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

[Skol600ml]

Anyone can help me pleaseeee, I met this problem:
I have 6 RX470, install Windows 10 1703 on gpt ssd. Driver I use is Crimson 16.9.2, windows detect all gpu as well.
First time i run tools, everything is ok, but after one times, i close Claymore tools by restart computer.
After this time, Claymore tools is working but only GPU 0 can display its temperature, fan, other GPU cannot display.
I use -di command but Claymore tools only can display only one GPU.
I don't know why, so i install windows many times, try newest AMD driver but, this problem still appears. Anyone met this help me pleaseeeeeee 

Usually this happens when crossfire is on.

[Skol600ml]

I have 11 rx 570. I have calculated that i use 1550w total setup in two rigs. Is this power usage high? Thats about 140w per card (including other components), or 135w without other components.

i have 3 rigs with 6x gtx 1060 3gb

1 rig with 4x rx 480 8gb

1 rig 6x 1060 3gb (samsung memory) mining at 143mh ( claymore 9.7 very good work is perfect now!!!! very stable!!!! better than ethminer now)
+100 core
+730 memory
tdp 55%
at wall kill a watt says 495 watt!!!!!!!!!
all this rigs using just one thermaltake 730w!!

1 rig 4x rx 480 bios mod (27+25+27+25 i cant get more) they mining 104mh power drain is 700w!!!!!!!! nvidia won!!!!
with this i am using corsair 850w

More like "i don't know how to mod AMD cards and dual mining"

[z0n0]

Does anyone know if claymore miner is a fork of CGminer or SGminer?

 

[scryptr]

Does anyone know if claymore miner is a fork of CGminer or SGminer?

 

Claymore's miner is based on proprietary code.  No doubt he looked at a lot of code over the years, but the code is his own.  Other coders try to take his stuff apart, some succeed.  If he copied code he would get caught.

He was caught once.  His Zcash miner had copied code, he had to remove it and he apologized.  There is no support for nVidia cards in his Zcash miner as a result.       --scryptr

[z0n0]

Does anyone know if claymore miner is a fork of CGminer or SGminer?

 

Claymore's miner is based on proprietary code.  No doubt he looked at a lot of code over the years, but the code is his own.  Other coders try to take his stuff apart, some succeed.  If he copied code he would get caught.

He was caught once.  His Zcash miner had copied code, he had to remove it and he apologized.  There is no support for nVidia cards in his Zcash miner as a result.       --scryptr

Thanks for your answer. Pls check PM.

[xxxmattion]

i`m testing smos and have a prop with overclocking
 
i`ve  1050 ti gpu

the problem is when i make a change on the website the miner repot to applay changes and it`s writing my oc settings
but the hash rate never changes

on windows the default hashrate is 11.5 - after over clock 13

on smos it`s only 11.5 and never changes

i downloaded the system 14-7-2017

with file name : simpleminer-NV-v1103-kernel4.11.4-nvidia375.66-i915fw.imgc

can some one help??

[NavyCuda]

Ain't gonna be Claymore - the new ISA may offer a few neat toys... but the REAL speed you'll need to work to get yourself by fucking with the HBM2.

Forgive me, but what is ISA?

I wish I could get the HBM2 memory clock over 1100MHz, but that seems to be the hard limit.  I can't get my vega cards to sit at a specified core clock.  It seems from my experimentation that the core clock needs to be equal to our slightly above the memory clock to achieve peak hashrate.  Any higher and it's wasted energy.  I've also noticed that GPU temp has a drastic effect on performance, so getting lower clocks, lower volts and high memory clocks seems to be the ticket.

[preda]

I have 11 rx 570. I have calculated that i use 1550w total setup in two rigs. Is this power usage high? Thats about 140w per card (including other components), or 135w without other components.

i have 3 rigs with 6x gtx 1060 3gb

1 rig with 4x rx 480 8gb

1 rig 6x 1060 3gb (samsung memory) mining at 143mh ( claymore 9.7 very good work is perfect now!!!! very stable!!!! better than ethminer now)
+100 core
+730 memory
tdp 55%
at wall kill a watt says 495 watt!!!!!!!!!
all this rigs using just one thermaltake 730w!!

1 rig 4x rx 480 bios mod (27+25+27+25 i cant get more) they mining 104mh power drain is 700w!!!!!!!! nvidia won!!!!
with this i am using corsair 850w

More like "i don't know how to mod AMD cards and dual mining"

Haahaha bro i know very well how to dual mining infact i am dual decred at 3200 mh with amd rig but is low earn

[Tharbo]

Hi, I just started using this version of Claymore and I get an error that doesn't let me mine ETH:

ETH: Received error: {"i":4,"result":null,"error":[24,"unauthorizedworker"]} and then only DCR keeps getting it's share normally. I have registered a worker correctly for both pools (ETH and DCR), any ideas? password is right for both too. ETH start mining normally and then suddenly this happens. I also have allpools set to 1

Also, any chance in the future to allow to mine only DCR for example?

[HarrisLam]

hey guys, new to the forums.  Been mining for about 1.5 months.

I have 2 PCs mining, 1 was my main rig (older) and the other was a somewhat dedicated miner.  Both on wifi adapters

I've been having a problem in this couple of days with my main PC where i get the "socket was closed remotely by pool error".

When it's not mining, the PC's internet connection is completely fine.  Once I open the miner, connection drops (but shows as connected at the icon), can't go nowhere with chrome and the error pops in the miner.

Once I close the miner, internet connection comes back after a while (or sometimes I need to restart, but point being that connection is ok as long as the miner isn't on)

A moment ago I experimented with a lan cable.  Wifi adaptor in, miner started, same old problem.  Unplugged wifi adaptor and plugged in cable while miner still running, error went away in like 20 seconds and it's happily mining.  This happened while the other PC is mining fine the whole time with a different wifi adaptor.

So according to above, I've seemed to pinpoint the problem to the wifi adaptor, a TP link TL-WN823N.  Anyone knows how this could happen though?  I've also experienced extremely poor performance in online gaming with this adaptor, but I mean the thing still works when I'm just surfing the web, even streaming videos in youtube so connection speed is out of the question.  

What makes the adaptor go bad when executing connection-dependant programs?  Is this adaptor trash or not?

[jstefanop]

Does anyone know the maximum safe voltage I can run HBM2 at?  I seem to have hit a wall at 1100MHz @ 1050mV.  I've risked as high as 1200mV but only managed an unstable 1125MHz on the memory clock.  1100 @ 1050 is rock solid for mining.

Your not changing HBM2 voltage. If your changing memory voltage via wattman your probably not doing anything(if anything your changing VDDCI). Vega's HBM2 voltage is locked at 1.35v for highest state. There is not much headroom to over volt it much more, since they are already over-volting it from the HBM2 spec of 1.2v.

[NavyCuda]

Does anyone know the maximum safe voltage I can run HBM2 at?  I seem to have hit a wall at 1100MHz @ 1050mV.  I've risked as high as 1200mV but only managed an unstable 1125MHz on the memory clock.  1100 @ 1050 is rock solid for mining.

Your not changing HBM2 voltage. If your changing memory voltage via wattman your probably not doing anything(if anything your changing VDDCI). Vega's HBM2 voltage is locked at 1.35v for highest state. There is not much headroom to over volt it much more, since they are already over-volting it from the HBM2 spec of 1.2v.

I did notice that wattman said 1000mV when I switched it off of auto.  I did need to go from 1000 to 1050mV to get 1100MHz stable, though it's possible other factors effected my results.

[NavyCuda]

Does anyone know the maximum safe voltage I can run HBM2 at?  I seem to have hit a wall at 1100MHz @ 1050mV.  I've risked as high as 1200mV but only managed an unstable 1125MHz on the memory clock.  1100 @ 1050 is rock solid for mining.

Your not changing HBM2 voltage. If your changing memory voltage via wattman your probably not doing anything(if anything your changing VDDCI). Vega's HBM2 voltage is locked at 1.35v for highest state. There is not much headroom to over volt it much more, since they are already over-volting it from the HBM2 spec of 1.2v.

You can't? *You* can't? Watch me.

Please share!  I would love to break the 1100MHz wall!

[scryptr]

hey guys, new to the forums.  Been mining for about 1.5 months.

I have 2 PCs mining, 1 was my main rig (older) and the other was a somewhat dedicated miner.  Both on wifi adapters

I've been having a problem in this couple of days with my main PC where i get the "socket was closed remotely by pool error".

When it's not mining, the PC's internet connection is completely fine.  Once I open the miner, connection drops (but shows as connected at the icon), can't go nowhere with chrome and the error pops in the miner.

Once I close the miner, internet connection comes back after a while (or sometimes I need to restart, but point being that connection is ok as long as the miner isn't on)

A moment ago I experimented with a lan cable.  Wifi adaptor in, miner started, same old problem.  Unplugged wifi adaptor and plugged in cable while miner still running, error went away in like 20 seconds and it's happily mining.  This happened while the other PC is mining fine the whole time with a different wifi adaptor.

So according to above, I've seemed to pinpoint the problem to the wifi adaptor, a TP link TL-WN823N.  Anyone knows how this could happen though?  I've also experienced extremely poor performance in online gaming with this adaptor, but I mean the thing still works when I'm just surfing the web, even streaming videos in youtube so connection speed is out of the question.  

What makes the adaptor go bad when executing connection-dependant programs?  Is this adaptor trash or not?

DON'T MINE WITH WIFI--

It really doesn't work well.  Wifi is OK while traveling, but at home I use cable.       --scryptr

[NavyCuda]

Two fun ways - VBIOS edit is simpler, direct I2C access via MMIO registers on Linux through /dev/mem is more fun.

Both my Vega FE's are in my gaming machine running win10.  My Linux box barely had room for my GTX 970.

I have zero experience modding video card bios, and limited experience editing the bios for overclocking a CPU.  I've always been a person to leave things stock unless I have a pressing need.  OC'ed my 3770k for Arma3... playing with my Vega's for mining purposes since that's why I bought them.

[AlainC]

Two fun ways - VBIOS edit is simpler, direct I2C access via MMIO registers on Linux through /dev/mem is more fun.

Is there place to look around that method on RX4/580 ? Is ther an I2C diag tool on Linux, similar to VRMTool...
atl least to learn who to put in VoltageObjectInfo....

[int64]

Can someone tell me some good advice about use of miner. example I use bat file only for start. all other stuf I do in afterburner, what can I beter do in bat or config file. thanks

[jstefanop]

Does anyone know the maximum safe voltage I can run HBM2 at?  I seem to have hit a wall at 1100MHz @ 1050mV.  I've risked as high as 1200mV but only managed an unstable 1125MHz on the memory clock.  1100 @ 1050 is rock solid for mining.

Your not changing HBM2 voltage. If your changing memory voltage via wattman your probably not doing anything(if anything your changing VDDCI). Vega's HBM2 voltage is locked at 1.35v for highest state. There is not much headroom to over volt it much more, since they are already over-volting it from the HBM2 spec of 1.2v.

You can't? *You* can't? Watch me.

Never said *I* can't ;P Already have it at 1.45v and pushing 1200. Just pointing out the other guy that thinks he's changing HBM2 voltage with the slider on watt man

[levistep]

Very low hash rate less than 3Mh/s per card! Offering .1 eth to solve problem.

My settings:
Windows 10 with latest updates
Nvidia 1070 Hybrids with latest updates (tried many different versions as well)
Synced full blockchain with Geth
Windows tweaks for 16gb virtual memory
Claymore 9.7 (also tried 9.5)


I have 4 GPUs running and they are pulling less than 9Mh/s in total. I have spent like 20 hours trying to fix this problem i'm just about ready to part out this whole rig. I even tried reinstalling windows and trying everything from scratch without any luck. Willing to pay .1 eth to whomever can help me get this to 25Mh/s+ per card like it's supposed to be.

[FFI2013]

I have a question Im loosing 20-35 mh from reported to calculated thats a whole card is this normal I no they should be different but I think this is high with my r9 cards usually everthing is close but this in on rx470 and 570 rigs Im on nanopool maybe I'll try switching pools

I can tell you about my case:

Dev fee = 2%
Pool fee = 1%
Stale shares/Rejected shares (due to network latency) + Incorrect shares (due to overclocking) = aprox 2%

So in my case, my average reported hashrate should be  5% lower than my reported hashrate and indeed it is.

Calculate those numbers in your case and if they do not sum up, then something is wrong with your pool.

I had 20% (!!!) lower shares from suprnova DCR pool for the last 2 weeks and I switched to CoinMine

Thank You that's what I figured something is wrong going to try miningpoolhub