Water cooled GPUs are a waste in a riser rig - they make sense for rigs in a case, or if you use those many-GPU-slot type motherboards due to the limited spacing between cards.
They can clock a HAIR higher (usually) than air cooled due to the better cooling, but generally not enough to make up for the extra cost.

 They CAN be very quiet, with proper design - and some folks need "quiet" to keep their Significant Other happy.

 1070 ti and low-end 1080 both have a 180 watt TDP, but if you run them at their EFFICIENT point around 106 watts - you'd still be pushing it a little with a 1000. Might be OK if you can build the REST of the system to only eat 50-60 watts or so.

 I run 5 cards on a 850 (mostly Seasonic X-850 but I have some EVGA G2 850s around too), but that's with left-over "designed for folding originally" systems that pull 100+ watts for the system while mining Monero on the FX8320E CPU.

 Currently 105 watts on each 1070 ti - you can fudge a few watts around that figure and still see very close efficiency numbers - running +200 core +700 clock on both the EVGA SC and the Zotac Mini models of the 1070 ti.

 1080 at that same power level is very very close on efficiency - they don't DROP as fast at the 1070 ti when you push them harder though.
 +100 core +100 clock on Gigabyte Windforce (2 fan OR 3 fan) or Zotac Mini models.
 1080 tends to cost around 10% more for the "same model" vs a 1070 ti, which gives the advantage to the 1070 ti unless you get a VERY good deal on a 1080 (or you get price gouged on the 1070 ti) as even pushed the 1080 doesn't do 10% higher hashrate.


 The blower style cards should be OK on cooling if run efficiently - their real advantage is for rigs built in a case, where they do a better job getting the hot air OUT of the case than almost any fan-type card does (Sapphire Nitro are an exception there, and the Zotac Mini cards push a lot of their air out of the case as well).

 They also are sometimes on sale for less than any other option, which can make them have a good hash/$ ratio especially if run for efficiency where the cooling isn't an issue.


 My only beef at this point with the ASRock H81 is the socket 1150 - which is getting harder to find CPUs for at all unless you're willing to depend on a used one.
 Consulted on a build with one a week and some back, seems to work well and the only issue was a bad set of risers (which certainly is NOT the fault of the MB).


 There has been wide reports of "ram production shortages" for many months now, and pricing HAS been climbing for a while.
 It also seems to be having some effect on video card pricing of late, though a small one so far.

Depends on the 1070 - I get 400 at about 102 watts TDP on my EVGA SC and Gigabyte ITX model 1070s (bit under 70% TDP for those models), none of which run AT ALL at +600 on memory clock (they CRASH in no time flat if they run at all) but do use +100 core +100 memory.

 Stock at stock TDP, most 1070 models should easily get to 430 sol/s and the higher-power ones (Gigabite Windforce and Gaming models, EVGA FTW, and OMG that crazy MSI Gaming model with the 240 watt default TDP. among others I've NOT personally used) should get somewhat more.

 DSTM for me is a mixed bag - a little faster at some settings, usually SAME or SLOWER at most "high efficiency settings", while eating more FEE % than EBWF can be set for.

 "Fixed 2%" fee is a RIPOFF when there are other miner programs with VERY similar capability, and IMO it's a ripoff in general.

On the subject of Gigabyte - at this point I am specifically recommending to AVOID them (except perhaps the Aorus line) due to their move to crappy junk DIE TOO YOUNG sleeve bearing fan designs sometime in the last generation or two.

 EVGA is all ball bearing and generally VERY good built quality, MSI appears to be all ball bearing.

 Zotrac it's impossible to tell for sure without taking one apart but appears to NOT use ball bearing fans on any card, Palit/Galax/KFA I can't speak to as they're not available in the USA.
 PNY, XFX, PowerColor - just generally crap build quality even *IF* a specific card uses ball bearing fans.
 HIS - used to have SOME models with ball bearing fans, but apparently have dropped the concept in recent models.

 On the AMD side, Sapphire is ALL ball bearing and very good designs in general.


 At this point, the only stuff I'm buying is EVGA, Sapphire, and will consider MSI if they're competative on cost.

 Per SLUSH himself, there is NO "backing from Slushpool" and he wants their name DISASSOCIATED from this project because of that.

 To me THAT RIGHT THERE is a major red flag warning, that the company was willing to use someone else's well known name WITHOUT PERMISSION.

 If they actually send a review unit (or more than one) to well known reviewers like HagisFin, then I would tend to start believing that they're at least a real company with a real product - however shady their advertising technique has been so far.

Gaming reviews have NOTHING to do with mining usage.
Try looking at the various build-up reviews for the Vega FOR MONERO MINING USAGE.

For a specific related example, most of the reviews I have seen put "high load" power usage of a GTX 1070 ti in the 180-220 watt range - yet the MOST EFFICIENT mining point for those cards is 106 watts give or take a couple, and even folks that "push" them for higher hashrate rarely go higher than 145-150 range because the gains above that are TINY compared to the increase in power usage (my specific 1070 ti mining cards are currently set to 105 watts to keep total system draw under 6 amps on the system they are in).

Also, the Vega 64 pulls a lot more power IN GAMING USAGE than the 56 does, the 56 was deliberately designed by AMD to have a much lower TDP limit

 YOU need to check your ASSUMPTIONS, that's where you are messing up.

 Floating Point operations are 100% WORTHLESS for cryptocoin usage, which is 100% INTEGER operations.
 Another BAD ASSUMPTION you make, not a mistake on MY part.
 Would you also care to explain how "LTE16x" cell phone interface helps mining? Just for ONE example of the "not useful stuff for mining" on a Snapdragon 835.
 
 My "under 300 watt" measurement was made AT THE WALL on a Brand power meter while the system was actively hashing at 1950+ hash on Monero mining - and that was on a system that is NOT "power optimised" well, as I've never figured out how to get bloody Wattman to do a lot of the stuff I can do routinely in Afterburner like UNDERVOLT (Vega 56 cards LOVE to be undervolted, they tend to clock HIGHER with some undervolt as it lets them stay below the TDP easier). It is running a severely overkill Gold-rated power supply (Seasonic X-850) because that's what I had available when I put the system together, but the system is based on a FM2 motherboard with am AMD A10-7890k (which is NOT a low power APU) with the iGPU running the graphics for Win10, an HGST 3TB hd (system is also doing BURST mining) so the actual power draw of the SYSTEM as a whole would probably be about 275 watts (That model of PS usually pulls about 92% efficiency in the 30-50% load range).
 220 watts draw for the GPU is a PESSIMISTIC estimate, as I'm pretty sure the rest of the system is pulling 80-100 watts total NOT 50-60.

 Presuming an optimistic 2 watts, can a Snapdragon manage 20 hash on Monero?
 8 cores at 2 Ghz (ballpark average, I saw the "big/little" core split) in theory should manage more, but how much CACHE do they have on them - Monero wants ballpark 2 MB of CPU CACHE MEMORY per thread to run efficiently (and I can't find a spec anywhere that shows the amount of CPU cache on a Snapdragon).

 Then figure in the COST of the things - even *IF* they can mine efficiently, is it worth the COST of the things for whatever hashrate they achieve?
 THAT is the primary reason pretty much any SoC setup gets ignored for mining - even if it IS efficient in hash/watt, the sheer COST makes the time to achieve ROI end up being measured in YEARS.

 Consumer-grade power supplies are designed to handle SHORT periods at high load, not continuous operation at full capacity or close.
 
 Many automobiles are designed to be able to go 150+ MPH - but most of them don't handle doing 150 MPH for 24 hours at a time.

 SAME ISSUE.


 For reference - the Distributed.Net client, which is a cryptographic program that loads a CPU or a GPU down very heavily like mining software (which does VERY SIMILAR cryptographic work) used to be used by many sites in their testing AS A TORTURE TEST because of the level of load it put on components in a system.

My current single Vega 56 system eats a hair under 300 watts for the COMPLETE SYSTEM (that's probably 200-220 for the Vega) while mining Monero at about 1960 hash/sec - and I've not had the time to work on power optimisation, and Wattman makes it a PITA to do so IMO (I had it working with Afterburner for a while, but had a power outage and haven't been able to get it working again with AB).

 I dunno where your "15 Vegas for 6 kilowatts" is out of, should be closer to 30 even with POOR power optimisation - and I've seen claims of managing close to 1900 with 120 watts out of a Vega 56 with a "flash it with a Vega 64 BIOS then well-optimise it" setup.

 250-300 watts for a Vega 56 is WELL above it's rated TDP, though they run cool enough that you might be able to get there if you push TDP hard enough - but a lot of that wattage will be waste heat, they run faster when undervolted and don't need tons of power to max out their core clock.


 Can an ARM even mine ETH at all?
 I would not hold out a lot of hope for it on ZEC, they're a LOT lower performance than the i7 hardware some folks used to CPU mine ZEC in the early days and the hashrates on THOSE was low double-digits.
 Monero - they might have a prayer against a Vega if they can manage better than 10 hash per watt used ...... hmm, given all the "extra stuff" on the Snapdragon 835 that doesn't DO computing, how much DOES that thing eat - and how bloody much does it COST?

 I go with the 9 to keep my OWN problems with breathing under control - especially this past summer of massive forest fires everywhere making Central Washington air far worse than Los Angeles ever gets.

 The miners would probably be plenty fine with those "fiberglass" type washable filters, or even no filtering at all.

One factor to keep in mind - ETH mining GOES AWAY sometime, when ETH moves to POS - and the last I remember seeing about it is that the "first move" in that direction is due very soon - though that "first move" was supposed to only be 1% into POS the rest still POW, and likely a few months at least before a bigger move towards POS happens.

 I'm also NOT buying "29 Mhash/sec" as a CONSERVATIVE ETH hashrate figure out of any RX 470/489/570/580 card that is dual mining, most of the figures I've seen reported were "27-29 SINGLE MINING" at best and 26-27 when dual-mining on that series of card - but that WOULD still leave AMD ahead at this point in the comparison.


 The main reason I moved to building NVidia rigs around the start of this year was to "dual fold" CureCoin and FoldingCoin - and for a lot of the last 8 months that combo was returning 50-100% better returns than ZEC - but the ability to SOAK more hashrate was so low I avoided talking about it at all.
 It was NICE seeing $8-$10 a DAY out of a single 1080, and some days hitting $15 on a 1080 ti for a brief while at the peak profitability about 4 months back - but then several other folks decided to jump in BIG time and profitability started dropping.
 At one point my "fairly small" farm was producing about 3% of the entire CureCoin team production - by the time I gave up I was stuggling to stay at 1% with TWICE AS MANY CARDS (and more high-end cards as a percentage).

 As of right now, my most recent calculation shows that "dual folding" is little if any more profitable than ZEC, at least in part due to abuse of a big commercial "render farm" one of the primary CureCoins developers has been using to deliberately depress earnings (and who has ticked me off with his MISUSE of his developer position to the point I refuse to have anything to do with Curecoin at all).


 I'm a bit shocked you didn't bring up Vega - though availability of Vega cards is almost NONE the past week (other than the crazy-high priced FE) and the cards that HAVE been available have been high priced enough that the 2000ish XMR hashrate they can achieve doesn't end up giving that high a value on return/investment right now (it WAS very quite high when you could get the Vega 56 for under $500).

Must be nice to have won the memory sweepstakes on so many cards.

Isn't that Delta PS a bit overkill though?

 9-)

 1070 ti followed very closely by the 1080, if operated at their most efficient spot both will hit 4.4 easily with peaks pushing 4.7 sometimes (I don't count "peaks" where the power is temporarily WAY below the set point as those are statistical abberations).
 1080 ti will top 4.1 easily and might hit 4.2-4.3 with a lot of tweeking on a good card.
 1070 struggles to get TO 4 in my experience with several models, but will usually manage it with enough settings work - this specificially includes the Zotac Mini and the Gigabyte ITX.


 When you factor in total system power draw, the 1080 ti closes the gap a hair.
 The 1080 has more "upside" than the 1070 ti when both are run above the "most efficient point" but both drop efficiency pretty fast above about 115-120 watts, and the 1080 normally costs ballpark 10% more for a "same model" card.
 The only advantage to the 1070 is the lower entry price - but it's not THAT much lower than the 1070 ti, and at "most efficient point" for each card the 1070 gives up close to 15% hashrate while only being ballpark 10% cheaper for a "same model" card.

 It also depends on what's on sale and what prices are like in your area.

 GSkill Ripjaws - never had an issue with them and they have better cooling that most.
 Price is usually competative as well.

 Think I have Crucial in a couple machines, but if so it's been reliable enough for me to forget about it (which is a GOOD thing).

 BURST

 Not high profits though, but them hard drive sure don't eat a lot of electric.

GND might be using custom-built cables, or they may have found a supplier for them - I've not looked all that hard for NEMA 6 splitters as the only NEMA 6 outlets I have had in the last 4 places I've been in were ones on subpanels I wired up myself, with more than one outlet per circuit.

 EVGA SC 1070 (also Gigabyte ITX 1070 but the EVGA was happier than the Gigabyte in the 90-100 watt range)

100 watts (appx 66% TDP, but this is in a LINUX system where the PL is set directly in watts) +100 core +100 memory for just over 400 sol/s.

 Lower than 100 watts was dropping the hashrate FAST and losing efficiency on both cards but more so on the Gigabyte.
 Higher wattage - efficiency drops slowly but hashrate went up fairly well.

 +150 core = unstable crashy card.
 Higher memory (up to +450 or so) gave NO hashrate improvement and past +300 gave hashrate DROP.
 Past +450 on memory = unstable crashy card.

 The "Black Edition" was the original SC model, the only real difference to the "non Black Edition" SC is lack of LED lighting other than the white "EVGA GTX 1070" edge light, and the color of the plastic on the fan mount/shroud.

 I'm not sure what memory brand the one I did the testing on has in it.

 I do have it at 60% fan but it's going to end up in a mining room that gets fairly hot in the summer, where I have it sitting right now 40% would be comfortable (though not 51 degree comfortable, it's running about 52 right now).
 Room temp it's in right now is 29-30C but the mining room in the summer gets up to about 38C.

 I greatly prefer the EVGA SC 1070 ti model, as it will push more like 440-450 sol/s at the SAME 100 watt power level, 460-470 at 106 watts, while the cost/sol is a hair better than the EVGA SC 1070 at recent pricing (more so at total system cost for a same-card-count rig).
 

MERV 9 filters out pollen and dust - it's enough.

 Most air compressors don't HAVE a filter, and it doesn't stop water from condensing as part of the compression cycle if they DO since the filter will normally be on the INTAKE to protect the compressor itself.
 Many but not all have a water trap on the output, but some of them only trap SOME of the condensed water (even the GOOD ones don't get it all, but they get enough to drop the RH of the output air enough to keep it from spewing water drops all over stuff).

 You did NOT take the total system cost into account.
 Try adding $40-$50 to the price of each card THEN compare - and that's being optimistic as I don't think you can build a viable *8 slot* system for $320. It's hard enough building a viable *6 slot* system for $320, if at all possible.
 
 Dual mine of DCR does in fact impact ETH hashrate - that's been WIDELY reported on the Claymore dual mining thread - though it's not a big impact, 10-20% per most of the reports.


 I do agree that relative profitability can vary - ETH in the last week has kicked up a lot more than ZEC has as a percentage, which is giving a bigger boost to ETH mining than to ZEC mining which is at least for the short term boosting the profits on AMD cards more than NVidia cards, at least 'till folks start building enough rigs to drive the diff up to match the price increase.


 It would have been interesting to see how the numbers compared a month ago, when both ETH and ZEC were a lot lower priced and fairly CLOSE on price (and both had a bit less network hashrate).

 Or better yet - NVidia GTX 750 ti (which is GENERATION OLD hardware on 28nm technology that's almost 5 years old now) pulls 250 hash/s or so on about 30 watts actual power usage.

The 3 cards should run together, but ANY mixed AMD+NVidia rig is going to have noticeably worse stability than an unmixed rig - which is the primary driving force behind my current "rationalization" process, getting the Nvidia cards OFF my FM2 based machines.

Waste of time.
Dogecoin is a Scrypt-based coin, mining for which got taken over by ASIC miners YEARS ago - then they went to "merged mined" and now most DOGE are mined by pools actively mining OTHER coins and doing merged mining on DOGE.