Win10 reserves something like 300-600MB VRAM, where as ethOS only reserves 8 MB. Win 10 rigs with 4gig cards will depending on card manufacturer and type, stop mining ETH some times around October. Linux based miners will run out og VRAM memory around Christmas.

You might be on to something there.
Until now, I have been running Claymore 15, in ethOS, but also tried ethminer, no difference.

Yes they are, 4 GB, Sapphire Nitro.

No out of VRAM alert.

My rigs stop mining today, when Epoch 352 went active.

After restart, 13 GPU are detected, but AMD driver is not capable of stopping fan on a single GPU anymore, indicating, no control.

When miner starts, miner is loading, but no hashing.

In "show miner" it say, all GPU hang and miner must be restarted.

Anyone have similar problem, or solution ?

Have tried ethOS update, no result.
Have tried miner update, no result.

Running ethOS 1.3.3, without trouble for month.

Any input appreciated.

My rigs running with 4GB Nitro GPU's are utilizing 3.4GB onboard RAM, and 4.8GB Vitual Memory each.

On my H110's I have Vitual Memory settings to 65536 MB, with 12 GPU.

Running fine on Claymore 11.5, don't work on 14.5.

I missed that you use 14.6 I can't get that to work on a mixed rig ( Asrock H110 ) either. Still need to figure out why it crashes when it try to load RXboost. Without RXboost, it runs perfect, 8 AMD + 5 GTX 1070.

Use GPU Tweak II, instead of command line in Claymore.

https://www.asus.com/us/site/graphics-cards/gpu-tweak-ii/

In my experience it work better than Claymore.
Set power to 65%.

Considder :
Create a *.bat file and load delay 90 sec GPU Tweak II.
Create a shortcut.

Ad to startup file, together with shortcut to start.bat for Claymore.

This creates auto start, makes life more easy.

During the last week or so the Hashing power x13 has jumped from a normal 2.5 TH/s to at times 5 TH/s on Nicehash. That has never happened before, even at high price when all where in, it never exceeded 3 TH/s.

I don't think that 2 TH/s X13 hashing power has suddenly been available, and joined NH.

It is more likely that some large mining operator has cracked the overclocking  on the giants.

Or am I overlooking something ?

Anyone know something ?

Exchange when ??

Nodes not working. Wallet no connection

additional nodes ??

This one works, but only for that pool. Have not found any global Explorer yet.

http://new.altcoinix.com/vunt/public/index.php?page=statistics&action=blocks

Try Nicehash or Prohash.
Then you got multicoin pool.

No power down required.

You can kill it anytime, and it will power up again.

Even the PSU is tolerant to voltage variations, but it will draw more Amps, in case voltage drops, so be sure your breaker is big enough.

Now, why would you want to do that ??
You have unstable grid or power cuts ?
The miner have auto reboot after power loss, so just plug it, and let it run.

I have been watching this forum from the outside for some time, and i feel it’s time to add a comment.

Those numbers on miners apparently sold, or up for sale, are in my opinion unrealistic. It would take a huge undertaking, production wise, to manufacture these units, not to speak of getting the chips, where Bitmain are in competition with all other manufacturers who use any kind of chips, to get a slot of production capacity. To my knowledge there are only 4 plants in the world that can manufacture 16 nanometer  technology, and they are constantly booked up.

Next, if Bitmain really should sell all those numbers of miners, I would not really worry about it. There will always be a balance between earnings and cost, and as soon as diff. rate increases so much that earnings go below cost, the first miners will be shut down. They will continue to shut down, in numbers, as diff. raise, until only those with very low (electricity ) cost are still mining.

So there will be a bottom in the marked, but only those who are in control of their cost will survive.