If they all factory clock the same, should be zero issue.

 If the factory clocks or "stock" TDP differ, they'll still work but you have to be extra careful to adjust them seperately to get optimal results.
 Stability should be the same as non-mixed if you run them at stock clocks or close.



 

 They haven't existed long enough to be sure.

 For perspective though, I bought 6 x HD 7750 back in Litecoin GPU mining days - ballpark 5 years back I think?
 5 are still running abet all of them needed fan transplants (the Sapphire JUST had it's fan flake out late last month, the HIS cards only lasted a few months on the original fans).

 One of my pair of HD 7870 dates from the same timeframe (HIS Ice-Q super-blower model), still running fine.
 The other one I bought about a year back used, no clue what it had been used for before I bought it.

 ALL of these cards have spent all of their time in my ownership either doing cryptocoin mining, or running the Distributed.net RC5-72 client (which is a different crypto program that puts a VERY similar load on a card to most cryptocoin mining) on a 24/7 basis with the only downtime being for fan replacement, system getting moved, power outages, moved to a different machine, or something ELSE in the system (usually HD) died.

 If you keep the cards fairly cool, they should last a VERY long time.



 If you run the Milkyway project on BOINC, you can earn GRC on a Tahiti card pretty efficiently.

Ryzen are seriously overpowered and rather overpriced for a mining rig.

 For AMD, a Sempron 145 (if you can still get one) is normally the best option.

 Big SSDs are also a waste on a pure mining rig.

Nicehash tends to bounce your miners around quite a bit due to the very nature of their service.
I'd not worry about hashrates bouncing a fair bit.

 Also, the chart is PROFITABILITY not hashrate - and that bounces around even more as folks start new orders, have orders run out, change payment on existing orders, etc.

 Based on them selling Baikal miners for LESS THAN BAIKAL DOES, scam.

 Baikal hasn't been able to keep their stuff in stock reliably for the last 2 months due to sales exceeding their available supplies....

*IF* you are using standard mechanical air conditioning to cool a room, you don't need to have ventilation.

 You DO need to have BTU capacity in the A/C to at least match the BTU heat generation of the gear in the room in such a case.


 It's possible to set up an A/C in such a way that it puts cooler air into the "cool side" of a ventilation system, but that's NOT the norm for standard mechanical A/C systems.


 If you are using Evaporative cooling type A/C, then you DO need to be able to vent the air out of the room.


 GPU mining can use just as much power and generate just as much heat as ASIC mining does - though most GPU rigs *PER RIG* use less power than most current ASIC miners, the TOTAL wattage/BTUs still adds up.

Mining is definitely speculative - like ANY investment, it is a gamble.

 The trick is to maximize your odds of achieving a positive net return.

 Z-axis is one of those "fancy" rifle-type sleeve bearing designs.
 Note the *recommended temperature range" on those.

 Many if not most GPUs see more than 40C at the intake (40C = 104F), which is why ball bearing is the way to go, they're INTENDED for use at higher temps.


 7V out of a molex as LOW currents is fine, just don't try to pull a LOT of current that way - couple amps per PS shouldn't be an issue though.


 I'm not sure if a single molex as used for PC drive/fan connections is specified to handle 80 watts at 12 VDC.
 I suspect it's probably OK though, that's less than 7 amps.

 ZEC isn't all that memory intensive, certainly not like ETH is.

 There were *2* versions of the 7870 - the original that was Pitcairn, and the "7870 XT" that was Tahiti (seems to have been a cut-down Tahiti chip with 1 or 2 CUs disabled).
 Arguably the "7870 XT" should have been named the "7930" or some such to avoid confusion.


 Turns out my 7850 wasn't broken - it was the PS I was trying to run it on was broken.

 There are adapters to go from a M2 socket to a PCI-E - thus the potential for the 14th card.

 Board is DESIGNED for 13 cards though.


 There is no reason that a chipset that supports both PCI-E and SATA can't have a M.2 socket - that socket didn't add any new electrical interface types, just packages existing stuff into a different form factor.

 You do realise that Qubit is mineable by the Baikal ASIC?

 Dunno if you're going to be profitable trying to compete against those with a GPU.

stratum+tcp://s1.theblocksfactory.com:9004   for DGB-scrypt.


 

 DGB has a skein option, if you mine it directly via a pool like theblocksfactory (I think suprnova also has a DGB-skein pool, but not sure there - I do recall at least 2 other DGB-skein pools from listings).

 Vortac does get one thing wrong in that article.

 The GIMPS project (Great Internet Mersenne Prime Search) of Prime95 fame predates SETI by about 2 years - founded in 1996 vs. SETI in 1998.

 The Distributed.net project predates SETI@Home as well (founded in early 1997), and was a replacement for an earlier project dating to 1996.

 SETI was NOT the first "use volunteers to do computation on their "spare" CPU cycles" project, nor is it the longest-lasting - both GIMPS and D.Net are still active (reference the Moo Wrapper and YOYO projects for GRC links to D.Net).

 SETI@Home did better "marketing" though and got a lot more widely known than any other distributed-type project - it MIGHT still be the most widely known such project, though I suspect Folding@Home may have passed it.


 Vortac's article applies better in some respects to Folding@Home and CURE/FLDC than it does to GRC, as a lot of GRC-supported work isn't involved in finding cures (Vortac himself, for example, is a major contributor to the Milkyway project, while I personally am a major Moo Wrapper contributor).

 Read speed on the Seagate Archive 8 TB is a little lower than most other 8TB drives, but we're talking a difference of 150 MB/s appx to perhaps 250 MB/s for the best other Hard Drives on a SATA 6GB interface (15k RPM drives excluded, those are CRAZY expensive).
 SSDs manage quite a bit higher of course, but are severely overpriced for their capasity and you don't NEED that kind of speed for BURST.


 The true key to BURST mining is "lowest cost per TB on a reasonably reliable drive with decent read performance" and having enough CPU or GPU to handle the capacity of your system fast enough to get the deadlines out on a consistent basis before the next block activates.

 That is where the Archive (and the other SMR drives on the market to a somewhat lesser degree as they COST more per TB) shines.

 GDDR 5x has higher latentcy than GDDR 5, which is a large part of the reason why the 1080 and 1080 ti don't perform much IF ANY better on ETH than the 1070 despite having higher memory bandwidth.

 There is no specific miner to dual-mine ETH + Skein, but you could always run separate miners for each to try it.

 Nicehash seems to have been doing a lot more "switching" this past few days vs it's norm - which impacts Claymore's miner more due to the overhead of setting up the SSL connection.


 R9 280x should run fine on Ubuntu 14.04 with the fglrx 15.12 drivers - but I think you have to manually download them and install, I don't remember them being an apt-get options anywhere offhand.

 There is ZERO reason to run "newer" drivers on that card, as long as you don't have anything more recent in the same machine with it.

I don't know if wplotgenerator is "up to date" but it works well enough - plotted out a Seagate Archive 8TB over the last 3.5 days while mining and doing other stuff on the same machine, AMD FX 8570 octocore.

 Probably would have managed it in 2 if I'd been doing nothing else on the machine.

 G1 clocks higher, and should have a 180 watt TDP vs 155 on the ITX.

 I like my ITX cards (just put another one on order yesterday intended to replace a 950), but they mostly shine in situations where you NEED a short card for thermal management - I don't recommend them for riser builds as you can get more performance out of full-length cards.