forrestv (OP)
|
|
January 27, 2012, 08:32:02 PM |
|
i am getting dead shares with the git version. which part of the code sets the difficulty lower for debugging?
The difficulty is always low whether you use --debug or not. The relevant code is on line 474 in main.py - "target = 2**256//2**32 - 1". If you want to change the minimum difficulty back to 8, you can replace it with "target = 2**256//2**32//8 - 1", though I'm not sure why you'd want to.
|
1J1zegkNSbwX4smvTdoHSanUfwvXFeuV23
|
|
|
Costia
Newbie
Offline
Activity: 28
Merit: 0
|
|
January 27, 2012, 08:34:46 PM |
|
the normal p2pool client was solving for the right difficulty ~200 the git one goes a lot faster and cgminer reports a lot of rejects p2pool reports just a few shares like before - but more of them are dead so i want to set the difficulty back to its proper value and see if cgminer still gets the rejects
|
|
|
|
forrestv (OP)
|
|
January 27, 2012, 08:35:28 PM |
|
the normal p2pool client was solving for the right difficulty ~200 the git one goes a lot faster and cgminer reports a lot of rejects p2pool reports just a few shares like before - but more of them are dead so i want to set the difficulty back to its proper value...
It going a lot faster is useful because you can see how well your miner is doing. If you a lot of rejects (>5%) on your miner, something is definitely wrong with it.
|
1J1zegkNSbwX4smvTdoHSanUfwvXFeuV23
|
|
|
Costia
Newbie
Offline
Activity: 28
Merit: 0
|
|
January 27, 2012, 08:36:20 PM |
|
i had 0 rejects with the old 0.8.1
|
|
|
|
Costia
Newbie
Offline
Activity: 28
Merit: 0
|
|
January 27, 2012, 08:38:47 PM |
|
yep
|
|
|
|
forrestv (OP)
|
|
January 27, 2012, 08:39:43 PM |
|
i had 0 rejects with the old 0.8.1
How many rejects do you have now?
|
1J1zegkNSbwX4smvTdoHSanUfwvXFeuV23
|
|
|
Costia
Newbie
Offline
Activity: 28
Merit: 0
|
|
January 27, 2012, 08:41:51 PM |
|
at the fast pace its ~10% in the cgminer ui i have ~200mhash, it will take a while before i can see anything useful at the low rate...
|
|
|
|
Costia
Newbie
Offline
Activity: 28
Merit: 0
|
|
January 27, 2012, 09:14:24 PM |
|
so. on the fast client: reduced the gpu threads from 4 to 2: 1/33 rejected reduced to 1 gpu thread 0/50 rejected I guess when the difficulty is lower there were fewer "races" between the threads
|
|
|
|
JWU42
Legendary
Offline
Activity: 1666
Merit: 1000
|
|
January 27, 2012, 09:29:53 PM |
|
Why 4 threads? cgminer I assume?
|
|
|
|
Ente
Legendary
Offline
Activity: 2126
Merit: 1001
|
|
January 27, 2012, 09:47:51 PM |
|
Finally I got around playing with p2pool. Friend and me have several miners, once I get p2pool running on one, I will merge them all. Windows machines, unfortunately.. It seems like bitcoin / bitcoind connect to the net and let p2pool connect. p2pool seems to be receiving work from bitcoin. poclbm seems to connect to p2pool too. However, no work seems to be actually done.I tried guiminer too, it says 286 mhash, no idea where these are going to.. Excerpts from what the shells tell: p2pool:c:\programme\python\python.exe c:\programme\p2pool\run_p2pool.py xxxxxxxxxx xxxxxxxxxx -a xxxxxxxxxx 22:29:17.321000 Pool: 97434MH/s in 17642 shares (9305/21649 verified) Recent: 0. 00% >0H/s Shares: 0 (0 orphan, 0 dead) Peers: 9 poclbm miner:c:\programme\p2pool\poclbm\poclbm.exe http://xxxxxxxxxx:xxxxxxxxxx@127.0.0.1: 9332 -d0 --verbose -f 100 -w 128 -v 127.0.0.1:9332 27/01/2012 22:29:33, long poll: new block 0000050c6c694815 Guiminer:2012-01-27 22:33:43: Miner "p2pool": 127.0.0.1:9332 27/01/2012 22:33:43, long poll: new block 0000050c6c694815 Glad for any hints.. And just so you know: Getting that stuff done on Windows was no pleasant experience so far. Will never ever give up my Debian here! :-) If someone asks, I'll paste my super-condensed step-by-step for p2pool with guiminer/poclbm on windows. Edit: full logs: http://pastebin.com/aBDP9dssGrateful for any ideas you have! Ente
|
|
|
|
Qoheleth
Legendary
Offline
Activity: 960
Merit: 1028
Spurn wild goose chases. Seek that which endures.
|
|
January 27, 2012, 09:57:18 PM |
|
Finally I got around playing with p2pool. Friend and me have several miners, once I get p2pool running on one, I will merge them all. Windows machines, unfortunately.. It seems like bitcoin / bitcoind connect to the net and let p2pool connect. p2pool seems to be receiving work from bitcoin. poclbm seems to connect to p2pool too. However, no work seems to be actually done.I tried guiminer too, it says 286 mhash, no idea where these are going to.. Excerpts from what the shells tell: p2pool:c:\programme\python\python.exe c:\programme\p2pool\run_p2pool.py xxxxxxxxxx xxxxxxxxxx -a xxxxxxxxxx 22:29:17.321000 Pool: 97434MH/s in 17642 shares (9305/21649 verified) Recent: 0. 00% >0H/s Shares: 0 (0 orphan, 0 dead) Peers: 9 poclbm miner:c:\programme\p2pool\poclbm\poclbm.exe http://xxxxxxxxxx:xxxxxxxxxx@127.0.0.1: 9332 -d0 --verbose -f 100 -w 128 -v 127.0.0.1:9332 27/01/2012 22:29:33, long poll: new block 0000050c6c694815 Guiminer:2012-01-27 22:33:43: Miner "p2pool": 127.0.0.1:9332 27/01/2012 22:33:43, long poll: new block 0000050c6c694815 Glad for any hints.. And just so you know: Getting that stuff done on Windows was no pleasant experience so far. Will never ever give up my Debian here! :-) If someone asks, I'll paste my super-condensed step-by-step for p2pool with guiminer/poclbm on windows. Edit: full logs: http://pastebin.com/aBDP9dssGrateful for any ideas you have! Ente How long did you leave it running before deciding that no work is being done? Shares are harder on p2pool than other pools; one is handed out about every 10 seconds, in much the same way Bitcoin makes a block every 10 minutes. So with 286Mhash, after one hour of running there's only a 65% chance that you'll have gotten your first share.
|
If there is something that will make Bitcoin succeed, it is growth of utility - greater quantity and variety of goods and services offered for BTC. If there is something that will make Bitcoin fail, it is the prevalence of users convinced that BTC is a magic box that will turn them into millionaires, and of the con-artists who have followed them here to devour them.
|
|
|
Ente
Legendary
Offline
Activity: 2126
Merit: 1001
|
|
January 27, 2012, 10:03:37 PM |
|
OH! No, I left it running for several minutes only! So I will solve shares in hours, and the stated mh/s is calculated by the solving-rate of shares?
So I "gave up" pretty close to the goal.. heh Will leave it runnign for a good while now.
Thank you a lot for your quick help, Qoheleth! :-)
Ente
|
|
|
|
Red Emerald
|
|
January 27, 2012, 10:07:55 PM |
|
So I will solve shares in hours, and the stated mh/s is calculated by the solving-rate of shares?
The MH/s is the total number of hashes (good or bad) per second. It has nothing to do with the solving rate.
|
|
|
|
Ente
Legendary
Offline
Activity: 2126
Merit: 1001
|
|
January 27, 2012, 10:13:19 PM |
|
Hmm.. So how long should it take for the p2pool stat to change? 22:29:17.321000 Pool: 97434MH/s in 17642 shares (9305/21649 verified) Recent: 0. 00% >0H/s Shares: 0 (0 orphan, 0 dead) Peers: 9 Especially the "0H/s". Will it update once I solved my first share, or a few seconds after I start working with a connected miner? Ente
|
|
|
|
Red Emerald
|
|
January 27, 2012, 10:15:01 PM |
|
Hmm.. So how long should it take for the p2pool stat to change? 22:29:17.321000 Pool: 97434MH/s in 17642 shares (9305/21649 verified) Recent: 0. 00% >0H/s Shares: 0 (0 orphan, 0 dead) Peers: 9 Especially the "0H/s". Will it update once I solved my first share, or a few seconds after I start working with a connected miner? Oh. I'm not sure how that one works. H/s is output in so many places lol
|
|
|
|
JWU42
Legendary
Offline
Activity: 1666
Merit: 1000
|
|
January 27, 2012, 10:18:09 PM |
|
Hmm.. So how long should it take for the p2pool stat to change? 22:29:17.321000 Pool: 97434MH/s in 17642 shares (9305/21649 verified) Recent: 0. 00% >0H/s Shares: 0 (0 orphan, 0 dead) Peers: 9 Especially the "0H/s". Will it update once I solved my first share, or a few seconds after I start working with a connected miner? Ente Took me about 10-15 minutes as I recall (with just shy of 4GH/s).
|
|
|
|
Costia
Newbie
Offline
Activity: 28
Merit: 0
|
|
January 27, 2012, 10:22:18 PM |
|
Why 4 threads? cgminer I assume?
yep. i think 4 threads gave me a bit higher hashrate
|
|
|
|
Costia
Newbie
Offline
Activity: 28
Merit: 0
|
|
January 27, 2012, 11:22:06 PM |
|
i think the split part of the pool found a block p2pool client: "Payout if block: 0.087618 BTC" http://blockchain.info/block-index/854582: 16fkg8GtdNbEvoJHRHJxhsNpTyxRbi8Twg 0.05413772 BTC
|
|
|
|
eja
Newbie
Offline
Activity: 39
Merit: 0
|
|
January 27, 2012, 11:24:03 PM |
|
Yes, seems so. Didn't get anything for this block I wonder, if this split will resolve itself eventually ?
|
|
|
|
Red Emerald
|
|
January 27, 2012, 11:25:59 PM |
|
If anyone does not have the technical skill, time, hardware, or whatever to run their own instance of p2pool, feel free to point your miner to "p2pool.stitthappens.com:8336" with your username set to a bitcoin address and the password being w/e you want. The more I look at it, the better I think p2pool is for bitcoin. Since it is non-trivial to run a p2pool node (although I wouldn't say it is hard), I'm willing to run this node for anyone. If you use my node and want to donate something, I appreciate it. 1GH2r1uELMdn5726zxufpi5A3KswWNKU6Y I've also setup a tor hidden service for access to p2pool: id7ibznt6zsfp4kr.onion Not sure if there are any uses for mining over tor or how the latency affects stale rate, but what better way to know than to test?
|
|
|
|
|