Claymore's Dual Ethereum AMD+NVIDIA GPU Miner v15.0 (Windows/Linux) - page 938.

Claymore

donator

Activity: 1610

Merit: 1325

Miners developer

Quote from: miner417 on April 20, 2017, 11:47:30 AM

Quote from: Claymore on April 20, 2017, 11:31:48 AM

Finally I got a good progress about the issue with floating speed, I was able to stabilize it, not 100% but close. I will release new version within 2-3 days.

Do you mean by this the issue that say we see 24 MH/s normal, and it drops to 18 or something, then spikes back up?

Exactly.

miner417

member

Activity: 84

Merit: 10

Quote from: Claymore on April 20, 2017, 11:31:48 AM

Finally I got a good progress about the issue with floating speed, I was able to stabilize it, not 100% but close. I will release new version within 2-3 days.

Do you mean by this the issue that say we see 24 MH/s normal, and it drops to 18 or something, then spikes back up?

fittsy

sr. member

Activity: 441

Merit: 252

Quote from: nikisev on April 20, 2017, 11:04:11 AM

Why confuse people?
1. lets presume that we always mine coins with very shirt block time and than maybe it's sort of true
2. Bandwidth needed for a single miner is in low kilobits. "10Mbs>x - can loose a lot" - total crap
3. Lets say your miner boots and starts mining in 2 minutes (normally should be faster than that) so 3 restarts a day will waste 0.16%

You still have to transmit your work. If you have 300 ms ping times then you are definitely going to have more rejected shares than someone on 60 ms ping times.

Claymore

donator

Activity: 1610

Merit: 1325

Miners developer

Finally I got a good progress about the issue with floating speed, I was able to stabilize it, not 100% but close. I will release new version within 2-3 days.

nikisev

full member

Activity: 266

Merit: 101

Im the very decease you pretend to be

Quote from: TechPark on April 19, 2017, 10:49:08 PM

Quote from: Kaller on April 19, 2017, 01:37:18 PM

Quote from: LJLT on April 19, 2017, 01:13:02 PM

@jackbox my problem is that i feel im loosing a complete GPU with those numbers, Claymore reports me 170 MHs but the pool 147 MHs,, as i say i feel im loosing one of my 6 GPUs

For my pool, I also just get about 151/162 = 93% of the hash rate reported by the pool.

Claymore fee 2%, Pool fee 1% makes a total of 3% lost right away. 170-3%=164

1. What is average ping response to your pool? ping poolipaddress -t
2. What is your internet connection bandwidth?
3. How many times per day your miner restarts?

Let us assume:
1. <100ms=you are fine. 100>x<300=1% lost. 300>x - can loose a lot
2. 20Mbsx>10=1% lost. 10Mbs>x - can loose a lot
3. 3 times=1% lost

Could be 5%+ lost easily just on your side.
Also Claymore miner may not calculate hash rate accurately, pool could have issues on their side...

Why confuse people?
1. lets presume that we always mine coins with very shirt block time and than maybe it's sort of true
2. Bandwidth needed for a single miner is in low kilobits. "10Mbs>x - can loose a lot" - total crap
3. Lets say your miner boots and starts mining in 2 minutes (normally should be faster than that) so 3 restarts a day will waste 0.16%

mafaz

newbie

Activity: 48

Merit: 0

Hi guys....

I have (had Sad

) 42 MSI rx 470
12 Sapphire Nitro+ 470
12 Sapphire reference 470 cards.

Latelly ... all off them started falling like leaves. I mean they either die or burn up.

6 msi - 3 sapphire nitro and 2 sapphire reference cards died all during past 3 weeks....

Anyone else have that many breakdowns lately?

I dont even know what to blame now. Bios mods... motherboards.... PSU or claymor v9

Sapphire reference cards all worked for 5 months with only 2 of them die before this nightmare....
Sapphire Nitro+ never broke for 5 months before this....
I recently purchased 30 MSI cards and 6 already dead in like 3 weeks.... they tend to bring out smoke while sapphire cards just die.

My rigs are 6xGPU all (used to be.... LOL) motherboards are BTC PRO and Z97 fatality.
PSU are Be quiet pro 2x 1200 and and Platimaxe's 1600.

I used to use -96 core woltage in afterburner.

I started thinking that might be the reason. But i changed it to 0 and 2 more cards died. SO im pretty much lost.

Any ideas are welcome.

Or maybe someone else has unusual high rate of failing cards recently?

xanx

newbie

Activity: 37

Merit: 0

Quote from: TechPark on April 20, 2017, 08:20:05 AM

Quote from: xanx on April 19, 2017, 11:13:10 PM

I have 5x rx480's and cant seem to get past 27ish on 4 of them and one is locked at 24. The 4 that are at 27 have 2175 memory clock w/ core clock at 1100 and the 5 is 1100/1750 ... so the question is- how can I modify them? Every time i try they revert back and dont want to budge.

Could you please explain step by step what you are doing so we can try to spot the error...

Solo mining ethereum using claymores v9.0 with: a b250f strix mobo and 1200w psu, 4gb RAM, 5 rx480's (I have 6 but apparently b250 chipset dosent allow 6 even though there are 6 pcie slots). In MSI, GPU's 0 through 3 are all synchronized at 1100 core/2176 memory and any I time I try to change theses values, even with small increments the GPU's stop and I have to restart PC. Same thing goes for GPU4 except, GPU4 is clocked at 1266/1750, any modifications results in the same thing and I have to restart only to come back to their reverted values. When I first had 2 GPU's I could get 29/30 no problem while over clocking in MSI but now GPU's 0-3 are stuck at 27ish and GPU 4 at 22-24ish mhz. My config is as follows:

setx GPU_FORCE_64BIT_PTR 0
setx GPU_MAX_HEAP_SIZE 100
setx GPU_USE_SYNC_OBJECTS 1
setx GPU_MAX_ALLOC_PERCENT 100
setx GPU_SINGLE_ALLOC_PERCENT 100
EthDcrMiner64.exe -epool eu1.ethermine.org:14444 -ewal

-epsw x -r 34

The restart timer is just an arbitrary value I play around with. In BIOS all I changed was PCIEXx16_2 configuration from auto to x2 (before this only 3 GPUS would catch in device manager), System agent configuration to Above 4D Decoding [Enabled] and changing the primary Display to CPU graphics. I have the 17.4.2 drivers installed for AMD and the latest/most up to date BIOS.

TechPark

full member

Activity: 238

Merit: 100

Quote from: philip7134 on April 20, 2017, 12:02:00 AM

Hi. I am having problems with some of my cards crashing when I try to overclock their memory. They sometimes go for a few minutes but then usually a card crashes and and the miner restarts as in the picture. I am using several different types of Polaris cards. 4 rx 480 and 2 rx 470. I am using crimson 16.9.1. Does anyone know how to fix this?
Pic of issue. http://imgur.com/a/5Vjy4

Could you please explain step by step what you are doing so we can try to spot the error...

TechPark

full member

Activity: 238

Merit: 100

Quote from: xanx on April 19, 2017, 11:13:10 PM

I have 5x rx480's and cant seem to get past 27ish on 4 of them and one is locked at 24. The 4 that are at 27 have 2175 memory clock w/ core clock at 1100 and the 5 is 1100/1750 ... so the question is- how can I modify them? Every time i try they revert back and dont want to budge.

Could you please explain step by step what you are doing so we can try to spot the error...

MrJaekin

sr. member

Activity: 275

Merit: 258

Quote from: winiu on April 20, 2017, 01:03:17 AM

Quote from: ralftoringo on April 18, 2017, 04:53:19 PM

I have the same issue as Coyn on 2x RX480, since upgrading from 9.0 to 9.1: Siacoin mining speed doubles, but Ethereum mining speed is very unstable and often drops. The cards do not appear to be throttling, temperatur is normal and there is sufficient power.

Adjusting dcri does not help, even with low numbers Ethereum mining speed is unstable. Currently I use -asm 0 to fix it, but of course that causes much lower Siacoin mining speed.

I have exacly the same problem on RX480 and RX470.
9.0 or -asm was much more stable.

Same problem here.. I've got a rig with 6 RX480s in it and no matter what dcri or asm setting I use, I can't get a decent ETH speed while dual mining. Stock BIOS. Mining ETH alone, I get 30.5 to 31MHS. If I dual mine with PASC, the highest I can get ETH is 25MHS. Can someone that is successfully getting full speed from ETH and PASL post their bat/config file?

nevermind41

full member

Activity: 186

Merit: 100

Quote from: Claymore on April 19, 2017, 07:43:51 AM

Quote from: yoshcat on April 19, 2017, 06:41:12 AM

Greetings, Claymore.
There is a suggestion - can you add to the dual-miner function - delayed launch of the second coin's mining? Preferably with a range of 0 - 60 sec.
Sometimes there are bugs (the dumping of video cards, a reset on power) at the start, there is an idea that this can happen because the second coin begins to dig before the dag is formed and the ether mining will start ...
Roll Eyes

Miner cannot mine second coin if DAG is not yet generated. Probably the reason is the power consumption in dual mode, check PSU and cables.

@Claymore
Can you add to your miner releases recommended drivers for rx 4xx cards?

winiu

newbie

Activity: 11

Merit: 0

Quote from: ralftoringo on April 18, 2017, 04:53:19 PM

I have the same issue as Coyn on 2x RX480, since upgrading from 9.0 to 9.1: Siacoin mining speed doubles, but Ethereum mining speed is very unstable and often drops. The cards do not appear to be throttling, temperatur is normal and there is sufficient power.

Adjusting dcri does not help, even with low numbers Ethereum mining speed is unstable. Currently I use -asm 0 to fix it, but of course that causes much lower Siacoin mining speed.

I have exacly the same problem on RX480 and RX470.
9.0 or -asm was much more stable.

noviceminer

newbie

Activity: 72

Merit: 0

Quote from: philip7134 on April 20, 2017, 12:02:00 AM

Hi. I am having problems with some of my cards crashing when I try to overclock their memory. They sometimes go for a few minutes but then usually a card crashes and and the miner restarts as in the picture. I am using several different types of Polaris cards. 4 rx 480 and 2 rx 470. I am using crimson 16.9.1. Does anyone know how to fix this?
Pic of issue. http://imgur.com/a/5Vjy4

Connect display using port on GPU to overclock then switch to VGA mode if possible. When you get your display back, start the miner. Use HDMI or DP port only to overclock and connect display to VGA. I faced same issue

cryptominer420

sr. member

Activity: 450

Merit: 255

Lower the memory overclock on GPU 4 you can find it by pressing 4 ing the miner to disable the GPU from mining then use msi to lower the mem by 50 MHz

philip7134

newbie

Activity: 5

Merit: 0

Hi. I am having problems with some of my cards crashing when I try to overclock their memory. They sometimes go for a few minutes but then usually a card crashes and and the miner restarts as in the picture. I am using several different types of Polaris cards. 4 rx 480 and 2 rx 470. I am using crimson 16.9.1. Does anyone know how to fix this?
Pic of issue. http://imgur.com/a/5Vjy4

xanx

newbie

Activity: 37

Merit: 0

I have 5x rx480's and cant seem to get past 27ish on 4 of them and one is locked at 24. The 4 that are at 27 have 2175 memory clock w/ core clock at 1100 and the 5 is 1100/1750 ... so the question is- how can I modify them? Every time i try they revert back and dont want to budge.

noviceminer

newbie

Activity: 72

Merit: 0

Quote from: TechPark on April 19, 2017, 10:49:08 PM

Quote from: Kaller on April 19, 2017, 01:37:18 PM

Quote from: LJLT on April 19, 2017, 01:13:02 PM

@jackbox my problem is that i feel im loosing a complete GPU with those numbers, Claymore reports me 170 MHs but the pool 147 MHs,, as i say i feel im loosing one of my 6 GPUs

For my pool, I also just get about 151/162 = 93% of the hash rate reported by the pool.

Claymore fee 2%, Pool fee 1% makes a total of 3% lost right away. 170-3%=164

1. What is average ping response to your pool? ping poolipaddress -t
2. What is your internet connection bandwidth?
3. How many times per day your miner restarts?

Let us assume:
1. <100ms=you are fine. 100>x<300=1% lost. 300>x - can loose a lot
2. 20Mbsx>10=1% lost. 10Mbs>x - can loose a lot
3. 3 times=1% lost

Could be 5%+ lost easily just on your side.
Also Claymore miner may not calculate hash rate accurately, pool could have issues on their side...

I just wanted to quote you since there is no option to give a THUMBS UP for a sensible technical post helping fellow miner understand the reason behind his issue. Kudos. Keep up the good work

TechPark

full member

Activity: 238

Merit: 100

Quote from: Kaller on April 19, 2017, 01:37:18 PM

Quote from: LJLT on April 19, 2017, 01:13:02 PM

@jackbox my problem is that i feel im loosing a complete GPU with those numbers, Claymore reports me 170 MHs but the pool 147 MHs,, as i say i feel im loosing one of my 6 GPUs

For my pool, I also just get about 151/162 = 93% of the hash rate reported by the pool.

Claymore fee 2%, Pool fee 1% makes a total of 3% lost right away. 170-3%=164

1. What is average ping response to your pool? ping poolipaddress -t
2. What is your internet connection bandwidth?
3. How many times per day your miner restarts?

Let us assume:
1. <100ms=you are fine. 100>x<300=1% lost. 300>x - can loose a lot
2. 20Mbsx>10=1% lost. 10Mbs>x - can loose a lot
3. 3 times=1% lost

Could be 5%+ lost easily just on your side.
Also Claymore miner may not calculate hash rate accurately, pool could have issues on their side...

adaseb

legendary

Activity: 3808

Merit: 1723

Quote from: adaseb on April 18, 2017, 07:03:04 PM

Looks like the EXP DEV fee pool was down or had issues and all my rigs stopped mining

Quote

17:52:11:400   9c4   DevFee: failed to mine for a long time, something is wrong, stop mining and reconnect.
17:52:11:400   9c4   Reconnect in 20 sec...
17:52:14:707   c8c   recv: 51
17:52:14:707   c8c   srv pck: 50
17:52:14:754   c8c   srv bs: 0
17:52:14:754   c8c   sent: 179
17:52:15:581   a68   checkread timeout
17:52:15:581   a68   DevFee: ETH: Job timeout, disconnect, retry in 20 sec...

It seemed to connect but not mine on the dev pool

Quote

17:52:35:595   a68   DevFee: ETH: Stratum - connecting to 'exp-eu.dwarfpool.com' <158.69.199.198> port 8018
17:52:35:642   a68   send: {"worker": "eth1.0", "jsonrpc": "2.0", "params": ["0xe8b6818cf0d24Bd0E7ded854b3D368662a150dAb", "x"], "id": 2, "method": "eth_submitLogin"}

17:52:35:642   a68   DevFee: ETH: Stratum - Connected (exp-eu.dwarfpool.com:8018)

Had no choice but to use -nofee 1

Claymore,

I am still getting this issue. Can't mine with any 280X anymore because the dev EXP pool seems to have issues.

AriesIV10

legendary

Activity: 1260

Merit: 1006

Mine for a Bit

I was waiting for this video to come out. Great presentation on the new RX5xx Series GPUs.

PCB Breakdown: Sapphire RX 580 Nitro +
https://youtu.be/eaE_6eFaYjs

I am sticking with the RX 470 4GB.

Topic: Claymore's Dual Ethereum AMD+NVIDIA GPU Miner v15.0 (Windows/Linux) - page 938. (Read 6590718 times)