

[Milkshanks]

are you running completely 24/7 or are you turning your rig off for a few hours? i leave mine running completely 24/7 unless i have to do video rendering projects or am playing a video game, for some reason i can still mine while i play SC2 though. the short fast rounds that last under an hour or last 2 hours are the make or breaks for whether or not your payouts are above normal or not.

Yup, running it 24/7 with -v -w128 with 2 5850s

[wtfman]

hey,

i am the operator of btcserv.net and wanted to let you know that now json stats are available @ http://btcserv.net/json/pool/ .. would be cool if dev team could update/add whatever necessary so the data is fetched from there.

greetings.

[Bloodred]

Thanks for the heads-up!

This .cfg should work:

Code:

;BTCSERV JSON stats
;Home Page: http:/btcserv.net
Type=PROP
Host=btcserv.net
Port=8335
JSON=http://btcserv.net/json/pool/
HashEx1="speed":\d+\.?\d*
HashEx2=\d+\.?\d*
SharesEx1="shares":\d+
SharesEx2=\d+
Div=1e9

This should help reduce traffic from miners polling stats.

[max in montreal]

took rfcpool out of the list, here is the mesage when you go to the site:

rfcpool has been shut down because the owner is a fucking idiot - that
should sum it up in brief.


If you were a miner, your money is all accounted for, and you will be paid as
soon as the most recent block matures. If you did not have a wallet address
in your account then email chris@rfcpool.com and after slinging your abuse
at me, tell me the username/email on account/wallet combination and you will
get paid.

So what happened? Well in short, running a pool and dealing with
people's money is a metric fucktonne harder than I ever imagined it would
be, and a series of fuckups relating to handling all said money has left me
about 150BTC out of pocket, so I'm pulling the plug before the situation
gets any worse.

- an early invalid cost us 30BTC
- the decision to go PPS was crazy and cost us 20BTC
- accidentally paying out double for a block reset in an attempt to turn PPS
  off again cost 100BTC

Good luck to all other pool ops.



---

And a note from Rennex: we've been sustaining losses from PPS and the
invalid block along the way, but this latest mess happened at almost
the worst possible moment - Chris was just about to leave on a trip, so
there was no time to start cleaning it up.

But things aren't all FUBAR, we'll take some time to examine the damage
and correct the errors. Some people may have been paid too much because
of auto payouts, and others' accounts got credited too much. We can fix
the latter and deal with the former - and if miners feel that they want
to return excess payouts once we determine the real balances, there will
be a way to do that too.

But for now, we'll cool off and examine the situation, and see if we
want to continue in the near future. Check back in a while

You can also contact us on IRC: #rfcpool on FreeNode.

Oh and if you don't know which pool to choose for now, I hear that
Ozcoin isn't half bad

[Bloodred]

Too bad about RFCPool, one less prop pool to mine.

Anyway, I've got something to show, from a testing version of v0.6.5:

Code:

* Updating pools
* polmine
* Pool update done
	Type: PROP
	Hash rate: 153.76 GH/s
	Monitoring mode: TIME
	Round time: 41254
	1.8184367533122552
* bitcoinpool
* Pool update done
	Type: PROP
	Hash rate: 178.0 GH/s
	Monitoring mode: TIME
	Round time: 8510
	0.4342485014006932