assuming that PoW is the decided mining method, you would need it to be an SHA-256 algorithm, to allow the Hundreds of Millions of dollars in existing equipment to be relevant. Interestingly, the brief paycoin rush demonstrated that with minimal co-ordination more than 20% of the bitcoin network moved over to an altcoin in the span of a week.

1minute blocks are the future, but come with a lot more blocks to store on drives, and i hope we see it become an issue by late 2015. right now the system works - improvement is second to security

There are two ways to provide <12V to the S5 unit without much difficulty:
1) use a 12V-9V regulator. these are about 90-93% efficient, atop efficiency losses of the 12V PSU powering it. probably $15/S5 to buy
2) buy a mains->9V power supply. most are about 80-85% efficiency and about $30/S5 to buy

IMO neither is a very good solution, but the S5 will be very useful at 0.5W/GH for quite some time anyways. IMO the next revision of the S5 (S5+, S5++) will simply include a few extra chips per chain on the board to reduce voltage and power draw of each individual chip. An interesting alternative is to build a small module of 4-8 chips that can fit inline with a PCIe connector and attach onto the IO chain to provide better efficiencies as time goes on.

5 antminer S1 units, cooled by a pair of Cougar 90cfm fans and running approx 460GH/430W  (still working on tuning, 2 units are only running 70GH). Runs off a single CS650M

ps: temps are up around 60-65C, and the hardware seems fine with that. I plan to get it in a cooler location soon though to drop temps 5-10C

5 antminer S1 units, cooled by a pair of Cougar 90cfm fans and running approx 460GH/430W  (still working on tuning, 2 units are only running 70GH). Runs off a single CS650M

ps: temps are up around 60-65C, and the hardware seems fine with that. I plan to get it in a cooler location soon though to drop temps 5-10C

Its possible it will, but i dont feel like it will. Lots of SP20s shipping out right now

1)  >250 from the GB and probably that many more Sp20 + Sp3X bought directly from SP-T. I assume they are selling 4-9 PH / difficulty period
2) Bitmain has begun shipping S5 units. once a few reviews come back in, they will probably sell 3-7 PH / difficulty period
3) Asicminer wont have much other than the AMHash farm for another month or two, so 1.5 PH/difficulty period
4) Private farms like KNC and Bitfury are the wild card. Both are probably a month or two away from the next chip or phase of expansion. If either one began deployment of a working next-gen chip they would certainly ramp up quickly to deploy 2-6 PH/difficulty period.

I think we will see around 9-18 Petahash added to the network per difficulty period for the next 2-3 jumps, increasing in relation to product price drops or BTC value increase. thats about 3-5% per jump (and about 3-5 Million Dollars worth of hardware)

bitmain approached me on the issue, likely due to participation in early batches of many previous units and some of the more useful portions of my posting history in thier threads. (whereas spondoolies effectively asked for legendary-only members to approach them instead). Either way, pretty interested to see what they produce, as I was a little cautious with this product since the fixed chip voltages limit modification without variable PSUs

+1. Bitfury was doing extremely good numbers (~1w/GH) on 55nm in the first version, and improved that by about 15% in the 55nm//Rev2. undervolting allows something around 0.85w/gh on those chips.

a 28nm chip with exact same design is theoretically capable of almost 4x the efficiency. that would achieve around 0.2w/gh on an underclocked chip. 20nm would achieve around 0.1w/GH with slight underclocking I expect. implementing improvements like chained power distribution or perhaps larger die sizes could also improve on the numbers.

(in contrast, the Bitmain 55nm chip from the S1 is not much different. 2w/GH at full steam, but with a bottom value of about 0.9w/GH if undervolted to about 40% of the full speed. Thier 28nm chips (S3 and S5) are capable of 0.5-0.8w/GH and 0.25-0.6w/GH power ranges also - clearly showing the improvements found by thier 28nm//Rev2 chip in the S5)

I got a ship notice for 3x SP20 from the GB today, and am anticipating an S5 demo unit delivery around the same time (@BMWarranty - i emailed yoshi today as i havent got any shipment or tracking details for the demo unit yet) hoping to do some nice side-by-side comparisons as both are competitively priced!

I thought the jump in hashrate would be more delayed this round. so instead of 2%, 7%, 4% i think it will now be: 4.25%, 5%, and 3%

lots of equiment coming onlin in the next 3 weeks as the SP20 and S5 compete for price dominance. however, BTC<$350 means that its unlikely to see anything capable of <0.6w/GH selling at <1BTC/TH necessary to be profitable for most users.

40F is pretty cold - not all equipment actually likes having <5C air at the inlet. Try:
1) open it up. reseat the power and data cables to all the hashing boards. A slightly loose connection that misses the occasional input/output data could trigger the board disabling.
2) visually inspect the boards make sure theres no missing capacitors or other damage (its unlikely but if the lid is off you may as well take 2 minutes with a bright light to check)
3) put the unit somewhere warmer. 10-20 degrees celcius is ideal
4) tune the unit. try:
4a) bump up hte voltage slightly, 10-20mv more might help
4b) bump down the frequency slightly. 25MHz of so might help

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

curious - any of the PSUs your boards work with run quietly? Most server supplies are only acceptable for datacenter use at anything over 40% load so over time ive tried to move to cheap secondhand 1200W+ gold supplies

what the f---? not remotely what i want to see at christmas, thats just wrong.

Its unlikely the S5 have been hashing for a month, at least not in large quantities. Likely bitmain had batch 1 (and maybe batch 2) in testing for som time, but likely only a small number of miners (<5PH) would be testing/burning-in at once. but the fact is that bitmain more likely runs an S4 farm or has/is developing an S6. Its insanely easier to monitor, install, and control a hundred devices than it is to control 4-5x as many.

diff is looking high on varience again (possibly due to short testing periods for new hardware, plus paycoin users switching back or turning off) - around 1% this jump seems likely, maybe 2%.

new batch in stock! including a new item!

24" 16awg PCIe-to-PCIe cable: great for bridging between PCIe sockets or using on compatible modular power supplies (such as the EVGA and some corsairs). Price is marginally more than the 24" PCIe leads (about $0.10/cable more)

I spoke with yoshi (by email) and they are considering an s1-s5 upgrade kit but not until they get some feedback on batch 1. Upgrade kits are not for inexperienced people and poor thermal paste application or nicking something with a screw driver could result in a few loud and disgruntled buyers who can't RMA a diy product that they damaged by improper installation.

My guess is upgrade kits will either have no warranty or be provided only in bulk packs of 5+ so that it isn't being sold to people incapable of correct use.

any chance of further discounts since target is already passed with several days to spare? Id add several more to my order if a better price was available (such as $2500/6)

The 6pin PCIe cable connects to the PSU with 2x12V and 2xGND pins. If one pin was broken, it means that twice the current was going through the other singular pin of the same type. this wire clearly swelled up and began melting at the PSU connection.

w/ regards to the SP20 end:  its likely that the pcie cable may drive different pins with different pins at the PSU connector. Thus if one pin was broken, at the PCIe connector 1-2 of the pins may not have been active, thus forcing large amperage through only 1 or two pins, causing a fire and risking shorts inside the SP20 as well.

Problem is the reasons listed a few posts back: it costs a lot of money and time to setup a warehouse in EU which will likely need 4-7 locals to staff it at minimum. Shipping from china is already available to them at a reasonable price and does not include the 12month contract risks and unknowns of a new office in a new country.

its a market suited to local resellers with a knowledge of the market and ideally the ability to run 40kW+ of equipment when waiting for sales orders to ship out.

very interesting. any better deals available if I want 5 units?