Ok, I got my SP20 units in and they were very quick to set up and get going.
PERSONAL OPINION: Either machine is a decent machine!!!
If sound is your deciding factor, the S5 has a little higher pitch than the SP20. My SP20 has a dull vibration sound from the other room. Reminds me of a washing machine noise or far off grain dryer.
By the time I played around with settings on both units, the $/gh for sound tolerance was nearly identical. The SP20 down clocked and the S5 over clocked. I know I won't be able to take the SP20 home because that dull sound resonates and will cause the wife to go postal. I'm going to try and sneak the S5 units home and see if she notices.
So after all is said and done, my choice will end up being which ever one I have enough power supplies for. I think I'm going to order more of both!
The S5 can be very managable if the fan is driven at 5V, and can still push 1.2TH at about 600W. The SP20 is about the same volume at ~10% fan speed and can drive about 1.25TH/600W in the same ambient settings. (I dont have a killawatt, but I know its not tripping my CS650M GOLD supplies (they trip at about 625W+ on 12V)) |
|
|
|
try running the included fan at 5V without PWM. The noise is managable (slightly louder than s3, but about 30% of the 12V volume), and the board temperatures should be about 54C in cold air or 63C in a warm room. Its still not something for a living room, but a basement or spare bedroom would likely contain the majority of noise
How would this be done? Direct to +5 on PSU? yup, feed 5V to the red wire, GND to the black. Even a 7V adapter (yellow and red wires from PSU) helps a lot to drop noise. ideally you should get it as slow as possible while achieving <62C |
|
|
|
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.  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) |
|
|
|
Still waiting....  My first set is suppose to show up today. Hope they give me enough time to test them and decide between the SP20 or S5. The S5 units that I received run very stable at stock and rather impressive overclock. Sound is going to be the determining factor I think. I have a feeling the S5 will be a winner for 0.5w/GH efficiency and low noise (with 5V adapter for the fan or revised PWM controller settings), but the SP20 has a lot more flexibility to stack and be driven down to better efficiencies without needing a custom <11V PSU to do so. |
|
|
|
sounds like the S5 has the SP20 beat for use at ~1250GH/630W or ~0.51w/GH
well, very close, but S5 has this crazy fan-my amazon fan supply for mod (got both Noctua and Silverstone) will be here only on Friday-I even had to get the prime trial which I never wanted in order to get it faster.. try running the stock fan at 5V - its actually a very manageable volume at that speed (~2200rpm approx) and is sufficienct to keep the unit around 60-65C at stock speeds. With some cold air at the intake I can keep under 62C at 375MHz |
|
|
|
it's the same format as s1, s3, sp20.... the more power it consumes the more ventilation it needs, resulting in higher decibels anything can be made to run silent, it only depends on your skill and the money available  Firstly, Kudos on a great post Philip! The form factor, the power cables still enter from the now open top. Ether on the end. With plastic sides, this makes, the form factor is rather different. Almost impossible to stack the s5's? The SP20 has the edge here, fully enclosed and easily stackable with cables all on one end. It sounds as though Bitmain and the s5 followed AM's one fan option to dump a lot of heat. We know that didn't work out too well long term. With the small price difference, lets be honest, I'd prefer something that is a Ferrari, over a plastic sided heater. An afterthought, cooling is easier also on a large scale with stackable single end entry. s5 has two things going for it: 1) it can operate at 0.5w/GH and up to 1300GH+ on 12V, whereas the SP20 is using closer to 0.6W/gh (or about 150W more) to achieve similar speeds 2) it should be cheaper to produce - if so it might be a viable farm option and a better fan control system would allow home use. you could stack these the same way as with the S1 or s3 - I often made 3x2 stacks with the units on thier sides and the bottoms of the units against each other or a flat surface. |
|
|
|
Awesome, thanks for the confirm.
Second question, for sure helpful for me, may be helpful for others. I plan on picking up several S5's to run at home. Have seen the Good News about overclocking potential of ~1400GH. All things considered, which power supply would you use to run your S5? 1st 2nd and 3rd choices, please. All Corsair PSUs:
1x HX850W
1x HX750W
3x CX750W
2x CX600
3x CX500M
Bitmain says 590W at the wall. As listed, they are pretty much in 1st 2nd and 3rd choice order now, yes?
Appreciate the feedback.
its unlikely 1400GH can be done at 12V, might require 13V and a lot of cold air. however, 1300GH is possible, and would draw something like 650-675W at the PSU. CX750W is bronze rated i think, so you would really cut it close. the HX650W is gold i think and should be a better option. I found 387.5MHz should draw around 610-630W, and was capable of tripping my CS650M PSU (612.5W of 12V rail) after about 10 minutes of mining. |
|
|
|
now you have seen the S5 naked From the consumer point of view, what would you expect from S5 or wish for? Would you rather pay $15-20 more on quieter fans (same hashing power but better fan or fans? Yes they were. And I've updated the original post to say a reboot seemed to have solved the issue. Thanks for the concern. Both units appear to be operating as expected, just not as quiet as hoped for and not a clear choice over the SP20 so far.
If you have a fan->molex adapter try running the fan at 7V, or even 5V (if you have cold air and <375MHz). Right now the PWM controller is way out of whack for the home user. Keeping the heatsinks at 48-52C at nearly 100% speeds is really loud when 40% speed can keep everything around 55-60C. If it was possible to implement a software control to change the PWM mode or set a fixed fan speed like on the SP20 home use would be very possible with this unit and achieve ~1100GH/550W or better at reasonable volume So out of 3 5 reviews 2 were duds?
nope, this user fixed it with a reboot. |
|
|
|
 The S5 has some serious overclocking potential! 1.3TH seems to not be a problem and 1.4TH may be possible. (and its likely that efficiency will not change significantly) With cold enough air intake (-7C) I could achieve 412.5MHz for around 1.36TH, but going beyond i lost hashrate even with lots of cold air, showing a limit at 12V. I've actually switched the fan to 5V w/o PWM and with 10C at the intake temps stay around 62C at 375MHz/1.22TH and noise is actually very managable, just a bit louder than a maxed-out S1. but the PWM controller drives it at twice that, and is too loud for home use. |
|
|
|
I would like to know what the dark number is, those miners using free (students at school or home, employees at work, play) or subsidized electricity-those with low or no cost that pass the buck to another entity-like tax payers (foreign governments, companies, corporations) . That would be interesting,
nothing compared to the 1MW+ installations that are being deployed worldwide. likely less that 25% of small-scale miners have free power I imagine. Even then, it may be limited to only a few kW of capacity |
|
|
|
|
sounds like the S5 has the SP20 beat for use at ~1250GH/630W or ~0.51w/GH
|
|
|
|
Cheap electricity is the key, the reality is that most people pay higher than .10 usd kw/hr
And the last 3 difficulty adjustments have been much below average, look for next diff jump around 10% or more as hashrate pushes 320-340 ph
+1 & +1. however, a 10% network growth would probably knock most remaining 1w-1.2w/GH miners off the network except in areas with cheap power (and those guys usually liquidate while they can). And there is a LOT of hardware in that efficiency range, likely 50PH of gear will leave the network in the time that it takes to add 100PH of new hardware |
|
|
|
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 |
|
|
|
Bitmain any info on how to change the webgui password and login via SSH?
SSH is root/admin because of the beaglebone. There is no asic-freq.conf file, instead the cgminer.conf includes the frequency instructions. I poked around a bit and there doesnt seem to be much to play with in SSH other than crontab |
|
|
|
bad idea. those things are fine for up to maybe 100CFM, but its gonna make the S5 fan work really hard and extremely loudly to pull lots of air through it. Honestly dust wont hurt the S5, I ran S1 units for months in a dusty area and despite some lint buildup that made them run louder, the ASICs were unnaffected |
|
|
|
|
try running the included fan at 5V without PWM. The noise is managable (slightly louder than s3, but about 30% of the 12V volume), and the board temperatures should be about 54C in cold air or 63C in a warm room. Its still not something for a living room, but a basement or spare bedroom would likely contain the majority of noise
|
|
|
|
|
swapped in an S1 fan, driven at 7V only. At this power the unit is a reasonable volume for home mining, but at 350MHz or 363.5Mhz requires some cool air intake to keep below 65C. at room temperature it would probably be at about 325MHz to maintain less than 65C
|
|
|
|
This unit can overclock, but needs some cool air.
-5C intake air, 412.5MHz, 53-56C temps, 1.32TH (1.35Th expected). Any higher than 412.5MHz seems to result in losing hashrate, even if temperatures are kept under 56C
-5C intake air, 400MHz, 51-54C temps, 1.32TH (1.35Th expected).
17C intake air, 400MHz, 61-65C temps, 1.32TH (1.33Th expected). almost no HW errors at this temperature.
This is with the stock fan connected to a 7V power, no PWM. Its about 3000RPM approx.
My next step will be to swap in an S1 fan at 7V (about 2/3 the airflow) and see what happens. Should be very manageable volume, but to keep temeratures under 65C it will need either a cold intake at 400MHz or warm intake at stock 350MHz
Do you have a kill-a-watt meter? If so, what is the power draw at the wall at these different settings? I don't, but it should be pretty linear as the chip voltage is no different, only frequency. (ie: if stock is 350MHz/560W, 400MHZ would use about 640W) One of the other reviews found that efficiency at stock was about 0.52w/GH whereas at 250MHz it was around 0.50w/gh. This is far more linear then the sp20 is.. about .49 to .51 from freq 225 to 356. I was wondering when it got to 400-412.5 what it was pulling at the wall where it would probably be not so linear at the peak. it really wouldnt change much. The SP20 changes voltage to change efficiency, whereas the S5 is fixed based on the input (12.05V). even at 412.5MHz its likely 0.54w/gh or better |
|
|
|
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 |
|
|
|
This unit can overclock, but needs some cool air.
-5C intake air, 412.5MHz, 53-56C temps, 1.32TH (1.35Th expected). Any higher than 412.5MHz seems to result in losing hashrate, even if temperatures are kept under 56C
-5C intake air, 400MHz, 51-54C temps, 1.32TH (1.35Th expected).
17C intake air, 400MHz, 61-65C temps, 1.32TH (1.33Th expected). almost no HW errors at this temperature.
This is with the stock fan connected to a 7V power, no PWM. Its about 3000RPM approx.
My next step will be to swap in an S1 fan at 7V (about 2/3 the airflow) and see what happens. Should be very manageable volume, but to keep temeratures under 65C it will need either a cold intake at 400MHz or warm intake at stock 350MHz
Do you have a kill-a-watt meter? If so, what is the power draw at the wall at these different settings? I don't, but it should be pretty linear as the chip voltage is no different, only frequency. (ie: if stock is 350MHz/560W, 400MHZ would use about 640W) One of the other reviews found that efficiency at stock was about 0.52w/GH whereas at 250MHz it was around 0.50w/gh. |
|
|
|
|
Show Posts
