Nvidia 2xxx or 3xxx series (+LHR) cards can dual-mine with Etherium using t-rex miner. Hardly any reduction in Eth hash rate.
I have used 2080ti, 2070, 2060KO, 1660ti all OK.

The 1xxx series cards don't dual mine well (low hash rate). I tried EVGA 1060.
You could use them to single mine just Alephium but newer series cards and dual-mining will be more profitable IMHO.

16h 15m
12.1% luck

how low will it go ?

Me 2.
I have noticed this.
Balance unchanged for ages today.
I have faith payments will accrue, eventually.
I hodl, so no sweat here.
1 day NMC will rain from the sky!
No worries.

I'm mining OK (using bfgminer) so it may just be you.

I use one of these http://www.tlc-direct.co.uk/Products/SLTD500SILENT.html silenced 6" duct fan to shift 500m3/h.
You could try using 3. They only draw 50W running flat-out.
They make them in a 8" version too. You might get away with using just two of those.
Good Luck !

Yea, my last NMC transaction from Eligius was on 09/03/2014.
But WK has been tied up fighting this massive long-running DDoS, which I don't understand why it has gone on so long for.
Anyway, I trust WK will get around to fixing NMC eventually but I understand that this in itself is a non-trivial fix so would take some time on its own even without the DDoS.
I just look forwards to a big NMC payout in the (hopefully) not too far distant future.
And qdos to WK for fighting the good fight aginst the mother of all DDoS's.

Actually, data center costs are all to do with location.
If you do as KNC have done and build your data centre on the edge of the arctic circle then there is unlimited free cooling by just opening the door.
They also built it next door to a hydro-electric dam, so plenty of cheap green electricity too.
And they rented an existing building, not built it from scratch, so start-up costs were kept low and operation start-up time was minimised.
Needless to say they made the ASIC mining machines at cost price and are now minting their own money big time.
Sadly with the cost of the S2's I can't afford to do the same thing.
I guess Bitmain doesn't want to move to Sweden either. 

Maybe they don't have any BTC they can use.
It could be that they cash out all mined BTC every day under contract to some big player/ exchange/ inverstor.
Maybe the Winkelvoss's buy it all.
Who knows?
KNC aren't saying.

SSH into miner, login as root.
Then just type POWEROFF and wait a minute or two for fan to slow down.
Then turn off PSU.
Simples.

Just an idea but, to stop the thieves getting that last 0.1 you could go in and change the donation % to 100%.
If you're lucky and they don't spot it and change it back then at least the pool would benefit instead of those scum bags.

Extrapolating from my November Jupiter, which uses 820W at wall using Platinum PSU to provide 760GH/s (190GH/s per board overclocked to 1000MHz) - so 18 boards = 3420GH/s @ 3690W.
Personally I would go with 4 x 1000W PSUs, but 3 x 1250W should be possible if you like running close to the wind. 

Yes, normal, I've had this several times.
It's where the estimate of your next payment varies slightly as time goes on and you are on the edge of the block, as I understand it.
Not to worry, if you don't get paid this block then you are well positioned to be paid, almost certainly, next block, assuming no untoward events, such as an orphan.

IMHO it is to KNC's advantage at this point to encourage refunds so that :-
1. They will have less future competition, as you do not end up mining against them with that hash unit.
2. They get more state of the art machines to stock their data centre(s) with from the initial production batches.
3. You may use the refund to purchase BTC on the market, i.e. from them potentially, and this will help maintain or rise the BTC/USD price, being advantageous to them as they are selling so much BTC daily.

Sad times.

The pool's been working fine (for me).
But the stats web site has been DDoS'ed to hell and back for the past couple of weeks, AFAIK.
WK has been so tied up mitigating the DDoS that other things have been taking a bit of a back-seat.
The last week's luck has not been so good but - that's luck for you !
I can take variable payouts and the pool has not failed me yet, so I'm staying around for a while.
I use BTCGuild as a failover because it is reliable (have used it for years for this reason) but I do notice the payments are smaller (if more regular).

Thanks very much for this info Luke-Jr.

Any additional merge-mined coins will be appreciated. 

+1

I just get a white page now when trying to bring up my stats.

Go get'em WK, don't let 'em grind you down.  

P.S. next time I can get into MyEligius I will up my donation rate.

Yes, pool has been working for me all morning.

Thanks for the offer but I don't think I can take you up on it this year (I would like to because that's what I spent most of my life doing, auditing DC's, it's almost like a hobby of mine now).
I'm sure you have a nice new "state of the art" DC but my point was that most are not.
I worked in a DC with 1.3 PUE, it was in Scotland. It was designed to 1.5KW/m2.
Credits to you running that PUE in Israel, where I imagine the average temperature is somewhat higher.
But you have mountains there so perhaps you could locate it at a cooler altitude?
Anyway, 30 of your SP10 units in a 45U rack would output ~36KW.
Even removing the doors and using a sealed hot/cold aisle with high-volume chilled air input you must be seeing exhaust temperatures in the 80C range.
And what's the benefit?
Most DC's are not that short of floor space.
You would run out of power input capacity at that power density before you ran out of floor space.
Nobody designs to 12KW/m2.
(IMHO).

IMHO putting 30 Spooners in one rack is a non-starter.
Because of the heat density you would have to spread them amongst racks.
Most ordinary racks can not dissipate more than about 10KW with the doors closed without some serious (i.e. expensive) cooling solurtion (water cooling, sealed hot/cold aisle, high volume air flow, etc.).
Also, most data centres are not designed to handle that kind of spot heat output (I worked in data centres for 36 years).
Sorry, big generalisation here as I do not know your precise circumstances, but when I watched their video when they said you could put 30 in one rack I split my sides laughing. 

1. the hashing unit will not draw exactly 1000W, probably a bit less.
2. they have chosen a "reliable" brand - Enermax - which is rated at 1000W continuous output (i.e. "load").
3. the draw at the wall will be load + efficiency, so say for example the hashing unit actually consumes 975W and the PSU is 90% efficient then it will draw about 1083W at the wall.
4. the PSU has been chosen because it has a long life which, although reduced because it is running close to 100%, will be longer than the hashing unit will remain profitable.

All IMHO as usual.