That article doesn't even mention BIOS at all.
 Do at least READ an article before you attribute things to it that are not there.

 Hint - LINUX "kernel" is NOT a BIOS on a motherboard.
 It's the core of the Operating System itself - kind of like DOS used to be in the Windows 3.x/Windows 95 days.

Try to get money from miners is a fail - they'll just shove them into OTHER ISP options, since we don't NEED high-speed Internet to mine with.

The ISPs are far more worried about the impact of HIGH VOLUME USAGE stuff that forces them to spend money for infrastructure upgrades - which have a MUCH MUCH HIGHER impact on their bottom line than trying to get a few $ out of "miners".

 Also keep in mind that outside the "3 cheap electric" counties in Central Washington, miner density is LOW in the US - there aren't enough "miners" using most ISPs outside of this area to make it profitable to TRY to chase forcing them to pay more - and there are MAJOR legal issues with the whole "special fee just for miners" concept that will probably COST an ISP that does that sort of stupid thing a ton more than they could make from it.

The bug is STILL PRESENT even with the "fix" installed, the "fix" is a work-around to AVOID the bug.

 A BIOS upgrade might be needed for Windows since Windows continues to use BIOS functions while it's running, but it's not going to do ANYTHING for LINUX since LINUX does not USE the BIOS for anything past very early initial boot steps, while the bug doesn't start cropping up 'till a much later stage.

 I'm really not sure WHY a BIOS update would be needed though, or how it would work to do anything about working around the bug involved.

If metering happens by any ISP, it is going to be HIGH BANDWIDTH USERS AND APPLICATIONS that get metered.

 They've done this to a degree ALREADY on many large ISPs - just not on an "application-specific" basis.

 An ISP has zero financial reason to meter a LOW BANDWIDTH usage application type, and several reasons to NOT do so.

The way ZEC works, sol/s *IS* hashrate for all practical purposes.

 Yes, that's the number you plug into whattomine or other calculators - always keeping in mind like ANY OTHER algorithm that the calculator can't account for stuff like stales....

 I have seen LINUX for the PS/4 mentioned on a few websites as a "done deal".
 Not sure on the XBox One though.

The iGPU in the XBox is going to perform closer to a RX 560 than a RX 580.
It is going to have serious thermal limits that won't let it clock as high as a discrete GPU with the same number of cores when under heavy load, so even though it has a similar core count to the 580 it's NOT going to be close on performance.
They may also be limited on how much memory they can access - the A10 series has a 1 GB limit on it's iGPU to date, even though some BIOS claim they can work with 2 they just CRASH if you try.

 For perspective - my A10-7860k and A10-7890k both have 512 cores running at about 800 Mhz - which is the SAME specs at the old HD 7750 - but for almost EVERY algorithm I've tried on both the HD 7750 blows those iGPUs out of the water on hashrate.
 Part of that is that the A10s are using GDDR3, but it's not enough to account for ALL of the difference.

 The ONLY thing I've run on those iGPUs that's close in performance to the HD 7750 is the Dnet client - but that client uses almost NO ram, so it's pretty much 100% core count/clock limited.


 In fairness, you SHOULD be able to mine on the things if you can get the software installed - just don't expect ANYWHERE NEAR RX 570 level hashrates, and I have some doubt they'll even match the RX 560 on a sustained basis.


 Also keep in mind when talking ETH that ETH is very much memory limited, more cores does NOT always help much if at all on hashrate.
 GTX 1070 has 1920 cores yet OUTMINES any Polaris based card AND the GTX 1080 on ETH despite the RX 470/480/570/580 and GTX 1080 all having more than 2000 cores.

 There are adapters that will let you run a high-wattage fan from a MOLEX for +12VDC power with a second header for the PWM control and RPM sense functions.

 They might fit, but the cooling is going to be terrible due to the VERY LOW space between the fans and the back of the next card.

ETHMonitor (not the control part, JUST the monitor part) works pretty well and is free to use.

 I don't bother with KVM switches any more, I just move the KB/Trackball/VGA connectors as needed with one set assigned to each "rack/shelf" unit.

 Nope. 12" maximum width on the MB.

 EATX uses a LONGER motherboard, not a wider one, so they can still fit into 4U cases with an ATX-type power supply "on it's side" without exceeding the max width of a 4U case or having to park the PS "over" something.

The "Meltdown" and "Spectre" vulnerabilities are a CPU issue, NOT a BIOS issue.

I'm fairly certain that the BIOS can't work around the issue - DEFINITELY it can't for LINUX as LINUX doesn't USE the BIOS after a very early part of the boot process.

Limitations on available PCI-E power connections on power supplies is why I prefer MOLEX powered risers - nothing to do with the power connector itself as both are plenty to handle riser power requirements.

I don't like using "splitters" when I can avoid them - and most splitters are too short to WORK anyway.

They're also an added expense, while the power supply comes with MOLEX power leads that generally don't get used otherwise.

Might be an addressing limitation, though I'm still inclined to think it's just ASUS doing wierd stuff with their bios or perhaps trying to push their P106 cards (there aren't a lot of folks that MAKE the "mining specific" cards).

 The PINOUT on modular cables from different PSUs is often not the same - even the "same Manufacturer" often has different cables on different models, as those different models are often not actually MADE by the same manufacturer.

 THAT is why you don't mix-and-match without doing careful research first.


 Draw from the PCI-E bus varies widely depending on the specific card.
 There have been tests that showed peak draw on some cards EXCEEDING the PCI-E spec by 5-10 watts - yes, that's with all of the PCI-E POWER connectors attached to the card at the time.
 The AMD RX 470/480 cards were NOTORIOUS this way, 'till AMD released a new driver version that apparently adjusted the "power split" the cards used to pull less from the bus.
 Some of the affected models were 150 watt or so TDP but had an *8 pin* PCI-E power connector.

I've never used an add2PSU on my own rigs (though the rig I consulted on for someone else used one) so I can't really speak to those.

 You can mix AMD and NVidia on Windows easily, at a cost in stability.
 The jury is out but indicating "not" on recent LINUX version - the NVidia drivers and the AMD drivers USED to be able to coexist but that capability seems to have been dropped by NVidia around drivers version 361.
 It MIGHT still be possible via workarounds, but I've yet to get it working on anything more recent to date.

 The 13 "real GPU" limit is DEFINITELY not a wanted feature - it may be a bug but I'm more inclined to think it's a "design decision for NO real or sensible reason".

 ASUS seems to be VERY VERY slow about getting an update out to fix the issue - they appear to think that forcing folks to use a specific TYPE of GPU is acceptable, or whatever is causing them to have set that limitation up is being more difficult for them to FIX then they expected.

 If you can get the board close to it's $150 MSRP, it's still a good option for a rig in the 8-13 card range, though perhaps somewhat overkill for the lower end of that range.
 The built-in PS management features are nice, and make the ACTUAL price a lot closer to all other options in that range then just the "raw price" compare would indicate.

50/50, given the LOW volume, that those "12nm" chips are going into Bitmain's new "deep learning AI" related project.

For reference, all of the "12nm" processes that has been announced to date are NOT on a "new node", they are 14/16nm node process refinements and the "12nm" is a marketing tag.

Similarly, the announced 7nm node stuff out of TSMC and GF *appear* to be about the same level on actual gate feature sizes as Intel's 10nm node (not sure where Samsung's 10nm node sits yet, but I suspect it's close to Intel's).

Given that every 14/16nm node production line is running at full capacity, I have no idea why TSMC and GF are worrying about "getting the most business" via markating exaggerations.


Anyone else remember "Web 2.0"?
I am starting to sense a similar stench of "Microsoft Marketing" getting even more contageous....

 Trying to pay Newegg from an exchange wallet has been VERY PROBLEMATICAL all along - the last couple months Bitpay has been having major issues even working with THEIR OWN SPECIFIED "works well" wallet options.

 2 of my last 3 Newegg purchases have been BORKED BY BITPAY, the last one is still "in process" of resolution.

 I'm about to change suppliers, if Bitpay RIPS ME OFF AGAIN ON REFUND of the current issue, as it was 100% CAUSED BY THEIR ISSUES I HAVE ALREADY NOTIFIED THEM OF AND THEY INSIST ON BLOWING OFF.

 At this point, I don't trust "Bitpay" as far as I can throw them.
 Newegg was a LOT better off before they switched to using this INCOMPETANT excuse for a payment processor.

 Does anyone know if "Bitpay" is owned by Paypal?
 It would explain a TON.

 Discussion of hardware options to mine altcoins certainly IS part of altcoin mining.

 It would apper you're looking for a more focused thread though, this series of threads does tend to be all-encompasing.