Author

Topic: [ANN][BURST] Burst | Efficient HDD Mining | New 1.2.3 Fork block 92000 - page 959. (Read 2170895 times)

legendary
Activity: 3248
Merit: 1070
regtable, are you using the java miner for window?

can you post a screen
hero member
Activity: 588
Merit: 500
I believe it`s smarter to have multiple smaller files then one giant file:
- smaller file can be enhanced with the merge utility so you can have a HUGE stagger (= number of nonces in your file) requiring just one disk read per file;
- smaller files can be moved around if you have multiple hdd of dif size/free space;
- you can automate the generation process through scripts and all the files will be completed (so can be enhanced with the merge utility)


example:

40G - 2 files
Code:
14643563564564646359_60000000_81920_4096
14643563564564646359_60081920_81920_4096

Totals:
0 nonces overlap (0 KB)
163840 nonces total (40 GB)
40 disk seeks per Scoop (0.3 seconds @ 8ms)

40G - 4 smaller files (same stagger)
Code:
14643563564564646359_60000000_40960_4096
14643563564564646359_60040960_40960_4096
14643563564564646359_60081920_40960_4096
14643563564564646359_60122880_40960_4096

Totals:
0 nonces overlap (0 KB)
163840 nonces total (40 GB)
40 disk seeks per Scoop (0.3 seconds @ 8ms)


40G - 4 smaller files (enhaced with merge utility)
Code:
14643563564564646359_60000000_40960_40960
14643563564564646359_60040960_40960_40960
14643563564564646359_60081920_40960_40960
14643563564564646359_60122880_40960_40960

Totals:
0 nonces overlap (0 KB)
163840 nonces total (40 GB)
4 disk seeks per Scoop (0 seconds @ 8ms)



my 2c

Well 3 of my disks are done plotting and 2 more are like 80% done. May as well just let them run their course now.
I have 1tb done ish a320gb a 250gb and a 500gb to do so il utilise this method on the 3  remaining drives to 1tb+ then switch over to re plot the tb
legendary
Activity: 1582
Merit: 1019
011110000110110101110010
I believe it`s smarter to have multiple smaller files then one giant file:
- smaller file can be enhanced with the merge utility so you can have a HUGE stagger (= number of nonces in your file) requiring just one disk read per file;
- smaller files can be moved around if you have multiple hdd of dif size/free space;
- you can automate the generation process through scripts and all the files will be completed (so can be enhanced with the merge utility)


example:

40G - 2 files
Code:
14643563564564646359_60000000_81920_4096
14643563564564646359_60081920_81920_4096

Totals:
0 nonces overlap (0 KB)
163840 nonces total (40 GB)
40 disk seeks per Scoop (0.3 seconds @ 8ms)

40G - 4 smaller files (same stagger)
Code:
14643563564564646359_60000000_40960_4096
14643563564564646359_60040960_40960_4096
14643563564564646359_60081920_40960_4096
14643563564564646359_60122880_40960_4096

Totals:
0 nonces overlap (0 KB)
163840 nonces total (40 GB)
40 disk seeks per Scoop (0.3 seconds @ 8ms)


40G - 4 smaller files (enhaced with merge utility)
Code:
14643563564564646359_60000000_40960_40960
14643563564564646359_60040960_40960_40960
14643563564564646359_60081920_40960_40960
14643563564564646359_60122880_40960_40960

Totals:
0 nonces overlap (0 KB)
163840 nonces total (40 GB)
4 disk seeks per Scoop (0 seconds @ 8ms)



my 2c

Well 3 of my disks are done plotting and 2 more are like 80% done. May as well just let them run their course now.
hero member
Activity: 588
Merit: 500

oh sorry... iam hooked up with that floppy
it just crash, i think someone just sent malformed request, and i dont check it
hahahahahahahahahaha my bad 0ool users.
hero member
Activity: 1400
Merit: 505

oh sorry... iam hooked up with that floppy
it just crash, i think someone just sent malformed request, and i dont check it
hero member
Activity: 588
Merit: 500
@coinits & @regtable69
you both are giving your harddrives a run for the money

Code:
Your results (assuming all plots are completed): 
10976174530077743023_507843_4507843_8191: Number of nonces (4507843) is not a multiple of stagger size (8191)
No overlapping nonces

Totals:
0 nonces overlap (0 KB)
4507843 nonces total (1100 GB)
550 disk seeks per Scoop (4.4 seconds @ 8ms)


Code:
Your results (assuming all plots are completed): 
4651495086655359966_19532000_307999_1000: Number of nonces (307999) is not a multiple of stagger size (1000)
I see your point smaller staggers/more plot files per hdd plus optimised stagger is more read writesper second thus better efficient mining. so currently I am not perform8ng at best. ffs a re plot again
No overlapping nonces

Totals:
0 nonces overlap (0 KB)
307999 nonces total (75 GB)
307 disk seeks per Scoop (2.4 seconds @ 8ms)
I will have 4 hdds going by the time im finished plotting. I k ow itsballs out during plottingbuti have been able to do 1tb in just over a day.

You missed my point... look at the disk reads per scoop needed with that settings.

Couple of posts back I`ve posted some comparative result with diff. settings.. have a look.

https://bitcointalksearch.org/topic/m.8713947
sr. member
Activity: 560
Merit: 250

Same question .. i only got this

Error: Unable to get mining info from wallet :http://burst-pool.cryptoport.io/burst?requestType=getMiningInfo
Error: Unable to get mining info from wallet :http://burst-pool.cryptoport.io/burst?requestType=getMiningInfo
hero member
Activity: 644
Merit: 500
hero member
Activity: 588
Merit: 500
I am buying few HDDs , is it better to get 10 1TBs or 5 2TBs , or either will not make any difference ?
either way is my guess. 10tb of 4gb memorysticks would do same job lmao

or mllions of floppy disk...
insert disk #2514 into drive A: to continue mining...

chalqnge accepted. my auto miner floppy setup. drawbacks are manual re cycle of disk in draw but it auto loads :-)

very cool, lot of people will want that, how much do you sell ? how much floppy / minute it can auto-reload ?
read and writes 1 floppy per sec and reloads at same speed if im on form. after a drink theyr all on the floor. it holds approx 60 floppys with room for a belt feed system similar well hacked from a browning. il sell this one of a kind rig for oooh  1000btc
sr. member
Activity: 458
Merit: 250
beast at work
@coinits & @regtable69
you both are giving your harddrives a run for the money

Code:
Your results (assuming all plots are completed): 
10976174530077743023_507843_4507843_8191: Number of nonces (4507843) is not a multiple of stagger size (8191)
No overlapping nonces

Totals:
0 nonces overlap (0 KB)
4507843 nonces total (1100 GB)
550 disk seeks per Scoop (4.4 seconds @ 8ms)


Code:
Your results (assuming all plots are completed): 
4651495086655359966_19532000_307999_1000: Number of nonces (307999) is not a multiple of stagger size (1000)
No overlapping nonces

Totals:
0 nonces overlap (0 KB)
307999 nonces total (75 GB)
307 disk seeks per Scoop (2.4 seconds @ 8ms)
I will have 4 hdds going by the time im finished plotting. I k ow itsballs out during plottingbuti have been able to do 1tb in just over a day.

You missed my point... look at the disk reads per scoop needed with that settings.

Couple of posts back I`ve posted some comparative result with diff. settings.. have a look.

https://bitcointalksearch.org/topic/m.8713947
hero member
Activity: 1400
Merit: 505
tell me please! after i try run DCCT miner... they say me


Plot directory 8124 does not exist
Plot directory 80 does not exist

my cmd

./mine burst-pool.cryptoport.io 80 /media/sglasio/mine4/plotgenerator/plots /media/sglasio/mine5/plots
./mine burst-pool.cryptoport.io 8124 /media/sglasio/mine4/plotgenerator/plots /media/sglasio/mine5/plots

remove the port number, this miner doesn`t need it

you need it for my pool, as dcct original miner target port is hard coded to 8125, and my pool use 8124 or 80, 8125 is for solo mining


actually he is not calling the miner directly, he is using mine.sh. So he don`t have to put the pool address as is already inside the script file... just use:

Code:
./mine /media/sglasio/mine4/plotgenerator/plots/ /media/sglasio/mine5/plots/

that's right, its ./mine.sh if i am not wrong
hero member
Activity: 1400
Merit: 505
I am buying few HDDs , is it better to get 10 1TBs or 5 2TBs , or either will not make any difference ?
either way is my guess. 10tb of 4gb memorysticks would do same job lmao

or mllions of floppy disk...
insert disk #2514 into drive A: to continue mining...

chalqnge accepted. my auto miner floppy setup. drawbacks are manual re cycle of disk in draw but it auto loads :-)

very cool, lot of people will want that, how much do you sell ? how much floppy / minute it can auto-reload ?
hero member
Activity: 588
Merit: 500
@coinits & @regtable69
you both are giving your harddrives a run for the money

Code:
Your results (assuming all plots are completed):
10976174530077743023_507843_4507843_8191: Number of nonces (4507843) is not a multiple of stagger size (8191)
No overlapping nonces

Totals:
0 nonces overlap (0 KB)
4507843 nonces total (1100 GB)
550 disk seeks per Scoop (4.4 seconds @ 8ms)


Code:
Your results (assuming all plots are completed):
4651495086655359966_19532000_307999_1000: Number of nonces (307999) is not a multiple of stagger size (1000)
No overlapping nonces

Totals:
0 nonces overlap (0 KB)
307999 nonces total (75 GB)
307 disk seeks per Scoop (2.4 seconds @ 8ms)
I will have 4 hdds going by the time im finished plotting. I k ow itsballs out during plottingbuti have been able to do 1tb in just over a day.
sr. member
Activity: 458
Merit: 250
beast at work
tell me please! after i try run DCCT miner... they say me


Plot directory 8124 does not exist
Plot directory 80 does not exist

my cmd

./mine burst-pool.cryptoport.io 80 /media/sglasio/mine4/plotgenerator/plots /media/sglasio/mine5/plots
./mine burst-pool.cryptoport.io 8124 /media/sglasio/mine4/plotgenerator/plots /media/sglasio/mine5/plots

remove the port number, this miner doesn`t need it

you need it for my pool, as dcct original miner target port is hard coded to 8125, and my pool use 8124 or 80, 8125 is for solo mining


actually he is not calling the miner directly, he is using mine.sh. So he don`t have to put the pool address as is already inside the script file... just use:

Code:
./mine /media/sglasio/mine4/plotgenerator/plots/ /media/sglasio/mine5/plots/
legendary
Activity: 3248
Merit: 1070
i am thinking maybe if u mine using java 32 bit it will cap the ram use ?

i'm just using the miner in the "burst-pool-miner r2" folder, dunno about the other pool miner

i noticed that the ram usage rise when too much shares are submitted at the same time, something wrong in the code

shhh you should fix and mine in secret LOL.

so whats' the cause?
i am not using java miner myself other than just for testing, so i dont know what happen after long run
i am using dcct miner it only use hundreds MB of mem, mining using linux on sempron 145 with 2 GB of RAM, running fine without any issues

I had to tame my miner bat because Java had memory leaks and caused the miners to crash.

I added this to the bat: C:\Windows\SysWOW64\java -Xmx1000m -cp pocminer_pool.jar to stop the Java.lang.OutOfMemoryError: Java heap space error. I am mining and generating at the same time with 32 GB of RAM.


whaaaaa I am on xmx8000m with 8gb of ram no issues stagger size of 8191 and 1 core less than my cpu hqs 7 not 8 as maxingto 8 makes miner crash maybe thats whats causing some issues. tryrunning1 less core so yourpc isnt completley lobotomised

you are not on windows apparently
lol I defiantly am on windows I assure you that
I dont use c:windows blah blah just
Code:
java -Xmx8100m -cp pocminer.jar;lib/*;lib/akka/*;lib/jetty/* pocminer.POCMiner generate 10976174530077743023 507843 4507843 8191 7

that's is for plotting, i'm talking about mining

plotting is fine for me no mem used at all
oh ok well I posted my mining bat above too. as I say im running little ram compared to some and suffering no adverse issuse mining and generating at same time

how many tera?
sr. member
Activity: 458
Merit: 250
beast at work
@coinits & @regtable69
you both are giving your harddrives a run for the money

Code:
Your results (assuming all plots are completed):
10976174530077743023_507843_4507843_8191: Number of nonces (4507843) is not a multiple of stagger size (8191)
No overlapping nonces

Totals:
0 nonces overlap (0 KB)
4507843 nonces total (1100 GB)
550 disk seeks per Scoop (4.4 seconds @ 8ms)


Code:
Your results (assuming all plots are completed):
4651495086655359966_19532000_307999_1000: Number of nonces (307999) is not a multiple of stagger size (1000)
No overlapping nonces

Totals:
0 nonces overlap (0 KB)
307999 nonces total (75 GB)
307 disk seeks per Scoop (2.4 seconds @ 8ms)
hero member
Activity: 588
Merit: 500
i am thinking maybe if u mine using java 32 bit it will cap the ram use ?

i'm just using the miner in the "burst-pool-miner r2" folder, dunno about the other pool miner

i noticed that the ram usage rise when too much shares are submitted at the same time, something wrong in the code

shhh you should fix and mine in secret LOL.

so whats' the cause?
i am not using java miner myself other than just for testing, so i dont know what happen after long run
i am using dcct miner it only use hundreds MB of mem, mining using linux on sempron 145 with 2 GB of RAM, running fine without any issues

I had to tame my miner bat because Java had memory leaks and caused the miners to crash.

I added this to the bat: C:\Windows\SysWOW64\java -Xmx1000m -cp pocminer_pool.jar to stop the Java.lang.OutOfMemoryError: Java heap space error. I am mining and generating at the same time with 32 GB of RAM.


whaaaaa I am on xmx8000m with 8gb of ram no issues stagger size of 8191 and 1 core less than my cpu hqs 7 not 8 as maxingto 8 makes miner crash maybe thats whats causing some issues. tryrunning1 less core so yourpc isnt completley lobotomised

you are not on windows apparently
lol I defiantly am on windows I assure you that
I dont use c:windows blah blah just
Code:
java -Xmx8100m -cp pocminer.jar;lib/*;lib/akka/*;lib/jetty/* pocminer.POCMiner generate 10976174530077743023 507843 4507843 8191 7

that's is for plotting, i'm talking about mining

plotting is fine for me no mem used at all
oh ok well I posted my mining bat above too. as I say im running little ram compared to some and suffering no adverse issuse mining and generating at same time
hero member
Activity: 1400
Merit: 505
tell me please! after i try run DCCT miner... they say me


Plot directory 8124 does not exist
Plot directory 80 does not exist

my cmd

./mine burst-pool.cryptoport.io 80 /media/sglasio/mine4/plotgenerator/plots /media/sglasio/mine5/plots
./mine burst-pool.cryptoport.io 8124 /media/sglasio/mine4/plotgenerator/plots /media/sglasio/mine5/plots

remove the port number, this miner doesn`t need it

you need it for my pool, as dcct original miner target port is hard coded to 8125, and my pool use 8124 or 80, 8125 is for solo mining
sr. member
Activity: 349
Merit: 250
Nicely job on the multipool.

Thanks. Just as a reminder make sure to promote the pool so that the hash keeps going up. The pool needs to grow to be successful.

burstmp.com

i'm promoting via twitter...
but plz... do not show "Hidden coin" on http://burstmp.com/workers.

just show which coin you are mining for greater transparency!!!

 Wink  Wink  Wink  Wink

I got a lot of requests for this. Disclosing will mean smaller profits for miners. I doubt miners want this.
sr. member
Activity: 458
Merit: 250
beast at work
tell me please! after i try run DCCT miner... they say me


Plot directory 8124 does not exist
Plot directory 80 does not exist

my cmd

./mine burst-pool.cryptoport.io 80 /media/sglasio/mine4/plotgenerator/plots /media/sglasio/mine5/plots
./mine burst-pool.cryptoport.io 8124 /media/sglasio/mine4/plotgenerator/plots /media/sglasio/mine5/plots

remove the port number, this miner doesn`t need it
Jump to: