Bitcoin Forum

Bitcoin => Hardware => Topic started by: Syke on December 30, 2014, 03:42:00 AM



Title: S5 review
Post by: Syke on December 30, 2014, 03:42:00 AM
Bitmain also offered me an S5, no strings attached. So here is my initial review.

Setup was easy. Plug it in and turn it on. DHCP grabs an IP address. My router identified it as "antMiner". Hit the IP with the browser and configure the web interface for pool credentials. Username and password to get into the web interface were root and root.

This new S5 is not quiet. The 120mm, 4200 rpm fan is loud. But sitting in a 40c room, the ASIC temps are reporting 62c, so it's looking nice and cool.

After running for an hour, it settled into:

Miner Speed: 1155 GH/s
BTCGuild Speed: 1145 GH/s
Power: 675 W (750 W Bronze PSU)

Power consumption is a little high due to the cheap Bronze PSU. Switching to my Platinum PSU should reach the advertised 590 W.

Initial opinion: Good efficiency, easy setup, meets advertised specs.

After some further testing I'll offer a comparison with the SP20.


Title: Re: S5 review
Post by: opieum2 on December 30, 2014, 03:56:50 AM
Thanks for the review!


Title: Re: S5 review
Post by: stan258 on December 30, 2014, 04:47:00 AM
Bitmain also offered me an S5, no strings attached. So here is my initial review.

Setup was easy. Plug it in and turn it on. DHCP grabs an IP address. My router identified it as "antMiner". Hit the IP with the browser and configure the web interface for pool credentials. Username and password to get into the web interface were root and root.

This new S5 is not quiet. The 120mm, 4200 rpm fan is loud. But sitting in a 40c room, the ASIC temps are reporting 62c, so it's looking nice and cool.

After running for an hour, it settled into:

Miner Speed: 1155 GH/s
BTCGuild Speed: 1145 GH/s
Power: 675 W (750 W Bronze PSU)

Power consumption is a little high due to the cheap Bronze PSU. Switching to my Platinum PSU should reach the advertised 590 W.

Initial opinion: Good efficiency, easy setup, meets advertised specs.

After some further testing I'll offer a comparison with the SP20.


Do you need to use all four PCIE cables or 2 for stock?


Title: Re: S5 review
Post by: klondike_bar on December 30, 2014, 05:47:15 AM

Do you need to use all four PCIE cables or 2 for stock?

at >560W its absolutely neessary to use all 4 I think, or have a high-end PSU, as 280W over a single PCIe cable will burn up any 18awg wires, and likely even get 16awg wiring warm


Title: Re: S5 review
Post by: stan258 on December 30, 2014, 05:52:14 AM
server psus and cables.....  Similar to SP20 draw per cable? I wonder if anybody has taken any readings yet. 


Title: Re: S5 review
Post by: siampumpkin on December 30, 2014, 06:02:04 AM
When you say it is loud, is it louder than an S3 or quieter? Thanks for your quick review.


Title: Re: S5 review
Post by: judypug1956 on December 30, 2014, 06:06:06 AM
When you say it is loud, is it louder than an S3 or quieter? Thanks for your quick review.

it is far louder then the s-3.

I am philipma1957.  I went to my other pc this only allows my backup account.

the s-3 is nothing compared to this.

if you want this in your house it will need fan mods.

In a shed or a barn  no big deal,

I boosted to freq 231  good solid numbers

552 watts for 1120 gh  that is better then .5 watt per gh

I don't know what these fans will allow safely but 1120 is good for now.

http://imagizer.imageshack.us/v2/1280x1024q90/912/HSm4aC.png


Title: Re: S5 review
Post by: stan258 on December 30, 2014, 06:08:58 AM
When you say it is loud, is it louder than an S3 or quieter? Thanks for your quick review.

it is far louder then the s-3.

I am philipma1957.  I went to my other pc this only allows my backup account.

the s-3 is nothing compared to this.

if you want this in your house it will need fan mods.

In a shed or a barn  no big deal

67 or so DB? Like a truck driving by or your boss telling you to get off Bitcointalk again. 


Title: Re: S5 review
Post by: Syke on December 30, 2014, 06:10:54 AM
server psus and cables.....  Similar to SP20 draw per cable? I wonder if anybody has taken any readings yet. 

Both miners have 4 power connectors. The SP20 can draw significantly more power per cable.


Title: Re: S5 review
Post by: judypug1956 on December 30, 2014, 06:14:07 AM
When you say it is loud, is it louder than an S3 or quieter? Thanks for your quick review.

it is far louder then the s-3.

I am philipma1957.  I went to my other pc this only allows my backup account.

the s-3 is nothing compared to this.

if you want this in your house it will need fan mods.

In a shed or a barn  no big deal

67 or so DB? Like a truck driving by or your boss telling you to get off Bitcointalk again. 


no 75db at one meter 
my 2 fan swap dropped it to 60db with better pitch.



Title: Re: S5 review
Post by: Bicknellski on December 30, 2014, 06:17:34 AM
WOW! Loud.

Interesting.

Thanks for the information. Does that mean by comparison the SP20 is quieter?

Edit from other thread:

Okay just got back home from a party. I need to find a few other fans.

 at freq 250 I do 824 gh and 404 watts that comes to .49 watts per gh.

 I bounce to 416 watts which is .504  watts per gh this is good.

It will run too hot with my current fan setup if I go above freq 250. temps are 60c and 57 c

It is too  loud stock!  it runs hot at stock speed with stock fan. stock fan ramps to 3800 rpm and temps were over 60c for each board.



 if you want db's 75 plus it is a nasty  pitch.

my complaint is really simple here 2 silverstone 140's would allow it to run max speed and quiet. A no brainer 1 minute mod.  instead I will need to fuck around with it.

If noise does not matter run it stock.  It is louder then the s1 or the s3 it is as loud as the asic miner tube.  So far at 225 ,250, 275, 300  and stock 350 it scales close to .51 or .52 watts

per gh. I am looking at the stock fan and it is a bizarre fan.  It is 38 mm yet has 1 inch deep blades I have never seen a fan like this. I have to go back and correct part of the review.

3 shots of the  my wife will kill me sounding  stock fan.  it has 38mm depth with 25mm blades then 13mm focus grill. it was pushing 3800 rpm at freq 350 worse sound of any fan I ever use for a miner.



Title: Re: S5 review
Post by: judypug1956 on December 30, 2014, 06:18:25 AM
server psus and cables.....  Similar to SP20 draw per cable? I wonder if anybody has taken any readings yet.  

Both miners have 4 power connectors. The SP20 can draw significantly more power per cable.

4 cables at the kwatt meter = 552 watts  so 552 x .92 = 508 watts / 4 = 127 watts at the cable.

it must use 4 pcie cables 2 are not safe.

I am at freq 231.  I will test faster in 7 hours or so .  time to crash. I am tired.

Would I buy these yes , but I know I am changing out fans.

If I do buy some.

@ bick

is the sp20 not as loud.  if you down clock to better power numbers  like 1000 to  1300gh  it is better for sound.  I can spend $0 extra money  on a sp20.  but I can't see that with the s-5.   

Now all of mine are in a home that I share with the wife.  We will be able to sleep tonight with no noise issues since I swapped out the fans on the s-5.  I am running 7 underclocked sp20's no extra money for sound mods need . Just down clocking.

  For most of us that buy it the game with the s-5 will be what is the cheapest fan replacement that allows a good clock and decent sound.  I spent 27 for the two fans.  10 on ebay for the delta and 16 on amazon for the  silverstone.


Title: Re: S5 review
Post by: Syke on December 30, 2014, 06:20:21 AM
WOW! Loud.

Interesting.

Thanks for the information. Does that mean by comparison the SP20 is quieter?

To an untrained ear I would say they are comparable. The S5 has a higher pitch.


Title: Re: S5 review
Post by: judypug1956 on December 30, 2014, 06:38:37 AM
WOW! Loud.

Interesting.

Thanks for the information. Does that mean by comparison the SP20 is quieter?

To an untrained ear I would say they are comparable. The S5 has a higher pitch.

yeah but the pitch is brutal on my ears.  

now when I underclocked to 250 the fan went to 3600rpm vs 3800 rpm.

maybe antminer can allow for lower fan speed if you clock to freq 250

I have  some more tests to do. full speed of freq 350

plus an over clock to  freq 375?

1 more fan due on the 31st

I have a 300 watt psu that allows for volts to be adjusted  to 10 volts.

I could do a board at freq 250 first at 12 volts then at 10 volts

compare to a board with a solid 12 volt atx eff of  90- 92

with careful testing and scaling the numbers I should be able to see how undervolting works.

still getting 1128 gh with freq 231


Title: Re: S5 review
Post by: MyRig on December 30, 2014, 08:20:10 AM
Hi Syke,

Thank you for the review posting!!   :D


Title: Re: S5 review
Post by: Guy Corem on December 30, 2014, 08:30:03 AM
Hi Syke,

Thank you for the review posting!!   :D
I understand you're imitating our marketing techniques, but why did you copy the miner loudness ?  ;D ;D ;D


Title: Re: S5 review
Post by: vortexz on December 30, 2014, 08:36:32 AM
Hi Syke,

Thank you for the review posting!!   :D
I understand you're imitating our marketing techniques, but why did you copy the miner loudness ?  ;D ;D ;D
ahaha
funny


Title: Re: S5 review
Post by: siampumpkin on December 30, 2014, 09:42:57 AM
Thanks for the info on the noise. Good information.  ;D


Title: Re: S5 review
Post by: philipma1957 on December 30, 2014, 01:21:08 PM
Syke thanks for review.  does your stock fan look like mine?


 case is 38 mm but blades are 25mm?


https://bitcointalk.org/index.php?topic=907219.msg9980310#msg9980310


Title: Re: S5 review
Post by: miter_myles on December 30, 2014, 01:32:53 PM
instead of having 1937109827981 S5 review threads.. can they all be consolidated into one???


Title: Re: S5 review
Post by: Bicknellski on December 30, 2014, 01:43:32 PM
server psus and cables.....  Similar to SP20 draw per cable? I wonder if anybody has taken any readings yet.  

Both miners have 4 power connectors. The SP20 can draw significantly more power per cable.

4 cables at the kwatt meter = 552 watts  so 552 x .92 = 508 watts / 4 = 127 watts at the cable.

it must use 4 pcie cables 2 are not safe.

I am at freq 231.  I will test faster in 7 hours or so .  time to crash. I am tired.

Would I buy these yes , but I know I am changing out fans.

If I do buy some.

@ bick

is the sp20 not as loud.  if you down clock to better power numbers  like 1000 to  1300gh  it is better for sound.  I can spend $0 extra money  on a sp20.  but I can't see that with the s-5.  

Now all of mine are in a home that I share with the wife.  We will be able to sleep tonight with no noise issues since I swapped out the fans on the s-5.  I am running 7 underclocked sp20's no extra money for sound mods need . Just down clocking.

  For most of us that buy it the game with the s-5 will be what is the cheapest fan replacement that allows a good clock and decent sound.  I spent 27 for the two fans.  10 on ebay for the delta and 16 on amazon for the  silverstone.

Thanks judypug... simple retro fit for S5s not a bad way or SP20s. I used to own 2 SP30's brilliant machines and sold them after they worked flawlessly for a month. Need to discuss this with my guy in Norway. Seeing as he will deal with this issue not me as the datacenter is on his turf.


Title: Re: S5 review
Post by: Syke on December 30, 2014, 04:47:24 PM
Syke thanks for review.  does your stock fan look like mine?


 case is 38 mm but blades are 25mm?


https://bitcointalk.org/index.php?topic=907219.msg9980310#msg9980310

Yeah, it's the same weird fan. Works great but sounds awful.

12 hrs in and it is rock solid at 350mhz doing 1155 GH/s with 120 V and a hot room. There is no difficulty running this thing at spec as advertised. Temps running a very cool 63c.

My SP20 peaks at about 1500 GH/s under the same conditions.


Title: Re: S5 review
Post by: MrTeal on December 30, 2014, 05:00:46 PM
The airguides on the fans aren't a sign of a crap fan. They are designed to help shape the airflow and make it more laminar. It gives some of the benefits of  a counter-rotating fan (like the Delta GFBs) without the added cost.

If you have time, use a smoke machine and look at the flow pattern of a standard fan vs one like this or the Delta FFC1212DE. It's pretty interesting.

Does anyone else here have a 50A+ adjustable power supply? I'd be interested to see some efficiency numbers in the range of 9V-14V.


Title: Re: S5 review
Post by: bgibso01 on December 30, 2014, 05:06:05 PM
Hi Syke,

Thank you for the review posting!!   :D
I understand you're imitating our marketing techniques, but why did you copy the miner loudness ?  ;D ;D ;D

So SP, does that mean since everybody is copying, you are going to lower your price and actually report your SP20 as a 1.1-1.3T miner? :)


Title: Re: S5 review
Post by: Syke on December 30, 2014, 05:07:36 PM
Does anyone else here have a 50A+ adjustable power supply? I'd be interested to see some efficiency numbers in the range of 9V-14V.

Now that they have a good ASIC, I would hope they have a board in design that can adjust voltages. This would be a perfect setup for the S6.


Title: Re: S5 review
Post by: MrTeal on December 30, 2014, 05:13:13 PM
Does anyone else here have a 50A+ adjustable power supply? I'd be interested to see some efficiency numbers in the range of 9V-14V.

Now that they have a good ASIC, I would hope they have a board in design that can adjust voltages. This would be a perfect setup for the S6.
I guess it will depend if they have spontaneous combustion issues with the S5. I don't imagine they'd want to go back to buying VRM components for the S6 unless they need to.
What they might do is include a server PSU where you can trim the output voltage through PMBus instead of an ATX supply, but I wouldn't hold my breath on that one. :P


Title: Re: S5 review
Post by: Guy Corem on December 30, 2014, 05:16:54 PM
... if they have spontaneous combustion issues with the S5 ...
One can only hope  ;D


Title: Re: S5 review
Post by: thomas_s on December 30, 2014, 05:17:09 PM
Hi Syke,

Thank you for the review posting!!   :D
I understand you're imitating our marketing techniques, but why did you copy the miner loudness ?  ;D ;D ;D

So SP, does that mean since everybody is copying, you are going to lower your price and actually report your SP20 as a 1.1-1.3T miner? :)
I've tested a bunch all 1.5-1.6 a few at 1.7, 1 at 1.4 due to a dead ASIC. It really depends on the environment they are in, if they are in a well cooled space they work at the advertised speeds. Well cooled isn't in most homes.


Title: Re: S5 review
Post by: Syke on December 30, 2014, 06:02:28 PM
Overclocked to 375 MHz. 1 hr in:

Ambient: 40c
ASIC temp: 69c
Board temp: 75c
Performance: 1250 GH/s
Power: 730 W
Efficiency: .58 J/GH (low efficiency Bronze PSU)


Title: Re: S5 review
Post by: MrTeal on December 30, 2014, 06:03:54 PM
ASIC temp: 69c
Board temp: 75c
That's interesting. Anyone know where the sensors are located?


Title: Re: S5 review
Post by: MyRig on December 30, 2014, 06:14:17 PM
Gotta learn from the master!   ;D

Bitcoin mining will come with noise hazard and heat!!!  for commercial mining operations you'll need the ear muff, no matter you use S5 or SP20.



Hi Syke,

Thank you for the review posting!!   :D
I understand you're imitating our marketing techniques, but why did you copy the miner loudness ?  ;D ;D ;D


Title: Re: S5 review
Post by: Xian01 on December 30, 2014, 06:20:52 PM
... if they have spontaneous combustion issues with the S5 ...
One can only hope  ;D
Dude. Not cool.


Title: Re: S5 review
Post by: Xian01 on December 30, 2014, 06:24:37 PM
Gotta learn from the master! 
You're not one to speak either, considering the launch issues beyond the S1.



Title: Re: S5 review
Post by: MyRig on December 30, 2014, 07:15:26 PM
 :'(
As usual, S5 will get better and better!


Title: Re: S5 review
Post by: Guy Corem on December 30, 2014, 07:22:55 PM
... if they have spontaneous combustion issues with the S5 ...
One can only hope  ;D
Dude. Not cool.
As long as BitcoinWarranty / Yoshi are ok with my humor (and they do), it's ok.


Title: Re: S5 review
Post by: Bicknellski on December 30, 2014, 07:26:05 PM
... if they have spontaneous combustion issues with the S5 ...
One can only hope  ;D
Dude. Not cool.
As long as BitcoinWarranty / Yoshi are taking it with Humor (and they do), it's ok.

Very healthy competition and healthy fun. Nice to see it between fabricators. Shills and fanbois should learn something.


Title: Re: S5 review
Post by: MyRig on December 30, 2014, 07:32:19 PM
Humor is a must okay!  

I don't think he really means to hope our hardware to catch on fire!  We know he is not that evil!  Good group of guys over there, too!!  

He has a bitter experience with that already...  

Bitcoin World is evolving so fast and most of us play works 24/7 non-stop with naps in between with super charged coffee or caffeine loaded drinks, we have to have humor and be positive to stay healthier  


YOSHI from Bitmain

*normally support staffs are monitoring this account...  

... if they have spontaneous combustion issues with the S5 ...
One can only hope  ;D
Dude. Not cool.
As long as BitcoinWarranty / Yoshi are taking it with Humor (and they do), it's ok.


Title: Re: S5 review
Post by: Guy Corem on December 30, 2014, 07:35:57 PM
Humor is okay! 

I don't think he really means to hope our hardware to catch on fire!  We know he is not that evil!  Good group of guys over there, too!! 

He has a bitter experience with that already... 

Bitcoin World is evolving so fast and most of us works 24/7 non-stop with naps in between with super charged coffee or caffeine loaded drinks, we have to have humor and be positive to stay healthier 


YOSHI from Bitmain

*normally support staffs are monitoring this account... 

... if they have spontaneous combustion issues with the S5 ...
One can only hope  ;D
Dude. Not cool.
As long as BitcoinWarranty / Yoshi are taking it with Humor (and they do), it's ok.
tushay  lol


Title: Re: S5 review
Post by: bgibso01 on December 30, 2014, 07:40:17 PM

Bitcoin mining will come with noise hazard and heat!!!  for commercial mining operations you'll need the ear muff, no matter you use S5 or SP20.


I think this is where I take both you and SP-T to task for implying the S5 and SP20 are 'home' miners.  I unconditionally agree regarding commercial operations in data centers, server rooms, garages, etc.  But 'home' miners are in the home.  At stock settings, I believe neither of these would fit that category very well.


Title: Re: S5 review
Post by: Biodom on December 30, 2014, 07:44:15 PM

Bitcoin mining will come with noise hazard and heat!!!  for commercial mining operations you'll need the ear muff, no matter you use S5 or SP20.


I think this is where I take both you and SP-T to task for implying the S5 and SP20 are 'home' miners.  I unconditionally agree regarding commercial operations in data centers, server rooms, garages, etc.  But 'home' miners are in the home.  At stock settings, I believe neither of these would fit that category very well.

exactly, although i am pretty sure that bitmain can put better fans on S5 (they already suggested that it might be possible through BitMW) and maybe include a pull fan.


Title: Re: S5 review
Post by: Xian01 on December 30, 2014, 07:52:22 PM
:'(
As usual, S5 will get better and better!
I appreciate your honesty this round, unlike the initial 28nm launch with the original 440GHs S3's.


Title: Re: S5 review
Post by: Meech on December 30, 2014, 08:03:57 PM

Bitcoin mining will come with noise hazard and heat!!!  for commercial mining operations you'll need the ear muff, no matter you use S5 or SP20.


I think this is where I take both you and SP-T to task for implying the S5 and SP20 are 'home' miners.  I unconditionally agree regarding commercial operations in data centers, server rooms, garages, etc.  But 'home' miners are in the home.  At stock settings, I believe neither of these would fit that category very well.
Agreed.  Home mining.... can barely support the noise/heat produced not to mention your only able to run a few of these without spending a thousand to upgrade your fuse box.  I used to enjoy the hum of the S1's in my living room and  even bedroom during last winter but alas that's not the case currently.  Healthy competition, maybe let's just hope were not caught in the middle.


Title: Re: S5 review
Post by: pekatete on December 30, 2014, 08:34:01 PM
The airguides on the fans aren't a sign of a crap fan. They are designed to help shape the airflow and make it more laminar. It gives some of the benefits of  a counter-rotating fan (like the Delta GFBs) without the added cost.

If you have time, use a smoke machine and look at the flow pattern of a standard fan vs one like this or the Delta FFC1212DE. It's pretty interesting.

<snip> ... </snip>

The point about the fans being crap was in relation to the noise they emit AND the S5 being touted as a home miner.
On their own, the V34809 (http://www.nidecpg.com/fanpdfs/va450dcf.pdf) (if that is the fan used in the S5) with its Advanced Vane Axial Design, can hold its own and is well suited to a commercial environment and NOT a home one; and bitmain probably chose that particular design because of its efficiencies without taking into account the home mining environment as a major target market (but targeting them anyway!).


Title: Re: S5 review
Post by: klondike_bar on December 30, 2014, 08:56:18 PM

Bitcoin mining will come with noise hazard and heat!!!  for commercial mining operations you'll need the ear muff, no matter you use S5 or SP20.


I think this is where I take both you and SP-T to task for implying the S5 and SP20 are 'home' miners.  I unconditionally agree regarding commercial operations in data centers, server rooms, garages, etc.  But 'home' miners are in the home.  At stock settings, I believe neither of these would fit that category very well.

as part of my thanks for a demo unit i advised bitmain several things to make this a "home miner" (as it is the S5 is excellent for large deployment i think). I suggested a S5-silent model that costs $20 more and includes:
1) top and bottom shrouds made with the same light durable plastic as the sides. nothing fancy or bolted down zealously.
2) a second fan, OR adjusting the temperature monitoring and PWM control to keep the fan speeds down (running at 3000RPM for 40C is silly, 2000RPM and 50C would be more like an S3)
3) default speed of 325MHz. a few percent cooler/slower/quieter and easily ovewrclocked at the expense of noise


Title: Re: S5 review
Post by: Syke on December 30, 2014, 09:02:09 PM
Overclocked to 400 MHz and still stable:

Ambient: 40c
ASIC temp: 69c
Board temp: 75c
Performance: 1320 GH/s
Power: 775 W
Efficiency: .59 J/GH (low efficiency Bronze PSU)

Over 1.3 TH/s easily achieved!


Title: Re: S5 review
Post by: IITravel01 on December 30, 2014, 09:04:13 PM

Bitcoin mining will come with noise hazard and heat!!!  for commercial mining operations you'll need the ear muff, no matter you use S5 or SP20.


I think this is where I take both you and SP-T to task for implying the S5 and SP20 are 'home' miners.  I unconditionally agree regarding commercial operations in data centers, server rooms, garages, etc.  But 'home' miners are in the home.  At stock settings, I believe neither of these would fit that category very well.

as part of my thanks for a demo unit i advised bitmain several things to make this a "home miner" (as it is the S5 is excellent for large deployment i think). I suggested a S5-silent model that costs $20 more and includes:
1) top and bottom shrouds made with the same light durable plastic as the sides. nothing fancy or bolted down zealously.
2) a second fan, OR adjusting the temperature monitoring and PWM control to keep the fan speeds down (running at 3000RPM for 40C is silly, 2000RPM and 50C would be more like an S3)
3) default speed of 325MHz. a few percent cooler/slower/quieter and easily ovewrclocked at the expense of noise

Agreed, looks like 311MHz default is more probable according to someone else's testing for a lower noise mode.  I would suggest instead of a silent version, just adding a button that would allow a silent mode so you can instantly up it if you're not in the room/area.  Also, simple request to move the network port a few mm higher as to allow easy use of the Silverstone 140mm fan in front.


Title: Re: S5 review
Post by: Biodom on December 30, 2014, 09:05:55 PM
Overclocked to 400 MHz and still stable:

Ambient: 40c
ASIC temp: 69c
Board temp: 75c
Performance: 1320 GH/s
Power: 775 W
Efficiency: .59 J/GH (low efficiency Bronze PSU)

Over 1.3 TH/s easily achieved!

these setting are almost like my SP20 in power, hash and temp. Care to report noise vs SP20?
Are you doing it on 750W bronze or higher wattage bronze?


Title: Re: S5 review
Post by: Syke on December 30, 2014, 09:12:27 PM
these setting are almost like my SP20 in power, hash and temp. Care to report noise vs SP20?
Are you doing it on 750W bronze or higher wattage bronze?

Noise is comparable to SP20 (both systems all stock hardware).

So far all my testing of this S5 is with the same 750W bronze PSU. Only 1 change - adjusting clock speed in the web UI dropdown menu. The dropdown lists speeds up to 500M, but I really don't expect it'll go that high. That would be over 1.6 TH/s!

Once I find the limits of this configuration, I'll try a platinum PSU to see if there's any more headroom for overclocking.


Title: Re: S5 review
Post by: philipma1957 on December 30, 2014, 09:31:33 PM
these setting are almost like my SP20 in power, hash and temp. Care to report noise vs SP20?
Are you doing it on 750W bronze or higher wattage bronze?

Noise is comparable to SP20 (both systems all stock hardware).

So far all my testing of this S5 is with the same 750W bronze PSU. Only 1 change - adjusting clock speed in the web UI dropdown menu. The dropdown lists speeds up to 500M, but I really don't expect it'll go that high. That would be over 1.6 TH/s!

Once I find the limits of this configuration, I'll try a platinum PSU to see if there's any more headroom for overclocking.

i topped out at 356 using the silverstone fans. i hated the sound the stock fan makes


Title: Re: S5 review
Post by: bgibso01 on December 30, 2014, 10:18:25 PM
Here's where I'm at with the 400 setting after 13min. Temp has been holding steady, but unit is in a room with the window partially open and 24F outside. :)  Don't want to push further today as I am getting ready to leave the office.


Elapsed   GH/S(5s)   GH/S(avg)   
13m52s
                1,490.62    1,368.15

these setting are almost like my SP20 in power, hash and temp. Care to report noise vs SP20?
Are you doing it on 750W bronze or higher wattage bronze?

Noise is comparable to SP20 (both systems all stock hardware).

So far all my testing of this S5 is with the same 750W bronze PSU. Only 1 change - adjusting clock speed in the web UI dropdown menu. The dropdown lists speeds up to 500M, but I really don't expect it'll go that high. That would be over 1.6 TH/s!

Once I find the limits of this configuration, I'll try a platinum PSU to see if there's any more headroom for overclocking.

i topped out at 356 using the silverstone fans. i hated the sound the stock fan makes




Title: Re: S5 review
Post by: Syke on December 30, 2014, 10:22:36 PM
I'm happy with it at 400M. Any higher starts getting hardware errors which doesn't improve performance.


Title: Re: S5 review
Post by: klondike_bar on December 30, 2014, 11:02:03 PM
I'm happy with it at 400M. Any higher starts getting hardware errors which doesn't improve performance.

i dont get much HW errors, but hashrate seems to peak at 412MHz, at 425MHz its back down around the same speeds as at 400MHz


Title: Re: S5 review
Post by: MrTeal on December 30, 2014, 11:03:53 PM
I'm happy with it at 400M. Any higher starts getting hardware errors which doesn't improve performance.

i dont get much HW errors, but hashrate seems to peak at 412MHz, at 425MHz its back down around the same speeds as at 400MHz
Can you try it at elevated voltage and see what happens?


Title: Re: S5 review
Post by: klondike_bar on December 30, 2014, 11:13:49 PM
I'm happy with it at 400M. Any higher starts getting hardware errors which doesn't improve performance.

i dont get much HW errors, but hashrate seems to peak at 412MHz, at 425MHz its back down around the same speeds as at 400MHz
Can you try it at elevated voltage and see what happens?

not sure how to easily do this without a PSU capable of 13V at >60A, or even colder intake or higher airflow. temps are 55/53C with the back end of the heatsink reading about 33C where it contacts the PCB. using a cheap infrared thermometer, readings on the PCB range from 45-65C


Title: Re: S5 review
Post by: IITravel01 on December 30, 2014, 11:55:40 PM
I'm happy with it at 400M. Any higher starts getting hardware errors which doesn't improve performance.

i dont get much HW errors, but hashrate seems to peak at 412MHz, at 425MHz its back down around the same speeds as at 400MHz
Can you try it at elevated voltage and see what happens?

not sure how to easily do this without a PSU capable of 13V at >60A, or even colder intake or higher airflow. temps are 55/53C with the back end of the heatsink reading about 33C where it contacts the PCB. using a cheap infrared thermometer, readings on the PCB range from 45-65C

I thinking he may be suggesting an adjustment in the software setting voltage (ie, as in the advanced settings on the S3+'s new firmware), I don't think the S5 has that option.


Title: Re: S5 review
Post by: MrTeal on December 30, 2014, 11:59:58 PM
I'm happy with it at 400M. Any higher starts getting hardware errors which doesn't improve performance.

i dont get much HW errors, but hashrate seems to peak at 412MHz, at 425MHz its back down around the same speeds as at 400MHz
Can you try it at elevated voltage and see what happens?

not sure how to easily do this without a PSU capable of 13V at >60A, or even colder intake or higher airflow. temps are 55/53C with the back end of the heatsink reading about 33C where it contacts the PCB. using a cheap infrared thermometer, readings on the PCB range from 45-65C

I thinking he may be suggesting an adjustment in the software setting voltage (ie, as in the advanced settings on the S3+'s new firmware), I don't think the S5 has that option.
No, I was suggesting upping the actual voltage. Klondike uses some server PSUs and many of them can trim the output voltage a bit.
I have a few big adjustable supplies, but there's probably some leeway there for adjustment.

I was actually blowing off my deck today and wondering what -20C air fed through a gas powered leaf blower would do. :)


Title: Re: S5 review
Post by: klondike_bar on December 31, 2014, 12:06:34 AM
I'm happy with it at 400M. Any higher starts getting hardware errors which doesn't improve performance.

i dont get much HW errors, but hashrate seems to peak at 412MHz, at 425MHz its back down around the same speeds as at 400MHz
Can you try it at elevated voltage and see what happens?

not sure how to easily do this without a PSU capable of 13V at >60A, or even colder intake or higher airflow. temps are 55/53C with the back end of the heatsink reading about 33C where it contacts the PCB. using a cheap infrared thermometer, readings on the PCB range from 45-65C

I thinking he may be suggesting an adjustment in the software setting voltage (ie, as in the advanced settings on the S3+'s new firmware), I don't think the S5 has that option.
No, I was suggesting upping the actual voltage. Klondike uses some server PSUs and many of them can trim the output voltage a bit.
I have a few big adjustable supplies, but there's probably some leeway there for adjustment.

I was actually blowing off my deck today and wondering what -20C air fed through a gas powered leaf blower would do. :)

Ive generally moved away from server gear, though you are probably right that some of the server supplies i have lying around can feed a bit higher if you throw a resister on the feedback pins. not a high priority though, as im aiming to get it as quiet as possible without needing subzero intake or having boards go past 65C


Title: Re: S5 review
Post by: Powell on December 31, 2014, 12:58:57 AM
I went to change the password to the webgui and got an invalid password (root/root which let me in).

I cannot login via SSH?  Am I missing something different on the S5s?


Title: Re: S5 review
Post by: Syke on December 31, 2014, 01:03:48 AM
I went to change the password to the webgui and got an invalid password (root/root which let me in).

I cannot login via SSH?  Am I missing something different on the S5s?

Yeah, I can't ssh in with root/root either.


Title: Re: S5 review
Post by: Powell on December 31, 2014, 01:09:17 AM
I can't even change the webgui password which was even weirder lol!


Title: Re: S5 review
Post by: s1gs3gv on December 31, 2014, 01:28:10 AM
Can anyone confirm the 0.2J/GH mining efficiency @ 9 volts made by Bitmain ?


Title: Re: S5 review
Post by: MrTeal on December 31, 2014, 01:43:46 AM
Can anyone confirm the 0.2J/GH mining efficiency @ 9 volts made by Bitmain ?
I can give you DC efficiency numbers, I have a nice 0-16C, 0-50A linear supply. It would be wildly inefficient compared to an ATX supply though, so for at the wall numbers you'd have to go based on your PSU efficiency.

Bitmain's not shipping my units until the 4th though.


Title: Re: S5 review
Post by: klondike_bar on December 31, 2014, 02:17:14 AM
I went to change the password to the webgui and got an invalid password (root/root which let me in).

I cannot login via SSH?  Am I missing something different on the S5s?

Yeah, I can't ssh in with root/root either.
root/admin, its a beaglebone like the S2/S4


Title: Re: S5 review
Post by: Syke on December 31, 2014, 02:21:35 AM
root/admin, its a beaglebone like the S2/S4

Yup, that does it.


Title: Re: S5 review
Post by: s1gs3gv on December 31, 2014, 03:31:27 AM
Can anyone confirm the 0.2J/GH mining efficiency @ 9 volts made by Bitmain ?
I can give you DC efficiency numbers, I have a nice 0-16C, 0-50A linear supply. It would be wildly inefficient compared to an ATX supply though, so for at the wall numbers you'd have to go based on your PSU efficiency.

Bitmain's not shipping my units until the 4th though.

Please do, when you can.


Title: Re: S5 review
Post by: Meech on December 31, 2014, 05:56:30 AM
I'm happy with it at 400M. Any higher starts getting hardware errors which doesn't improve performance.

i dont get much HW errors, but hashrate seems to peak at 412MHz, at 425MHz its back down around the same speeds as at 400MHz
Can you try it at elevated voltage and see what happens?

not sure how to easily do this without a PSU capable of 13V at >60A, or even colder intake or higher airflow. temps are 55/53C with the back end of the heatsink reading about 33C where it contacts the PCB. using a cheap infrared thermometer, readings on the PCB range from 45-65C

I thinking he may be suggesting an adjustment in the software setting voltage (ie, as in the advanced settings on the S3+'s new firmware), I don't think the S5 has that option.
No, I was suggesting upping the actual voltage. Klondike uses some server PSUs and many of them can trim the output voltage a bit.
I have a few big adjustable supplies, but there's probably some leeway there for adjustment.

I was actually blowing off my deck today and wondering what -20C air fed through a gas powered leaf blower would do. :)
Don't try outdoor fed air, too much moisture unless filtered somehow.  May work for awhile but eventually leads to corrosion.


Title: Re: S5 review
Post by: MrTeal on December 31, 2014, 07:14:33 AM
Don't try outdoor fed air, too much moisture unless filtered somehow.  May work for awhile but eventually leads to corrosion.
There's very little humidity in the outside air here. There's as much actual moisture in the air at -20C and 90%RH as there is at 10C and 10% RH.


Title: Re: S5 review
Post by: pak13 on December 31, 2014, 07:41:47 AM
Hi Syke,

Thank you for the review posting!!   :D
I understand you're imitating our marketing techniques, but why did you copy the miner loudness ?  ;D ;D ;D

so both of you could afford a beer at the end of the day :P A way to cut costs... hearing is overrated from a designer/engineer perspective. Cost from a managerial perspective. 


Title: Re: S5 review
Post by: pak13 on December 31, 2014, 07:44:47 AM
I'm happy with it at 400M. Any higher starts getting hardware errors which doesn't improve performance.

i dont get much HW errors, but hashrate seems to peak at 412MHz, at 425MHz its back down around the same speeds as at 400MHz
Can you try it at elevated voltage and see what happens?

not sure how to easily do this without a PSU capable of 13V at >60A, or even colder intake or higher airflow. temps are 55/53C with the back end of the heatsink reading about 33C where it contacts the PCB. using a cheap infrared thermometer, readings on the PCB range from 45-65C

I thinking he may be suggesting an adjustment in the software setting voltage (ie, as in the advanced settings on the S3+'s new firmware), I don't think the S5 has that option.
No, I was suggesting upping the actual voltage. Klondike uses some server PSUs and many of them can trim the output voltage a bit.
I have a few big adjustable supplies, but there's probably some leeway there for adjustment.

I was actually blowing off my deck today and wondering what -20C air fed through a gas powered leaf blower would do. :)
Don't try outdoor fed air, too much moisture unless filtered somehow.  May work for awhile but eventually leads to corrosion.

Unless of course you live where the humidity is typically <30% on average. Temps year average 12c. Hint... take a look at where all the new (google, facebook, etc) data centres/centers are located.


Title: Re: S5 review
Post by: dogie on December 31, 2014, 08:31:11 AM
Does anyone else here have a 50A+ adjustable power supply? I'd be interested to see some efficiency numbers in the range of 9V-14V.

Now that they have a good ASIC, I would hope they have a board in design that can adjust voltages. This would be a perfect setup for the S6.
I guess it will depend if they have spontaneous combustion issues with the S5. I don't imagine they'd want to go back to buying VRM components for the S6 unless they need to.
What they might do is include a server PSU where you can trim the output voltage through PMBus instead of an ATX supply, but I wouldn't hold my breath on that one. :P

It will be interesting how the 9V or variable PSU thing works out. At the moment there's no supply, model, infrastructure or experience with them - but now there's a market.


Title: Re: S5 review
Post by: MrTeal on December 31, 2014, 04:52:01 PM
Does anyone else here have a 50A+ adjustable power supply? I'd be interested to see some efficiency numbers in the range of 9V-14V.

Now that they have a good ASIC, I would hope they have a board in design that can adjust voltages. This would be a perfect setup for the S6.
I guess it will depend if they have spontaneous combustion issues with the S5. I don't imagine they'd want to go back to buying VRM components for the S6 unless they need to.
What they might do is include a server PSU where you can trim the output voltage through PMBus instead of an ATX supply, but I wouldn't hold my breath on that one. :P

It will be interesting how the 9V or variable PSU thing works out. At the moment there's no supply, model, infrastructure or experience with them - but now there's a market.
Or if you want to get really creative (and adventurous), wire 3 12V supplies in series similar to what the RC guys do, and then string together 4 S5s. ;)

Edit: Don't actually do this. Things will blow up.


Title: Re: S5 review
Post by: Biodom on December 31, 2014, 04:58:06 PM
Hi Syke,

Thank you for the review posting!!   :D
I understand you're imitating our marketing techniques, but why did you copy the miner loudness ?  ;D ;D ;D

so both of you could afford a beer at the end of the day :P A way to cut costs... hearing is overrated from a designer/engineer perspective. Cost from a managerial perspective.  

only for the nearsighted people. I can afford (electricity wise) many more miners that i will NOT employ simply because of the noise and my house configuration-no basement. How many people are in the same category-I suspect many, but I am 90-99% sure that nobody at SPT and BMT had bothered to investigate.


Title: Re: S5 review
Post by: klondike_bar on December 31, 2014, 04:58:36 PM
Does anyone else here have a 50A+ adjustable power supply? I'd be interested to see some efficiency numbers in the range of 9V-14V.

Now that they have a good ASIC, I would hope they have a board in design that can adjust voltages. This would be a perfect setup for the S6.
I guess it will depend if they have spontaneous combustion issues with the S5. I don't imagine they'd want to go back to buying VRM components for the S6 unless they need to.
What they might do is include a server PSU where you can trim the output voltage through PMBus instead of an ATX supply, but I wouldn't hold my breath on that one. :P

It will be interesting how the 9V or variable PSU thing works out. At the moment there's no supply, model, infrastructure or experience with them - but now there's a market.
Or if you want to get really creative (and adventurous), wire 3 12V supplies in series similar to what the RC guys do, and then string together 4 S5s. ;)

Edit: Don't actually do this. Things will blow up.

I dont think you can safely drive that much wattage through the S5 units like that. (if you wanted to try, you could run both blades of an S5 in series on 24V)


Title: Re: S5 review
Post by: MrTeal on December 31, 2014, 05:09:09 PM
Does anyone else here have a 50A+ adjustable power supply? I'd be interested to see some efficiency numbers in the range of 9V-14V.

Now that they have a good ASIC, I would hope they have a board in design that can adjust voltages. This would be a perfect setup for the S6.
I guess it will depend if they have spontaneous combustion issues with the S5. I don't imagine they'd want to go back to buying VRM components for the S6 unless they need to.
What they might do is include a server PSU where you can trim the output voltage through PMBus instead of an ATX supply, but I wouldn't hold my breath on that one. :P

It will be interesting how the 9V or variable PSU thing works out. At the moment there's no supply, model, infrastructure or experience with them - but now there's a market.
Or if you want to get really creative (and adventurous), wire 3 12V supplies in series similar to what the RC guys do, and then string together 4 S5s. ;)

Edit: Don't actually do this. Things will blow up.

I dont think you can safely drive that much wattage through the S5 units like that. (if you wanted to try, you could run both blades of an S5 in series on 24V)
From the point of view of the S5, it wouldn't be any different. Ideally you'd still have 12V across each one and the same current flowing through them. The problem is sequencing and making sure you don't get much more than 12V across any one unit.


Title: Re: S5 review
Post by: Meech on December 31, 2014, 05:39:46 PM
Don't try outdoor fed air, too much moisture unless filtered somehow.  May work for awhile but eventually leads to corrosion.
There's very little humidity in the outside air here. There's as much actual moisture in the air at -20C and 90%RH as there is at 10C and 10% RH.
Yes but I live in the Pacific NW.  My bad for assuming these conditions are the norm.


Title: Re: S5 review
Post by: MrTeal on December 31, 2014, 05:49:57 PM
Don't try outdoor fed air, too much moisture unless filtered somehow.  May work for awhile but eventually leads to corrosion.
There's very little humidity in the outside air here. There's as much actual moisture in the air at -20C and 90%RH as there is at 10C and 10% RH.
Yes but I live in the Pacific NW.  My bad for assuming these conditions are the norm.
LOL... No, not everywhere. Here in the prairies we have to be a little more careful. Those days you look out the window and its gloriously sunny and it seems you can see so far you can see the curvature of the Earth; put on some long johns before you head outside. :)


Title: Re: S5 review
Post by: Syke on December 31, 2014, 10:37:04 PM
15A circuit, 1300W PSU, 2 miners living in perfect harmony:

Ebony (Overclocked S5): 1240 GH/s
Ivory (Underclocked SP20): 1140 GH/s

Each miner is drawing 700 W (+/- 2%) at the wall.


Title: Re: S5 review
Post by: Biodom on December 31, 2014, 10:59:40 PM
15A circuit, 1300W PSU, 2 miners living in perfect harmony:

Ebony (Overclocked S5): 1240 GH/s
Ivory (Underclocked SP20): 1140 GH/s

Each miner is drawing 700 W (+/- 2%) at the wall.

ha ha, I have almost exactly the same setup for a pair and a similar thought  ;D.
https://bitcointalk.org/index.php?topic=902305.msg9993723#msg9993723
Once I get quiet S5 fans-will do a second pair, otherwise it would cause some stress during the NY holidays.


Title: Re: S5 review
Post by: klondike_bar on December 31, 2014, 11:54:07 PM
15A circuit, 1300W PSU, 2 miners living in perfect harmony:

Ebony (Overclocked S5): 1240 GH/s
Ivory (Underclocked SP20): 1140 GH/s

Each miner is drawing 700 W (+/- 2%) at the wall.

ha ha, I have almost exactly the same setup for a pair and a similar thought  ;D.
https://bitcointalk.org/index.php?topic=902305.msg9993723#msg9993723
Once I get quiet S5 fans-will do a second pair, otherwise it would cause some stress during the NY holidays.

for the S5 fans, try driving them at 5V if its cold intake, or 7V if you want to overclock. One of these http://www.moddiy.com/product_images/j/547/T1JRxJXglbXXce7cAT_011207__14471_zoom.jpg will work, or makeshift your own (its just a few watts). The PWM revving to 80-100% is terribly loud.

as for the SP20 - you should be able to drive it a bit higher, to 1200Gh or so, by setting each PCIe to 155W and the voltage range to 0.638-0.655. Should settle in around 1200GH/0.646V and draw about equal to the S1 (this is with fans way down, which might free up 5-10W)


Title: Re: S5 review
Post by: Saturn7 on January 05, 2015, 07:45:13 PM
Would this work?

http://www.jameco.com/webapp/wcs/stores/servlet/Product_10001_10001_2104719_-1 (http://www.jameco.com/webapp/wcs/stores/servlet/Product_10001_10001_2104719_-1)

It's 9 volts with 16.7 amps output but only rated at 150 watts.


Title: Re: S5 review
Post by: Syke on January 05, 2015, 11:50:57 PM
Would this work?

http://www.jameco.com/webapp/wcs/stores/servlet/Product_10001_10001_2104719_-1 (http://www.jameco.com/webapp/wcs/stores/servlet/Product_10001_10001_2104719_-1)

It's 9 volts with 16.7 amps output but only rated at 150 watts.

Interesting. Although you'd probably need 4 of them per S5.


Title: Re: S5 review
Post by: Syke on January 27, 2015, 04:50:30 PM
Sorry to report that half my SP20 has kicked the bucket. Cycling the power and upgrading the firmware has not helped.

Code:
2.6.1
Uptime:153 | FPGA ver:100 | BIST in 60
-----BOARD-0-----
PSU[UNKNOWN]: 0->(127w/127w)[127 127 127] (->127w[127 127 127]) (lim=155) 0c 252GH cooling:0/0x0
-----BOARD-1-----
PSU[UNKNOWN]: 0->(140w/140w)[140 140 140] (->140w[140 140 140]) (lim=155) 0c 277GH cooling:0/0x0
-----BOARD-2-----
PSU[UNKNOWN]: 0->(5w/5w)[5 5 5] (->5w[5 5 5]) (lim=155) 0c 0GH cooling:0/0x0
-----BOARD-3-----
PSU[UNKNOWN]: 0->(5w/5w)[5 5 5] (->5w[5 5 5]) (lim=155) 0c 0GH cooling:0/0x0
LOOP[0] ON TO:0 (w:6)
 0: DC2DC/1/:[vlt1:625 vlt2:629(DCl:794 Tl:794 Ul:629) 54W  87A  50c] ASIC:[ 85c (125c) 710hz(BL: 710)    3 (E:192) F:0 L:0]
 1: DC2DC/1/:[vlt1:626 vlt2:629(DCl:794 Tl:794 Ul:629) 46W  73A  56c] ASIC:[ 85c (125c) 600hz(BL: 600)    3 (E:193) F:0 L:0]
LOOP[1] ON TO:0 (w:14)
 2: DC2DC/1/:[vlt1:626 vlt2:629(DCl:794 Tl:794 Ul:629) 57W  91A  62c] ASIC:[ 85c (125c) 740hz(BL: 740)    3 (E:193) F:0 L:0]
 3: DC2DC/1/:[vlt1:626 vlt2:629(DCl:794 Tl:794 Ul:629) 53W  85A  64c] ASIC:[ 85c (125c) 700hz(BL: 700)   11 (E:193) F:0 L:0]
LOOP[2] OFF TO:0 (test serial failed or something)
4: disabled (i2c BAD, btw!)
5: disabled (i2c good, no OC)
LOOP[3] OFF TO:0 (test serial failed or something)
6: disabled (i2c BAD, btw!)
7: disabled (i2c BAD, btw!)


Title: Re: S5 review
Post by: Guy Corem on January 27, 2015, 05:20:46 PM
Sorry to report that half my SP20 has kicked the bucket. Cycling the power and upgrading the firmware has not helped.

Code:
2.6.1
Uptime:153 | FPGA ver:100 | BIST in 60
-----BOARD-0-----
PSU[UNKNOWN]: 0->(127w/127w)[127 127 127] (->127w[127 127 127]) (lim=155) 0c 252GH cooling:0/0x0
-----BOARD-1-----
PSU[UNKNOWN]: 0->(140w/140w)[140 140 140] (->140w[140 140 140]) (lim=155) 0c 277GH cooling:0/0x0
-----BOARD-2-----
PSU[UNKNOWN]: 0->(5w/5w)[5 5 5] (->5w[5 5 5]) (lim=155) 0c 0GH cooling:0/0x0
-----BOARD-3-----
PSU[UNKNOWN]: 0->(5w/5w)[5 5 5] (->5w[5 5 5]) (lim=155) 0c 0GH cooling:0/0x0
LOOP[0] ON TO:0 (w:6)
 0: DC2DC/1/:[vlt1:625 vlt2:629(DCl:794 Tl:794 Ul:629) 54W  87A  50c] ASIC:[ 85c (125c) 710hz(BL: 710)    3 (E:192) F:0 L:0]
 1: DC2DC/1/:[vlt1:626 vlt2:629(DCl:794 Tl:794 Ul:629) 46W  73A  56c] ASIC:[ 85c (125c) 600hz(BL: 600)    3 (E:193) F:0 L:0]
LOOP[1] ON TO:0 (w:14)
 2: DC2DC/1/:[vlt1:626 vlt2:629(DCl:794 Tl:794 Ul:629) 57W  91A  62c] ASIC:[ 85c (125c) 740hz(BL: 740)    3 (E:193) F:0 L:0]
 3: DC2DC/1/:[vlt1:626 vlt2:629(DCl:794 Tl:794 Ul:629) 53W  85A  64c] ASIC:[ 85c (125c) 700hz(BL: 700)   11 (E:193) F:0 L:0]
LOOP[2] OFF TO:0 (test serial failed or something)
4: disabled (i2c BAD, btw!)
5: disabled (i2c good, no OC)
LOOP[3] OFF TO:0 (test serial failed or something)
6: disabled (i2c BAD, btw!)
7: disabled (i2c BAD, btw!)

Was it modded ?
Please contact remo@ for diagnostic.


Title: Re: S5 review
Post by: Syke on January 27, 2015, 05:31:13 PM
Was it modded ?
Please contact remo@ for diagnostic.

All stock. I'll email.