Topic: [1200 TH] EMC: 0 Fee DGM. Anonymous PPS. US & EU servers. No Registration! - page 34. (Read 499781 times)

Actual share are not important as it is not much useful to the user/miner.

I think what juhakall wants is a SUM of total shares submitted in diff 1 format

I feel like you don't really understand what I'm saying. I want to see diff1 shares in that column, because that's the actual measure of work done. I thought you would understand that as a pool op. It's currently showing the raw number of shares, which is very uninformative when the difficulty of those shares is not known.

I don't have a problem with that... does anyone else reading this still want "actual" shares vs diff1 shares in that column?  If so, why?

I mean the total amount of work that's actually used in reward calculations. If I have submitted 1000 diff1 shares, 1000 diff2 shares and 1000 diff4 shares, it doesn't make much sense to tell me that I've submitted 3000 shares. Total accepted difficulty in this case would be 1000+1000*2+1000*4=7000. Or 3000 shares times the average difficulty of 2.333, just in a more useful format.

Hmm... ok, so when you say accepted difficulty, what do you mean specifically?  Like an aggregate or average?  I'm certainly willing to change it.

Knowing the amount of shares I submitted during a round isn't useful, when the shares have dynamic difficulty. For example, some shares may have been submitted at diff 2, others at diff 4, especially if my hashrate isn't constant. Total accepted difficulty would be a more meaningful stat. You already calculate this in a similar way on the "Diff1 Shares (Actual)" column - I presume the number in parentheses is the number of shares accepted, while the bigger number represents total accepted difficulty.

I'm not really sure what you mean.  How is it any less or more useful than before?

"My Shares" column in block history is displaying the amount of shares submitted, which is not very useful when using vardiff. It should be changed to show the combined difficulty of those shares.

Yeah, US1 was down.  I really need to move that server.  It should be back up now.

Seconded the thought US1 is down.

I think US1 is down...

Ok... US2 should be catching up now.  I am going to pull that server and try to figure out what's wrong with it.  I suspect a bad stick of memory at this point.

I have been waiting to get a new US1 server installed as well as bringing back EU and PACRIM... I will try to make that a priority this week.

-EDIT-

*BANGS HEAD* ... the server tech just took the wrong server offline.  It will be back up shortly.

us2, they are connecting fine now but stats are not updating.  It shows last activity of 12hrs.  

edit: switched to us3 and stats updated.

One of my machines is mining stratum on US3 with no problems.  What server are you on?

Anyone else having trouble connecting to stratum this morning?  I noticed none of my miners were connected and they all switched to the backup pool.  Any ideas?

I've updated cgminer, just to find that a new version came out, so I need to update again =)). Now it seems to run smooth, rejects are at usual level. Its sad to hear that nmc is falling apart. They could've used it for website certification, digital documents signing and just the obvious ID verification. But as we know, some of these can be provided by bitcoin. Its strange that your pool has lots of invalids for nmc, as bitminter doesn't. But bitminter experiences regular namecoind crashes, which affect bitcoin mining.

Try the latest version of CGMiner (or BFGMiner) and let me know if you're still getting stales.  Use US2 or US3 as well, since those are closer to your boxes than US1 is.

As far as NMC goes, Namecoin is falling apart... it's going to be disabled soon.  If you look at the Namecoin list, they are almost all showing up as invalid.  The whole Namecoin protocol for merged mining is a shambles and it's more trouble than it's worth.

Actually not, its 2.9.4.  I think I need to update  . Also, I've noticed that I'm not getting nmc anymore.

BlackPrapor, are you running an up to date cgminer or bfgminer?

Not me! Stratum works great. Under .02% rejects usually.

 Last night I had an issue. I assumed because Bitpenny(4) (my backup backup backup) kept working that my 4 pools before it had all failed. All but one marked dead. The not dead one was not actually working for me. I think there might be a slight issue still but Con will fix it I have faith. Since I could post here and had some dynamic data working I Assumed (bad idea) that Eclipse was the problem as hashpower(0) comes and goes, US1(1) was marked alive but 3s hashrate was pinned at 0 and my Single was quiet. US2(2), US3(3) are marked dead. Had been alive earlier and have used them as backups. Cgminer switched from hashpower to US1 earlier US1 had an issue (remote failure), switched to US2 and shortly after remote failure, Switched to US3 and next thing was remote failure. Bitpenny marked not providing work fast enough did eventually get marked dead I am sure. US3 GBT was also marked dead its pool 5 and I try to use it as a LP source for bitpenny.

Normally 100% of my connctions work part of the day, 80% of them work nearly 100% of the time. It isn't that Eclipse is a bad pool. It isn't. Inaba keeps it up very nearly 100% of the time. I have never had anything over 10% rejects at eclipse I don't belive. Then again at 10% give or take cgminer marks the pool as rejecting and moves on to someone else for a while. This should remove the overloaded condition that usually causes higher rejects.

Possibly your rejects are because of variable difficulty? Check your logs for share blah blah blah (below target). If it says below target difficulty jumped up on you and you are submitting work that is too low. You get paid for anything above the target so assuming you spent some time at 2 you would have some rejects from any diff 1 shares you send in but any accepted 2diff shares are paid 2x as much. Assuming in that you on average every other work unit get a solution of 2diff you will get paid 2x for 2x work but any sub 2x is rejected.