ANTMINER S3+ Discussion and Support Thread

[Nite]

Anybody find their S3's like cold weather?

-5c outside and my ants in the shed are performing better than ever! hash rate has increased board temps are low 30's,
All the outside miners are hashing at or above 520ghs at 243m. Impressive

becareful with that, if the miners or PSU's get too cold they can get moisture around them, I wouldn't depend on the heat that they put off to keep them warm enough.

[1714636195]

1714636195

[1714636195]

1714636195

[1714636195]

1714636195

[Soros Shorts]

Anybody find their S3's like cold weather?

-5c outside and my ants in the shed are performing better than ever! hash rate has increased board temps are low 30's,
All the outside miners are hashing at or above 520ghs at 243m. Impressive

becareful with that, if the miners or PSU's get too cold they can get moisture around them

Only if they get colder than the ambient air temperature, which will never happen unless you bring them back inside the house while they are icy cold..

[Xian01]

I'm interested in tinkering with turning a couple S3's into space heaters now that it's getting colder.

The aim would get to get two of them quiet enough to run in a bedroom or office.

Could someone experienced offer any insight if stacking a few non PWM low speed fans in sequence, on each end of the miner, work to push enough air through the unit more quietly ? Any recommendations for fans to that end ?

[klondike_bar]

Anybody find their S3's like cold weather?

-5c outside and my ants in the shed are performing better than ever! hash rate has increased board temps are low 30's,
All the outside miners are hashing at or above 520ghs at 243m. Impressive

becareful with that, if the miners or PSU's get too cold they can get moisture around them, I wouldn't depend on the heat that they put off to keep them warm enough.

thats not how condensation works. the gear is still hotter than the surrounding air - ergo will not condensate water.

[hurricandave]

Isn't that like asking a mechanic if you could get twice the horsepower from your car if he would just put two or three sparkplugs in each cylinder instead of just the one.

I'm interested in tinkering with turning a couple S3's into space heaters now that it's getting colder.

The aim would get to get two of them quiet enough to run in a bedroom or office.

Could someone experienced offer any insight if stacking a few non PWM low speed fans in sequence, on each end of the miner, work to push enough air through the unit more quietly ? Any recommendations for fans to that end ?

[FlensGold]

Maybe it is a stupid question, but I might be too blind to find it:
Where is the "reset to factory defaults" (Reset to defaults) option on the WebIF? I am quite sure it has been under System -> Backup / Flash Firmware, but it is no longer there when using the current firmware.
Where can I find it now?

[luckypyrate]

I seem to be having some significant efficiency reduction after upgrading my remaining S1's to S3+.  

Noting the new units all came with the latest OCT242014 firmware I decided to upgrade the rest of my units as well.  So now all units are exactly the same except for IP, obviously.  Is there any known issues with the latest firmware that could cause my efficiency to drop 20% since upgrading?  (I host my own P2Pool node that is how I see the efficiency drop).

Also:

1)I read somewhere that someone had each miner depositing into it's own wallet address...is there any merit to this as being better?

2)The new S3+ firmware has a new setting option for voltage under the advanced tab where the overclocking options are.  Does anyone know what these settings should be?  Does the voltage change if I overclock the unit?  The new ones came with "075" in this box by default.

[pekatete]

I seem to be having some significant efficiency reduction after upgrading my remaining S1's to S3+.  

You need to clarify (even qualify!) that statement. You SHOULD get a significant improvement either way since an S1 consumes 400 watts for ~200GH/s while an S3+ consumes as much for ~500GH/s. So the question is, what exactly do you mean by a reduction in efficiency?

Is there any known issues with the latest firmware that could cause my efficiency to drop 20% since upgrading?  (I host my own P2Pool node that is how I see the efficiency drop).

It has been said in this thread and a multitude others, if it ain't broke, don't fix it!

[Xian01]

Not sure that is a proper analogy. The possibly ignorant question on my part remains.

Isn't that like asking a mechanic if you could get twice the horsepower from your car if he would just put two or three sparkplugs in each cylinder instead of just the one.

I'm interested in tinkering with turning a couple S3's into space heaters now that it's getting colder.

The aim would get to get two of them quiet enough to run in a bedroom or office.

Could someone experienced offer any insight if stacking a few non PWM low speed fans in sequence, on each end of the miner, work to push enough air through the unit more quietly ? Any recommendations for fans to that end ?

[moss]

I'm interested in tinkering with turning a couple S3's into space heaters now that it's getting colder.

The aim would get to get two of them quiet enough to run in a bedroom or office.

Could someone experienced offer any insight if stacking a few non PWM low speed fans in sequence, on each end of the miner, work to push enough air through the unit more quietly ? Any recommendations for fans to that end ?

Not sure about going for non PWM fans - surely they'll just run flat out and be even more noisy?
On my S3s, I use a Corsair PWM SP120 High Performance 120mm on the inlet and a Thermalright PWM TY143 140mm on the outlet (screw holes fit 120mm spacing).
This makes the S3s much quieter than with the standard fans.  Temps are up by 1 or 2 degrees (those standard fans are powerful things) but it's an acceptable trade-off.
Mine are in my office and this has been well worthwhile for me.  Not sure if I would want them in my bedroom though - any fan will sound mighty loud in the wee small hours!
You could always try doing just the one S3 to start with, to see if it gives the level of noise reduction that you want.

[Xian01]

Not sure if I would want them in my bedroom though

TBH, I have a 500GHs Bitfury rig running in my bedroom that is super quiet. Looking for something more efficient (and quiet) to replace it.

[luckypyrate]

Not sure if I would want them in my bedroom though

TBH, I have a 500GHs Bitfury rig running in my bedroom that is super quiet. Look for something more efficient (and quiet) to replace it.

How much does that run nowadays?

[Xian01]

TBH, I have a 500GHs Bitfury rig running in my bedroom that is super quiet. Look for something more efficient (and quiet) to replace it.

How much does that run nowadays?

I'm not sure they are officially available any more. Might get lucky on Fleabay or the Hardware sales section of these forums. I've had the rigs for over a year now they're still chugging along.

[pekatete]

TBH, I have a 500GHs Bitfury rig running in my bedroom that is super quiet. Look for something more efficient (and quiet) to replace it.

How much does that run nowadays?

I'm not sure they are officially available any more. Might get lucky on Fleabay or the Hardware sales section of these forums. I've had the rigs for over a year now they're still chugging along.

500GH/s bitfury rig? How much wattage does that beast consume, bearing in mind an OC'd S3+ @ 500+ GH/s consumes ~400w?

[Xian01]

500GH/s bitfury rig? How much wattage does that beast consume, bearing in mind an OC'd S3+ @ 500+ GH/s consumes ~400w?

It's over 1W/GHs by a bit, which is why I'm looking to create a couple quiet S3 space-heaters, and retire the current Bitfury space-heaters.

[luckypyrate]

I seem to be having some significant efficiency reduction after upgrading my remaining S1's to S3+.  

You need to clarify (even qualify!) that statement. You SHOULD get a significant improvement either way since an S1 consumes 400 watts for ~200GH/s while an S3+ consumes as much for ~500GH/s. So the question is, what exactly do you mean by a reduction in efficiency?

Is there any known issues with the latest firmware that could cause my efficiency to drop 20% since upgrading?  (I host my own P2Pool node that is how I see the efficiency drop).

It has been said in this thread and a multitude others, if it ain't broke, don't fix it!

Okay so what is going on is my pool that I host here is saying my efficiency is ~80% where it was ~115% before the upgrade. Seems more than suspicious to me...I do see the improvement in hash rate, but with my efficiency tanking I am seeing my deposits drop .02 satoshis or so instead of increasing.  Killing me, smalls!

[pekatete]

I seem to be having some significant efficiency reduction after upgrading my remaining S1's to S3+.  

You need to clarify (even qualify!) that statement. You SHOULD get a significant improvement either way since an S1 consumes 400 watts for ~200GH/s while an S3+ consumes as much for ~500GH/s. So the question is, what exactly do you mean by a reduction in efficiency?

Is there any known issues with the latest firmware that could cause my efficiency to drop 20% since upgrading?  (I host my own P2Pool node that is how I see the efficiency drop).

It has been said in this thread and a multitude others, if it ain't broke, don't fix it!

Okay so what is going on is my pool that I host here is saying my efficiency is ~80% where it was ~115% before the upgrade. Seems more than suspicious to me...I do see the improvement in hash rate, but with my efficiency tanking I am seeing my deposits drop .02 satoshis or so instead of increasing.  Killing me, smalls!

That means you are either getting a lot more stales than before, or you have set your timeout LOW such that the "dark" HW errors are not showing up in your GUI and your "upgraded"(?)  rigs are thus submiting less shares. You could try increasing the timeout on one unit and see how that affects your earnings, else I'd press the hard reset and revert to the stock firmware. Having said that, (and this is by no means authoritative), did you ever hear of those fabled old tales about the S3 not being optimised for P2Pool, or am I mistaken?
I'll have to ask again, why on earth did you choose to "upgrade" the firmware?

EDIT: I had to say timeout set LOW rather than HIGH

[pekatete]

500GH/s bitfury rig? How much wattage does that beast consume, bearing in mind an OC'd S3+ @ 500+ GH/s consumes ~400w?

It's over 1W/GHs by a bit, which is why I'm looking to create a couple quiet S3 space-heaters, and retire the current Bitfury space-heaters.

If it is space heaters that you REALLY want, AND you do not mind maintaining / reducing your hash-rate, you could be better served by acquiring a set of S1's, underclock & undervolt them (to output ~125 GH/s @ - and don't ask me for a link, search the forum!) and replace the fan with a couple of quieter ones in a push pull layout at full pelt, aka blue wire hack; For a tad less power for both compared to your bitfury rig, you'd be laughing!

EDIT: I am currently running a couple of bare S1 boards (neither underclocked  nor undervolted) off cgminer with a fairly loud fan blowing over them and they hold their own as space heaters! Did my laundry drying rather quickly today too, so I can vouch for a couple of these archaic beasts to do the job well.

[luckypyrate]

I seem to be having some significant efficiency reduction after upgrading my remaining S1's to S3+.  

You need to clarify (even qualify!) that statement. You SHOULD get a significant improvement either way since an S1 consumes 400 watts for ~200GH/s while an S3+ consumes as much for ~500GH/s. So the question is, what exactly do you mean by a reduction in efficiency?

Is there any known issues with the latest firmware that could cause my efficiency to drop 20% since upgrading?  (I host my own P2Pool node that is how I see the efficiency drop).

It has been said in this thread and a multitude others, if it ain't broke, don't fix it!

Okay so what is going on is my pool that I host here is saying my efficiency is ~80% where it was ~115% before the upgrade. Seems more than suspicious to me...I do see the improvement in hash rate, but with my efficiency tanking I am seeing my deposits drop .02 satoshis or so instead of increasing.  Killing me, smalls!

That means you are either getting a lot more stales than before, or you have set your timeout LOW such that the "dark" HW errors are not showing up in your GUI and your "upgraded"(?)  rigs are thus submiting less shares. You could try increasing the timeout on one unit and see how that affects your earnings, else I'd press the hard reset and revert to the stock firmware. Having said that, (and this is by no means authoritative), did you ever hear of those fabled old tales about the S3 not being optimised for P2Pool, or am I mistaken?
I'll have to ask again, why on earth did you choose to "upgrade" the firmware?

EDIT: I had to say timeout set LOW rather than HIGH

I upgraded the firmware to the latest so all my units would be uniform (the 4 new units I ordered came with 24OCT14 firmware where as the others had the august firmware release).  If I revert them all back and see improvement I will post here and let you know.

[Nite]

has anyone ever had the issue where half the mining power turns off after running fine for about a week solid? this is the second time its happened to me, I have to keep turning off the PSU and turn it back on to get it back to full speed again.

temps were only around 36 - 39c, only have 1 unit running off a coolermaster 1000w silent pro gold PSU, and I'm not overclocked.