I suppose they could but are there any issues with multiple psu's power an avalon 6.
we may find 2 ¼ is better then 2 or 2 ½
one avalon 6 would use a pcie from 4 psu's
rather
then one avalon 6 using a pcie from 2 psu's
I think a lot depends on the owner of the avalon 6. If these prove to be real flexible in over clock or under clock and up volt or down volt.
we may find 2 ¼ is better then 2 or 2 ½
one avalon 6 would use a pcie from 4 psu's
rather
then one avalon 6 using a pcie from 2 psu's
I think a lot depends on the owner of the avalon 6. If these prove to be real flexible in over clock or under clock and up volt or down volt.
Phillip,
2500W DC (2.5 Avalon6) is well within the capability of the IBM 2880W PSU's, for long-term reliability and efficiency. These are nothing like ATX PSU's as far as durability, 87% steady load is a walk in the park. However, as far as safety is concerned it is naturally up to the hosting facility to decide, as they are the ones with the liability. 2 1/4 miners wouldn't work as the Avalon 6's use only 2 hashboards, and mfr's don't recommend having multiple PSU's on one board.
I've been lurking for a while and this urges me to add my voice here 
Avalon6 and BitMain's S7 are overpriced, too much, they're actually priced for 2-3c/kWh electricity.
I mean, BitMain in China does not have access to such cheap electricity (I don't have any proof, just long reads here about cheapest electricity in China) so they have to be sure that nobody can earn more than themselves when selling an S7.
BTW, they sell it on Hashnest as well, so Hashnest's price is a little lower since you're paying .10c/kWh there.
What we should find, IMHO, is cheap electricity, like .05c or .04c and a bunch of second hand S5s or SP20s, which are overpriced as well (from the WTS offers I see here on bitcointalk), but not as much as a new S7 and should possibly ROI no further than April next year and be able to be marginally profitable (thanks to cheap electricity) beyond next halving.
Spending 1600 USD/unit now to earn 98 USD 400 or more days from now... well, is not that thrilling as a proposition, not to mention that a single failing unit wipes out the earnings of 16 more units.
spiccioli
Avalon6 and BitMain's S7 are overpriced, too much, they're actually priced for 2-3c/kWh electricity.
I mean, BitMain in China does not have access to such cheap electricity (I don't have any proof, just long reads here about cheapest electricity in China) so they have to be sure that nobody can earn more than themselves when selling an S7.
BTW, they sell it on Hashnest as well, so Hashnest's price is a little lower since you're paying .10c/kWh there.
What we should find, IMHO, is cheap electricity, like .05c or .04c and a bunch of second hand S5s or SP20s, which are overpriced as well (from the WTS offers I see here on bitcointalk), but not as much as a new S7 and should possibly ROI no further than April next year and be able to be marginally profitable (thanks to cheap electricity) beyond next halving.
Spending 1600 USD/unit now to earn 98 USD 400 or more days from now... well, is not that thrilling as a proposition, not to mention that a single failing unit wipes out the earnings of 16 more units.
spiccioli
all profitability numbers are bogus anyway with btc being $380 one day and $295 the next.
the only play here is for halving-associated btc expected jump.
So, you accumulate coins like there is no tomorrow, then sell some.
If no jump, then profits would be marginal, if any.
Did ROI calculators a year ago predict that the S5's would still be worth what they are today? They haven't really depreciated much at all in fairness. For quite a few generations now skeptics have doubted ROI, assuming that difficulty would drive hardware obsolescence. And yet, those who have bought hardware for the past 12 months have done considerably better than those who have bought coin. We have reached the point where you can factor equity into the purchase price of miners, unlike early generations of ASICs, and that is something that calculators do not include.
Now, with the halving coming, revenue will be cut in half. However, I don't believe that the interest in bitcoin will be cut in half. In between the lowering of difficulty (due to many miners becoming un-profitable) and the raise in price due to similar demand and lowered supply, I don't believe anything drastic will happen in the world of mining, and that current gen miners will stay profitable for a long time to come. Just my BTC