[XPM] [ANN] Primecoin High Performance | HP14 released!

[Armchair Miner]

I pushed some big optimizations for -hp10 to github so that more people can test them.

Any chance of getting a compiled Windows binary? I'd like to test on a 32 core Opteron rig and a 3930k.

I second that request for the compiled Windows binary... Or I'll have buy more RAM and go linux!

[1503100553]

1503100553

[1503100553]

1503100553

[1503100553]

1503100553

[arnuschky]

I second that request for the compiled Windows binary... Or I'll have buy more RAM and go linux!

What, more RAM for Linux??

[Tamis]

It will come, give the man some time.
Why more ram ?

[Armchair Miner]

Yeah, I'm giving the man the time..  More RAM because I will still do VM linux inside Windows. Any preferences for the good VM?

[shazbits]

Running in vmware

Getting
"chainspermin" : 10
"chainsperday" : 1.14342394
"primespersec" : 1157

With
"genproclimit" : 4

How do these numbers look to you?

[Tamis]

blockchain download is VERY slow right now, my guess is that a lot of people just compiled and we are all downloading at the same time.

I'm sure there is a way to keep the already downloaded blockchain with hp9, but I wasn't sure so started fresh after compiling.

edit :

Here is what I get with HP10 on one server :

CPD went from 1.31 to 2.75
PPS went from 2k to 2.7k

Next is tweaking sieveextensions...

[mikaelh]

-hp10 released! Downloads are on the first page as usual.

Changes in -hp10:
 * Major rewrite of the sieve to support extending the sieve (originally implemented by jh000).
 * Added a new parameter 'sieveextensions' which controls how many times the sieve is extended.
 * Removed the problematic and unnecessary fast-division test.
 * Changed the value of nL1CacheElements to 224000 which is slightly faster (thanks to nushor on IRC).
 * The 'roundsievepercentage' parameter is gone. The round primorial is now automatically adjusted.
 * New RPC command 'listtopprimes' by Sunny.

[Trillium]

-hp10 released! Downloads are on the first page as usual.

Changes in -hp10:
 * Major rewrite of the sieve to support extending the sieve (originally implemented by jh000).
 * Added a new parameter 'sieveextensions' which controls how many times the sieve is extended.
 * Removed the problematic and unnecessary fast-division test.
 * Changed the value of nL1CacheElements to 224000 which is slightly faster (thanks to nushor on IRC).
 * The 'roundsievepercentage' parameter is gone. The round primorial is now automatically adjusted.
 * New RPC command 'listtopprimes' by Sunny.

Thanks for the update!

[reb0rn21]

gr8 work as usual !

I presume we gona see a lot more development in future, looks like there is always something new to add and optimize even just for CPU

[Entz]

Great work indeed, thanks!

CPD is more than double what it was before.

[atariguy]

blockchain download is VERY slow right now, my guess is that a lot of people just compiled and we are all downloading at the same time.

I'm sure there is a way to keep the already downloaded blockchain with hp9, but I wasn't sure so started fresh after compiling.

If you're using Windows, simply replace primecoind and primecoin-qt and start it.

[Tamis]

i7 3370k @ 4400 :

{
"blocks" : 132521,
"chainspermin" : 38,
"chainsperday" : 3.16242444,
"currentblocksize" : 1000,
"currentblocktx" : 0,
"difficulty" : 9.78874415,
"errors" : "",
"generate" : true,
"genproclimit" : -1,
"primespersec" : 3267,
"pooledtx" : 0,
"sieveextensions" : 6,
"sievepercentage" : 10,
"sievesize" : 1000000,
"testnet" : false
}

Very nice work mikaelh !

[atariguy]

Great work indeed, thanks!

CPD is more than double what it was before.

Just hope the same thing doesn't happen to the difficulty now... 

[ivanlabrie]

I start primecoin and it crashes...  

Used to work fine and suddenly started acting out...replaced with the new hp10 and same thing.
Gonna have to fiddle some more with it. ARGH

[Tamis]

Difficulty is clearly on the rise again !

[redphlegm]

Great work indeed, thanks!

CPD is more than double what it was before.

Just hope the same thing doesn't happen to the difficulty now... 

Pretty sure this will push us into 10-chain territory.

[cabin]

Is there a guide on how to compile this on windows?

[manawenuz]

I pushed some big optimizations for -hp10 to github so that more people can test them.

it was one hell of a optimiziation on my 2650s i went from ~2.6 to 5.9-6.0 ch/d
i got a meaningful boost on my 5636 and 5620s as well , but on my Pentium something something (socket 1156) i didn't get any boost.
i guess it's either because it's only dual core or that it lacks some instructions (i doubt it )

[jjiimm_64]

If you want to get an estimation of your block/day :
Take one minus the decimal part of the difficulty and multiply it with your chain/day.

block/day = (1 - dec(difficulty) ) * chainsperday

From your example :
    "chainsperday" : 0.28633614,
    "difficulty" : 9.68146938,
block/day = (1 - 0.68146938) * 0.28633614 = 0,091206828 = around one block every 11 days !

firstly, finally a forum search where i found what I was looking for

secondly..  since we dont use the whole number of the diff, can I assume that the chainsperday calculation takes into account the whole number?

with the example

From your example :
    "chainsperday" : 0.28633614,
    "difficulty" : 9.68146938,
block/day = (1 - 0.68146938) * 0.28633614 = 0,091206828 = around one block every 11 days !

what about "difficulty" : 10.68146938,?   what am i missing?

[glongsword]

...
secondly..  since we dont use the whole number of the diff, can I assume that the chainsperday calculation takes into account the whole number?
...
what about "difficulty" : 10.68146938,?   what am i missing?

I believe the chainsperday is the number of integer-part-of-the-difficulty-length chains the miner expects to find each day.  I have no idea how it actually works, so maybe this is a gross simplification, but I believe it finds about that number of 10 length chains in your example, but only (1-.68146938)*100% of those chains that it finds are going to be accepted by the other miners (they have to have some property that only that fraction of the chains satisfy).  At least that makes sense if the formula is correct, which I'm not sure of.