joehoebot
|
|
October 27, 2014, 01:07:49 PM |
|
ok thx. question: if when i start up the miner, i am getting shares accepted very very fast, and then it gradually slows down, does that maybe mean that suprnova is auto-tuning to give me only the more difficult shares after a couple of mins?
|
|
|
|
djm34
Legendary
Offline
Activity: 1400
Merit: 1050
|
|
October 27, 2014, 01:08:36 PM |
|
so... how long until the end of POW? I have no idea. The developer can answer this. But i think he is sleeping right now! https://bitcointalk.org/index.php?topic=735170.msg9310611#msg9310611Assuming I understood what dev meant (let say there is 50% chance I did... but from recent experiences I am rather pessimistic at that level and I guess I could add a 2 dice rolls and remove the numbers from 50% , so it 50% - 2 random numbers between 1 and 6 ) I would say never (or in about a year at pow phase-II) (however since it is below 50% because of the dice roll... the most probable answer is "yes sometimes between next week and a year")
|
djm34 facebook pageBTC: 1NENYmxwZGHsKFmyjTc5WferTn5VTFb7Ze Pledge for neoscrypt ccminer to that address: 16UoC4DmTz2pvhFvcfTQrzkPTrXkWijzXw
|
|
|
joehoebot
|
|
October 27, 2014, 02:52:16 PM |
|
cmon juicy block!
|
|
|
|
Spexx
Sr. Member
Offline
Activity: 350
Merit: 250
Mining Co-operative
|
|
October 27, 2014, 03:40:06 PM |
|
Spexx's, why your 2 new miners (32bit & 64bit) is not loading fully processor core: 1 core only ~24% (-t 1); 2 core only ~49% (-t 2); 3 core only ~51-65% (-t 3); 4 core only ~47-65% (-t 4). hashrate 2core > hashrate 4core > hashrate 3core - Why? problems of parallelization process? especially bad in 64-bit versions...
I have been through this before. The problem is something to do with the Cygwin compiler not handling pthreads very well so I do not recommend using my minerd with --threads greater than 1. The trick is to start multiple instances of minerd instead i.e. if you have 4 cores then start minerd.exe four times with --threads=1 and then you will find the CPU is 100% used. My Windows batch file "wrapper" automine.bat does this for you.
|
|
|
|
joelao95 (OP)
Legendary
Offline
Activity: 1190
Merit: 1009
Coin of the Magi!
|
|
October 27, 2014, 03:51:31 PM |
|
so... how long until the end of POW? I have no idea. The developer can answer this. But i think he is sleeping right now! https://bitcointalk.org/index.php?topic=735170.msg9310611#msg9310611Assuming I understood what dev meant (let say there is 50% chance I did... but from recent experiences I am rather pessimistic at that level and I guess I could add a 2 dice rolls and remove the numbers from 50% , so it 50% - 2 random numbers between 1 and 6 ) I would say never (or in about a year at pow phase-II) (however since it is below 50% because of the dice roll... the most probable answer is "yes sometimes between next week and a year") DJM, that's so nice of you that recall us the link for updates, thanks.
|
|
|
|
|
joelao95 (OP)
Legendary
Offline
Activity: 1190
Merit: 1009
Coin of the Magi!
|
|
October 27, 2014, 04:20:02 PM |
|
ok thx. question: if when i start up the miner, i am getting shares accepted very very fast, and then it gradually slows down, does that maybe mean that suprnova is auto-tuning to give me only the more difficult shares after a couple of mins? hmm, won't be possible.
|
|
|
|
|
Spexx
Sr. Member
Offline
Activity: 350
Merit: 250
Mining Co-operative
|
|
October 27, 2014, 04:38:36 PM |
|
ok thx. question: if when i start up the miner, i am getting shares accepted very very fast, and then it gradually slows down, does that maybe mean that suprnova is auto-tuning to give me only the more difficult shares after a couple of mins? I spotted this problem on both pools long ago and solved it by 1) Starting multiple instances of minerd with a single thread 2) Pointing each minerd at its own unique worker name on the pool(s) Doing this appears to keep the miners/pools happy and shares are accepted at a fairly constant rate. It is way more efficient than having a single minerd with multiple threads pointing at a single worker on the pool. I do not know exactly why this should be so, but would be fascinated by comments from the pool operators about it
|
|
|
|
joehoebot
|
|
October 27, 2014, 04:54:57 PM |
|
ah ok so i just have to create about 1000 workers
|
|
|
|
MarcusDe
|
|
October 27, 2014, 05:00:54 PM |
|
Hi,
Something fucked up on nounce-pool? I have all 'booooo' from miners. After restart there is few 'yay' and again only 'boooo' then. All worked fine few hours ago.
|
|
|
|
ocminer
Legendary
Offline
Activity: 2688
Merit: 1240
|
|
October 27, 2014, 05:02:01 PM |
|
ok thx. question: if when i start up the miner, i am getting shares accepted very very fast, and then it gradually slows down, does that maybe mean that suprnova is auto-tuning to give me only the more difficult shares after a couple of mins? I spotted this problem on both pools long ago and solved it by 1) Starting multiple instances of minerd with a single thread 2) Pointing each minerd at its own unique worker name on the pool(s) Doing this appears to keep the miners/pools happy and shares are accepted at a fairly constant rate. It is way more efficient than having a single minerd with multiple threads pointing at a single worker on the pool. I do not know exactly why this should be so, but would be fascinated by comments from the pool operators about it It is totally correct and normal that the pool varies your diff to get your miner send a share every x seconds, its not necessary to send a low diff every second if you can also work on a share 10 times that hard which gives you 10 times the credit but you only submit every 10 seconds which lowers the load on the pool greatly. try it - your payout estimates will be the same.
|
suprnova pools - reliable mining pools - #suprnova on freenet https://www.suprnova.cc - FOLLOW us @ Twitter ! twitter.com/SuprnovaPools
|
|
|
Bojcha
|
|
October 27, 2014, 05:40:18 PM |
|
ok thx. question: if when i start up the miner, i am getting shares accepted very very fast, and then it gradually slows down, does that maybe mean that suprnova is auto-tuning to give me only the more difficult shares after a couple of mins? I spotted this problem on both pools long ago and solved it by 1) Starting multiple instances of minerd with a single thread 2) Pointing each minerd at its own unique worker name on the pool(s) Doing this appears to keep the miners/pools happy and shares are accepted at a fairly constant rate. It is way more efficient than having a single minerd with multiple threads pointing at a single worker on the pool. I do not know exactly why this should be so, but would be fascinated by comments from the pool operators about it It is totally correct and normal that the pool varies your diff to get your miner send a share every x seconds, its not necessary to send a low diff every second if you can also work on a share 10 times that hard which gives you 10 times the credit but you only submit every 10 seconds which lowers the load on the pool greatly. try it - your payout estimates will be the same. Ofcourse it will not be same. What if block is found in 5, 10 or 20 sec.. your shares will not be sent on time so you wont get paid. Other story is that suprnova server is slow.
|
|
|
|
Spexx
Sr. Member
Offline
Activity: 350
Merit: 250
Mining Co-operative
|
|
October 27, 2014, 06:10:32 PM |
|
ah ok so i just have to create about 1000 workers If you are running 1000 CPU cores then that is what I am suggesting - yes I found that setting up 32 workers did not take too much time, but 1000 is another matter Try setting up a few extra workers first and point some single-threaded minerds at them. You will see what I mean - you should see a greater rate of accepted shares compared to having multi-threaded minerds pointing at a single worker. It is up to you to decide if you think it worthwhile to invest some time in setting up multiple workers on the pools, based on your results of testing with your particular hardware/software environment.
|
|
|
|
djm34
Legendary
Offline
Activity: 1400
Merit: 1050
|
|
October 27, 2014, 06:13:54 PM |
|
ok thx. question: if when i start up the miner, i am getting shares accepted very very fast, and then it gradually slows down, does that maybe mean that suprnova is auto-tuning to give me only the more difficult shares after a couple of mins? I spotted this problem on both pools long ago and solved it by 1) Starting multiple instances of minerd with a single thread 2) Pointing each minerd at its own unique worker name on the pool(s) Doing this appears to keep the miners/pools happy and shares are accepted at a fairly constant rate. It is way more efficient than having a single minerd with multiple threads pointing at a single worker on the pool. I do not know exactly why this should be so, but would be fascinated by comments from the pool operators about it It is totally correct and normal that the pool varies your diff to get your miner send a share every x seconds, its not necessary to send a low diff every second if you can also work on a share 10 times that hard which gives you 10 times the credit but you only submit every 10 seconds which lowers the load on the pool greatly. try it - your payout estimates will be the same. Ofcourse it will not be same. What if block is found in 5, 10 or 20 sec.. which isn't the case here. Actually it is better to solve high diff share rather than small, miners are more stable and work at full capacity. If you solve very often low diff, share then the miner lose lots of time in transactions... and mostly for nothing....
|
djm34 facebook pageBTC: 1NENYmxwZGHsKFmyjTc5WferTn5VTFb7Ze Pledge for neoscrypt ccminer to that address: 16UoC4DmTz2pvhFvcfTQrzkPTrXkWijzXw
|
|
|
111magic
Legendary
Offline
Activity: 1750
Merit: 1005
|
|
October 27, 2014, 08:15:41 PM |
|
getting a mingw assertion error when trying to open the new wallet, anyone else?
No problem with the wallet. Do you have the latest version? What kind of error?
|
bitcoin: bc1qyadvvyv29z08ln2ta7g3uqwzkscr7wq4p09wuz
|
|
|
qiwoman2
Legendary
Offline
Activity: 2114
Merit: 1023
Oikos.cash | Decentralized Finance on Tron
|
|
October 27, 2014, 08:27:20 PM |
|
|
|
|
|
111magic
Legendary
Offline
Activity: 1750
Merit: 1005
|
|
October 27, 2014, 08:47:59 PM |
|
Thank you qiwoman2! Looks very good. Nice job and welcome in this great community!
|
bitcoin: bc1qyadvvyv29z08ln2ta7g3uqwzkscr7wq4p09wuz
|
|
|
aceoyame
Jr. Member
Offline
Activity: 51
Merit: 1
|
|
October 28, 2014, 01:43:43 AM |
|
The algo switch has been giving me a hell of a time on nonce.
None of my shares are counting after awhile
|
|
|
|
joelao95 (OP)
Legendary
Offline
Activity: 1190
Merit: 1009
Coin of the Magi!
|
|
October 28, 2014, 01:52:16 AM |
|
The algo switch has been giving me a hell of a time on nonce.
None of my shares are counting after awhile
Which miner you're using, vps, windows?
|
|
|
|
|