Topic: ANTMINER S5: 1155GH(+OverClock Potential), In Stock $0.319/GH & 0.51W/GH - page 264. (Read 451048 times)

That's typically what the second revision of a generation is - more chips and less extreme clock rates. It wouldn't make sense for the first product in a generation though as the W/GH is already sufficiently low, compared with the potential loss of GH to get down further.

Wire transfer confirmations can take a few days to show, you'll probably get a dispatch confirmation have the product dispatched at the same time. You can email info@ with your transfer details to help match it up to you.

What's the usual confirmation time on an order getting switched from unpaid to paid after a wire transfer, or does it never update? My recollection from the S1/S2 days is that the website often never gets updated with that kind of thing, but it'd be nice to make sure the transfer made it through.

Chips are not much more than the cost of sand if you own the manufacturing and intellectual property.. clearly a growth path to sell miners with more and more chips per unit with low power requirements one the tech is proven to be stable to the community I would think.   It seems like it would be easy to double the hashing output with similar cooling and power consumption when the timing is right for bitmain to do so.

I'm not an EE genius.. but I don't understand why they wouldnt eventually just use different DC-DC modules internally to provide 9v or 12v w/ standard PSU.  Possibly saving that for the next miner with the BM1384?  I also don't fully understand the string vs chain discussions going on..  I guess stringing the chips make other chips dependant on one another for hashing output.. My mind draws analogies to a string of christmas lights where the rest go out when a bulb blows, or running a bunch of hard drives in raid 0 (other drives in the array have useless data with the missing member).  Of course I haven't scoured the discussions on these matters, and have only gleaned over the facts... but there sure is a lot of crazy heated discussion going on.  Anyone have any link backs to messages in this thread or other threads which may help to learn me up.   When I get a minute I'll try to read over the BM1384 whitepaper, I assume it might be the source of a lot of this conjecture.

So far MOST of the bitmain stuff I have purchased has been pretty on point and meets the specs which they claim their miners should meet*, and have publicly and quickly adjusted those specs when they were not accurate..  Ie. Antminer S3 launch 480GH/s -> 441GH/s and even gave refunds for the hashing difference if you purchased when they claimed it would be 480GH/s

(*Its important to note that honesty is good business, but bitmain isn't in the business of charity, you would be lucky to mine back the BTC you spend buying any miner.. so do this for the fun of it.. Don't buy miners to profit.. companies which make miners certainly aren't selling them so you can make a profit.. you WON'T make a profit mining.. Expect 60% of your BTC to come back that you spend AFTER you pay the electric bill. the shrewd miners who do still mine do it for the love of mining and they will sell their equip to recoup and break even, and buy back in on the next most efficient equipment.. even though they know even that isnt the sane investment to make.. The smartest miners probably rebuy their BTC so they are spending fiat not crypto, so at the end of the day they can see some justifiable profit.. even though buying BTC would have been much smarter.. but not nearly as tangible, or fun.)

As far as spondoolies goes.  I love my Sp20's.... great miners.  However the Spondoolie guys clearly overstated its hashing power at 1.7TH.  Honest marketing here would have been 1.333TH/s @ .55J/Gh at the wall...  with overclocking to 1.55TH/s (1.7TH/s being possible but not guaranteed)..  This is the real specs of the Sp20 miner...  And what about those refunds they said they would be sending for slipped delivery dates.. I never got one?  Did anyone?

Anyway.. I can understand concerns with chip failures.. they happen.. and if it takes out a whole string permanantly if a chip dies permanantly.. thats bad bad bad.. but so far every other antminer I have owned (well over 30 so far) have had exceedingly resilient asics... if one did overheat it has always come back after a reboot if it dropped off because of an overclock, and always behaved well when it was clocked at the advertised frequencies... When deployed sanely the built in safeties seem to do their job.. and allow an adjustment to consistently stable operation.

I'm just sharing some thoughts.  

Rich

That's all the S2 is, is sorely underclocked S1 - even used the same VRM hardware. Twice as many chips per bank, half the speed and half the power. Underclocking new hardware is really a discussion in arbitrary stock setpoints - when a new ASIC is released, they could say "the stock point is 0.5W/GH but it underclocks to 0.2W/GH" or they could say "the stock point is 0.2W/GH but it overclocks to 0.5W/GH" and nothing really changes (except maybe the retail price). All mining will go the way of economically-efficiently-clocked ASICs; whether that's "underclocked" or not is mostly semantics.

yeah after I have run my sp20's from spondoolies   and I get

.47 at 1000 gh = 470 watts
or .756  at 1620 gh =  1225 watts      


I think these will be the same    hard to knock the power savings from underclocking

Something about this chip gives me the feeling that all mining will eventually go the way of underclocked ASICS.  Its a lot less GH but its so much lower J/G.

OK. Sorry, I skipped a couple of beats. Now I see the 9v 0.26w/GH @ 560GH.

Carry on

these are all possible down clock ideas.  

the s-1 went from 180gh to 140gh  and from 360 watts to  170 watts  or 2 watts a gh to 1.2 watts

so if the ideas being tossed about can work ..

we may go from 1100 gh and 600 watts = .54  per gh to   950 gh and 400 watts =  .42 per gh

Also, klondike_bar, pretty sure it's a 15-segment string.

Groovy man, go hug a rainbow.

i couldnt have put it any better myself.

6 months of being fkd around over a grand is more than enough time to waste .

i know now that bitmain will lie, twist, promise & deny any and all he chooses to, to save a few pennies.

hes a typical stingy, crooked penny-pinching liar.

and the turds he employs are either fully aware of this and dont care or are ignorant and willfully so.

all that is left to do is to warn others as often as i feel like doing so.

anything less than actual compensation is a waste of my time & energy as im not getting paid to annoy/lie/confuse people.

not that i would lower myself into such a role.

dogie & his wannabe gang of timewatster shillettes deserve all they have brought upon themselves and more.

!!BUYERS BEWARE!!

So what's all the hoo haa about 0.2j/GH?

It seems that the 30A for 100% duty cycle will be sufficient if the miner will only pull 16amps at 9v?

  yeah I found I have enough on hand to fully test this.

 I advise no one to try it if they do not have the psu in hand.  If I recall the 2 psu's I found are only 80% so savings are not much power wise.    9/12  volts is 75 percent less   but an 80 percent power supply vs a 92% atx loses a lot of power

example  600 watts x .75 = 450 watts / .8 = 562 watts

vs  600 watts / .92 =  652 watts.

 So  the savings of 150 watts  because  volt reduction of  reduces to  savings of 90 watts due to the psu lacking good efficiency .

I will do testing because  I have spoken with a member about an adapter for a psu like the evga 1300 g2.  It would make the psu maybe go from 92 % to 87 %   and give the same volt reduction to  9 volts..

This translates to 450 watts/ .87 =  517 watts   
 
which is  better then the 562 watts possible  with the cheap megawatt psu's

All of this is theory  not tested.  So I advise  all to delay buying the megawatt psu's for just a bit of time.

And if the price of xbt skyrockets they can let the coupons expire without being accused of hiking prices.

problem is its rated at 30A for a 100% duty cycle (like bitcoin mining is) so the S5 would pull more power than is available for most voltages. Bitmain says that 10A+ will be sufficient for 9V but i imagine that only applies to a single blade

12v/14 chips = 0.86V (stock is 0.8V?)
10v/14 chips = 0.71V or ~0.34w/GH or ~780Gh = 265W (26.5 A)
9v/14 chips = 0.64V or ~0.26w/GH or ~560Gh = 145.6W (16 A)   

plus, at 9V draw, the linked power supply is likely <~85% efficient so youll be drawing more power at the wall

Beats me for the rationale, but there's no way I can use over 500 coupons for the various units.

I would say the massive amount of coupons bitman was giving away was a way for them to clear out their inventory. I had contacted them asking for coupons and they responded by asking if I would be willing to buy ~4x as many S4s as I had asked for coupons.

They probably were anticipating that more efficient miners would be released soon (for example the s5) which would result in lower selling prices for their machines. Also giving away coupons like this means that their customers can have the physiological feeling that they are getting a good deal

FYI, website is now showing tracking number for batch 1. Yeah!!!