Author

Topic: Low Hash score and wu on S9 (Read 1177 times)

full member
Activity: 125
Merit: 100
Canadian Miner Passing Through~~
April 17, 2017, 08:39:18 AM
#12
Pushed receive, will do.  Grin
Thanks
hero member
Activity: 756
Merit: 560
April 13, 2017, 05:34:05 PM
#11
well look at it this way.

3 boards not hashing for a month

or

1 board not hashing for 12 months+

It will take 9 months to cover that month of downtime but the S9 is assuredly going to be on the market as the top dog for at least another year. There is nowhere for the tech to go right now as 10nm is 2+ years out for a boutique manufacturer to even tape out and test.

Personally I would warranty it.
full member
Activity: 125
Merit: 100
Canadian Miner Passing Through~~
April 12, 2017, 03:44:46 PM
#10
Its a pretty simple thing to figure out on your own.

1. Are you under warranty?

2. If not, are you willing to pay like $300+ to fix the board?





Yes i am under warranty but it is worth it of losing 1 month of mining for 1 downed hashboard?
I am more on the yes side of sending it to be repaired but still need a little push XD
hero member
Activity: 756
Merit: 560
April 12, 2017, 03:40:39 PM
#9
Its a pretty simple thing to figure out on your own.

1. Are you under warranty?

2. If not, are you willing to pay like $300+ to fix the board?



full member
Activity: 125
Merit: 100
Canadian Miner Passing Through~~
April 12, 2017, 03:36:46 PM
#8
So i found the defective board and unplugged it. so i save 400watts  Smiley

http://imgur.com/a/aJMnk

But not sure if i should get it repaired or not. Should i just keep it mining or sent it to be repaired?

Thanks

Bump for advise?
full member
Activity: 125
Merit: 100
Canadian Miner Passing Through~~
April 10, 2017, 10:46:54 AM
#7
So i found the defective board and unplugged it. so i save 400watts  Smiley

http://imgur.com/a/aJMnk

But not sure if i should get it repaired or not. Should i just keep it mining or sent it to be repaired?

Thanks
full member
Activity: 125
Merit: 100
Canadian Miner Passing Through~~
April 06, 2017, 01:13:13 PM
#6
Wow i never noticed the 28c  Grin Grin
Well now i know which board to start with.
Thanks
hero member
Activity: 756
Merit: 560
April 06, 2017, 11:49:39 AM
#5
Well if you take a look at chip temps the problem looks like it is board 1 aka chain 6. You can see that on the chip 1 temperature that is only at 28c so obviously there is something funky with that board.
full member
Activity: 125
Merit: 100
Canadian Miner Passing Through~~
April 06, 2017, 09:59:32 AM
#4
Ok will start that when i arrive home tonight. But i though maybe it's a firmware thing or something but will try that.
Thanks
legendary
Activity: 4326
Merit: 8950
'The right to privacy matters'
April 06, 2017, 09:57:52 AM
#3
Just an FYI the link below shows a different miner running along side it

http://imgur.com/a/Hydsl


board a --------  has 3 pcie jacks
board b --------  has 3 pcie jacks
board c -------- has  3 pcie jacks


on board a  pull the 3 pcie cables   run for a day  see if you do  say 8.5-9  or  4.2 -4.5

repeat  the next day  with board b

repeat the next day with board c

my guess is one of the 3 boards is sick  the 3 day test above should let you which one is sick

good luck
full member
Activity: 125
Merit: 100
Canadian Miner Passing Through~~
April 06, 2017, 09:35:06 AM
#2
Just an FYI the link below shows a different miner running along side it

http://imgur.com/a/Hydsl
full member
Activity: 125
Merit: 100
Canadian Miner Passing Through~~
April 06, 2017, 09:08:06 AM
#1
Hi, well as can be seen in the images linked below.

http://imgur.com/a/qUiNA

The hash rate is at 12.57th/s but the average is almost always below 9th/s. I tried to change the difficulty to a 2048 and 10200 but the problem still persists. When i reboot the miner it hashes perfectly at the 12.5th/s but gradually goes down to the 8.5th/s avg range but the entire time the rt/s is always at 12.5th/s. I read a few threads and they say that it may be because of the HW errors but it is at only at 0.0001%. Here's me hoping there's a simple solutions to this.

Thanks
Jump to: