Bitcoin Forum
May 17, 2024, 03:13:59 PM *
News: Latest Bitcoin Core release: 27.0 [Torrent]
 
   Home   Help Search Login Register More  
Pages: « 1 ... 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 99 100 101 102 103 104 105 106 107 ... 115 »
  Print  
Author Topic: [ANN] Bitfury ASIC sales in EU and Europe  (Read 250419 times)
MXRider
Sr. Member
****
Offline Offline

Activity: 466
Merit: 250



View Profile
September 13, 2013, 08:21:06 PM
 #1121

Thank you guys for testing. Seems like these instructions are working well.

Those limits need some fine tuning so you can try these limits too:

Noncerate    Speed
<0.3    OFF
0.3-0.6    52-53
0.6-0.9    53-54
>0.9    55
btc4life
Newbie
*
Offline Offline

Activity: 18
Merit: 0


View Profile
September 13, 2013, 08:53:04 PM
 #1122

I have about 15 chips that seem to fluctuate and I am not quite sure yet if I should turn them off yet.  I am going to check the stats periodically and keep track of the noncerate and disable them if the hash rate stays consistently low.   A few of them may remain enabled as they tend to spike from the less than .3 range to the .3 to .7 range.
arorts
Sr. Member
****
Offline Offline

Activity: 408
Merit: 250


View Profile
September 13, 2013, 09:10:34 PM
 #1123

Thank you guys for testing. Seems like these instructions are working well.

Those limits need some fine tuning so you can try these limits too:

Noncerate    Speed
<0.3    OFF
0.3-0.6    52-53
0.6-0.9    53-54
>0.9    55

What about knowing when to increase chips to something higher than 55?
When should we turn the fans on? At some point high temp starts to neutralize those gains based on my experience.
Foofighter
Hero Member
*****
Offline Offline

Activity: 882
Merit: 547


BTC Mining Hardware, Trading and more


View Profile
September 13, 2013, 09:12:01 PM
 #1124

its all here:
https://bitcointalk.org/index.php?topic=287590.0

ex official Canaan Distributor (Cryptouniverse)
pbigmoon
Newbie
*
Offline Offline

Activity: 59
Merit: 0


View Profile
September 13, 2013, 09:12:06 PM
 #1125

Can anyone tell me if this is normal?

No matter which h-card i put first it always runs slow around 13gh/s while the second one runs at a solid 20gh/s.

I have swapped the cards around  tried different slots on the m-board but same thing so they are both capable of running at least 20gh/s.

It is it worth my while trying to optimize the first card or is the m-board broke?
juhakall
Sr. Member
****
Offline Offline

Activity: 657
Merit: 250


View Profile WWW
September 13, 2013, 09:14:18 PM
 #1126

Thank you guys for testing. Seems like these instructions are working well.

Those limits need some fine tuning so you can try these limits too:

Noncerate    Speed
<0.3    OFF
0.3-0.6    52-53
0.6-0.9    53-54
>0.9    55

What about knowing when to increase chips to something higher than 55?

After you have done the tuning MXRider already mentioned, increase all chips that perform normally from 55 to 56. Give it a couple 5 minute runs to get multiple samples of .stat.log, and note the chips that still don't produce errors. Then drop all the others back to 55. I tried this with my starter kit, and found 3 chips that run fine at 56. But at least for me, the effect on hashrate was so small that I can't distinguish it from normal variance.

I'm currently developing an experimental social AI platform
achimsmile
Legendary
*
Offline Offline

Activity: 1225
Merit: 1000


View Profile
September 13, 2013, 09:28:21 PM
 #1127

Can anyone tell me if this is normal?

No matter which h-card i put first it always runs slow around 13gh/s while the second one runs at a solid 20gh/s.

I have swapped the cards around  tried different slots on the m-board but same thing so they are both capable of running at least 20gh/s.

It is it worth my while trying to optimize the first card or is the m-board broke?

Had the same problem here. Did yo try slot B1 + C1 instead of A1 + B1?

The config that works best with my boards is A1 + C1, don't ask me why, I tried all possibilities but hashrate kept falling sub 20gh/s on one board.

Now it's 30gh/s + 28gh/s steady, pencil modded (this needs a fan or you get too many errors).
pbigmoon
Newbie
*
Offline Offline

Activity: 59
Merit: 0


View Profile
September 13, 2013, 09:31:17 PM
 #1128

Can anyone tell me if this is normal?

No matter which h-card i put first it always runs slow around 13gh/s while the second one runs at a solid 20gh/s.

I have swapped the cards around  tried different slots on the m-board but same thing so they are both capable of running at least 20gh/s.

It is it worth my while trying to optimize the first card or is the m-board broke?

Had the same problem here. Did yo try slot B1 + C1 instead of A1 + B1?

The config that works best with my boards is A1 + C1, don't ask me why, I tried all possibilities but hashrate kept falling sub 20gh/s on one board.

Now it's 30gh/s + 28gh/s steady, pencil modded (this needs a fan or you get too many errors).

Currently running b1 and c1  also tried a few different ones when i have the time ill try all different combinations and write them down see if any is better Smiley
MXRider
Sr. Member
****
Offline Offline

Activity: 466
Merit: 250



View Profile
September 13, 2013, 10:11:53 PM
 #1129

Can anyone tell me if this is normal?

No matter which h-card i put first it always runs slow around 13gh/s while the second one runs at a solid 20gh/s.

I have swapped the cards around  tried different slots on the m-board but same thing so they are both capable of running at least 20gh/s.

It is it worth my while trying to optimize the first card or is the m-board broke?

Had the same problem here. Did yo try slot B1 + C1 instead of A1 + B1?

The config that works best with my boards is A1 + C1, don't ask me why, I tried all possibilities but hashrate kept falling sub 20gh/s on one board.

Now it's 30gh/s + 28gh/s steady, pencil modded (this needs a fan or you get too many errors).

Auto tune on? Try my instructions for manual tuning and paste the .stat.log to pastebin after 15 minutes of hashing.
pbigmoon
Newbie
*
Offline Offline

Activity: 59
Merit: 0


View Profile
September 13, 2013, 10:26:33 PM
 #1130

Can anyone tell me if this is normal?

No matter which h-card i put first it always runs slow around 13gh/s while the second one runs at a solid 20gh/s.

I have swapped the cards around  tried different slots on the m-board but same thing so they are both capable of running at least 20gh/s.

It is it worth my while trying to optimize the first card or is the m-board broke?

Had the same problem here. Did yo try slot B1 + C1 instead of A1 + B1?

The config that works best with my boards is A1 + C1, don't ask me why, I tried all possibilities but hashrate kept falling sub 20gh/s on one board.

Now it's 30gh/s + 28gh/s steady, pencil modded (this needs a fan or you get too many errors).

Auto tune on? Try my instructions for manual tuning and paste the .stat.log to pastebin after 15 minutes of hashing.

Is there a way to set all chips aIfDSo 55 at once or do i have to do 1 chip at a time?
Trupik
Member
**
Offline Offline

Activity: 136
Merit: 13



View Profile
September 13, 2013, 10:40:07 PM
 #1131

Is there a way to set all chips aIfDSo 55 at once or do i have to do 1 chip at a time?
You can use vim, and column mode. If you are not familiar with vim, then you are better off with downloading the file to your computer with WinSCP, modify it in some more user friendly GUI editor (I like Geany myself), and upload it back with WinSCP.
goxed
Legendary
*
Offline Offline

Activity: 1946
Merit: 1006


Bitcoin / Crypto mining Hardware.


View Profile
September 13, 2013, 11:35:57 PM
 #1132

Is there a way to set all chips aIfDSo 55 at once or do i have to do 1 chip at a time?
You can use vim, and column mode. If you are not familiar with vim, then you are better off with downloading the file to your computer with WinSCP, modify it in some more user friendly GUI editor (I like Geany myself), and upload it back with WinSCP.
some quick vim commands for best.cnf will change all lines

Replace the A with a
:%s/\tA/\ta/g

change speed from 57 to 56
:%s/\t57/\t56/g

Revewing Bitcoin / Crypto mining Hardware.
btc4life
Newbie
*
Offline Offline

Activity: 18
Merit: 0


View Profile
September 14, 2013, 03:14:58 AM
 #1133

I just spent a couple of hours fine tuning and I still get the best hash rate letting auto tune take care of everything.   I am getting 440 to 450 noncerate on a full rig with my workers set to a diff of 64.
goxed
Legendary
*
Offline Offline

Activity: 1946
Merit: 1006


Bitcoin / Crypto mining Hardware.


View Profile
September 14, 2013, 05:20:37 AM
 #1134

I just spent a couple of hours fine tuning and I still get the best hash rate letting auto tune take care of everything.   I am getting 440 to 450 noncerate on a full rig with my workers set to a diff of 64.
Autotune is the best for most cases. These chips are very interesting physically, they have certain pareto-optimal points of operation which includes Voltage, Temperature, and Probably some other random events Wink

Revewing Bitcoin / Crypto mining Hardware.
Micky25
Legendary
*
Offline Offline

Activity: 974
Merit: 1000



View Profile
September 14, 2013, 05:38:00 AM
 #1135

MXRider
Sr. Member
****
Offline Offline

Activity: 466
Merit: 250



View Profile
September 14, 2013, 07:11:08 AM
 #1136

I just spent a couple of hours fine tuning and I still get the best hash rate letting auto tune take care of everything.   I am getting 440 to 450 noncerate on a full rig with my workers set to a diff of 64.

Pencil mod? R01F changed?
spiccioli
Legendary
*
Offline Offline

Activity: 1378
Merit: 1003

nec sine labore


View Profile
September 14, 2013, 07:16:30 AM
 #1137

I just spent a couple of hours fine tuning and I still get the best hash rate letting auto tune take care of everything.   I am getting 440 to 450 noncerate on a full rig with my workers set to a diff of 64.


I agree, I've spent more than 10 hours until now, the problem is sometimes the unit spirals down for unknown reasons.

I've made a couple changes to the miner, but they're not enough for it not to happen.

spiccioli
jlsminingcorp
Hero Member
*****
Offline Offline

Activity: 493
Merit: 500


Hooray for non-equilibrium thermodynamics!


View Profile
September 14, 2013, 07:17:27 AM
 #1138

Is there a way to set all chips aIfDSo 55 at once or do i have to do 1 chip at a time?
You can use vim, and column mode. If you are not familiar with vim, then you are better off with downloading the file to your computer with WinSCP, modify it in some more user friendly GUI editor (I like Geany myself), and upload it back with WinSCP.
some quick vim commands for best.cnf will change all lines

Replace the A with a
:%s/\tA/\ta/g

change speed from 57 to 56
:%s/\t57/\t56/g

And if you don't like vi for some reason (it is arguably the best linux text editor, if not the most user friendly) then you can use sed from the command line e.g.
Code:
sed -i 's/AI/aI/g' best.cnf
Code:
sed -i 's/  57/  55/g' best.cnf
Just be careful to set your sed search string to something that will only match what you want to replace (e.g. don't replace every letter "a" in the file with "A").

Bitcoin can only be understood backwards, but it must be lived forwards.
BitFury ASIC miner hosted group buy [DONE MINING]
puffmais
Sr. Member
****
Offline Offline

Activity: 404
Merit: 250


View Profile
September 14, 2013, 07:40:46 AM
 #1139

Hi,
every night my bitfurys stopped working on eligius pool...but in the web interface stand the correct hashing rate...
everyone else has this problems? i must shut down, restart it some time than it works again...
pbigmoon
Newbie
*
Offline Offline

Activity: 59
Merit: 0


View Profile
September 14, 2013, 08:53:19 AM
 #1140

Hi,
every night my bitfurys stopped working on eligius pool...but in the web interface stand the correct hashing rate...
everyone else has this problems? i must shut down, restart it some time than it works again...

Has happened to me couple of times on Bitminter.
Pages: « 1 ... 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 99 100 101 102 103 104 105 106 107 ... 115 »
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!