I've had to replace almost all of my Gigabyte board fans - most of them have NMB 92mm case fans, just wired to 12VDC since the power consumption is low enough to not bother with PWM - those cards commonly run 50C or less even in a room in the low 80s F ambient temp.
A few have left over Delta 80mm fans from various retired CPU cooling setups, though I'm not going to waste the pair of 68CFM screamers on that duty.

Corsair AX (i) series is junk - the AX series WITHOUT the (i) are good units, Seasonic SnowSilent (which is the Platinum version of the Seasonic X series) clones except they use a Corsair-specific pinout on the PS side of the cables and a Corsair-specific ball-bearing fan.
I have no clue why they use the AX name on 2 such totally different power supply designs - it's misleading and VERY confusing.

I'm making quite a bit more than "a nickle per day per GPU" - except on the R9 290s (they're still underwater right now, but it's close) and possibly the pair of 960s I still have mining.
On the other hand, I'm not paying 8 cents/KWH.
I also do NOT expect to see $1/day/GPU back anytime soon, if ever.

GPU and S9 mining is still profitable here - but the profitability IS a lot worse, I was making more on perhaps 1/3d the hardware 2 years ago vs what I'm making now.
I DID have to shut down some of my older rigs (R9 290s are history, and I'd already shut down the smaller R9 series and HD-7xxx stuff months ago) due to them going underwater, and the D3 units are just a little underwater right now (but they work well as winter heat).

I have been switching out of "pure crypto" the last couple months as a workaround for the "crypto rate increase" due soon - a rate increase that hits small farms a LOT worse than big ones, and is therefore very STUPIDLY set up and does not address the "why we are doing this" AT ALL.

And was unilaterally created with NO consensus that it was a good choice.
Without Bitmain forcing it to be used to back their miners for quite a while, it would have never had ANY traction and would have dropped down to pennies (or perhaps single-digit dollars at best) within the first 6 months.

I am quite sure that most of it is being driven by Jihan Wu financing his "War on BCH-SV" coupled with Bitmain cashflow issues and having to dump a ton of their BCH (and probably other coins as well) horde.

I have never understood the whole "BCH" junkcoin push by Wu - it's not like cryptocoin NEEDED yet another junkcoin, and the only reason BCH ever got any traction at all was Bitmain FORCING it to be used to buy their miners for quite a while.

The facility they are opening is in a strange spot though - it's NOT in the land of Super Low Cost Power (which might be going away for most crypto anyway, given the recent pushback by the 3 county PUDs involved) - the rate they'll be paying there is fairly high for a large crypto operation, as it's NOT all that close to major hydropower in large quantities.

If they were running S9s there right now, they'd be LOSING money to do so - and even the T15 wouldn't be highly profitable.

6-7 cent industrial rate as I recall, might be 5.something at best.

I'd not bet on that - they're probably in the 3 cent OR LESS range on their power cost.

---

An IPO that has already had WIDELY talked about as having a lot of unrealistic expectations and some outright LIES in it.
I'd not bet on that IPO doing well if it actually makes it to market.

Worthless report, way too much indication they did NOT actually study anything.

This one phrase is enough to indicate how WORTHLESS this "report" is.

"Currently, some of the prominent players operating in the global cryptocurrency market are BitFury Group Limited, Microsoft Corporation, Alphapoint Corporation, NVIDIA Corporation, and Intel Corporation."

They left out Bitmain and AMD (both of which make a LOT more from cryptocurrency than Nvidia does).
Microsoft has almost NO presence in Cryptocoin at this time.
Never heard OF Alphapoint but they're definitely NOT a "prominent player".
Intel has ZERO presence in cryptocoin other than indirect as a fairly common choice for CPUs to put on motherboards for GPU rigs.

If the rest of their "research" is as shoddy at THAT statement, their report is 100% worthless.

I'm more inclined to think that the apparent Bitmain cash crunch is a bigger reason.

Technical Analysis isn't voodoo - but it does tend to be meaningless on something that is immature, as external forces will have a LOT more affect then market forces most of the time.

I'm still mining - but have shut down my last pair of R9-based rigs as they went under water this past week - while my Nvidia-based rigs continue to transition out what they've been doing most of this year into something more profitable.

And it got worse in the last 24 hours.

Gotta wonder if Bitmain finally sold out on their remaining stock of S9 units and decided to pull the trigger on another round of "coin dump" to make more of a market for the S15/T15 as one of the very few profitable miners.

On the other hand, a lot of miners are at 5 cent OR LESS power cost, pretty sure even at current market conditions the S9 is still profitable for the low-cost miners.

I still think "Bitmain cash crunch" is the primary reason for the massive sell pressure on Bitcoin right now - had ANOTHER volume spike that seems to have caused the second "price dump".

They tend, from what I've seen and many other reports, to have a lot more issues with "infant mortality" than death over time - like most electronic gear, especially when it's pushing a heavy load.

I actually am starting to think, based on Bitcoin VOLUME jumping bigtime just as the drop hit, that this might be Bitmain dumping some of their coin horde to pay their bills.

That doesn't make sense - it would tend to drop BCH and aid other coins.

What apology?
I was explaining why your comment was not only wrong but rude.

Odd concept since they've had original Bitcoin support longer than anything else.

There is no new "more efficient" node anticipated to BUILD miners on for years.
Is that hard for you to understand?

In fact, there is some question how much longer semiconductor node CAN continue to progress - GMO's over-optimistic projections to the contrary.
Intel and IBM in particular have both stated that "pure silicon" has reached the end of the road - IBM's current "next gen" process design work uses a MIXED silicon/germanium die while Intel has been quite mum on their plans for anything past their current 10nm node.

The old "new miner model every 6 months" routine died when ASIC mining hardware caught up with semiconductor state-of-the-art at the 14nm node (S9 and such), and CAN'T ever return.

---

Just noticed something I find a bit iffy about the S15/T15.

"exposed die" design on the chips.

Does anyone remember the massive FAIL issues associated with that concept on the Athlon Thunderbird/Palomino and same timeframe Pentium 3 generation CPUs that ALSO used an exposed die?

Probably less of an issue with the "glue-on" heatsinks Bitmain uses, but still a bit iffy - and definitely NOT THE NORM that Bitmain claims in that part of their "overview."

Given the existance of an ASIC for Siacoin, there's no point in updating GPU miners for it.
Technically there are multiple ASIC but the devs have announced (or implimented by now) a fork to disable all but the Obilisk (sp?) ASIC that they apparently have a stake in.

Much higher efficiency, which has always been a good justification for a new miner model costing more per hash.
S15 will be profitable for years, the S9 will probably be a money loser at 3 cent electric by this time next year.

I've never understood why Coinbase chose to support BCH.
Without Bitmain pushing it so hard for so long, it would be just another common junk coin with nothing in particular going for it, or more likely would not exist at all.

Poloniex supports quite a few junk coins, though they do weed out a few occasionally - and have also dropped some coins that have a REAL reason to exist (like CureCoin and Foldingcoin that support medical research).