ANTMINER S3+ Discussion and Support Thread

[jpchrist]

Anyone want to speculate which hardware manufacturer was testing harder to the tune of 50 phash for a day or so to give us such a pretty spike in difficulty?

Hopefully it stays gone....

http://www.coinwarz.com/network-hashrate-charts/bitcoin-network-hashrate-chart

Dive difficulty Dive!!

Right, because turning off 50PHs (just for the hell of it) makes a lot of sense.  

"Nah, we don't need the money, let's just turn them off and let them collect dust. But we'll continue to pay the lease on the 2 massive datacenters that they are in."

If there is a sudden large spike in the network hashrate and it suddenly disappears, it's variance... not someone having a play with literally tons of equipment.

lol 50 ph in variance


Just to make it clear for you how many times has the network gone to say 50k from 100k. If you think the actual hashrate is varying? ( yes it will vary if someone brings new equipment online or takes it off, or chips vary slightly. )  That isn't luck lol.

You are not going to see for example your home hardware spiking from 440 ( on an s3 ) to 660 and to 220 for any length of time, much less everyone's at once.

Made me smile though.

Edit-

I am very curious though who/what/where it was.  Will probably find out soon if it wasn't any of the major equipment vendors testing.   Would be interesting to see it stay up for a while then we start to get a slew of complaints x vendor just shipped me dusty used equipment.  Then again it could be some cloud mining operation setting up in Iceland finally and getting it right.

Bitfury datacenter (google it)

[philipma1957]

here is a little reminder for us     well  for me.

vi /etc/config/asic-freq

cause I got a new one to try!!


I got a working freq!!!! 


231.25     just started testing it

lets hope it works for some:

[squashpile]

Let me know how that Freq goes. Mine never averaged over 440 at 231.25.
225 was actually higher GH/s for me. Yeah.. it was strange. No HW errors, but 225 ended up being better. I may try it again.

here is a little reminder for us     well  for me.

vi /etc/config/asic-freq

cause I got a new one to try!!


I got a working freq!!!!  


231.25     just started testing it

lets hope it works for some:

[bluebottle73]

Way to go BITMAIN.

You sent me an S1 controller board instead of an S3.  Unless there's something I need to do to get this working.

I got in to configure, and also updated the firmware to the latest you have posted.  However on the status screen, it says the frequencies are 350 (like an S1, not 218.75 like it should be for an S3).  It's reporting twice the number of ASICs, but half of them are x's.  And hash speed is all 0.

Does anyone have any idea how to reprogram this so it'll work as an S3??  Or am I going to have to wait another 2 weeks while they send the RIGHT part for the unit that wasn't working right out of the box??

Another 2 weeks.  Another diff.  Sheesh.

At least my other 2 units are working reasonably well (although they're now hashing at around 220 GH/s - consistently.  Well below what was promised.)

[opieum2]

Way to go BITMAIN.

You sent me an S1 controller board instead of an S3.  Unless there's something I need to do to get this working.

I got in to configure, and also updated the firmware to the latest you have posted.  However on the status screen, it says the frequencies are 350 (like an S1, not 218.75 like it should be for an S3).  It's reporting twice the number of ASICs, but half of them are x's.  And hash speed is all 0.

Does anyone have any idea how to reprogram this so it'll work as an S3??  Or am I going to have to wait another 2 weeks while they send the RIGHT part for the unit that wasn't working right out of the box??

Another 2 weeks.  Another diff.  Sheesh.

At least my other 2 units are working reasonably well (although they're now hashing at around 220 GH/s - consistently.  Well below what was promised.)

I suggest emailing them. In my own experience I have had great success with replies from Bitmain. They might even have a solution for your issues. So far so good with my S3s (got 8 with 3 more on the way) +2 S2s in my small but growing farm. They have been consistently reliable with shipping sometimes shipping before the promised date. Their support generally is good. Just a thought. Of course speaking about my own experience. That might not be the same for everyone else. Just speaking for myself here.

[itod]

Anyone want to speculate which hardware manufacturer was testing harder to the tune of 50 phash for a day or so to give us such a pretty spike in difficulty?

Hopefully it stays gone....

http://www.coinwarz.com/network-hashrate-charts/bitcoin-network-hashrate-chart

Dive difficulty Dive!!

Right, because turning off 50PHs (just for the hell of it) makes a lot of sense.  

"Nah, we don't need the money, let's just turn them off and let them collect dust. But we'll continue to pay the lease on the 2 massive datacenters that they are in."

If there is a sudden large spike in the network hashrate and it suddenly disappears, it's variance... not someone having a play with literally tons of equipment.

lol 50 ph in variance


Just to make it clear for you how many times has the network gone to say 50k from 100k. If you think the actual hashrate is varying? ( yes it will vary if someone brings new equipment online or takes it off, or chips vary slightly. )  That isn't luck lol.

You are not going to see for example your home hardware spiking from 440 ( on an s3 ) to 660 and to 220 for any length of time, much less everyone's at once.

Made me smile though.

Edit-

I am very curious though who/what/where it was.  Will probably find out soon if it wasn't any of the major equipment vendors testing.   Would be interesting to see it stay up for a while then we start to get a slew of complaints x vendor just shipped me dusty used equipment.  Then again it could be some cloud mining operation setting up in Iceland finally and getting it right.

Of course that can't be variance, but why all those conspiracy theories? Isn't it obvious who turned those 50PH off?
Hints:
* not all mining is done in large datacenters;
* there are (were) tons of Avalons, AsicMiners, etc, miners out there who mine(d) with some combination of free/low cost electricity
* what can you do if you have those dinosaur, near extinct, miners when one large hashrate increase signals you it's just not worth it to run them any more?

Why search for conspiracy theories which datacenter turned off 50PH, when there is simple logical explanation?

[bluebottle73]

Way to go BITMAIN.

You sent me an S1 controller board instead of an S3.  Unless there's something I need to do to get this working.

I got in to configure, and also updated the firmware to the latest you have posted.  However on the status screen, it says the frequencies are 350 (like an S1, not 218.75 like it should be for an S3).  It's reporting twice the number of ASICs, but half of them are x's.  And hash speed is all 0.

Does anyone have any idea how to reprogram this so it'll work as an S3??  Or am I going to have to wait another 2 weeks while they send the RIGHT part for the unit that wasn't working right out of the box??

Another 2 weeks.  Another diff.  Sheesh.

At least my other 2 units are working reasonably well (although they're now hashing at around 220 GH/s - consistently.  Well below what was promised.)

I suggest emailing them. In my own experience I have had great success with replies from Bitmain. They might even have a solution for your issues. So far so good with my S3s (got 8 with 3 more on the way) +2 S2s in my small but growing farm. They have been consistently reliable with shipping sometimes shipping before the promised date. Their support generally is good. Just a thought. Of course speaking about my own experience. That might not be the same for everyone else. Just speaking for myself here.

Oh, I've already emailed them.  This post was more of a rant/call for help in case there is anyone there who has a suggestion.  I won't hear from Bitmain until later tonight given the time difference.

And I supposed to other miner manufacturers their support could be considered good.  However, my unit's network interface was dead right out of the box and unlike manufacturers of other high-end cost-sensitive and mission critical hardware, they do not have the option to advance ship the replacemnt, resulting in more downtime while waiting for the faulty part to get to them (I won't rehash that here any more, my opinion in earlier posts is well known).

I'm just a lot frustrated because it's more than 2 weeks now and this thing is still not earning for me.

I suspect the controller boards are quite similar (looks practically identical), so I am guessing it might just be a matter of changing configuration files.  But I'm not that experienced yet with configuring miners, so I'd rather defer to more experienced forum members or Bitmain themselves if it's possible to just update some files and get it working without having to send it back AGAIN.

[philipma1957]

Let me know how that Freq goes. Mine never averaged over 440 at 231.25.
225 was actually higher GH/s for me. Yeah.. it was strange. No HW errors, but 225 ended up being better. I may try it again.



new set of numbers:

here is a little reminder for us     well  for me.

vi /etc/config/asic-freq

cause I got a new one to try!!


I got a working freq!!!!  


231.25     just started testing it

lets hope it works for some:

[Rabinovitch]

At least my other 2 units are working reasonably well (although they're now hashing at around 220 GH/s - consistently.  Well below what was promised.)

So if you have 2 "other units" (I suggest you have tho S3s working normally, right?) why don't try to clone the SD card content from normal S3 to SD card of S1's controller? Maybe it will work? 

[squashpile]

Hmm
I'm averaging 453 on two@ 225 over 23hrs. Need to get the Kill A Watt back on it to check at the wall and compare to 231 if I can average higher.
I'll try 231 again to see what happens. Like I said it was strange. No issues on your HW errors for sure..lol Mine are bad above 231 so far. Need to test more.

Let me know how that Freq goes. Mine never averaged over 440 at 231.25.
225 was actually higher GH/s for me. Yeah.. it was strange. No HW errors, but 225 ended up being better. I may try it again.



new set of numbers:

here is a little reminder for us     well  for me.

vi /etc/config/asic-freq

cause I got a new one to try!!


I got a working freq!!!!  


231.25     just started testing it

lets hope it works for some:

[bluebottle73]

At least my other 2 units are working reasonably well (although they're now hashing at around 220 GH/s - consistently.  Well below what was promised.)

So if you have 2 "other units" (I suggest you have tho S3s working normally, right?) why don't try to clone the SD card content from normal S3 to SD card of S1's controller? Maybe it will work? 

That's the kind of thing I'm thinking of, but not sure of the procedure to do that.  I'm not put off by a challenge, just don't know where to start..

[philipma1957]

and I got freq 243.75 to work now 


option  'freq_value'    '1306' #243.75M
option  'chip_freq'      '243.75'
option  'timeout'        '16'


option  'freq_value'    '1206' #231.25M
option  'chip_freq'      '231.25'
option  'timeout'        '17'

both extra ones to try.

so we have

250
243.75
237.50
231.25
225
218.75   > stock should do 441gh
212.5
206.25
200

Hmm
I'm averaging 453 on two@ 225 over 23hrs. Need to get the Kill A Watt back on it to check at the wall and compare to 231 if I can average higher.
I'll try 231 again to see what happens. Like I said it was strange. No issues on your HW errors for sure..lol Mine are bad above 231 so far. Need to test more.

Let me know how that Freq goes. Mine never averaged over 440 at 231.25.
225 was actually higher GH/s for me. Yeah.. it was strange. No HW errors, but 225 ended up being better. I may try it again.



new set of numbers:

here is a little reminder for us     well  for me.

vi /etc/config/asic-freq

cause I got a new one to try!!


I got a working freq!!!!  


231.25     just started testing it

lets hope it works for some:

[bitcoin_miner]

bad news caused my S3 delayed.. Scheduled Delivery: 18-August-2014

Shenzhen, China   13/08/2014   6:07   Departure Scan
13/08/2014   4:05   Severe weather conditions have delayed delivery.
13/08/2014   0:27   Export Scan
Shenzhen, China   11/08/2014   21:57   Origin Scan
11/08/2014   19:28   Collection Scan
China   11/08/2014   2:37   Order Processed: Ready for UPS

I checked in http://cuaca.info world wide weather online. Shenzhen, China good weather Partly Cloudy only. but UPS delayed delivery  

[dsheaffer]

I have an "ROI" question.  When you all calculate your return, are you considering the fair market value of your miners/PSU's etc.  The reason I ask is that I have purchased 15 miners (1 Avalon, 7 S1s and 7 S3s) and sold 8 (just have the S1s  Edit: I meant I just have the S3s now). After electricity cost, PSU cost, WiFi antenna cost, and everything else I can think of, I have an ROI of ~13% in 5 months (~31% annualized) when I consider the current fair market value of the hardware I currently own.  To me, this is a GREAT ROI...what am I missing?

[squashpile]

I would if you know you're actually going to put it up for sale. I'll get lazy and end up with a pile of say ummm Antminer U2's I for sure was going to sell..lol. Now I just give them away.
But hell yeah your're not missing anything but more free time.

I have an "ROI" question.  When you all calculate your return, are you considering the fair market value of your miners/PSU's etc.  The reason I ask is that I have purchased 15 miners (1 Avalon, 7 S1s and 7 S3s) and sold 8 (just have the S1s now). After electricity cost, PSU cost, WiFi antenna cost, and everything else I can think of, I have an ROI of ~13% in 5 months (~31% annualized) when I consider the current fair market value of the hardware I currently own.  To me, this is a GREAT ROI...what am I missing?

[smilingloki]

So some of my miners require a restart about every 12-17 hours - they occasionally show up with the dreaded "x" but then upon a restart they are all good to go,

I'm really tired of having to restart them with the gui,  is there anyone out there that can help me with an automated reboot option in the

"Scheduled Tasks
      This is the system crontab in which scheduled tasks can be defined."

 that i can save and autorestart the miners - once or twice a day lets say

12midnight and 12 midday?

Regards

loki.

[BITMAIN]

Way to go BITMAIN.

You sent me an S1 controller board instead of an S3.  Unless there's something I need to do to get this working.

I got in to configure, and also updated the firmware to the latest you have posted.  However on the status screen, it says the frequencies are 350 (like an S1, not 218.75 like it should be for an S3).  It's reporting twice the number of ASICs, but half of them are x's.  And hash speed is all 0.

Does anyone have any idea how to reprogram this so it'll work as an S3??  Or am I going to have to wait another 2 weeks while they send the RIGHT part for the unit that wasn't working right out of the box??

Another 2 weeks.  Another diff.  Sheesh.

At least my other 2 units are working reasonably well (although they're now hashing at around 220 GH/s - consistently.  Well below what was promised.)

Hi bluebottle73,
May you PM your email box to look into, Our truly apologies for the inconvenience.

Thank you!

[grn]

So some of my miners require a restart about every 12-17 hours - they occasionally show up with the dreaded "x" but then upon a restart they are all good to go,

I'm really tired of having to restart them with the gui,  is there anyone out there that can help me with an automated reboot option in the

"Scheduled Tasks
      This is the system crontab in which scheduled tasks can be defined."

 that i can save and autorestart the miners - once or twice a day lets say

12midnight and 12 midday?

Regards

loki.

0 */12  *   *   *     /sbin/reboot

[smilingloki]

So some of my miners require a restart about every 12-17 hours - they occasionally show up with the dreaded "x" but then upon a restart they are all good to go,

I'm really tired of having to restart them with the gui,  is there anyone out there that can help me with an automated reboot option in the

"Scheduled Tasks
      This is the system crontab in which scheduled tasks can be defined."

 that i can save and autorestart the miners - once or twice a day lets say

12midnight and 12 midday?

Regards

loki.

0 */12  *   *   *     /sbin/reboot

You sir are a GOD ! thankyou, you wouldn't believe how annoying looking at 12miners can be!

[Darthswan]

Received my 2 S3s yesterday.  They have been happily hashing away at 440 and 460 stock.  Guess I need to overclock and see what happens.  I've seen a range of opinions on overclocking these.  How have the later batches been?  These are Batch 5.