philipma1957
Legendary
Offline
Activity: 4746
Merit: 11343
'The right to privacy matters'
|
 |
January 23, 2016, 03:22:39 PM |
|
We are running about 100 of these units and the GLOBAL hw error state is of about 75%. We have a huge number of hardware errors, on batch 6/7 miners at stock clock. Not sure where to go from that, these numbers can't be healthy. Example of 1 miner:  And another one:  Don't know what to do with this so far, other than downclocking but such high numbers aren't normal. Try freq 693. You will burn less watts and get higher hash. The problem can also be fixed with higher volts if your psu,s are doing 12.0. And can be bumped to 12.1 the errors will drop lower. With 100 s-7s. If I were you freq 693 would be my choice
|
|
|
|
generalt
Legendary
Offline
Activity: 1096
Merit: 1021
|
 |
January 23, 2016, 03:51:48 PM |
|
We are running about 100 of these units and the GLOBAL hw error state is of about 75%.
We have a huge number of hardware errors, on batch 6/7 miners at stock clock. Not sure where to go from that, these numbers can't be healthy.
Don't know what to do with this so far, other than downclocking but such high numbers aren't normal.
Wasn't batch 6 supposed to run at 600 mhz and batch 7 at 625 mhz default?
|
BTC: 1GENERALrtBAjEv2Ps5cmEW1FADnXh1bCZ
|
|
|
fr4nkthetank
Legendary
Offline
Activity: 2294
Merit: 1182
Now the money is free, and so the people will be
|
 |
January 23, 2016, 03:59:47 PM |
|
I dont get it, global hw rate? makes no sense, you dont add up the %. under 0.01% is fine
|
|
|
|
|
philipma1957
Legendary
Offline
Activity: 4746
Merit: 11343
'The right to privacy matters'
|
 |
January 23, 2016, 04:01:10 PM |
|
We are running about 100 of these units and the GLOBAL hw error state is of about 75%.
We have a huge number of hardware errors, on batch 6/7 miners at stock clock. Not sure where to go from that, these numbers can't be healthy.
Don't know what to do with this so far, other than downclocking but such high numbers aren't normal.
Wasn't batch 6 supposed to run at 600 mhz and batch 7 at 625 mhz default? frankly I lost track they have a different number for every one  and since they never compensated me for the low hashing batch 1's i received I do not buy any for myself anymore. The last one I purchased for my self was a batch 2 that is rated for freq 575 I still feel they owe me about 150 usd in compensation or maybe 134 usd. I asked for coupons not coins, They blew me off so I no longer buy for myself.. I also ended my mining in China over this. Frankly I don't understand why anyone not in China mines on a Chinese pool when kano.is has proven to crush the Chinese pools.
|
|
|
|
not.you
Legendary
Offline
Activity: 1726
Merit: 1018
|
 |
January 23, 2016, 04:19:50 PM |
|
We are running about 100 of these units and the GLOBAL hw error state is of about 75%. We have a huge number of hardware errors, on batch 6/7 miners at stock clock. Not sure where to go from that, these numbers can't be healthy. Example of 1 miner:  And another one:  Don't know what to do with this so far, other than downclocking but such high numbers aren't normal. Those aren't terrible. I think you meant .0075% not 75%. I usually like to keep mine under .01%. When these things first start from a power-on they tend to throw a lot of errors. If you go into the miner config and do a "save and apply" without changing anything just to get cgminer to restart it tends to start with far fewer errors. Also I personally think you would get better results overall if you raised the temperature on those.
|
|
|
|
|
philipma1957
Legendary
Offline
Activity: 4746
Merit: 11343
'The right to privacy matters'
|
 |
January 23, 2016, 05:05:46 PM |
|
We are running about 100 of these units and the GLOBAL hw error state is of about 75%. We have a huge number of hardware errors, on batch 6/7 miners at stock clock. Not sure where to go from that, these numbers can't be healthy. Example of 1 miner:  And another one:  Don't know what to do with this so far, other than downclocking but such high numbers aren't normal. Those aren't terrible. I think you meant .0075% not 75%. I usually like to keep mine under .01%. When these things first start from a power-on they tend to throw a lot of errors. If you go into the miner config and do a "save and apply" without changing anything just to get cgminer to restart it tends to start with far fewer errors. Also I personally think you would get better results overall if you raised the temperature on those. his numbers suck only because he is counting 100 x 0.0157 = 0.75 ie he is losing about .75 x 4800 = 3600 gh which is nutty math as he is losing 1.57 x 4800 = 7536gh due to error I ran enough of these to know he should clock them all at 693 his errors will drop to 0.0020 or less. his power use will drop 1% he will end up with maybe .20 x 4800 = 960 gh lost not 3200gh lost and his power will drop 1% every s-7 has a 0.0020 or better freq if he has to drop to 650 from 700 to drop from .0157 to .0020 it does not pay so he should change everyone to 693 and check his 0.0157 numbers great s-7's will do under 0.0020 at stock he won't lose by doing two 1 day tests one at freq 693 one at freq 687
|
|
|
|
Pissenlit
Newbie
Offline
Activity: 39
Merit: 0
|
 |
January 23, 2016, 06:52:34 PM |
|
Is it better to run the S7s at 12.75v instead of 12.5v from the 2880w psu? Not overclocked B9s. Can it hurt the miners? Thank you.
|
|
|
|
|
philipma1957
Legendary
Offline
Activity: 4746
Merit: 11343
'The right to privacy matters'
|
 |
January 23, 2016, 06:57:32 PM |
|
Is it better to run the S7s at 12.75v instead of 12.5v from the 2880w psu? Not overclocked B9s. Can it hurt the miners? Thank you.
no if you are providing 12.5 volts at stock freq you should have almost 0.0005 errors what do you get with your 12.5 volt setting at stock freq?
|
|
|
|
Pissenlit
Newbie
Offline
Activity: 39
Merit: 0
|
 |
January 23, 2016, 07:07:42 PM |
|
Is it better to run the S7s at 12.75v instead of 12.5v from the 2880w psu? Not overclocked B9s. Can it hurt the miners? Thank you.
no if you are providing 12.5 volts at stock freq you should have almost 0.0005 errors what do you get with your 12.5 volt setting at stock freq? I will receive them on monday, I'm preparing the rig to receive them. The 2880w psu can be jumped to raise the voltage 0.25v more to 12.75v. Normal voltage is 12.5v. My overclocked SP20s runs at 12.75v without issues and they hash really strong with minimal errors but the S7 seems to be less reliable. Thank you!
|
|
|
|
|
thedreamer
Legendary
Offline
Activity: 1694
Merit: 1002
Go Big or Go Home.....
|
 |
January 23, 2016, 07:16:35 PM |
|
and since they never compensated me for the low hashing batch 1's i received I do not buy any for myself anymore.
The last one I purchased for my self was a batch 2 that is rated for freq 575
I still feel they owe me about 150 usd in compensation or maybe 134 usd. I asked for coupons not coins,
They blew me off so I no longer buy for myself..
I also ended my mining in China over this. Frankly I don't understand why anyone not in China mines on a Chinese pool when kano.is has proven to crush the Chinese pools.
Me and at least 2 others have not received any compensation still either for the Batch 1 fiasco...
|
Go Big or Go Home.
|
|
|
Biodom
Legendary
Online
Activity: 4382
Merit: 5809
|
 |
January 23, 2016, 07:32:47 PM |
|
frankly I lost track they have a different number for every one  and since they never compensated me for the low hashing batch 1's i received I do not buy any for myself anymore. The last one I purchased for my self was a batch 2 that is rated for freq 575 I still feel they owe me about 150 usd in compensation or maybe 134 usd. I asked for coupons not coins, They blew me off so I no longer buy for myself.. I also ended my mining in China over this. Frankly I don't understand why anyone not in China mines on a Chinese pool when kano.is has proven to crush the Chinese pools. You are right about being upset about this, of course. That said, maybe look at it from a karmic sense/perspective. You got low hashing B1, then you got very high hashing B2 (I remember that you said they could do 625 instead of 575). Maybe it was the universe smiling at you for B2 or maybe a friendly product manager at Bitmain- who knows? In essence, these two purchases are balanced, don't you think? Of course, coupons would be a nice gesture. I ordered pretty much one of each batch (stopped at 7) and my B1 performs very slightly below spec, but B5 can do 625 easily (5Th). So, averaging gave me what one should expect-average hashing in comparison with what was expected.
|
|
|
|
|
|
hawkfish007
|
 |
January 23, 2016, 08:45:43 PM |
|
Can't be more happier with my S7s. B1 2 board S7 beating Avalon 6 at kano.is. The other board was on RMA with Bitmain. 
|
|
|
|
|
nhando
|
 |
January 23, 2016, 09:12:29 PM |
|
Anyone have different batches that can compare temperatures? Does Batch 8/9 run much hotter due to the higher overclocking or does the lower total chip count kind of balance things out? I want to see if someone who have the 162 Chip S7 compare temperature against Batch 8/9 with 135 Chips.
|
Just "Mining" my own business.
|
|
|
opentoe
Legendary
Offline
Activity: 1274
Merit: 1000
Personal text my ass....
|
 |
January 23, 2016, 09:19:27 PM |
|
I also ended my mining in China over this. Frankly I don't understand why anyone not in China mines on a Chinese pool when kano.is has proven to crush the Chinese pools.
Prior to S7 sales Antpool was approx %18-%19 of network. Thanks to you guys that support diversity, they jumped to almost %23-%25 in just 3 months. Keep buying those S7's. Can't wait to see the growth after the S8/S9 sales. Good job guys.  Yes, very sarcastic post NOT meant to break or shatter ones delicate nature. Disclaimer: My numbers could be off by %50. I still think Pluto is a planet.
|
|
|
|
tutorialevideo
Legendary
Offline
Activity: 1161
Merit: 1001
Don`t invest more than you can afford to lose
|
 |
January 23, 2016, 09:21:28 PM |
|
First antminer s7 board is dead... batch 8
|
|
|
|
|
hawkfish007
|
 |
January 23, 2016, 09:28:36 PM Last edit: January 23, 2016, 11:04:05 PM by hawkfish007 |
|
Anyone have different batches that can compare temperatures? Does Batch 8/9 run much hotter due to the higher overclocking or does the lower total chip count kind of balance things out? I want to see if someone who have the 162 Chip S7 compare temperature against Batch 8/9 with 135 Chips.
Here you go nhando, B7 and B8 screenshots, they are at sidehack's place and I am guessing they are on the same rack, fan speed is manual 60%. B7  B8  Edit: my B9s haven't been shipped, order date Jan 16.
|
|
|
|
VirosaGITS
Legendary
Offline
Activity: 1302
Merit: 1069
|
 |
January 23, 2016, 09:30:30 PM |
|
I also ended my mining in China over this. Frankly I don't understand why anyone not in China mines on a Chinese pool when kano.is has proven to crush the Chinese pools.
Prior to S7 sales Antpool was approx %18-%19 of network. Thanks to you guys that support diversity, they jumped to almost %23-%25 in just 3 months. Keep buying those S7's. Can't wait to see the growth after the S8/S9 sales. Good job guys.  Yes, very sarcastic post NOT meant to break or shatter ones delicate nature. Disclaimer: My numbers could be off by %50. I still think Pluto is a planet. People are kind of oblivious, everyone flock to the bigger agglomeration, its a survival instinct. By default the miners are set to Antpool, too and then they advertise 0% fee. So people thing the less = better. But Antpool keep the txfee, so the fee is much higher than Kano's 0.9%. The other thing is that Bitmain might be keeping a lot of S7 for themselves. It would be rather profitable for them to do so.
|
|
|
|
Tupsu
Legendary
Offline
Activity: 1204
Merit: 1003
|
 |
January 23, 2016, 09:48:13 PM |
|
People are kind of oblivious, everyone flock to the bigger agglomeration, its a survival instinct. By default the miners are set to Antpool, too and then they advertise 0% fee. So people thing the less = better. But Antpool keep the txfee, so the fee is much higher than Kano's 0.9%.
The other thing is that Bitmain might be keeping a lot of S7 for themselves. It would be rather profitable for them to do so.
Antpoool last month Luck 24 Hours 3 Day 1 Week 2 Weeks 1 Month 106.00% 108.00% 111.00% 107.00% 103.00% https://www.antpool.com/poolStats.htm
|
|
|
|
|
philipma1957
Legendary
Offline
Activity: 4746
Merit: 11343
'The right to privacy matters'
|
 |
January 23, 2016, 10:08:31 PM |
|
People are kind of oblivious, everyone flock to the bigger agglomeration, its a survival instinct. By default the miners are set to Antpool, too and then they advertise 0% fee. So people thing the less = better. But Antpool keep the txfee, so the fee is much higher than Kano's 0.9%.
The other thing is that Bitmain might be keeping a lot of S7 for themselves. It would be rather profitable for them to do so.
Antpoool last month Luck 24 Hours 3 Day 1 Week 2 Weeks 1 Month 106.00% 108.00% 111.00% 107.00% 103.00% https://www.antpool.com/poolStats.htmand it claims lifetime to be 100% lets say that is true kano.io claims lifetime to be 104.33% lets say that is true. so an antpool miner that earns 100 coins before fees before fees = 97.5 coins after 2.5 % fee and no transaction fees so total of 97.5 coins a miner that mined with kano would be getting 104.33 coins = 103.39 coins after .9% fee and about 4 blocks worths of fees say 1 more coin 104.39 coins so a big time miner not in China is getting 97.5 coins after fees at antpool and a bigtime miner with kano would be getting 104.39 coins So I get if you are in China why mining at antpool makes sense due to firewall issue. but 97.5 coins vs 104.39 coins no contest where anyone should mine.
|
|
|
|
VirosaGITS
Legendary
Offline
Activity: 1302
Merit: 1069
|
 |
January 23, 2016, 10:13:42 PM |
|
People are kind of oblivious, everyone flock to the bigger agglomeration, its a survival instinct. By default the miners are set to Antpool, too and then they advertise 0% fee. So people thing the less = better. But Antpool keep the txfee, so the fee is much higher than Kano's 0.9%.
The other thing is that Bitmain might be keeping a lot of S7 for themselves. It would be rather profitable for them to do so.
Antpoool last month Luck 24 Hours 3 Day 1 Week 2 Weeks 1 Month 106.00% 108.00% 111.00% 107.00% 103.00% https://www.antpool.com/poolStats.htmand it claims lifetime to be 100% lets say that is true kano.io claims lifetime to be 104.33% lets say that is true. so an antpool miner that earns 100 coins before fees before fees = 97.5 coins after 2.5 % fee and no transaction fees so total of 97.5 coins a miner that mined with kano would be getting 104.33 coins = 103.39 coins after .9% fee and about 4 blocks worths of fees say 1 more coin 104.39 coins so a big time miner not in China is getting 97.5 coins after fees at antpool and a bigtime miner with kano would be getting 104.39 coins So I get if you are in China why mining at antpool makes sense due to firewall issue. but 97.5 coins vs 104.39 coins no contest where anyone should mine. Thats correct math, except Antpool has a 0% fee. So 100% luck would be 100 coins. And Kano after 0.9% fee is indeed at 104~105BTC. So even with Antpool's 0% fee, you lose more using Antpool than Kano. Even if Kano would be at 100% luck overall, Kano would be superior, because the TX fee account to more than the 0.9% we pay Kano pool. And also Kano has a great backend, which help eliminating Orphans.
|
|
|
|
|