[Available Plugin] XMRig - Monero (XMR) and AEON Cryptocurrency Miner

Joshua Parker Ruehlig

Hall of Famer
Joined
Dec 5, 2011
Messages
5,949
Hmm. Could be, I just rebooted but I am still capped at 180. Care to share your settings? Are you saying that when you rebooted that you went up to 280 but then dropped again after a while?
basically all my settings were default except 'cpu affinity = 0x55'.
I used algorithm variation 1(Hardware AES). I'm not sure if you tried forcing different algorithm variations, but that might change things as well.

So for my e3-1240v2 system, it is actually a FreeBSD server I use to host some stuff, including my security camera system (zoneminder), in a jail. Which uses alot of cpu cycles since it is analyzing 5 HD camera streams. I temporarily disabled zoneminder to test my max hashrate on the system. I noticed I was getting 280h/s on the system, but when I ran xmrig in a jail I got 160-180h/s max. I was confused, tweaking everything, but after rebooting I got 280h/s in a jail as expected.
That was at least a week ago, unfortunately I don't know what hashrate the system can currently max out at, since I normally run zoneminder which keeps the system is busy, and I get 100h/s or less.

It doesn't sound like rebooting does anything in your case, my case might have been something else unrelated, or specific to my setup.
 
Joined
Apr 9, 2015
Messages
1,258
EDIT: after playing around with settings, it seems it is not recognizing my 8 threads. It is only using 4 of them (as is show by my hash rates per thread in the settings page of the plug). Strange. If I change it to 2 threads, it uses 2. If I change it to 4, it uses 4. If I change it to 8, it uses 4. Hmmm?


Remove the 0X55 setting. That is forcing it to only use 4 cores. If you want it to use more cores you have to either remove it or increase the number of cores used.
 

rplucker

Dabbler
Joined
Jul 10, 2016
Messages
30
Hmmmm. So I've had mine running for almost 24 hours and nicehash seems to report 0 hashes/second about 50% of the time, even though I can see 350H/S locally consistently. Any thoughts, anyone?
Capture.PNG
Capture2.PNG
 

Joshua Parker Ruehlig

Hall of Famer
Joined
Dec 5, 2011
Messages
5,949
Hmmmm. So I've had mine running for almost 24 hours and nicehash seems to report 0 hashes/second about 50% of the time, even though I can see 350H/S locally consistently. Any thoughts, anyone?
View attachment 22378 View attachment 22379
I haven't tried out nicehash yet, but maybe it is only accepting your hashes in bursts? I'd check to see what your average hashrate is on nicehash's site and make sure it makes sense with what your miner is submitting.
 

Joshua Parker Ruehlig

Hall of Famer
Joined
Dec 5, 2011
Messages
5,949
Thanks for the plugin works nice for me :)
Would GPU version be possible? i have few spare recent amd gpus in my freenas box
https://github.com/xmrig/xmrig-amd
I'm not sure, i think if freenas had the right driver for it, it could possibly work. if someone could port this to freebsd, i could try to make a plugin for it. though i don't have any amd GPUs to even test.
 

Joshua Parker Ruehlig

Hall of Famer
Joined
Dec 5, 2011
Messages
5,949
Could you make freeNAS plugin based on this one? https://github.com/fireice-uk/xmr-stak
In my testing it has better performance and nice webUI and also really good GPU support for the future
I'm sure i could build the plugin but someone has to port it to freebsd. also for gpu mining to work the video card drivers need to be available in the plugin jail. i don't have time to do all that any time soon, but can help if others do the effort.
 

nojohnny101

Wizard
Joined
Dec 3, 2015
Messages
1,478
Since this seems to be a good place to share. Is everyone else in this thread who is currently mining with this plugin seeing wide variations over time of hashrates? I don't know if this is the nature of mining cryptocurrency (my first time doing it) or if it has something to do with my system. Screenshot for reference:
hashrate.png
 

Nucshuco

Explorer
Joined
Aug 22, 2014
Messages
52
Yes I am
xmrigs.PNG
 
Joined
Apr 9, 2015
Messages
1,258
Since this seems to be a good place to share. Is everyone else in this thread who is currently mining with this plugin seeing wide variations over time of hashrates? I don't know if this is the nature of mining cryptocurrency (my first time doing it) or if it has something to do with my system. Screenshot for reference:
hashrate.png
Yes that is how ALL mining works. Your system will have variations on the speed depending on the difficulty that is given to you. The remote system will also have windows where they average everything together so the chart is just fine. The number you really want to look at is your 24 hour average when you start figuring how much you make.
 

Jailer

Not strong, but bad
Joined
Sep 12, 2014
Messages
4,974
@Jailer I see that we have the same CPU. What are you mining? Did you try Monero and what kind of hashrates were you able to get when maxing everything out?

EDIT: after playing around with settings, it seems it is not recognizing my 8 threads. It is only using 4 of them
I get the same thing on my desktop only using about 45% of the CPU. It is due to the amount of cache on the CPU according to some reddit posters who know a lot more than I do. I have not tried Monero, I just use nicehash.

I've included the CPU benchmark from my desktop (i7 6700K @4.5Ghz) that was in the config file for nicehash. If you'll notice the highest hash rate comes from using 3 threads. I'm consistently getting around 196H/s with the 6700K and around 178H/s on the E3-1240 v3.

Code:
"AlgorithmStringID": "XmrStackCPU_CryptoNight",
			"Enabled": true,
			"BenchmarkModes": [
				{
					"BenchmarkID": "t-1",
					"ExtraLaunchParameters": "-t 1",
					"Speed": [
						73.1,
						0
					],
					"MinerBinaryHashID": "ec90dcf49a849199aa1f8c368d5962fa871624d3"
				},
				{
					"BenchmarkID": "t-2",
					"ExtraLaunchParameters": "-t 2",
					"Speed": [
						140.8,
						0
					],
					"MinerBinaryHashID": "ec90dcf49a849199aa1f8c368d5962fa871624d3"
				},
				{
					"BenchmarkID": "t-3",
					"ExtraLaunchParameters": "-t 3",
					"Speed": [
						199,
						0
					],
					"MinerBinaryHashID": "ec90dcf49a849199aa1f8c368d5962fa871624d3"
				},
				{
					"BenchmarkID": "t-4",
					"ExtraLaunchParameters": "-t 4",
					"Speed": [
						188.4,
						0
					],
					"MinerBinaryHashID": "ec90dcf49a849199aa1f8c368d5962fa871624d3"
				},
				{
					"BenchmarkID": "t-5",
					"ExtraLaunchParameters": "-t 5",
					"Speed": [
						175.7,
						0
					],
					"MinerBinaryHashID": "ec90dcf49a849199aa1f8c368d5962fa871624d3"
				},
				{
					"BenchmarkID": "t-6",
					"ExtraLaunchParameters": "-t 6",
					"Speed": [
						175.4,
						0
					],
					"MinerBinaryHashID": "ec90dcf49a849199aa1f8c368d5962fa871624d3"
				},
				{
					"BenchmarkID": "t-7",
					"ExtraLaunchParameters": "-t 7",
					"Speed": [
						176.2,
						0
					],
					"MinerBinaryHashID": "ec90dcf49a849199aa1f8c368d5962fa871624d3"
				},
				{
					"BenchmarkID": "t-8",
					"ExtraLaunchParameters": "-t 8",
					"Speed": [
						170.3,
						0
 

raidflex

Guru
Joined
Mar 14, 2012
Messages
531
Nice guide I just wish power costs weren't the second highest in the country where I live, otherwise I would mine. Running a server 24/7 idle will still use significantly less power then being 50-100% usage. Idle power draw on a E3-1230V3 is around 20W vs 90W+ under load.
 

Joshua Parker Ruehlig

Hall of Famer
Joined
Dec 5, 2011
Messages
5,949
Nice guide I just wish power costs weren't the second highest in the country where I live, otherwise I would mine. Running a server 24/7 idle will still use significantly less power then being 50-100% usage. Idle power draw on a E3-1230V3 is around 20W vs 90W+ under load.
70W+ seems like a huge delta to me for mining with XMRig. If that is true, it seems like the v3 Xeons have much better idle power consumption, but significantly increase power consumption when actually being stressed, as compared to the v2 Xeons.

I just tested two of my systems to double check my earlier power consumption tests that were taken with a UPS involved.
All of the below results were gathered using a Kill-A-Watt meter, without a UPS involved, mining Monero (XMR).

FreeNAS System
CPU - e3-1220v2
Other Power Users - 10GbE NIC + LSI Card + 10xHDD + 2xSSD
Idle (Not Mining) - 104W
Low Power Mode (using 2 threads) - 122W (18W delta) > 218H/s
Standard Mode (using 4 threads) - 127W (23W delta) > 265H/s

FreeBSD System
CPU - e3-1240v2
Other Power Users - 10GbE NIC + 2*SSD + 2*NVMe
Idle (Not Mining) - 44W
Low Power Mode (using 2 threads) - 74W (30W delta) > 249H/s
Standard Mode (using 4 threads) - 84W (40W) > 295H/s

A few thoughts I had. I have a feeling my PSUs are very inefficient at lower wattages, and increase in efficiency as they get loaded up. The same can probably be said for my UPS (OR1500PFCRT2U) which didn't show as much of a wattage increase when I tests between that and the wall previously. Since my cost is only 0.12cents/KW-hr it pretty much always makes sense to increase H/s at the cost of more Watts, but that probably is not true for areas with higher power rates.
 

raidflex

Guru
Joined
Mar 14, 2012
Messages
531
70W+ seems like a huge delta to me for mining with XMRig. If that is true, it seems like the v3 Xeons have much better idle power consumption, but significantly increase power consumption when actually being stressed, as compared to the v2 Xeons.

I just tested two of my systems to double check my earlier power consumption tests that were taken with a UPS involved.
All of the below results were gathered using a Kill-A-Watt meter, without a UPS involved, mining Monero (XMR).

FreeNAS System
CPU - e3-1220v2
Other Power Users - 10GbE NIC + LSI Card + 10xHDD + 2xSSD
Idle (Not Mining) - 104W
Low Power Mode (using 2 threads) - 122W (18W delta) > 218H/s
Standard Mode (using 4 threads) - 127W (23W delta) > 265H/s

FreeBSD System
CPU - e3-1240v2
Other Power Users - 10GbE NIC + 2*SSD + 2*NVMe
Idle (Not Mining) - 44W
Low Power Mode (using 2 threads) - 74W (30W delta) > 249H/s
Standard Mode (using 4 threads) - 84W (40W) > 295H/s

A few thoughts I had. I have a feeling my PSUs are very inefficient at lower wattages, and increase in efficiency as they get loaded up. The same can probably be said for my UPS (OR1500PFCRT2U) which didn't show as much of a wattage increase when I tests between that and the wall previously. Since my cost is only 0.12cents/KW-hr it pretty much always makes sense to increase H/s at the cost of more Watts, but that probably is not true for areas with higher power rates.

Yeah that is not bad power draw and 0.12cents/KW-hr is great, mine is 0.20cents/KW-hr here.
 

nojohnny101

Wizard
Joined
Dec 3, 2015
Messages
1,478
@Joshua Parker Ruehlig those are good numbers.

I just checked my power draw again through my UPS (APC BR1000G) and this is what I'm getting:

CPU: E3-1240 v3
at idle: 75w
with 6 threads at 75% CPU utilization: 115w
Hashrate at 75% CPU utilization: 180 - 200H/s
 

Joshua Parker Ruehlig

Hall of Famer
Joined
Dec 5, 2011
Messages
5,949
@Joshua Parker Ruehlig those are good numbers.

I just checked my power draw again through my UPS (APC BR1000G) and this is what I'm getting:

CPU: E3-1240 v3
at idle: 75w
with 6 threads at 75% CPU utilization: 115w
Hashrate at 75% CPU utilization: 180 - 200H/s
are you specifying 6 threads? i thought 4 threads (each 2MB cache) would optimal for your processor.
 

nojohnny101

Wizard
Joined
Dec 3, 2015
Messages
1,478
I am specifying 6 threads. Strange thing was is that I was running 4 threads at 50% MAX CPU utilization. I want to use 75% CPU utilization but when I just raise that metric and leave it at 4 threads, the CPU utilization stays at 50%. So I decided to utilize 6 threads with the setting set at 75% utilization which gave me 75% utilization real world. This is what I wanted.

I'm sure the return is maybe not economical as I'm only getting 20-30H/s increase going from 50% to 75% it seems. Hard to tell and I'm just roughly guessing because the hashrate bounces around a lot as I posted above.
 
Joined
Apr 9, 2015
Messages
1,258
The easiest way to figure is look at your average after 24 hours on the site if it has ran constantly. Or look at the 15 min average on the webpage that shows the stats. Then use that based off of your power usage to find your return vs your power costs.

For instance, with my GPU miner I use approx 800 watts (via Kill-a-Watt, 3 GTX1080's) at 0.10 per KWH (actually a little less as once I reach 600 KWH of use it drops to around 0.07 but I don't really care) over 24 hours it costs me approx $1.92 to run. At current spot rates that earns me about 12.21 per day based on a 1.7KH/s average on ZEC. So I am earning quite a bit more than I am spending on electricity.
 

Joshua Parker Ruehlig

Hall of Famer
Joined
Dec 5, 2011
Messages
5,949
I am specifying 6 threads. Strange thing was is that I was running 4 threads at 50% MAX CPU utilization. I want to use 75% CPU utilization but when I just raise that metric and leave it at 4 threads, the CPU utilization stays at 50%. So I decided to utilize 6 threads with the setting set at 75% utilization which gave me 75% utilization real world. This is what I wanted.

I'm sure the return is maybe not economical as I'm only getting 20-30H/s increase going from 50% to 75% it seems. Hard to tell and I'm just roughly guessing because the hashrate bounces around a lot as I posted above.
Ahh, I didn't know increasing more than 1thread/2MB cache could increase hashrate. I might experiment with this a bit more too.
 
Top