[XPM] [ANN] Primecoin Release - First Scientific Computing Cryptocurrency - page 170.

twobits

sr. member

Activity: 574

Merit: 250

Quote from: 8bitPunk on July 08, 2013, 04:42:21 PM

Quote from: ReCat on July 08, 2013, 04:04:16 PM

Quote from: Vorksholk on July 08, 2013, 03:10:06 PM

I'll put up one Bitcoin in bounty for a CPU-miner (standalone) that will run on both windows and linux and outperform the built-in miner.

Recompile the existing one with some more optimization flags or with the ICC compiler. Tongue

Anyone compiled a win64 optimized binary yet?

Yeah, at least going by primespersec figures, it did not give the boost that it did for scrypt.

Tuck Fheman

sr. member

Activity: 363

Merit: 250

Quote from: MaGNeT on July 08, 2013, 04:38:25 AM

Awful release of what could have been a nice coin.

- lack of nodes in OP
- lack of info in OP (diff change, blocks/hour, blockreward, etc)
- lack of mining-info in OP

Yeah, I know, we can figure it out... But not everyone is an experienced miner or wants to read a whole topic...

Why spend so much time on creating a new coin and blow it on the release?

I must have missed all that. Everything worked fine for me ("setgenerate true") and I felt more informed beforehand about this coin than any other; but I did read the prerelease thread thoroughly.

Vorksholk

legendary

Activity: 1713

Merit: 1029

I have 5 i7-3770s and 2 i7-2600s mining, most with 3 cores. One of them running with 4 cores has gotten 8 blocks. The rest have gotten two and three, with one getting zero. All centos-6-based (x64).
My i5-2500k (at 3.3GHz) has gotten 5 blocks. ( Windows 8 ).
My Toshiba with an integraded AMD APU has gotten two blocks. No idea how. Mines very slowly. (Windows 8 again).
My 8350s have gotten no blocks at all. ( Windows 7 and Windows 8 )

My digitalocean tiny instances (any larger than tiny and they don't get performance worth the penny per hour) have gotten, out of 15 instances running for 6 hours, a whopping three blocks. However, being that each instance costs $0.007 (yes, less than a penny) per hour, and they've been running for around 8 hours, it's not bad. (Ubunt u10.04 x32)

Basically, i7-3770s are really good at mining, but luck is a crazy thing. AMD chips I have had little luck with, even though they do have high PPS. Cloud instances that are of any girth seem to provide very low hash rates. EC2 CPU XL provided around 24PPS with a max around 28PPS, at a cost of $0.50 per hour! No good.

scootero1211

newbie

Activity: 4

Merit: 0

Quote from: human on July 08, 2013, 06:01:10 PM

I have been mining with about 180 pps since 3 hours after the release (more than 24 hours now) but so far not got a single coin Sad

Seems to be sheer luck at this point. I've been mining with 50pps for not even 24 hours and got a block within 5 hours of starting. But nothing since.

human

full member

Activity: 197

Merit: 100

I have been mining with about 180 pps since 3 hours after the release (more than 24 hours now) but so far not got a single coin Sad

Stinky_Pete

hero member

Activity: 560

Merit: 500

At last! The dual Xeons have found their first block.

Lauda

legendary

Activity: 2674

Merit: 2965

Terminated.

Quote from: zeroblock on July 08, 2013, 05:36:34 PM

After your Primecoins mature, feel free to PM me with offers. Will pay for in BTC. Nothing ridiculous though. I'm not paying 0.01 BTC per XPM. Looking for bulk purchase.

It will be certainly more worth than 0.01btc.

zeroblock

full member

Activity: 238

Merit: 100

After your Primecoins mature, feel free to PM me with offers. Will pay for in BTC. Nothing ridiculous though. I'm not paying 0.01 BTC per XPM. Looking for bulk purchase.

eule

hero member

Activity: 756

Merit: 501

Quote from: nmersulypnem on July 08, 2013, 05:29:00 PM

Why not use "native" flag?

Umm i have no clue tbh, compiler noob here. Prior to optimizing i just cloned in the git and let the makefile do it's job and got 120 - 150, now 170-200pps atm. Try the native flag and tell us if it further improves things, I'm happy for now. Grin

edit: Ofcourse it still dips down to ~140 sometimes, very variable "primerate".

nmersulypnem

full member

Activity: 238

Merit: 100

Quote from: eule on July 08, 2013, 05:23:39 PM

lol... people, don't sell your precious primecoins for pocket money. Tongue

Quote from: ig0tik3d on July 08, 2013, 04:14:36 PM

QMAKE_CXXFLAGS += -O3 -march=core-avx-i
QMAKE_CFLAGS += -O3 -march=core-avx-i

Very good clue. Gained ~33% by doing this (using primecoind on linux though, just changed the xCXXFLAGS to -O3 -march=corei7 etc in the makefile). Before my i7 930 peaked around 150, highest so far is 200 (7 cores). Grin

Look for your cpu type to put after march here: http://gcc.gnu.org/onlinedocs/gcc/i386-and-x86_002d64-Options.html

Why not use "native" flag?

eule

hero member

Activity: 756

Merit: 501

Op of "WTB Primecoins but not for reasonable prices" too: https://bitcointalksearch.org/topic/buying-your-xpm-primecoin-for-ltc-251946
Grin

mr_random

legendary

Activity: 1344

Merit: 1001

Quote from: lunchb0x on July 08, 2013, 05:15:58 PM

Quote from: Online24o0n on July 08, 2013, 05:01:06 PM

Buying PrimeCoin, PM me with your offers.

Wait, aren't you the one that called primecoin "another thread cluttering coin" at the beginning of this thread?

Yup. He is. Everybody get your pitchforks out!!

eule

hero member

Activity: 756

Merit: 501

lol... people, don't sell your precious primecoins for pocket money. Tongue

Quote from: ig0tik3d on July 08, 2013, 04:14:36 PM

QMAKE_CXXFLAGS += -O3 -march=core-avx-i
QMAKE_CFLAGS += -O3 -march=core-avx-i

Very good clue. Gained ~33%, maybe less since it's difficult to compare, by doing this (using primecoind on linux though, just changed the xCXXFLAGS to -O3 -march=corei7 etc in the makefile). Before my i7 930 peaked around 150, highest so far is 200 (7 cores). Grin

Look for your cpu type to put after march here: http://gcc.gnu.org/onlinedocs/gcc/i386-and-x86_002d64-Options.html

lunchb0x

newbie

Activity: 20

Merit: 0

Quote from: Online24o0n on July 08, 2013, 05:01:06 PM

Buying PrimeCoin, PM me with your offers.

Wait, aren't you the one that called primecoin "another thread cluttering coin" at the beginning of this thread?

com911

hero member

Activity: 518

Merit: 500

Who knows, what was start prime number on XPM? 1 or biggest available ATM?

teknohog

sr. member

Activity: 520

Merit: 253

555

Quote from: azwccc on July 08, 2013, 03:34:35 PM

Quote from: teknohog on July 08, 2013, 03:00:09 PM

Gentoo ebuild of the daemon here, for the true believers. Modified from bitcoind 0.8.2 ebuild.

https://github.com/teknohog/ebuilds/tree/master/net-p2p/primecoind

Any guide on how to use it?

You use Gentoo and you need to ask? Perhaps start with these (the first two I googled)

http://linuxreviews.org/gentoo/ebuilds/
https://www.gentoo.org/doc/en/handbook/handbook-x86.xml?part=3&chap=5

Quote from: ReCat on July 08, 2013, 04:04:16 PM

Quote from: Vorksholk on July 08, 2013, 03:10:06 PM

I'll put up one Bitcoin in bounty for a CPU-miner (standalone) that will run on both windows and linux and outperform the built-in miner.

Recompile the existing one with some more optimization flags or with the ICC compiler. Tongue

Hence the ebuild Cheesy

(I do have a standalone build script for a few coins, but I've had issues with it recently, and the ebuild way was easier to fix.)

nmersulypnem

full member

Activity: 238

Merit: 100

Quote from: pgbit on July 08, 2013, 04:53:50 PM

Quote from: nmersulypnem on July 08, 2013, 04:48:09 PM

Quote from: pgbit on July 08, 2013, 04:46:18 PM

"There are a few who seek primes just for the money. There are prizes for the first prover ten-million digit prime ($100000), the first hundred-million digit prime ($150000), and the first billion digit prime ($250000)."

primecoin is not designed to find the largest prime (millions of digits long). It will not attempt to look for the ones listed in that competition.

If it was possible computationally to achieve this with Primecoin, why not? A small donation to the humble miner(s) helping to find the block...

primes at that size are too computationally intensive. Remember, every other node on the network needs to validate the work, so it becomes prohibitive to look for very large primes. The "value" of this program is to look for chains of primes at smaller values. I'm not sure if it will have any real value though since they will be a random smattering of chains... Specifically Cunningham chains which have already been described by mathematicians. What's the value in finding more of them? It's sort of like finding your second starfish on the beach.

Online24o0n

member

Activity: 98

Merit: 10

"Ignorance never settles a question."

Buying PrimeCoin, PM me with your offers.

Aggrophobia

legendary

Activity: 1106

Merit: 1001

Quote from: laughingbear on July 08, 2013, 04:48:40 PM

Quote from: pgbit on July 08, 2013, 04:46:18 PM

"There are a few who seek primes just for the money. There are prizes for the first prover ten-million digit prime ($100000), the first hundred-million digit prime ($150000), and the first billion digit prime ($250000)."

Sunny King you brilliant bastard!

If he were brilliant he would integrate GIMPS in it to get the possiblity to achieve some of these prices

~~ten-million digit prime ($100000)~~ was achieved 2009

The primes we calculate will never reach the ones proven by GIMPS

to prove that 2^57,885,161-1 is prime it took 88 hours with a GTX Titan, 200 hours with a 3570k

TheMightyX

sr. member

Activity: 350

Merit: 250

Vires in Numeris

is anyone working on a mining pool?
almost 24 hours and nada on my i7 920, yet some shmuck's mined 3 blocks on his crap laptop? ><

Topic: [XPM] [ANN] Primecoin Release - First Scientific Computing Cryptocurrency - page 170. (Read 688812 times)