ANTMINER S3+ Discussion and Support Thread

[-ck]

After what I learnt from the S4 investigation, I checked the S3s, and they have a very diminished version of the problem with pools that start at low diffs. Here's an updated binary for the S3s.

http://ck.kolivas.org/apps/cgminer/antminer/s3/4.6.1-141009/cgminer

This might make startup/pool switching a bit smoother, especially if your pools start at diff 1 and take a while to increase their vardiff.

EDIT: Hashrates even seem a little faster after a few minutes...

EDIT2: No it's not any faster after longer running, it's the same overall performance. Only the startup won't overload the S3 controller on low diff starting pools.

[1714641872]

1714641872

[1714641872]

1714641872

[1714641872]

1714641872

[1714641872]

1714641872

[dankefoss]

This always happens when a company opens cloud mining. Their prices need to raise to make the cloud mining look like a better deal. Still, with the current price of BTC, the S3+ is cheaper if you consider the cost in fiat.

There are much better deals currently available: Rockminer and ASICMiner Prisma are two.

I use the following spreadsheet to compare apples to apples. I hesitate to share because well, people will freak out because it is not completely accurate in its forecast, but it does use an industry scaling factor, which will probably be overrun unless BTC goes up. I will probably adjust the scaling factor for myself.

You need to understand excel to use it, but essentially, when I go in, I update the difficulty, day and price, unless your electricity rate matches mine, you'd also want to update electricity $/kwh. I then click on the name of each miner, and it takes me to their website where I double check the price. If the principle cost is fiat, I list that bolded, if BTC, I list that bolded. I also use a calculation for the s&h if in BTC. The last three columns are where I compare apples to apples. The notes explain much of what I was thinking when I programmed it the way I did. You'd want to update costs for PSU and other. I have my preferred units that are essentially priced here.

http://www.hundredpalms.com/profitability_comparison.xlsx

Dan

[mdude77]

After what I learnt from the S4 investigation, I checked the S3s, and they have a very diminished version of the problem with pools that start at low diffs. Here's an updated binary for the S3s.

http://ck.kolivas.org/apps/cgminer/antminer/s3/4.6.1-141009/cgminer

This might make startup/pool switching a bit smoother, especially if your pools start at diff 1 and take a while to increase their vardiff.

EDIT: Hashrates even seem a little  faster after a few minutes...

Thanks CK.  If you can fix this, maybe you can fix the problem with the S2s not being able to respond quick enough to work restarts? 

M

[whonesta]

Price stays the same, BUT now shipping is NOT included, ugh...  guess BitMainTech's new owners have a mental problem. No sales...  Raise the price..  No sales..  Add shipping.. THUD!

1 less site to check now.

[IITravel01]

1.74 for 3 S3+ B10 + .52 shipping = 2.26 BTC ($790) shipped to west coast USA.

[cloh76]

off topic, but since the S3 is so much more with the shipping cost, Im considering the rockminer T1.  How loud are the fans on the rockminer T1?  Is it just as quiet as the S3s?

[MariaQin]

Sorry if this question has already been asked.
But is there a way to completely control the miner and see the stats with only using ssh?
I'm placing the miners somewhere I won't be able to reach them in a short notice. But SSH is possible.
So what would the commands be to check status, change pools, update firmware, etc.?

Hi,
SSH is possible to meet your above requirements. We are asking for commands from engineers, reply you tomorrow morning.
We would recommend you to visit miner via webGUI, which is much more convenient than SSH.

PMed.

[aclass]

This could at least be priced in USD so the total price in BTC goes down as BTC goes up.

Naah... **** the miners.

[MasterRadix]

This costs are rediculouse!
Booooh Bitmain, booooooh!!
With just 15% of difficulty increase and without VAT and electricity cost you NEVER ROI!
This is so weak. I loose so much respect to you!!

[bspurloc]

Price stays the same, BUT now shipping is NOT included, ugh...  guess BitMainTech's new owners have a mental problem. No sales...  Raise the price..  No sales..  Add shipping.. THUD!

1 less site to check now.

is the no sales part of this derived from the SOLD OUT portion? I really wanted to buy some S3's a week ago and more but there were none to buy.

[pekatete]

Price stays the same, BUT now shipping is NOT included, ugh...  guess BitMainTech's new owners have a mental problem. No sales...  Raise the price..  No sales..  Add shipping.. THUD!

1 less site to check now.

is the no sales part of this derived from the SOLD OUT portion? I really wanted to buy some S3's a week ago and more but there were none to buy.

It is the classic "chickens coming home to roost" situation. I am sure it made sense to bitmain to offer their hardware in batches (possibly to judge demand before going to the forge?), but that we are where we are now, and that on batch 10, should have been ample time and sales for them to switch to a "rolling" shop, with minimal delivery wait. Sure enough they have cottoned onto it and done that with B10 (till they discontinue S3+ ?), but then there is the S4 problem (and the lately depressed / plunging bitcoin to fiat rate), which has forced their hand in effectively overpricing B10 by adding shipping to the B9 price!
And yes, I really wanted to buy a couple of S1 upgrade kits last week too! But those, unlike the S3+ are discontinued.

[MissouriMiner]

After what I learnt from the S4 investigation, I checked the S3s, and they have a very diminished version of the problem with pools that start at low diffs. Here's an updated binary for the S3s.

http://ck.kolivas.org/apps/cgminer/antminer/s3/4.6.1-141009/cgminer

This might make startup/pool switching a bit smoother, especially if your pools start at diff 1 and take a while to increase their vardiff.

EDIT: Hashrates even seem a little faster after a few minutes...

EDIT2: No it's not any faster after longer running, it's the same overall performance. Only the startup won't overload the S3 controller on low diff starting pools.

Any chance you will build 4.6.x for the S2?

[captjrc]

This always happens when a company opens cloud mining. Their prices need to raise to make the cloud mining look like a better deal. Still, with the current price of BTC, the S3+ is cheaper if you consider the cost in fiat.

There are much better deals currently available: Rockminer and ASICMiner Prisma are two.

I use the following spreadsheet to compare apples to apples. I hesitate to share because well, people will freak out because it is not completely accurate in its forecast, but it does use an industry scaling factor, which will probably be overrun unless BTC goes up. I will probably adjust the scaling factor for myself.

You need to understand excel to use it, but essentially, when I go in, I update the difficulty, day and price, unless your electricity rate matches mine, you'd also want to update electricity $/kwh. I then click on the name of each miner, and it takes me to their website where I double check the price. If the principle cost is fiat, I list that bolded, if BTC, I list that bolded. I also use a calculation for the s&h if in BTC. The last three columns are where I compare apples to apples. The notes explain much of what I was thinking when I programmed it the way I did. You'd want to update costs for PSU and other. I have my preferred units that are essentially priced here.

http://www.hundredpalms.com/profitability_comparison.xlsx

Dan

On your Excel spreadsheet you show Rockminer R3 Box (4) having a total hash rate of 2.7 THs (or 675 GHs per unit), but Rockminer's site shows their hash rate to be 450 - 480 GHs each.  Multiply this by 4 and you get 1,800 -  1920 GHs, or 1.8 - 1.92 THs.  Do these actually Overclock up to 2.7 THs?  That would be a 40% hashing power increase over stock (480 GHs).
Secondly, you show the Rockminers as being overclocked at the astounding rate of 2.7 THs, yet the other miners are shown stock.  This isn't Apples to Apples.
Your work looks great, I'm just not sure it's completely accurate or unbiased.

[cloh76]

With BTC price value improving ove the past day, it might just be better to buy BTC and hold then sell at a high

[CHAOSiTEC]

After what I learnt from the S4 investigation, I checked the S3s, and they have a very diminished version of the problem with pools that start at low diffs. Here's an updated binary for the S3s.

http://ck.kolivas.org/apps/cgminer/antminer/s3/4.6.1-141009/cgminer

This might make startup/pool switching a bit smoother, especially if your pools start at diff 1 and take a while to increase their vardiff.

EDIT: Hashrates even seem a little faster after a few minutes...

EDIT2: No it's not any faster after longer running, it's the same overall performance. Only the startup won't overload the S3 controller on low diff starting pools.

well, seems it did some magic to one of my S3s, before it was stable at 478 (250 MHz) now, its at 501 after almost 1 hour running...

[jelin1984]

how do the installation???
478 or 250

?
for 500 gh???

[dankefoss]

On your Excel spreadsheet you show Rockminer R3 Box (4) having a total hash rate of 2.7 THs (or 675 GHs per unit), but Rockminer's site shows their hash rate to be 450 - 480 GHs each.  Multiply this by 4 and you get 1,800 -  1920 GHs, or 1.8 - 1.92 THs.  Do these actually Overclock up to 2.7 THs?  That would be a 40% hashing power increase over stock (480 GHs).
Secondly, you show the Rockminers as being overclocked at the astounding rate of 2.7 THs, yet the other miners are shown stock.  This isn't Apples to Apples.
Your work looks great, I'm just not sure it's completely accurate or unbiased.

If you click to the order page where the specs are shown, they advertise the 2700 GHs. I only bumped values if the manufacturer made a claim based upon buyers experience, or if I saw that users had reported that this was more typical. For instance, my S3+ machines work at 503 GHs, so that is how I listed it. Keep in mind this is for me. Utilize it if you desire and change it to suit your needs.

Again, no need to complain about the accuracy. I am not selling anything. I am sharing it. Use it or don't. Show me a better tool you developed or found and maybe I will use that instead.



Dan

[faetos]

how do the installation???
478 or 250

?
for 500 gh???

On web UI go to Miner Configuration > Advanced Settings

Select 250M and click apply

[cloh76]

After what I learnt from the S4 investigation, I checked the S3s, and they have a very diminished version of the problem with pools that start at low diffs. Here's an updated binary for the S3s.

http://ck.kolivas.org/apps/cgminer/antminer/s3/4.6.1-141009/cgminer

This might make startup/pool switching a bit smoother, especially if your pools start at diff 1 and take a while to increase their vardiff.

EDIT: Hashrates even seem a little faster after a few minutes...

EDIT2: No it's not any faster after longer running, it's the same overall performance. Only the startup won't overload the S3 controller on low diff starting pools.

well, seems it did some magic to one of my S3s, before it was stable at 478 (250 MHz) now, its at 501 after almost 1 hour running...

Lucky.  Mine actually went the opposite direction.  My 3 S3+ Antminers were steady at 478-500GH/s and after the update consistently below or just under 450GH/s

[lowbander80]

HA HA HA

just saw it now they ask shipping for two miners for EU 0.419 BTC

Yes am going to proceed with my order now bitmain  just wait for me

Sorry to see this company going down....