[ANN][GAP] Gapcoin - Prime Gap Search - New Math Algo - CPU / GPU - Zero Premine

[Supercomputing]

This seems to be normal for an optimized miner.

The miner does not stop mining the sieve until the original one is verified by a slow  mpz_nextprime() call.  When your miner is significantly faster than the PoW function you tend to find lots of shares before the PoW can stop the sieve from mining.

Regards,

--
bsunau7

@bsunau7

Thanks for the info and I will implement a small sieve for validation and use a single exponentiation test at each end point; it should be at least 100x faster and less than 1/(2¹²⁸) probability of it being a larger gap.

[bsunau7]

@bsunau7

Thanks for the info and I will implement a small sieve for validation and use a single exponentiation test at each end point; it should be at least 100x faster and less than 1/(2¹²⁸) probability of it being a larger gap.

There was talk on GMP mailing lists about speeding up gmp_nextprime in a very similar manner, sieve the start/end gap before running expensive MR tests.    As the gap has been pre-sieved you might just be able to MR test the non-composite in the mining sieve and get the same result.

I personally didn't bother with the PoW validation code as for larger gaps it just isn't called enough to matter.

Also you can tweak the pprocessor->process() to terminate the sieve early which will cure the stales and get you to the next "block" a few seconds faster at the risk of having a slightly wrong difficulty (in effect a non issue).

PS.  Just pushed testnet to >8 might slow down your stale shares some (I am only getting only 75-90% stale at this difficulty).

Regards,

--
bsunau7

[qwep]

Dev Is it possible to reduce the load on the processor

[drumingspz]

hi guys, can you please post a sample gapcoin.conf please?

also, has anyone with a radeon 270x tried the gpu miner? if yes, care to share your results?

Code:

rpcuser=user
rpcpassword=password
server=1
addnode=212.227.143.155
addnode=104.131.69.82
addnode=95.215.44.9
addnode=155.254.36.179
addnode=199.127.226.192

thanks!

i get the following when running gpu miner with stratum(both 32&64):

Code:

E:\gap\windows\64\gpu>gapminer-gpu.exe -o mine3.gap.nonce-pool.com -p 3385 -u user.1 -x password -t 1 --stratum --use-gpu
[2014-11-07 23:24:48] Found platform[0] name = AMD Accelerated Parallel Processi
ng
[2014-11-07 23:24:48] Found 1 device(s)
[2014-11-07 23:24:48] Loaded kernel binary size = 872620 bytes
[2014-11-07 23:24:56] Using GPU 0 [Pitcairn]: which has 20 CUs
[2014-11-07 23:24:56] Requesting work failedsend: The file handle supplied is no
t valid
[2014-11-07 23:24:56] Error receiving message form server: read_until: The file
handle supplied is not valid

getwork mines ok at ~141000 pps

I see a similar issues - Win7x64 64 bit miner (stratum connection)

Code:

D:\Downloads\Mining\GAPCoin\GPU Miner\windows\64\gpu>gapminer-gpu -o mine3.gap.n
once-pool.com -p 3385 -u username -x password -t 1 -g --stratum
[2014-11-07 15:53:02] Found platform[0] name = AMD Accelerated Parallel Processing
[2014-11-07 15:53:02] Found 1 device(s)
[2014-11-07 15:53:02] Loaded kernel binary size = 872620 bytes
[2014-11-07 15:53:02] Using GPU 0 [Tahiti]: which has 32 CUs
[2014-11-07 15:53:02] Requesting work failedsend: The file handle supplied is not valid
[2014-11-07 15:53:02] Error receiving message form server: read_until: The file handle supplied is not valid

Same system - longpoll connection.  Miner looks like its beginning to work, then dies...

Code:

D:\Downloads\Mining\GAPCoin\GPU Miner\windows\64\gpu>gapminer-gpu -o mine3.gap.n
once-pool.com -p 4200 -u username -x password -t 1 -g
[2014-11-07 15:54:41] Found platform[0] name = AMD Accelerated Parallel Processing
[2014-11-07 15:54:41] Found 1 device(s)
[2014-11-07 15:54:41] Loaded kernel binary size = 872620 bytes
[2014-11-07 15:54:41] Using GPU 0 [Tahiti]: which has 32 CUs
[2014-11-07 15:54:42] Server supports longpoll
[2014-11-07 15:54:42] Got new target: 13.0000000 @ 22.4626094

I've tried increasing my CPU threads, as I only have 1 GPU in this machine, but that hasn't helped.  The system has 12GB of RAM, so I don't think that's the issue either.

[q327K091]

ok got passed OpenCL compiler error, now this

( replaced

// uint4 q[2] = {0, 0};
    uint4 q[2] = {0, 0, 0, 0};
)

yep that's experimental   initializing uint4 as uint2 is indeed experimental
(however it isn't obvious that opencl and cuda use the same definition for these structures...)

so NVIDIA users are shut out from this software for the time being ?  

[Palmdetroit]

Good job on the gpuminer

Tested GPU miner 7950s, worked great

tried old ubuntu box with a 6870 and:

Code:

[2014-11-07 20:14:01] pps: -2147483648 / -2147483648  10g/h -2147483648 / -21474                                                  83648  15g/h -2147483648 / -2147483648
[2014-11-07 20:14:11] pps: 0 / 0  10g/h 0 / 0  15g/h 0 / 0

Seemed to always be 0 and never submit share, issue with legacy gpu?

Thanks will test some other cards this weekend if able.

[dcct]

Tested GPU miner 7950s, worked great

How many pps?

[Palmdetroit]

Tested GPU miner 7950s, worked great

How many pps?

~150k-200k each, without OC , Windows, I would imagine that can be tweaked higher with current miners and improvements, also on a relative slow CPU - 8150

Miner seems CPU heavy

Didn't try to tweak it much.

[nonce-pool]

Until stratum works for windows miners we have opened another longpoll port on 4205 for gpu miners.

It has a higher difficulty for better pps and less stales.

[qwep]

Tested GPU miner 7950s, worked great

How many pps?

240k

[antonio8]

I'll admit that I am not always the brightest and this has been asked a ton.

I created 2 different folders one 280x (device 0) and one 270x (device 1). I have in 280x folder --use-gpu 0. I have in 270x folder --use-gpu 1.

The 280x runs fine. I start the bat in 270x (gpu 1) and the bat runs gpu 0, the 280x.

Can someone enlighten me please.

[qqqq]

I'll admit that I am not always the brightest and this has been asked a ton.

I created 2 different folders one 280x (device 0) and one 270x (device 1). I have in 280x folder --use-gpu 0. I have in 270x folder --use-gpu 1.

The 280x runs fine. I start the bat in 270x (gpu 1) and the bat runs gpu 0, the 280x.

Can someone enlighten me please.

-d 0 and -d 1 and --use-gpu without any numbers

[qqqq]

I got 3x290 and it seems work fine but than after some time the speed starts decreasing in 400 times. It's solo mining.

[antonio8]

I'll admit that I am not always the brightest and this has been asked a ton.

I created 2 different folders one 280x (device 0) and one 270x (device 1). I have in 280x folder --use-gpu 0. I have in 270x folder --use-gpu 1.

The 280x runs fine. I start the bat in 270x (gpu 1) and the bat runs gpu 0, the 280x.

Can someone enlighten me please.

-d 0 and -d 1 and --use-gpu without any numbers

Awesome man. I really appreciate that.

Got them up and running.

[ivcelmik]

I'll admit that I am not always the brightest and this has been asked a ton.

I created 2 different folders one 280x (device 0) and one 270x (device 1). I have in 280x folder --use-gpu 0. I have in 270x folder --use-gpu 1.

The 280x runs fine. I start the bat in 270x (gpu 1) and the bat runs gpu 0, the 280x.

Can someone enlighten me please.

-d 0 and -d 1 and --use-gpu without any numbers

Awesome man. I really appreciate that.

Got them up and running.

Can someone help me i get this eror http://postimg.org/image/i3evcadmn/

Thx in advance

[xen0s]

I'll admit that I am not always the brightest and this has been asked a ton.

I created 2 different folders one 280x (device 0) and one 270x (device 1). I have in 280x folder --use-gpu 0. I have in 270x folder --use-gpu 1.

The 280x runs fine. I start the bat in 270x (gpu 1) and the bat runs gpu 0, the 280x.

Can someone enlighten me please.

-d 0 and -d 1 and --use-gpu without any numbers

Awesome man. I really appreciate that.

Got them up and running.

Can someone help me i get this eror http://postimg.org/image/i3evcadmn/

Thx in advance

delete stratum option and change port

[alibabacool]

Until stratum works for windows miners we have opened another longpoll port on 4205 for gpu miners.

It has a higher difficulty for better pps and less stales.

Some miners stop work after 5 minutes on 4205 port, on 4200 this is not observed.

[ivcelmik]

I'll admit that I am not always the brightest and this has been asked a ton.

I created 2 different folders one 280x (device 0) and one 270x (device 1). I have in 280x folder --use-gpu 0. I have in 270x folder --use-gpu 1.

The 280x runs fine. I start the bat in 270x (gpu 1) and the bat runs gpu 0, the 280x.

Can someone enlighten me please.

-d 0 and -d 1 and --use-gpu without any numbers

Awesome man. I really appreciate that.

Got them up and running.

Can someone help me i get this eror http://postimg.org/image/i3evcadmn/

Thx in advance

delete stratum option and change port

it's working thx and how can in use all my 5 cards because it's working just Gpu[0]

[j0nn9]

I got 3x290 and it seems work fine but than after some time the speed starts decreasing in 400 times. It's solo mining.

The Miner currently uses the cpu for sieving, and is very memory hungry,
maybe you run out of memory and parts of the program where moved onto the disk (pagefile).

The memory consumption is so high, because a very lage sieve is used, with many primes,
and also a queue is used to keep the cpu busy while the gpu is processing the fermat tests.

You can alter a few parameters to reduce the memory load:

Code:

  -s  --sieve-size        the prime sieve size

  -w  --work-items        gpu work items (default 2048)

  -a  --max-primes        maximum sieve primes (for use with gpu)

  -z  --queue-size        the gpu waiting queue size (memory intensive)

The Sieve dynamically adjusts the sieve primes,
so that it opimaly generaets a new set of prime candidates while the gpu is testing a previous set.
The queue must have a minimum size of 2.

[bsunau7]

Hi Guys,

Here is a little fix which increased my block count (got your attention)?

Added TCP keep alives with a sub 5 minute period to the wallet.  Three lines of code in net.cpp and netbase.cpp and a tweak of the Linux kernel.

net.cpp diff:

Code:

--- net.cpp_orig        2014-11-07 19:38:26.941369345 +0100
+++ net.cpp     2014-11-07 19:42:11.077371850 +0100
@@ -1594,6 +1594,7 @@
     // Different way of disabling SIGPIPE on BSD
     setsockopt(hListenSocket, SOL_SOCKET, SO_NOSIGPIPE, (void*)&nOne, sizeof(int));
 #endif
+    setsockopt(hListenSocket, SOL_SOCKET, SO_KEEPALIVE, (void*)&nOne, sizeof(int));

 #ifndef WIN32
     // Allow binding if the port is still in TIME_WAIT state after

netbase.cp diff:

Code:

--- netbase.cpp_orig    2014-11-07 19:38:21.365369283 +0100
+++ netbase.cpp 2014-11-07 19:41:02.549371084 +0100
@@ -329,6 +329,8 @@
     int set = 1;
     setsockopt(hSocket, SOL_SOCKET, SO_NOSIGPIPE, (void*)&set, sizeof(int));
 #endif
+    int set = 1;
+    setsockopt(hSocket, SOL_SOCKET, SO_KEEPALIVE, (void*)&set, sizeof(int));

 #ifdef WIN32
     u_long fNonblock = 1;

Add into /etc/sysctl.conf

Code:

net.ipv4.tcp_keepalive_time = 240
net.ipv4.tcp_keepalive_intvl = 15
net.ipv4.tcp_keepalive_probes = 9

Then "sysctl -p /etc/sysctl.conf" to make it take effect.

I went from spending half my time in 45 minute limbo land mining dead blocks due to the wallet hanging to no issues in the last 24 hours (no wasted mining).

Just quick and dirty hack (this is code for "It could have been done in a nicer way") which seems to have sorted out a big source pain for my little 4-core mining rig.

Oh, this might also address stratum mining issues this coin seems to have.  Monotonic coins don't go 15+ minutes without a transactions (i.e network traffic) so they would not see this issue.

Regards,

--
bsunau7