Claymore's Dual Ethereum AMD+NVIDIA GPU Miner v15.0 (Windows/Linux)

[IncludeBeer]

I finally got my voltmeter and took some measurements.  My rig:
H81 PRO BTC
2x 1k watt EVGA Supernova PSUs
4x 390x
2x 280x
tiny intel cpu

I configured my gpus with

• 1x 390x + 2x 280x + MB on 1st PSU
• 3x 390x on 2nd PSU

Using Add2PSU to connect the 2.

Using dual mode mining at stock speeds, I'm pulling ~1040 watts from first PSU and ~1100 watts from 2nd PSU

Is that safe to do? I didn't know a 1000 watt PSU could pull more than that. 

PSUs are able to deliver more wattage in peaks, but not at constant load.
I would buy a proper wattage meter and check then.
I dont think that your readings are correct, your mining would not be stable with this kind of load.

900W on each is certainly doable with your setup and I would not stress the PSUs more than that.
Undervolt if possible, it can do miracles.

Ok, thanks for the advice.  It's a Kill-a-Watt voltmeter, which I've been told are quite reliable.  To be fair though, I only ran dual mining a couple minutes to take measurements.  I've been mining in eth-only mode since because I was afraid of what it might do to my rig with constant mining.  I bought some more fans for my rig, so when I get those connected I'll take some time optimizing (undervolting/underclocking) and then take some more measurements.

[1713571953]

1713571953

[1713571953]

1713571953

[1713571953]

1713571953

[pacolito]

@Claymore

Any idea why this doesn’t work on my end?
Really hope you can help me out.
Thanks.

EthDcrMiner64.exe -epool us1.ethermine.org:4444 -esm 1 -ewal 0x362613b3bb5fb1e2bcd6b015674ed7a5877c90cb.rig1 -epsw x

I keep getting this message:

ETH: Stratum - Cannot connect to us1.ethermine.org:4444
ETH: Stratum - Failed to connect, retry in 20 sec...

[Gagana]

@Claymore, I'm testing -tstop option.

Why does it stop monitoring the temperature of the stopped GPU? I thought it continued mining when it cooled down... Can you add the option -tstart? ) If the temp decreases to this limit, it starts mining again.

And... Is it possible to finish the calculation of the current share even if the temp has reached the -tstop limit?

You are correct about monitoring temp after stop, I will fix it. But -tstart is not a good idea, if GPU reached -tstop value, something is wrong.

Well, if I host 50+ rigs and I have a temporary problem with the cooling, I don't want to restart every rig after the problem is solved. And I even don't want to stop completely. I just don't want to overheat the cards. So, even if my rigs will work 1/2 or 1/3 of the usual time, it's much better then the full stop. Please, think it over again )

[elzium]

@Claymore

Any idea why this doesn’t work on my end?
Really hope you can help me out.
Thanks.

EthDcrMiner64.exe -epool us1.ethermine.org:4444 -esm 1 -ewal 0x362613b3bb5fb1e2bcd6b015674ed7a5877c90cb.rig1 -epsw x

I keep getting this message:

ETH: Stratum - Cannot connect to us1.ethermine.org:4444
ETH: Stratum - Failed to connect, retry in 20 sec...

Nothing wrong with command line, copied into my rig and works fine, just renamed worker. Getting u started, check stats

Maybe connect ur rig to internetz?

[pacolito]

That's really weird, I am baffled.
Thank you for trying this but my internet is connected as I'm able to use nanopool without any issues.
There was another guy who claimed to have the same issue so maybe I am missing something.
Anyway, I doubt I can figure this one out so hopefully someone who's had this problem before could shed some light.
Thanks again for your help.

[pikachuy]

That's really weird, I am baffled.
Thank you for trying this but my internet is connected as I'm able to use nanopool without any issues.
There was another guy who claimed to have the same issue so maybe I am missing something.
Anyway, I doubt I can figure this one out so hopefully someone who's had this problem before could shed some light.
Thanks again for your help.

Try -esm 0 or port 14444 or see if your firewall is not blocking the miner.

[pacolito]

tried all those without success.
When I change the 4444 to 3333 I'm able to mine but after googling it, it seems I am mining on ethpool with that.
Someone was saying how some ISPs have blocked that port so at this point it looks like I cant mine on ethermine.org
Maybe I'll try a more distant port just to see if that works but probably not the best since there will be a big distance.

[pikachuy]

I say something is wrong with ethermine.org, mine was acting all weird starting yesterday, like it reported 0 hash and so forth, but is back to normal today. Nothing was changed on my end. So most likely something wrong with their servers because it went from working perfectly fine, then derp mode for the whole day, then back to perfectly fine.

[merc84]

I finally got my voltmeter and took some measurements.  My rig:
H81 PRO BTC
2x 1k watt EVGA Supernova PSUs
4x 390x
2x 280x
tiny intel cpu

I configured my gpus with

• 1x 390x + 2x 280x + MB on 1st PSU
• 3x 390x on 2nd PSU

Using Add2PSU to connect the 2.

Using dual mode mining at stock speeds, I'm pulling ~1040 watts from first PSU and ~1100 watts from 2nd PSU

Is that safe to do? I didn't know a 1000 watt PSU could pull more than that. 

PSUs are able to deliver more wattage in peaks, but not at constant load.
I would buy a proper wattage meter and check then.
I dont think that your readings are correct, your mining would not be stable with this kind of load.

900W on each is certainly doable with your setup and I would not stress the PSUs more than that.
Undervolt if possible, it can do miracles.

Ok, thanks for the advice.  It's a Kill-a-Watt voltmeter, which I've been told are quite reliable.  To be fair though, I only ran dual mining a couple minutes to take measurements.  I've been mining in eth-only mode since because I was afraid of what it might do to my rig with constant mining.  I bought some more fans for my rig, so when I get those connected I'll take some time optimizing (undervolting/underclocking) and then take some more measurements.

I was getting whacked reading from my watt meter when it was plugged into a power board, plugging it (the watt meter) in at the wall made the reading much more realistic. Also the rating on psu is for deliverable power for example depending on efficiency of psu a 1000w psu can deliver 1000w but will consume more at the wall depending on how efficient it is.

[fr4nkthetank]

Hey everyone, hope mining is good.  My garage rig = 2x390, 2x380, pulls 1140w at the wall, about 15% overclock, slight undervolt.  could be better but hey power is cheap here


Anyways, since I added my 4th gpu, I got these weird reboots once in a while, error is win bug recheck.  so thats hardware I suppose.  any ideas ?  its a z97 ann edition with 4 slots (3 x1 and 1x16) and I run 3 gpus off usb extenders.  never rebooted before.  psu is fine, its a 1600w evga I had laying around...

edit: for fun, im getting 108 on eth and 2.15 on dcr

[pacolito]

go to bios and make sure your pci-e gen is 2 and not 3
This worked for me.

[fr4nkthetank]

go to bios and make sure your pci-e gen is 2 and not 3
This worked for me.

ah thanks, I literally just thought of that.  didnt really want to bring my monitor all the way down there lol but hey what can you do.  thanks ill give it a try.

[Docnaster]

Hey everyone, hope mining is good.  My garage rig = 2x390, 2x380, pulls 1140w at the wall, about 15% overclock, slight undervolt.  could be better but hey power is cheap here


Anyways, since I added my 4th gpu, I got these weird reboots once in a while, error is win bug recheck.  so thats hardware I suppose.  any ideas ?  its a z97 ann edition with 4 slots (3 x1 and 1x16) and I run 3 gpus off usb extenders.  never rebooted before.  psu is fine, its a 1600w evga I had laying around...

edit: for fun, im getting 108 on eth and 2.15 on dcr

What is the capacity of the power supply. If it is at the limit of the PSU, the system will reboot some times. If that is not a single rail, you need to balance the output.

[Patrice7]

That's really weird, I am baffled.
Thank you for trying this but my internet is connected as I'm able to use nanopool without any issues.
There was another guy who claimed to have the same issue so maybe I am missing something.
Anyway, I doubt I can figure this one out so hopefully someone who's had this problem before could shed some light.
Thanks again for your help.

It's most likely just the port number 4444.  My ISP has blocked that port number because the W32.Blaster.Worm from years ago. Check with your ISP.

[Alehandros]

love the mining manager! thanks Claymore

the webpage...its ripe for some pimpin, has anyone had any luck?

I suppose a Chrome plugin could work, a sparkline chart would be nice

[fr4nkthetank]

Hey everyone, hope mining is good.  My garage rig = 2x390, 2x380, pulls 1140w at the wall, about 15% overclock, slight undervolt.  could be better but hey power is cheap here


Anyways, since I added my 4th gpu, I got these weird reboots once in a while, error is win bug recheck.  so thats hardware I suppose.  any ideas ?  its a z97 ann edition with 4 slots (3 x1 and 1x16) and I run 3 gpus off usb extenders.  never rebooted before.  psu is fine, its a 1600w evga I had laying around...

edit: for fun, im getting 108 on eth and 2.15 on dcr

What is the capacity of the power supply. If it is at the limit of the PSU, the system will reboot some times. If that is not a single rail, you need to balance the output.

seems fixed now   Anyways, I would recommend every one to simply put a shortcut to your .bat file into your startup folder (if you run windows), so when the computer reboots, the miner starts automatically.  Also I put a .txt to pop up so I know it rebooted.  Oh the psu....its a beast, evga g2 1600w, has a massive 133A on a single rail...

[tolazy]

someone should write a website like the cryptoglance bundel for asics miners.
i know i cant do it but must be somone that can and might have some time over for it

[MacT]

I say something is wrong with ethermine.org, mine was acting all weird starting yesterday, like it reported 0 hash and so forth, but is back to normal today. Nothing was changed on my end. So most likely something wrong with their servers because it went from working perfectly fine, then derp mode for the whole day, then back to perfectly fine.

 Those guys at ethermine/ethpool.org have had a rough 10 days of one thing after another making it so payouts didnt happen or underpaid or hashrates went to half for a hour or two multiple days, payout modules going rogue, servers busting, and on and on. There was some mention of someones rig using Claymore not showing up at all but their genoil ones being ok. Starts back at May 17 and continues through today. https://forum.ethereum.org/discussion/5532/ethermine-org-the-fastest-way-to-mine-ether/p9

[IncludeBeer]

I finally got my voltmeter and took some measurements.  My rig:
H81 PRO BTC
2x 1k watt EVGA Supernova PSUs
4x 390x
2x 280x
tiny intel cpu

I configured my gpus with

• 1x 390x + 2x 280x + MB on 1st PSU
• 3x 390x on 2nd PSU

Using Add2PSU to connect the 2.

Using dual mode mining at stock speeds, I'm pulling ~1040 watts from first PSU and ~1100 watts from 2nd PSU

Is that safe to do? I didn't know a 1000 watt PSU could pull more than that. 

PSUs are able to deliver more wattage in peaks, but not at constant load.
I would buy a proper wattage meter and check then.
I dont think that your readings are correct, your mining would not be stable with this kind of load.

900W on each is certainly doable with your setup and I would not stress the PSUs more than that.
Undervolt if possible, it can do miracles.

Ok, thanks for the advice.  It's a Kill-a-Watt voltmeter, which I've been told are quite reliable.  To be fair though, I only ran dual mining a couple minutes to take measurements.  I've been mining in eth-only mode since because I was afraid of what it might do to my rig with constant mining.  I bought some more fans for my rig, so when I get those connected I'll take some time optimizing (undervolting/underclocking) and then take some more measurements.

I was getting whacked reading from my watt meter when it was plugged into a power board, plugging it (the watt meter) in at the wall made the reading much more realistic. Also the rating on psu is for deliverable power for example depending on efficiency of psu a 1000w psu can deliver 1000w but will consume more at the wall depending on how efficient it is.

Ok, makes sense.  I live in an older building, so figured it was best to not have both PSUs at the same outlet (e.g. with an extender).  So both are of course connected directly to wall.

[Charloz24]

Updated all my miners to 4.4 and set up remote manager and must say IT'S REALLY NICE!

Suggestion: adding e-mail notification on manager when a problem is detected