 Show Posts
|
|
Pages: [1] 2 »
|
It might have been answered already, or it might have not been;
On rigs with 9+ cards, how do I turn off card nr 10 and above? Even from command line when I use -di 0123456789, how to turn on or off those cards that are numbered higher than 9?
abcdef cheers thanks |
|
|
|
|
It might have been answered already, or it might have not been;
On rigs with 9+ cards, how do I turn off card nr 10 and above? Even from command line when I use -di 0123456789, how to turn on or off those cards that are numbered higher than 9?
|
|
|
|
|
That's what dynamic DNS services are for.
Also there is no practical limit to how many masternodes you can run on a single public IP address especially since all the masternodes (at least that I had the opportunity to run) do not need to be exposed to the internet at all (ie can be behind NAT, no need for any port forwards).
Again, my example might not be the right one for everyone since I have several servers and do stuff that are not always crypto related but I definitely prefer this type of setup simply because I'm in control over basically everything.
|
|
|
|
|
There is a whole saga on wireless. The decision on an Access Point is just one part of the equation. If done right, it will work flawlessly. Latency on correctly done wireless is 1-3ms which is negligible since the connection to the mining pool will probably be 30-50ms or more.
If you don't know how to troubleshoot network issues and don't want to invest time and effort into learning, I'd recommend to skip wireless and just use copper.
|
|
|
|
you can but don't do it.
Most PCIe cables are AWG16 (better ones are AWG14) and they're rated at 13 amps. 13A * 12V = 156W. That's the maximum power that's safe to draw using those cables. If you use 1 cable for 2 cards you're asking for trouble, those cables might catch fire.
not like this. PEIe cables with 8Pin was 3x12v, so 13*12*3=450w. Are you sure AWG16 can run with 13A? I check some data , AWG16 max is about 6A. so one PEIe with 8pin can only drive 6*12*3=216w. so this is why 1080ti must 8+6pin or 8+8pin You're right and wrong and I was wrong  AWG16 max current is 3,7A as per https://www.powerstream.com/Wire_Size.htmMy first post was from my head and you're right I didn't take the 3 power leads into consideration. So the correct math is 3,7*3*12=133W per cable. |
|
|
|
|
you can but don't do it.
Most PCIe cables are AWG16 (better ones are AWG14) and they're rated at 13 amps. 13A * 12V = 156W. That's the maximum power that's safe to draw using those cables. If you use 1 cable for 2 cards you're asking for trouble, those cables might catch fire.
|
|
|
|
An issue with 2.6; A machine that works as expected with 2.5d, with 2.6 after it creates the DAG file and starts mining the main thread is somehow reported as not responding although it is still mining. I've setup the reboot.bat file so the machine goes into reboot loop after that. 17569:04:49:32.675: eths Eth: Received: {"id":6,"jsonrpc":"2.0","result":true}
17569:04:49:37.711: wdog Main thread not responding
17569:04:49:37.711: wdog Thread(s) not responding. Restarting.
17569:04:49:38.355: eths Eth: New job #df7f78ef from eu1.ethermine.org:4444; diff: 4000MH
17569:04:49:39.648: main Eth speed: 172.288 MH/s, shares: 3/0/0, time: 0:00
17569:04:49:39.648: main GPUs: 1: 28.326 MH/s (0) 2: 29.184 MH/s (0) 3: 28.627 MH/s (1) 4: 29.190 MH/s (0) 5: 28.579 MH/s (2) 6: 28.382 MH/s (0)
17569:04:49:39.648: main GPU1: 51C 70%, GPU2: 50C 70%, GPU3: 50C 70%, GPU4: 51C 70%, GPU5: 58C 70%, GPU6: 49C 70%
Same machine with the same config file works as expected on 2.5d. Might be worth mentioning that all mining machines have only access to the DNS servers and the mining pools, everything else is blocked. |
|
|
|
|
I'm back to Claymore. The stability issues along with the reconnection issues made me lose more than I gained from increased speed.
IMHO this is not a production ready product, sorry. I gave it a good 2 weeks testing over 3 different versions but the issues make it just not worth it.
Good luck with the future development as it seems to be a worthy contender but it's not really production ready yet.
|
|
|
|
I`m just wondering - my BB has 6 pins, but PICO input connector 4 pins - is that ok? My mobo has 8 pin power connector, but PICO 4 pin. Do you have same and it works?
My PICOs don't have an input connector like that, I have to modify them to work with a 6-pin PCIe connector. It's okay as long as it's a 12V input PICO and you don't swap the polarity. Make sure you don't just look at the wire colors but actually check polarity with a multimeter because I actually got a number of PCIe power cables that have their wires swapped. In what sequence you turn your psu`s on ?
I turn them on at the same time. I actually have this all on a switched PDU. Depending on how long your mainboard POSTs you should be good even if you don't power them on at the same time. It's safe to turn the card-only PSUs first as well. |
|
|
|
|
I run a very similar setup.
Depending on the algo and the card optimization (OC/DC) it should run that no problem.
On 220V that PSU runs 3x1080Ti + mainboard no issue. Don't know if you have 220V but I think it's 900W max on 110V which should be enough in theory if you optimize your cards.
|
|
|
|
|
yes, it's pretty much the same.
On AMD there's a noticeable difference (cca 1MH/s higher). Besides that I can see a much more consistent submitted share count compared to Claymore. The graph is not so choppy but almost a nice straight line. This goes for nVidia cards as well.
|
|
|
|
|
a pleasant surprise I must say.
besides lower fee, AMD cards actually hash 1Mh/s more, didn't see any increase on nVidia
All things considered, 1,5%-2% better mining performance on AMD systems.
thanks a lot!
|
|
|
|
|
The only issue I had with this board was pre-skylake processor support. Once that was sorted everything works as expected.
|
|
|
|
|
I bought a used Dell 2U server with plenty of RAM (64GB) and run pretty much everything on it.
Using KVM hypervisor and over 20 OS instances, most of them are not crypto related.
That's 500$ worth of hardware, no rent, just electricity cost and know how.
|
|
|
|
|
hello boys,
First thanks to Fullzero and everyone else who contributed, love the project!
Maybe someone has an idea why my rig is rebooting:
- It's a 12-card GTX 1060 rig powered by a 2880W PSU
- changes in in 1bash:
COIN="ETH"
LOCALorREMOTE="REMOTE"
SSH="YES"
ETH address
ETH pool us1.ethpool.org:3333
So the rig boots, runs 1bash and after it sets the OC settings 1bash seems to start from the beginning and after 3 or 4 loops the rig restarts.
anyone has a clue what's up with that?
thank in advance!
EDIT: fixed it, was running LOCALorREMOTE="REMOTE" with a monitor attached
|
|
|
|
|
Yes and if I do that then you will ask for me to bring the card to you and let you test it yourself. And when I do that then you will ask for proof of purchase and will want me to rip the card apart so that you can see for yourself the print on the chip.
Believe it or don't believe it dear but this is all you're getting
|
|
|
|
Is the vega rx 64 for press review
Or so you say  yes master next time we will ask for your permission |
|
|
|
|
It is true, it can only do 37MH/s stock. Maybe after driver optimization and other mining tricks applied we will see better but so far it's only 37 MH/s
|
|
|
|
|
