technobitVesi
Newbie
Offline
Activity: 65
Merit: 0
|
|
October 05, 2014, 06:00:14 AM |
|
HI, Today is Sunday and election day here . I'll be able to update tomorrow.
Regards: Vesi
|
|
|
|
guyver
|
|
October 05, 2014, 02:59:47 PM |
|
HI, Today is Sunday and election day here . I'll be able to update tomorrow.
Regards: Vesi
vesi enjoy your election and i would like an update on my order too. like everyone else who has been waiting along time
|
|
|
|
y_boonstra
|
|
October 05, 2014, 09:42:33 PM |
|
What got changed on the boards we are getting and the one user 2good received ? For one im seeing a lot less capacitors on the board and im tsting the board now on all diff kind of clock setings and voltage settings.. but im not nearly getting the "Good" efficientcy number as 2good was getting with undervolting!! Marto did you reduce your BOM to save money with a result a less efficient miner.. the most efficient im getting now is 1,08 W/GH and thats 660mV 400Mhz or 1,13W/GH 760mV 700MHz that no where near the 0,78 2good was getting !!! Hi, There is no difference in terms of power efficiency between the bom with a lot of caps ( it was intended for 8 chip version) and this one . IF you use bigger PSU you can see about 15% difference. Als check again 2 good measurements @ the wall . Best Vesi Im using a IBM DPS-2000BB (2000W PSU). im measuring the power draw from the wall using a voltcraft energy logger 4000. Ill post some results im getting later.. but till now the efficientcy is quite dissapoiting.. o ps my psu is around 0.96 Cos Phi regarding efficientcy of the unit it a realy good PSU and not putting out a lot of heat so that means its efficient!
|
|
|
|
Vcore
Newbie
Offline
Activity: 28
Merit: 0
|
|
October 06, 2014, 04:13:06 AM |
|
Im using a IBM DPS-2000BB (2000W PSU). im measuring the power draw from the wall using a voltcraft energy logger 4000.
Ill post some results im getting later.. but till now the efficientcy is quite dissapoiting.. o ps my psu is around 0.96 Cos Phi regarding efficientcy of the unit it a realy good PSU and not putting out a lot of heat so that means its efficient!
Are you using it for a single board? It is a great psu and I have 8 of them waiting but a single undeclocked board might be a too small load, see here (it is for the 2500W one but they are similar): http://www.rcgroups.com/forums/showpost.php?s=ff664b0f8735ce2a9c40796f40b57c49&p=18117345&postcount=611It would be great if you could measure volts and amps at the 12V rail or if you could add more load to the psu if you are using one board. If that is not the case it is strange, crappy voltage regulators should get way more inefficient as the load increases exagerating the advantage of the underclocking so it can't be that. Having less capacitors might increase the ripple, overshoot and undershoot but it shouldn't wreak havok with the efficiency. The only way I can picture the dc-dc converters drawing a lot of extra power due to having less capacitors and it being more noticeable with more underclocking is if marto increased that dc-dc converter frequency a lot to compensate for the loss of capacitance and that forced him to increase the gate drive of the mosfets to insane levels wasting a ton of power driving them. I have done that wasting up to 20W per mosfet in a high voltage generator but it required a discrete gate drive circuit, I have never seen anything like that in computer hardware and it would be insane if you cared about efficiency or even cost as caps are cheaper than building discrete gate drives.
|
|
|
|
sorehammer
|
|
October 06, 2014, 05:19:04 PM |
|
There was some info in the review for undervolting it drop the mv in the tplink just play with the settings.
|
|
|
|
technobitVesi
Newbie
Offline
Activity: 65
Merit: 0
|
|
October 06, 2014, 06:52:49 PM |
|
Hi, I'll try to put lsit of good working setups clock/voltage tommrow The best balance by our opinion is 930/760-780 mv
Regards: Vesi
|
|
|
|
y_boonstra
|
|
October 06, 2014, 09:09:12 PM |
|
Technobit, What happend to the boards? efficiency is terrible.. Setup: - PSU = IBM DSP-2000BB - 2x HEX4m board (1 connected and running) - Voltcraft energylogger 4000 - TP-link - 2x 40mm 13000 rpm fans for cooling the PSU (rpm at about 4000) - Freezer 7 pro heatsinks for the minion chips. my results so far: Efficiency [W/GH] Hashrate [GH/s]Power usage [W] or [J/s] Error rate [%]0 = NOT STABLE or NOT WORKING I have run 2 board from the PSU to put more load on it to check if my prefomance would increase do to better PSU efficiency but i only got a 3% better result !!
|
|
|
|
Vcore
Newbie
Offline
Activity: 28
Merit: 0
|
|
October 06, 2014, 11:46:51 PM |
|
Interesting spreadsheet. Can you try 400mhz at 860mv?
It should burn about 294W by my numbers. It should also hash at the same speed and the efficiency would be shit but if it burns way more or way less than that it might tell us where to look.
|
|
|
|
technobitVesi
Newbie
Offline
Activity: 65
Merit: 0
|
|
October 07, 2014, 07:17:00 AM |
|
Hi, We just made mesurment with 2 boards - small and big caps Result @ 930/780 mV is exactly 397-403 W per board with HP 1000W server PSU. Got the same result with 1200 W standard PC ATX PSU. Mesurments made with killawat @ the wall 355-365 GHs each board Regards: Vesi
|
|
|
|
sorehammer
|
|
October 07, 2014, 08:14:04 AM |
|
Any chance of the stock settings at the wall please.
|
|
|
|
technobitVesi
Newbie
Offline
Activity: 65
Merit: 0
|
|
October 07, 2014, 11:07:07 AM |
|
Hi, WE tested especially for you : setting 1016/805 - 405 Ghs - 438 W @ the wall
|
|
|
|
technobitVesi
Newbie
Offline
Activity: 65
Merit: 0
|
|
October 07, 2014, 11:23:26 AM |
|
|
|
|
|
sorehammer
|
|
October 07, 2014, 12:20:11 PM |
|
Hi, WE tested especially for you : setting 1016/805 - 405 Ghs - 438 W @ the wall
Cheers will try that been running at 1024/820 since I got it.
|
|
|
|
sorehammer
|
|
October 07, 2014, 12:58:14 PM |
|
tried 1016/805 and am only getting 235ghs are you sure of those settings.
|
|
|
|
technobitVesi
Newbie
Offline
Activity: 65
Merit: 0
|
|
October 07, 2014, 01:57:36 PM |
|
tried 1016/805 and am only getting 235ghs are you sure of those settings.
Yes it happens depends on chips try 1016/810 or 815 2 of your chips need more voltage
|
|
|
|
kea808
Member
Offline
Activity: 115
Merit: 10
“Create Your Decentralized Life”
|
|
October 07, 2014, 03:45:25 PM |
|
tried 1016/805 and am only getting 235ghs are you sure of those settings.
Yes it happens depends on chips try 1016/810 or 815 2 of your chips need more voltage Is there a way to set freq/voltage per-board? I see extreme differences between boards and it would be great to tweak each one separately.
|
|
|
|
kea808
Member
Offline
Activity: 115
Merit: 10
“Create Your Decentralized Life”
|
|
October 07, 2014, 04:11:16 PM |
|
We have it as an option in the FW, but not enabled for now We are busy with production flow now. May be we'll be able to include it on time, but no promises there
Will you be enabling this temperature option anytime soon?
|
|
|
|
y_boonstra
|
|
October 07, 2014, 06:30:43 PM |
|
Hi, WE tested especially for you : setting 1016/805 - 405 Ghs - 438 W @ the wall
Those settings are not stable with my board.. and im getting 50+ watt more than you !? undervolting aint working im not getting below 1W/GH ?? how come Is there any way to optimise the power usage? its worthless now.. edit: @ 1016Mhz and 815mV it running +- 400GH/s and using 482W
|
|
|
|
technobitVesi
Newbie
Offline
Activity: 65
Merit: 0
|
|
October 07, 2014, 06:56:02 PM |
|
Hi, Most probably one or two of the chips on your board ask for more voltage to work stable. If so in a mode with less voltage per Mhz you get 2/3 or 1/2 of the expected hash rate roughly. This is coming with chips. What we can test about the chips is if they have all engines working, and then the board if it is giving at least 400 GHs at 1024/820 If it passes this then we ship it. One more thing , sometimes it can be cooling try adding additional or faster fan on the bottom and remove and place again coolers (clean termo greese from both chip and cooler and reaply it)
Regards: Vesi
|
|
|
|
sorehammer
|
|
October 07, 2014, 06:58:20 PM |
|
Well I personally don't think the fault lies with technobit's boards but with the crappy chips supplied by BA.
|
|
|
|
|