Given AMD issues with AMD GPU-PRO from day one AND their long-running LINUX driver issues in general, I'm more inclined to blame AMD's "lazy" attitude towards LINUX over some BS conspiracy theory.

 DO note that the "AMD/Intel" partnership appears to have been at the request of APPLE, and therefore would tend to be ANTI-Microsoft.

BTW - don't believe that $5/day out of a Vega mining XMR - though I suppose $4.50 has been pretty close on average the last couple days IF you can get a Win10 system to stay up reliably and don't have power glitches that bork it.

 9-)

 In THEORY the 17.40 LINUX drivers should work - but I've yet to see anyone post that they actually DO work, much less give the high hashrate out of the blockchain drivers WHEN WORKING CORRECTLY on Win10.

 I plan to work on a LINUX system sometime this week to see if AMD finally got VEGA support working AT ALL.

 The Neptune was a fiasco at least in part because it burnt connectors on a frequent basis due to overloading them.

 One "delay" of "one batch" on a miner that has been shipping for many months and mostly working reliably is not even on the same PLANET, much less in the same ballpark.

 well over 5000 posts later, and you ask this?

 Thread would not have lasted this long if this was a virus.

I have one of those ASUS "Turbo" blower 1080 ti models. Runs a bit on the hot side, removing the "mounting bracket" near the air exhaust area helps some but it STILL runs a little hotter than my EVGA SC 1080ti model and a LOT hotter than the Aorus.
 Good choice for a "in a case" build, but I'm not sure it's worth the low price for a riser build.


 Did some testing on my Zotac GTX 1080 cards today - anything over +100 on core OR memory cost hashrate , and it was most efficient down around 106 watts, pretty much matching the EVGA SC 1070 ti cards I have (but keep in mind I'm running EBWF on both).
 Testing done on the XUbuntu machine I currently have my pair of 1070 ti cards in, so EVERYTHING was the same except the GPU itself.

 If the PRICE was a match, this would be a good choice - but 10%+ or so HIGHER cost for the same hashrate and tossup efficiency is a LOSS in my book.

 I should probably get around to testing out DSTM soon - have seen conflicting results on "better than" or "worse than" EBWF but they seem to be "model dependent" in part.

 I have yet to find evidence that the mainstream press thinks AT ALL.

 Very rare exceptions once in a while on an individual article.

 It's harder to factor in how much you use due to Nicehash switching your hashrate around a lot (and the very short-term losses from that factor).
 They definitely don't have an 8% overall advantage - I'm more inclined to figure that after all costs are taken into account they're lucky if they are even, unless you're running Windows and using their profit-switching software (which make up in part for switching you to more profitable algorithms by costing you MORE outage time in the switching).

With THAT kind of an electric bill, you should consider moving to somewhere with lower cost electric.

 20 cent per KWH is prohibitive for long-term mining profits.

 My last bill was $550, and while I was "only" generating 15k-16k sol/s during that month I also had a ton of power feeding the A2 Scrypt farm (MORE than the ZEC farm I think, though it's gotten quite close) + some feeding my DNet/Moo Wrapper/BOINC farm, which includes all of my older powerhog HD7xxx and R9 280x/290 cards (along with some RX 470s and A10 iGPUs), and some feeding the CPUs I have mining Monero on most of the GPU-centric machines.


 Oh yeah - some of it was also going into the 1 Evap cooler I still have running, and a little into the A/C unit that keeps the seperate "office" type space cool.

 Depends in part on where it sets the STOCK clock of the memory - part of the reason my EVGA 1070 ti SC cards overclock memory so well is that the BASE MEMORY CLOCK is underclocked when they are running a mining program.

 Card models VARY, and even individual CARDS of the SAME model can vary - memory brand seems to make a big difference, and EVGA (like ALL card makers) tends to use whatever memory it can get ahold of to make a batch of cards, even if it's not the SAME brand as what went into a previous batch of the same card model.


 MSI Armor Gaming 1070 card is a VERY GOOD example of the "variation in TDP between models" issue - 240 watt TDP by factory default, where MOST 1070 models are 151 or 180 watts.


 For the record, I've NEVER had a 1070 model that could do "+600" in afterburner settings AT ALL, most were unstable at +550 and at least a couple unstable at +500 - even on models other folks reported being able to STABLE overclock more.
 Like I said, cards VARY.

 Under 3 all costs considered, not under 2.
 Might be under 2.5 though if you're on the scale that Gigawatt is building at.



 MOST of Europe has very high energy costs - there are a few areas that have significant hydropower, but only a few small ones.

 US is NOT an "unstable country" especially by EUROPEAN standards - consider events in places like Spain, Russia, Greece, France, and the like before you go trying to tar the US with THAT brush.
 We're also STILL the strongest economy in the world by a WIDE margin, and our power infrastructure in MOST areas is quite good.


 BTW - if you get away from Wenatchee specifically, there is a significant number of existing buildings that are FINE for use by miners.
 It you're getting in to the Megawatt and up range, you are probably going to have to build infrastructure almost ANYWHERE you go - I'd really like to see these alleged "warehouses (PLURAL)" in KY that have multi-megawatt "ready for you", I'd bet that's an abberation NOT the norm.

 You might also want to look at the reasons Gigawatt moved AWAY from the "big warehouse" type mining setup - starting with "can't get enough airflow through the building" to keep stuff cool.


 Finding industrial-level rates in the under 5c/kwh range is NOT all that easy, if you count in ALL of the costs of that power. Under 5c BASE rate, sure - but then factor in load fees, taxes, meter charges, transmission fees, fuel surcharges, the other stuff many or most utilities add in on top of the base rate and you find the REAL rate goes up a lot in most cases.

 

Light can only "travel around the earth" a bit over 7 times in a second.

 There is a REASON all Sat internet services (except Iridium) have an absolute minimum latency of close to a half second, AND usually more like 600-700 ms for North America - the "double bounce" out to Geosynch that takes about 120ms PER LEG (2 legs per bounce) even if you're directly under the sat on the Equator - longer as you get further north AWAY from the Equator, and even longer as the sat is usually somewhat "east" or "west" from you as well.

 186,280 (appx) miles per second vs earth being right about 25,000 miles AROUND (varies some, pole-to-pole is a hair under).

 Any MB that suppports Skylake should be able to work with Kaby Lake - though many of the older 1151 boards need a BIOS update to do so, which can be a Catch-22 type mess.

 Kaby Lake was not a noticeable upgrade from Skylake - VERY small gains in general - and the Mickey$loth refusal to support Kaby Lake on anything older than Win10 makes Kaby Lake somewhat problematical (LINUX supports Kaby Lake, and many MB MANUFACTURERS support Kaby Lake on 7 despite Mickey$loth because THEY know where the market is at).



 That vegamining blog has quite a bit of useful information, and some links to more. Not a bad place to start.

 When you factor in "pool" AND "exchange" fees, and the transaction fees, you quickly find that Nicehash fees aren't as bad as you first think they are.

 The big issue right now with Vega is that AMD has taken forever to get WORKING drivers available for LINUX - so you had to use Win10 if you wanted to use the Blockchain drivers (needed for high XMR mining rates).

 In THEORY, the 17.40 drivers for LINUX have fixed that - but I've not had the chance to check that out (it's on my agenda for this week) and also include the Blockchain "fixes".

 Given that AMD CLAIMS the 17.30 "Blockchain" drivers are supported for Windows 7 WHEN IN ACTUAL FACT THEY DON'T WORK AT ALL UNDER 7, I am not holding my breath.



 AMD is doing a fairly good job of shooting themselves in the foot with their last 2.5 years of borked LINUX drivers, and their drivers that have been frequently total junk in general.

 My long term intent is to build up to probably 17 cards (1070 ti), but using 3 850 watt power supplies split between 2 circuits and running the cards in their "high efficiency" range at 100-110 watts per card.
 2 x 6 running by themselves on each of 2 PS, then 5 + MB/RAM/HD on a third - though I might go with a 1250/1300 (I've got some X1250 Seasonic and 1300 EVGA G2 not in use) to power 7 cards, that PS would be on it's own circuit that would probably have a box fan and monitor and nothing else.

 And certain areas lots of Vets and Navy/Marine types.

 San Diego manages to be overall quite conservative despite being in California - though it's the only LARGE chunk of Conservative in California.
 Pendelton, 32'nd Street, Coronado, Miramar, Point Loma, MCRD (part of which used to be NTC), and all the smaller bases and the folks that RETIRE to San Diego after having served at one or more of those bases distort the demographics of the area a LOT.

 8-)



 I AM aware that there are large mining companies with HQ in California (the late Zoomhash comes to mind) that had mining operations in Central Washington - I just find it funny that LA is mentioned by the article, seems to have been yet another case of "poor fact checking".


 As far as money transfer goes - you people aren't thinking.
 They're MINERS, they just don't cash out a large chunk of their Bitcoin/Litecoin/ETH/other cryptocoin into yuan, and use THAT to fund their "other country" expansion by cashing it out into THAT COUNTRY'S currency.
 No bank needed AT ALL.

 They DO have to worry about "immigration" restrictions in at least some cases though - but if they demonstrate they're bringing in a large net worth of cryptocoin with the intent of building a large "local business", that tends to get rather easier.
 Most countries LIKE having "rich" folks move in.

"Factory" Boost Clock" rating is a "we guarentee the card will boost AT LEAST this high" statement, almost all cards will boost quite a bit higher than that in usual usage.
 For example, my Gigabyte Aorus (NOT the extreme) was typically boosting to well over 1900 Mhz and sometimes over 2000 depending on what I was testing on it at the time - even at "stock" clock settings.

 Very good cards if you have the SPACE for the massive cooling setup.

 Asus "Turbo" blower card doesn't boost all that high, and runs quite a bit hotter.
 Removing the mounting bracket helps on temps noticeably as it unblocks the primary exhaust some.
 This also seems to be the ONLY non-founders edition 1080 ti blower card that DOES NOT PUT A MASSIVE DVI CONNECTOR RIGHT IN THE EXHAUST OF THE CARD BLOCKING HALF OF IT.
 Best card for an "in case" setup IMO at this time, but I can't really recommend it for a riser-type setup as there are better options for those (though at it's usual cost, I don't recommend AGAINST it for a riser setup).

 Gigabyte "Windforce" models stay fairly cool, but I can no longer recommend them due to the junk fans they use.
 I didn't realise Gigabyte had DROPPED usage of ball bearing fans in thelr mid-range models 'till recently - and have had *2* fan deaths in Gigabyte cards in the last 2 weeks, AND an ITX outright CARD death that might be fan-caused overheating caused.

 EVGA SC stays fairly cool, but doesn't boost quite as high as the Aorus.
 Good solid card and the 8+6 power connection makes it a LOT easier to use several in one rig than the 8+8 many "higher level" cards use.



 Hash difference between cards is mostly going to happen at the higher "push them" TDP settings - all of my GTX 1080 ti cards seem to settle into about the SAME hashrate once they get under 70% and they've very close at 80%, in the limited testing I've done so far on all of them.
 Temperature is one of the things the card takes into account when deciding how high to push the boost, the higher-end massive cooling cards that eat 2.5-3 slot width cool better so they boost higher at high power levels.

 No - it just means you didn't get lucky in the silicon sweepstakes on your memory or your core - OR you have poor cooling for the card and you managed to overheat it.

 What temp does the card get up to after 5 minutes or so?

 I have no idea what memory type my EVGA SC 1070 ti cards have, as I don't run them on Windows so no GPU-Z to easily tell without disassembly.


 Not sure why "in game" was mentioned in some of the replies, OP didn't say anything about gaming on the card.

 8 gauge - NO BLOODY WAY. Only rated for 30 to 55 amps depending on specific wire type - and in a hot mining environment you need to DERATE for temperature.

 6 gauge - probably OK depending on the wire type.

 http://www.usawire-cable.com/pdfs/nec%20ampacities.pdf

 Thickness of the jacket is actually a MINUS for it's ability to carry current without overheating, as the jacket is an insulator that RETAINS heat.

 DO NOT WIRE BREAKERS UP IN PARALLEL UNDER ANY CIRCUMSTANCES.
 THAT IS AN EXTREMELY UNSAFE CONDITION.


 220 does not "need" neutral to feed pure 220 volt devices - the only time a 220 circuit uses a "neutral" is if it is ALSO feeding 110 volt devices, or if it is feeding "dual voltage" devices like a lot of ranges that use 110 for the clock/timer stuff.


 A breaker on a circuit should ALWAYS have the same or LOWER amperage rating of the WIRE used on a circuit - this makes the BREAKER the point that will "fail" first.
 Doing it the other way around is a major safety hazard, as it can allow the WIRE to overheat and fail before the breaker pops open.