Bitcoin Forum
January 18, 2019, 07:00:27 PM *
News: Latest Bitcoin Core release: 0.17.1 [Torrent]
 
   Home   Help Search Login Register More  
Pages: « 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 [52] 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 ... 133 »
  Print  
Author Topic: [XPM] [ANN] Primecoin High Performance | HP14 released!  (Read 396703 times)
tyrion70
Legendary
*
Offline Offline

Activity: 934
Merit: 1000



View Profile
July 21, 2013, 08:55:39 PM
 #1021

-hp6 released!

Lots of changes in this release. Here's a list of notable changes:
 * Added fast divisibility tests before doing the expensive Fermat's test
 * Lots of other small optimizations
 * Introduced a new experimental tuning parameter "sievepercentage". Default value is 10, minimum is 1, maximum is 100.
 * Added new RPC command "getchainspermin"
 * Added the following information to "getmininginfo": chainspermin, difficulty, sievepercentage, and sievesize
 * Attempt to fix the random crash while mining

My PPS on mainnet went up 23% with this release with difficulty >= 9 on mainnet. This release attempts to fix the random crash while mining. I'm not sure if it's actually fixed because I have heard both positive and negative reports. The crash is more likely with high number of cores. 16 and 32 core machines are suffering the most.

The "sievepercentage" parameter is there for people to experiment with. It limits how many base primes are used to filter out candidate multipliers in the sieve. 8% seems optimal for testnet but I have no idea about mainnet.
This build is doing horrible for me. Dropped down to 9 blocks last hour, and 5 this one (almost at the end of the hour). Was getting between 8 and 20 the last build, even after diff 9. Gonna give it one more hour of variation, then switch back. My servers restart primecoind upon crashing, so could the reduction be from multiple crashes? Any explanation?
It seems that the total number of blocks found per hour changed drastically since diff 9. It now is around 3-4 blocks per minute as far as I can tell, while it was 5-6 per minute a few hours ago.. It seems difficulty is now getting where it should be.. PPS went down as well it seems; used to have 400k pps / 100 machines, now its 200k pps / 100, but still finding more than 3 blocks per 200k pps new style.. Definately getting harder though Smiley

1547838027
Hero Member
*
Offline Offline

Posts: 1547838027

View Profile Personal Message (Offline)

Ignore
1547838027
Reply with quote  #2

1547838027
Report to moderator
1547838027
Hero Member
*
Offline Offline

Posts: 1547838027

View Profile Personal Message (Offline)

Ignore
1547838027
Reply with quote  #2

1547838027
Report to moderator
1547838027
Hero Member
*
Offline Offline

Posts: 1547838027

View Profile Personal Message (Offline)

Ignore
1547838027
Reply with quote  #2

1547838027
Report to moderator
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction. Advertise here.
1547838027
Hero Member
*
Offline Offline

Posts: 1547838027

View Profile Personal Message (Offline)

Ignore
1547838027
Reply with quote  #2

1547838027
Report to moderator
1547838027
Hero Member
*
Offline Offline

Posts: 1547838027

View Profile Personal Message (Offline)

Ignore
1547838027
Reply with quote  #2

1547838027
Report to moderator
1547838027
Hero Member
*
Offline Offline

Posts: 1547838027

View Profile Personal Message (Offline)

Ignore
1547838027
Reply with quote  #2

1547838027
Report to moderator
notlist3d
Legendary
*
Offline Offline

Activity: 1470
Merit: 1000



View Profile
July 21, 2013, 09:00:31 PM
 #1022

-hp6 released!

Lots of changes in this release. Here's a list of notable changes:
 * Added fast divisibility tests before doing the expensive Fermat's test
 * Lots of other small optimizations
 * Introduced a new experimental tuning parameter "sievepercentage". Default value is 10, minimum is 1, maximum is 100.
 * Added new RPC command "getchainspermin"
 * Added the following information to "getmininginfo": chainspermin, difficulty, sievepercentage, and sievesize
 * Attempt to fix the random crash while mining

My PPS on mainnet went up 23% with this release with difficulty >= 9 on mainnet. This release attempts to fix the random crash while mining. I'm not sure if it's actually fixed because I have heard both positive and negative reports. The crash is more likely with high number of cores. 16 and 32 core machines are suffering the most.

The "sievepercentage" parameter is there for people to experiment with. It limits how many base primes are used to filter out candidate multipliers in the sieve. 8% seems optimal for testnet but I have no idea about mainnet.
This build is doing horrible for me. Dropped down to 9 blocks last hour, and 5 this one (almost at the end of the hour). Was getting between 8 and 20 the last build, even after diff 9. Gonna give it one more hour of variation, then switch back. My servers restart primecoind upon crashing, so could the reduction be from multiple crashes? Any explanation?
It seems that the total number of blocks found per hour changed drastically since diff 9. It now is around 3-4 blocks per minute as far as I can tell, while it was 5-6 per minute a few hours ago.. It seems difficulty is now getting where it should be.. PPS went down as well it seems; used to have 400k pps / 100 machines, now its 200k pps / 100, but still finding more than 3 blocks per 200k pps new style.. Definately getting harder though Smiley

I'm in PPS envy.
CoinBuzz
Sr. Member
****
Offline Offline

Activity: 476
Merit: 250



View Profile
July 21, 2013, 09:07:13 PM
 #1023

did you hard coded that path of primecoin.conf ?!

i changed the path of data folder in my run.bat so the blockchain and wallet is in some other places ( in my D drive ).

I make a primecoin.conf there and changed the sievesize to 2milion, but when i run the program, sievesize is still 1milion.

Tuck Fheman
Sr. Member
****
Offline Offline

Activity: 363
Merit: 250


View Profile WWW
July 21, 2013, 09:15:10 PM
 #1024

It's looking like mining is getting significantly less profitable without a gpu miner.

says the guy with 400-500k pps. lulz

notlist3d
Legendary
*
Offline Offline

Activity: 1470
Merit: 1000



View Profile
July 21, 2013, 09:17:36 PM
 #1025

did you hard coded that path of primecoin.conf ?!

i changed the path of data folder in my run.bat so the blockchain and wallet is in some other places ( in my D drive ).

I make a primecoin.conf there and changed the sievesize to 2milion, but when i run the program, sievesize is still 1milion.


Same here mine says the default.
maco
Sr. Member
****
Offline Offline

Activity: 294
Merit: 250



View Profile
July 21, 2013, 09:42:20 PM
 #1026

I still don't understand how I overcame this error few days ago and it keeps coming back -- any ideas?  I tried swap commands but I guess the server I am has it locked and does not allow swapping, but it has enough memory to processor the make command: make -f makefile.unix USE_UPNP=-
I even tried make clean command and still no progress when I launch this. Any updates or ideas?

g++: internal compiler error: Killed (program cc1plus)
Please submit a full bug report,
with preprocessed source if appropriate.
See <file:///usr/share/doc/gcc-4.6/README.Bugs> for instructions.
make: *** [obj/alert.o] Error 4
Tuck Fheman
Sr. Member
****
Offline Offline

Activity: 363
Merit: 250


View Profile WWW
July 21, 2013, 09:43:45 PM
 #1027

Does reducing the # of threads increase chainspermin for anyone else?

On my i7 I started using 6 threads and saw 8-12 CPMs.

When I switched to -1 it dropped to 3-6 CPMs.

I changed back to 6 and it increased again.

GSnak
Full Member
***
Offline Offline

Activity: 182
Merit: 100



View Profile
July 21, 2013, 09:44:52 PM
 #1028

It's looking like mining is getting significantly less profitable without a gpu miner.

says the guy with 400-500k pps. lulz

If everyone had a GPU miner... you know how that would go.
mikaelh
Sr. Member
****
Offline Offline

Activity: 301
Merit: 250


View Profile
July 21, 2013, 10:00:26 PM
 #1029

difficulty 9 already, that's the reason

What's the reason it is only 9?
I'm trying to get my head around the idea that Quarkcoin launched a few hours ago and is already 57 difficultly.*

Is that an apples to oranges comparison? Surely Quarkcoin can't have 6 times the network strength of Primecoin.

As Primecoin difficulty rises, the supply of coins really gets choked off - at difficultly of 32 . . . that's 999/(32*32)= .975 coins per block.

*Oh, it's gone up to 63 since I wrote those two sentences.

It is very much like comparing apples and oranges. For Primecoin, the integer part of difficulty determines the required probable prime chain length. E.g. if difficulty is 9.07233632, a probable prime chain of length 9 is required. The fractional part is used to determine how many length-9 chains are accepted. It scales logarithmically so that going from 9.8 to 9.9 is a lot harder than going from 9.1 to 9.2. The details for the fractional part are in Sunny King's paper.

The current algorithm also does more work for longer target chain lengths. That's why PPS numbers are also dropping.
rethaw
Sr. Member
****
Offline Offline

Activity: 378
Merit: 250



View Profile
July 21, 2013, 10:08:19 PM
 #1030

Does reducing the # of threads increase chainspermin for anyone else?

On my i7 I started using 6 threads and saw 8-12 CPMs.

When I switched to -1 it dropped to 3-6 CPMs.

I changed back to 6 and it increased again.

I wasn't able to replicate this. Maybe this has something to do with hyperthreading on your processor. Is it a quad-core?

ivanlabrie
Hero Member
*****
Offline Offline

Activity: 812
Merit: 1000



View Profile
July 21, 2013, 10:21:44 PM
 #1031

Does reducing the # of threads increase chainspermin for anyone else?

On my i7 I started using 6 threads and saw 8-12 CPMs.

When I switched to -1 it dropped to 3-6 CPMs.

I changed back to 6 and it increased again.

I wasn't able to replicate this. Maybe this has something to do with hyperthreading on your processor. Is it a quad-core?

It's a well known fact that you see a boost in flops when disabling HT...but I'm not sure if that would be a good idea or not in this case. (intel burn test surely likes that)
redphlegm
Sr. Member
****
Offline Offline

Activity: 246
Merit: 250


My spoon is too big!


View Profile
July 21, 2013, 10:32:10 PM
 #1032

I still don't understand how I overcame this error few days ago and it keeps coming back -- any ideas?  I tried swap commands but I guess the server I am has it locked and does not allow swapping, but it has enough memory to processor the make command: make -f makefile.unix USE_UPNP=-
I even tried make clean command and still no progress when I launch this. Any updates or ideas?

g++: internal compiler error: Killed (program cc1plus)
Please submit a full bug report,
with preprocessed source if appropriate.
See <file:///usr/share/doc/gcc-4.6/README.Bugs> for instructions.
make: *** [obj/alert.o] Error 4


RAM.

Whiskey Fund: (BTC) 1whiSKeYMRevsJMAQwU8NY1YhvPPMjTbM | (Ψ) ALcoHoLsKUfdmGfHVXEShtqrEkasihVyqW
FreeTrade
Legendary
*
Offline Offline

Activity: 1162
Merit: 1010



View Profile
July 21, 2013, 10:41:03 PM
 #1033

It is very much like comparing apples and oranges. For Primecoin, the integer part of difficulty determines the required probable prime chain length. E.g. if difficulty is 9.07233632, a probable prime chain of length 9 is required. The fractional part is used to determine how many length-9 chains are accepted. It scales logarithmically so that going from 9.8 to 9.9 is a lot harder than going from 9.1 to 9.2. The details for the fractional part are in Sunny King's paper.

The current algorithm also does more work for longer target chain lengths. That's why PPS numbers are also dropping.

Thanks Mikael, I read that it takes a 30 fold increase in processing power to increase difficulty by 1 . . . if so difficult to imagine difficultly ever climbing beyond 12 or 13.

999/(13*13) =  5.9 per block

HODLCoin ANN - 5% Interest. No Staking Req. Term Deposits 10%. Solo Mining.http://hodlcoin.com/
Tamis
Sr. Member
****
Offline Offline

Activity: 476
Merit: 250



View Profile
July 21, 2013, 10:45:51 PM
 #1034

difficulty is rising fast ! gpu miner out ?
maco
Sr. Member
****
Offline Offline

Activity: 294
Merit: 250



View Profile
July 21, 2013, 10:46:13 PM
 #1035

HP6 dropped me to 800 pps instead of 3k pps normally.
jubalix
Legendary
*
Offline Offline

Activity: 1946
Merit: 1002


View Profile WWW
July 21, 2013, 10:48:23 PM
 #1036

did we all see the PPS drop

I went from 14k to 8k....or is that hp6 v hp5 effect?

Admitted Practicing Lawyer::BTC/Crypto Specialist. B.Engineering/B.Laws

https://www.binance.com/?ref=10062065
redphlegm
Sr. Member
****
Offline Offline

Activity: 246
Merit: 250


My spoon is too big!


View Profile
July 21, 2013, 10:49:08 PM
 #1037

I feel like I'm sort of flying blind on sievepercentage. I don't know what it's supposed to be changing so I don't know whether to tweak it up or down. Comparatively speaking, I don't know if running default hp5 is better / worse than default hp6 for me since pps isn't a great metric to use.

Any suggestions / insight?

Whiskey Fund: (BTC) 1whiSKeYMRevsJMAQwU8NY1YhvPPMjTbM | (Ψ) ALcoHoLsKUfdmGfHVXEShtqrEkasihVyqW
jubalix
Legendary
*
Offline Offline

Activity: 1946
Merit: 1002


View Profile WWW
July 21, 2013, 11:19:00 PM
Last edit: July 21, 2013, 11:32:37 PM by jubalix
 #1038

just tried to compile hp6, go error


Code:
:~$ cd ~/primecoin/src
:~/primecoin/src$ sudo make -f makefile.unix USE_UPNP=-
Building LevelDB ...
make[1]: Entering directory `/home/ubuntu/primecoin/src/leveldb'
g++ -I. -I./include -fno-builtin-memcmp -pthread -DOS_LINUX -DLEVELDB_PLATFORM_POSIX -O2 -pthread -Wall -Wextra -Wformat -Wformat-security -Wno-unused-parameter -g -DBOOST_SPIRIT_THREADSAFE -D_FILE_OFFSET_BITS=64 -I/home/ubuntu/primecoin/src -I/home/ubuntu/primecoin/src/obj -DUSE_IPV6=1 -I/home/ubuntu/primecoin/src/leveldb/include -I/home/ubuntu/primecoin/src/leveldb/helpers -DHAVE_BUILD_INFO -fno-stack-protector -fstack-protector-all -Wstack-protector -D_FORTIFY_SOURCE=2  -c db/._builder.cc -o db/._builder.o
db/._builder.cc:1:1: warning: null character(s) ignored [enabled by default]
db/._builder.cc:1:2: error: stray ‘\5’ in program
db/._builder.cc:1:2: error: stray ‘\26’ in program
db/._builder.cc:1:2: error: stray ‘\7’ in program
.....

db/._builder.cc:1:9: error: ‘Mac’ does not name a type
db/._builder.cc:1:158: error: expected unqualified-id before numeric constant
db/._builder.cc:1:167: error: ‘Firefox’ does not name a type
db/._builder.cc:1:3827: error: ‘This’ does not name a type
make[1]: *** [db/._builder.o] Error 1
make[1]: Leaving directory `/home/ubuntu/primecoin/src/leveldb'
make: *** [leveldb/libleveldb.a] Error 2


any ideas on how to fix this???

skay fixed

Admitted Practicing Lawyer::BTC/Crypto Specialist. B.Engineering/B.Laws

https://www.binance.com/?ref=10062065
dudeguy
Member
**
Offline Offline

Activity: 182
Merit: 10



View Profile
July 21, 2013, 11:25:26 PM
 #1039

-hp6 released!

Lots of changes in this release. Here's a list of notable changes:
 * Added fast divisibility tests before doing the expensive Fermat's test
 * Lots of other small optimizations
 * Introduced a new experimental tuning parameter "sievepercentage". Default value is 10, minimum is 1, maximum is 100.
 * Added new RPC command "getchainspermin"
 * Added the following information to "getmininginfo": chainspermin, difficulty, sievepercentage, and sievesize
 * Attempt to fix the random crash while mining

My PPS on mainnet went up 23% with this release with difficulty >= 9 on mainnet. This release attempts to fix the random crash while mining. I'm not sure if it's actually fixed because I have heard both positive and negative reports. The crash is more likely with high number of cores. 16 and 32 core machines are suffering the most.

The "sievepercentage" parameter is there for people to experiment with. It limits how many base primes are used to filter out candidate multipliers in the sieve. 8% seems optimal for testnet but I have no idea about mainnet.
This build is doing horrible for me. Dropped down to 9 blocks last hour, and 5 this one (almost at the end of the hour). Was getting between 8 and 20 the last build, even after diff 9. Gonna give it one more hour of variation, then switch back. My servers restart primecoind upon crashing, so could the reduction be from multiple crashes? Any explanation?

Can we change "sievepercentage" from debug console? My primecoin.conf is not working no matter what I try. I've literally tried everything.
LlamaMaster
Newbie
*
Offline Offline

Activity: 51
Merit: 0


View Profile
July 21, 2013, 11:37:03 PM
 #1040

My pps went down after I changed the sievesize from 1M to 2M.  Should I be worried, or is this "normal"?
Pages: « 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 [52] 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 ... 133 »
  Print  
 
Jump to:  

Bitcointalk.org is not available or authorized for sale. Do not believe any fake listings.
Sponsored by , a Bitcoin-accepting VPN.
Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!