Pages:
Author

Topic: bitHopper: Python Pool Hopper Proxy - page 62. (Read 355813 times)

bb
member
Activity: 84
Merit: 10
August 07, 2011, 08:24:19 PM
Yeah, I was really surprised that bitclockers worked at all after all the trouble, but now it's lagging out...
sr. member
Activity: 476
Merit: 250
moOo
August 07, 2011, 08:13:28 PM
fun signing up at itzod.. but thanks i got it working.

would be nice if on bitclockers it was just hardware issues.. be real nice.

cool c00w.. as so far I havent sent a share to them yet... so does startLP work with alt slicer or default or both?

still not sure how mainframe is lyign but they are lying.. i guess they wait to find the next block and then split off the shares of the one they found. either way i guess it dont matter they are still liars. Smiley
full member
Activity: 196
Merit: 100
August 07, 2011, 08:08:21 PM
I think I fixed LP based hopping of deepbit. At the bare minimum I fixed a bunch of LP bugs. Um, it only works with the default scheduler for now. And it mines for the entire period of a block. So I'm not sure if its that great of a system.
hero member
Activity: 798
Merit: 1000
August 07, 2011, 07:53:34 PM
So, on the Bitclockers fiasco, this was just posted:

"you can thank all the cpuminers. should be fixed by Wednesday when new servers get in place"

https://bitcointalksearch.org/topic/m.438806

legendary
Activity: 3108
Merit: 1359
member
Activity: 78
Merit: 10
August 07, 2011, 07:31:34 PM
Experimented a bit with "mine" and "mine_slush" setting for Slush pool:

setting: "mine"  -         7197    2011-08-07 13:55:57    0:28:23    756490    0.03079831    139980    99 confirmations left
setting: "mine"  -         7173    2011-08-06 17:54:24    0:36:02    973207    0.07741029    139872    confirmed
setting: "mine"  -         7172    2011-08-06 17:18:22    0:29:47    800452    0.07217588    139869    confirmed

setting "mine_slush" -   7193    2011-08-07 08:58:18    0:39:06    1027781    0.00533275    139950    69 confirmations left
setting "mine_slush" -   7189    2011-08-07 07:59:48    0:29:45    791033    0.00554819    139938    57 confirmations left
setting "mine_slush" -   7181    2011-08-07 03:21:55    0:32:52    856621    0.00898394    139919    38 confirmations left

If I see those scores after testing, why use "mine_slush"  Huh

Detail: I always have 1 card @ 120Mhash/s pointed to the same worker @ Slush.



Now do the same tests with rounds @ +3 mill shares, and see how you will make make next to nothing (or possibly exactly nothing), on both "mine" and "mine_slush". ie you will have "wasted" 4 times as many shares on "mine".
hero member
Activity: 798
Merit: 1000
August 07, 2011, 07:14:20 PM
Um, I didn't write altslicer. But the guy who wrote it doesn't appear to be maintaining it. So I'll take a look and try and fix it ignoring pools. The default slicer reslices properly.

Aaah, I see, thanks c00w.   I'll switch back for now.
full member
Activity: 196
Merit: 100
August 07, 2011, 07:13:33 PM
Um, I didn't write altslicer. But the guy who wrote it doesn't appear to be maintaining it. So I'll take a look and try and fix it ignoring pools. The default slicer reslices properly.

EDIT: Oh and I'm debugging the whole deepbit hopping not working. Stay tuned.
hero member
Activity: 798
Merit: 1000
August 07, 2011, 07:09:34 PM
Ok, so altslicer.  It seems, if it's currently working on one or more slices, then it'll wait until those have finished even if another pool reset to zero.   Right now Slush and bclc just reset (just about), but altslicer seems to be ignoring them (presumably until the 350ish slices for polmine finishes....

Is that intended behaviour?

EDIT: Yep, once the polmine slices ran out it recalculated to include bclc (Slush had already past the 11%).   I suppose, ideally, the altslicing should recalculate immediately on any enabled pool resetting to zero.   Does that make more sense?
sr. member
Activity: 476
Merit: 250
moOo
August 07, 2011, 07:04:57 PM
you can delete x8s from the user.cfg. default they are closed down.. and of course the info it is sending is wrong.



Quote
Does anyone fancy coding in a quick line at the bottom of the stats page that just totals all your entered Payout figures?
i'd love to have someone code a est total and payout total at the bottom.

Quote
OMG, I don't check the thread for 4 days and there are 30 new pages!

cant break from this thread.  and yeah I think this needs it own subforum here., cause you are right a lot of this doesnt help c00w.


I'm sold on mine_slush so far

bitcoin.lc ... up and down.. they lie to us but seem to make over 100% per share

bb
member
Activity: 84
Merit: 10
August 07, 2011, 07:00:00 PM
OMG, I don't check the thread for 4 days and there are 30 new pages! Smiley

Just fought my way through them.

So, 1) I know we used to hop bitcoins.lc (back in the old days ^^) but stopped for a reason. It appears as though some people are back hopping them?

2) Is everybody using backup_latehop??

3) How are your hooping efficiencies on slush turning out?

Also, the payload dropped on this thread is ridiculous. Half of the last 30 pages was "pool xy (not) working / doing blah", another 25 - 30% was user support, maybe 10% dev talk.

This could easily be split over multiple threads (hden inv any1?).
hero member
Activity: 798
Merit: 1000
August 07, 2011, 06:50:34 PM
Does anyone fancy coding in a quick line at the bottom of the stats page that just totals all your entered Payout figures?
legendary
Activity: 924
Merit: 1004
Firstbits: 1pirata
August 07, 2011, 06:46:14 PM
mining mainframe appears to be lying about stats anyways.
digbit says they just  found a blockabout 30 minutes ago


and miningmainframe agrees on their stats page

but the info on the stats page says 139.50 percent and has for quite some time now... can probably disable them if you have them feeding you info.,

edit.. they seem to be doing the hour delay.. should go to zero shortly... they do appear to report the block on time.. on the stats page but then report shares for the last block everywhere else for an hour after.

nicely spotted sir, they have a real BOFH in there I think
sr. member
Activity: 476
Merit: 250
moOo
August 07, 2011, 06:40:53 PM
Quote
It isnt very pretty, i know.
If you want to use this set output destination, open itzod and btcmp in ie on mining rig with windows and start the script. I grab output via http server, you can probably find one.

very cool rino.. I will try this.. looks like it will work... would hate to do too many sites this way but for a few good ones that is a great idea.
sr. member
Activity: 476
Merit: 250
moOo
August 07, 2011, 06:34:48 PM
mining mainframe appears to be lying about stats anyways.
digbit says they just  found a blockabout 30 minutes ago


and miningmainframe agrees on their stats page

but the info on the stats page says 139.50 percent and has for quite some time now... can probably disable them if you have them feeding you info.,

edit.. they seem to be doing the hour delay.. should go to zero shortly... they do appear to report the block on time.. on the stats page but then report shares for the last block everywhere else for an hour after.
newbie
Activity: 38
Merit: 0
August 07, 2011, 06:34:05 PM
I finished writing some ... things for hopping itzod, btcmp and possibly other pools with harder to reach round shares.
I used ahk script to grab their shares directly from browser.
Code:
sleep, 1000
;MouseClick, left, 100, 750, 1, 10
sleep, 1000
itzodshares=2000000
btcmpshares=2000000
loop                                           ;-test
{
IfWinExist, pool.itzod.ru (under construction) - Windows Internet Explorer
WinActivate
else
Exit
loop
{
sleep, 2000
MouseClick, left, 685, 37, 1, 10                 ;-test
sleep, 10000                                      ;-test
;shares
imagesearch, six, siy, 0, 0, 1024, 768, sharespic2.png
if errorlevel=0
{
break
}
}
MouseClick, left, six+150, siy, 2, 10
sleep, 500
Send, {CTRLDOWN}c{CTRLUP}
sleep, 500
tooltip, %clipboard%
sleep, 100
sleep, 50
itzodshares=%clipboard%
if itzodshares=%btcmpshares%
{
msgbox, something broke during itzod
}
FileDelete, Z:\_Soft\_Bitcoin\bitHopper\http\poolshares
FileAppend, itzod=%itzodshares% btcmp=%btcmpshares%, Z:\_Soft\_Bitcoin\bitHopper\http\poolshares




IfWinExist, bitcoin minepool - Windows Internet Explorer
WinActivate
else
Exit
sleep, 2000
MouseClick, left, 685, 37, 1, 10                
sleep, 15000                                      
;shares
loop
{
sleep, 2000
MouseClick, left, 685, 37, 1, 10                
sleep, 15000                                      
;shares
imagesearch, six, siy, 0, 0, 1024, 768, bitminepool.png
if errorlevel=0
{
break
}
}
MouseClick, left, six+85, siy, 2, 10
sleep, 200
Send, {CTRLDOWN}c{CTRLUP}
sleep, 500
tooltip, %clipboard%
sleep, 100
btcmpshares=%clipboard%
if btcmpshares=%itzodshares%
{
msgbox, something broke during btcmp
}
FileDelete, Z:\_Soft\_Bitcoin\bitHopper\http\poolshares
FileAppend, itzod=%itzodshares% btcmp=%btcmpshares%, Z:\_Soft\_Bitcoin\bitHopper\http\poolshares


sleep, 305000
}

Code:
[btcmp]
name: btcmp
mine_address: http://rr.btcmp.com:8332
api_address: http://localhost:8080/poolshares
api_method:re
api_key:btcmp=([0-9]+)
role:mine
user: z.worker
pass: z
[itzod]
name:itzod
mine_address:lp1.itzod.ru:8344
api_address:http://localhost:8080/poolshares
api_method:re
api_key:itzod=([0-9]+)
role:mine
user: z
pass: z
image for itzod: http://i.imgur.com/jG5Va.png

image for btcmp: http://i.imgur.com/ZCNbY.png

It isnt very pretty, i know.
If you want to use this set output destination, open itzod and btcmp in ie on mining rig with windows and start the script. I grab output via http server, you can probably find one.
sr. member
Activity: 476
Merit: 250
moOo
August 07, 2011, 06:09:59 PM
Quote
the newest block 140015 they say is 96% chance it is deepbit.. mind you they have accuracy issues as well... but this is just a place holder to see how well they predicted those two.

14016 ...98% chance deepbit

both of these turned out to be deepbit


mining with nothign but bithopper.py --startLP

havent submitted a single share to deepbit or btcguild.. i was using altslice when they found those blocks but deepbit apparently found another while i wasnt using altslice



and on a back burner note..




Mining mainframe why do you have to mess up my page? You are set to info.. NOT MINE.. go back down there with the rest of the infos. GRRRRRR
hero member
Activity: 504
Merit: 502
August 07, 2011, 05:52:22 PM
That could work as well. People have been kicking it around for a while and well nobody ever did anything. If someone could figure out how hashrates scales to shares/second I could probably do it myself.


that reminded me.. I have seen several places in this forum where people wonder if their is an actual hard relationship between the two.

There was one post where a guy discovered that he got more shares from NOT overclocking his card than when he did. I dismissed it as heat.. and maybe it is still the issue.. as a hot laptop will go slow even if it has a lot of GHZ

I cant find the one post I wanted but I can find a supporting post with zero replies


as you can see this person complains that theri card with less mHs constantly has more shares.


Depending on your requested shares and time to solve the share (some shares take longer than other) you can effectively get more than default unlucky streak since you could get a batch of 10k shares only being solved at 300mh even though your card actually process them at 400mh.

The tests you speak of here I rate doesnt apply well to what he tried to figure out, he could have had a good share sample that solved fast with his first test run using underclocked card as compared to his overclocked results. Only way to truely test what he had in mind is to run a much larger sample over a couple of blocks. True is, overclocked gpu will outperform slower clocked gpus over the long haul , allways.
sr. member
Activity: 476
Merit: 250
moOo
August 07, 2011, 05:45:35 PM
That could work as well. People have been kicking it around for a while and well nobody ever did anything. If someone could figure out how hashrates scales to shares/second I could probably do it myself.


that reminded me.. I have seen several places in this forum where people wonder if their is an actual hard relationship between the two.

There was one post where a guy discovered that he got more shares from NOT overclocking his card than when he did. I dismissed it as heat.. and maybe it is still the issue.. as a hot laptop will go slow even if it has a lot of GHZ

I cant find the one post I wanted but I can find a supporting post with zero replies


as you can see this person complains that theri card with less mHs constantly has more shares.
member
Activity: 84
Merit: 10
August 07, 2011, 05:35:52 PM
you think you do better with cherry picker just with bitclocker?.. i still make decent over there. but i dont get to mine as much as i would like.
c00ws seems to have faster development and more pools and I havent quite figured out making my own cherry pickers pool configs too well.,

anyways you know if deepbit works.. cherry wont be far behind adding it.


Well today alone I have found 4 blocks at bitclockers for a total of 0.63544748 BTC with 690Mhash/s.  So yea, its awesome.  With bithopper I MAY have made .3 BTC out of those 4 blocks if I was lucky.  Cherrypickers has taken my estimated payout from .03BTC/round while mining to .2 BTC a round while mining.  Also, for some odd reason I have NEVER seen my hash rate reported on the site below 750Mhash/s and its usually above 800Mhash/s.  No idea why that is but I am not complaining.

Can't wait to see how this LP hopping works out.  Has a lot of potential to cut through all of the BS that pools do to stop hoppers.  And to your last point, I'd rather bithopper figure out bitclockers so that I could come back here because the entire website monitoring thing is awesome c00w is a mad genius.
Pages:
Jump to: