[SUPPORT] beeeeer.org - tech support

xolokram · November 19, 2013, 8:02am

read the second line of your user stats (the one in the brackets)
pps is always a problematic measure, at first i would say that a factor of 4 is really good (compared to the original client), i dont know if this is what you should expect, it’s a while ago i tested the original client the last time (maybe other users can help?). on the other hand, people like jh00 & co. (tandyuk,rbe…) are using a different measure. e.g. their PPS measure is counting every iteration of the mining process (giving them 20k “PPS”) even when the iteration hasn’t found a primechain, while my miner only counts iterations finding at least a primechain of length 1 (giving me only 3k “PPS”).

unfortunately people only see 20k PPS vs 3k PPS and going crazy about it…

xolokram

ps. some hints:

tandyuk/jhPrimeminer/blob/master/src/primecoinMiner/prime.cpp#L948


   break;
}


nTests++;
mpzChainOrigin = mpzHash * mpzFixedMultiplier * nTriedMultiplier;		
nChainLength = 0;		
	ProbablePrimeChainTestFast(mpzChainOrigin, testParams);
nProbableChainLength = nChainLength;
sint32 shareDifficultyMajor = 0;


primeStats.primeChainsFound++;


if( nProbableChainLength >= 0x06000000 )
{
   shareDifficultyMajor = (sint32)(nChainLength>>24);
}
else
{
   continue;
}

github.com

thbaumbach/primecoin/blob/master/src/prime.cpp#L802


if (unlikely(mpzR == 1 || mpzR == mpzNMinusOne))
    return true;
if (likely(fFastFail))
    return false;
// Failed test, calculate fractional length
mpzR2 = mpzR * mpzR;
mpzR = mpzR2 % n; // derive Fermat test remainder


mpzFrac = n - mpzR;
mpzFrac <<= nFractionalBits;
mpzFrac /= n;
unsigned int nFractionalLength = mpzFrac.get_ui();
if (unlikely(nFractionalLength >= (1 << nFractionalBits)))
    return error("FermatProbablePrimalityTest() : fractional assert");
nLength = (nLength & TARGET_LENGTH_MASK) | nFractionalLength;
return false;
}


// Test probable primality of n = 2p +/- 1 based on Euler, Lagrange and Lifchitz
// fSophieGermain:
//   true:  n = 2p+1, p prime, aka Cunningham Chain of first kind

ksolomon · November 19, 2013, 1:30pm

I have a general question about using the miner. On all my machines I’m using -genproclimit=4, however, most of my machines only have 2 cores. Is there a downside to using more workers than you have cores? There was a message in the announce thread that seemed to indicate you should set this option to match your core count, but I figured it was worth a post just to ask…

icedaddy · November 19, 2013, 4:25pm

[quote=“xolokram, post:1112, topic:342”]

xolokram

ps. updated first post
pps. port 443 & 8080 will be added soon[/quote]

Are there still plans to return port 443 support? Any mention of it has disappeared.

melds39 · November 24, 2013, 12:25pm

/stats not working for me. Just get a blank page. It was fine until recently.

*edit: It’s working now!

jbacchus · November 24, 2013, 1:33pm

Hey what happened? From 0400 to 0700 it is showing that nothing was mined (which is just not true!!) Please fill us in !!

ivanlabrie · November 24, 2013, 1:47pm

Pool seems to be down, I’m mining pts meanwhile…anyone know how to set up a failover in linux? (like cgminer’s option)

melds39 · November 24, 2013, 2:05pm

It’s working fine for me. I’m connected to the US server.

xolokram · November 24, 2013, 4:30pm

main thread

Samegutt · November 25, 2013, 9:43am

In my (limited) experience, I am getting optimal performance by using as many threads as number of cores OR as many as hyperthreading allows, if the CPU has HT… I don’t think you can gain anything by using more than the number of cores. It will cause the processes to get cycled CPU-time. Theoretically, HT should also be less efficient due to some cycling, but on my stats it shows slightly better performance going by HT core-count than true core-count.

d2rhino · November 26, 2013, 4:04am

Newbie here: I’m using the Xolo miner (x64). Working fine on my test laptop (~2.2k primes/sec). Should I be trying to make jhprimeminer work? Are there any profit calculators?

Thanks!

lumierre · November 26, 2013, 7:03am

I can’t view my earnings.

http://www.beeeeer.org/user/ just shows a blank page. Anyone else having the same problem?

Dejmien · November 26, 2013, 7:17am

same problem here.

Happific · November 26, 2013, 7:57am

I’m getting the same thing, just a blank page on the user page.

xolokram · November 26, 2013, 9:03am

@happific+djmien+lumierre+iheartcryptocoin+defragger:
yes, backend was stuck again >:(
BUT i was able to restart everything without losing the shares of the affected hours
blocks & payout pages are available

i disabled the web sharelog/sharehistory information as this is using the same database as the share submission/payouts (and i dont want them to be affected)
i guess i have to split / duplicate all stored shares: one for the pool, one for the backend

@d2rhino:
the primes/ses measure is not really comparable between xolominer & jhprimeminer
jhprimeminer is currently not usable on beeeeer anyway, this will probably change in the future

xolokram

oleg · November 26, 2013, 11:16am

I can’t connecto to the pool !!!

*** ptsminer - Pts Pool Miner v0.7 RC2
*** by xolokram/TB - www.beeeeer.org - glhf

*** press CTRL+C to exit

using AVX
spawning 8 worker thread(s)
[WORKER1] Hello, World!
[WORKER0] Hello, World!
[WORKER[WORKER3] Hello, World!
2] Hello, World!
[WORKER4] Hello, World!
[WORKER5] Hello, World!
[WORKER6] Hello, World!
[WORKER1] GoGoGo!
[WORKER3] GoGoGo!
[WORKER0] GoGoGo!
[WORKER7] Hello, World!
[WORKER2] GoGoGo!
[WORKER4] GoGoGo!
[WORKER5] GoGoGo!
[WORKER6] GoGoGo!
[WORKER7] GoGoGo!
connecting to 54.201.26.128:1337
no connection to the server, reconnecting in 10 seconds
connecting to 54.201.26.128:1337
no connection to the server, reconnecting in 10 seconds

And it continues, but if i do a telnet to the server and the port is open. Any ideas ? I’m running on Win 7 Pro Core i7

Xaero · November 26, 2013, 5:24pm

Are the sharelogs/payhistory working properly? I believe I had a 9-ch share earlier this morning, and haven’t seen it added yet.

porzione · November 26, 2013, 6:29pm

Yesterday earned ~10 xpm with thbaumbach miners and only 2.9 today in same time - is it ok?

bearmeat · November 26, 2013, 6:45pm

WOW ~10 xpm a day. Care to share how fast/how many machines you are running on?

Side now I went to my user page and saw this. “SHARELOG/HISTORY CURRENTLY DISABLED, UNDER MAINTENANCE!” But it has been like that for some time.

porzione · November 26, 2013, 6:50pm

WOW ~10 xpm a day. Care to share how fast/how many machines you are running on?[/quote]
I have 144 cores in total (xeon X5660 + i7) but need to spend some cpu ticks for the other little tasks, OS debian linux amd64.

xolokram · November 26, 2013, 7:27pm

@oleg:
you’re in the wrong forum!
this is the peercoin (ppc) / primecoin (xpm) forum
everything is alright on my side, i can connect, and the server’s looking good, please check your connection/setup or write in the pts forum

@porzione+bearmeat+Xaero:
welcome to the forum

the sharelog/history is disabled currently on the webserver, due to some major performance issues related with the database and i dont wanted to damage the productive system
they will be back, when i have a smart solution for this issue

i highly recommend to use Sy’s stats meanwhile: http://xpm.syware.de/

the share reward is currently fluctuating due do the increasing amount of miners and the long confirmation time of mined blocks (2.5 days) resulting in a piled up sharelog
please get your head around CPPSRB to understand the payout system

it’ll increase again within the next days

- xolokram