Claymore's Dual Ethereum AMD+NVIDIA GPU Miner v15.0 (Windows/Linux) - page 974.

LJLT

newbie

Activity: 48

Merit: 0

Quote from: adly3000 on April 04, 2017, 08:51:27 AM

Quote from: matmator on April 04, 2017, 03:56:08 AM

Quote from: adly3000 on April 03, 2017, 06:00:03 PM

Quote from: matmator on April 03, 2017, 03:57:36 PM

Quote from: adly3000 on April 02, 2017, 10:05:30 PM

I been mining on 2 rigs for 2 months using:
windows 7 ulti.
h81 bro btc + Ram 8 GB + Intel Pentium Processor G3240
every rig have 5 cards of Sapphire nitro+ 480 8G 11260-01-20G set at 1342/2200 hashing 27.4-27.5
I saw people saying that it can be raises to 28-29 (without any mod or bios flash) only using overclock
I would like to say the rigs are stable and working good with 16.11.5 (pow lim 0 & mv -24)
The cards have samsung rams with asics (75-85) but all same hash 27.4-27.5
any help to raise would be appreciated even 0.5 for 10 cards it will be 5 Mh/s

My batch is:
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:4444 -ewal mywallet.Rig2 -epsw x -ethi 16 -esm 2 -mode 1 -tstop 80,75,75,75,75 -allcoins 1
pause

Hello

1342 it's too high!

1200/2200 or 1150/2150.

and -48mv

With bios mod you can have 30mh

Thank you so much. Ireduced to 1266/2200 same production. noticed on 1266/2250 showing 28 Mh/s but 2250 is the max on the msi afterburner which I dont like as it is the maximum, so returned to 1266/2200. I dont like any mod it might destroy card.

If your gpu blocks the memory but not the coreclock, on these cards you can leave the coreclok at 1200.

But your gpu 4 lowers the memory to 2150.

Thank you it was a new valuable info for me after 10 months of mining. I reduced GPU4 to 2150 and the situation is too much better but I am not sure what is the accepted errors limit. here is the current situation:

GPU0 - errors: 1 - Asic:79.5 @1266/2200 -24mv
GPU1 - errors: 774 - Asic:85.6 @1180/2150 -24mv
GPU2 - errors: 1315 - Asic:73.6 @1266/2200 -24mv
GPU3 - errors: 4 - Asic:75.8 @1266/2200 -24mv
GPU4 - errors: 34 - Asic:75 @1180/2150 -24mv
GPU5 - errors: 295 - Asic:71.1 @1266/2200 -24mv

Hi, how do you get this kind of report? i mean this: GPU0 - errors: 1 - Asic:79.5 @1266/2200 -24mv

AzzAz

legendary

Activity: 1030

Merit: 1006

Quote from: fr4nkthetank on April 04, 2017, 09:10:13 AM

reduce mem clocks on those gpu's that have many incorrect shares. just calculate the % of incorrect shares vs real shares and you'll see you lose a lot of hashrate.

Or you can see that rising mem clock does not rise hash on particular card(s) so lower clk there. As for memory errors best is none because temps also can rise and get errors back on.

fr4nkthetank

legendary

Activity: 2294

Merit: 1182

Now the money is free, and so the people will be

reduce mem clocks on those gpu's that have many incorrect shares. just calculate the % of incorrect shares vs real shares and you'll see you lose a lot of hashrate.

adly3000

newbie

Activity: 55

Merit: 0

Quote from: matmator on April 04, 2017, 03:56:08 AM

Quote from: adly3000 on April 03, 2017, 06:00:03 PM

Quote from: matmator on April 03, 2017, 03:57:36 PM

Quote from: adly3000 on April 02, 2017, 10:05:30 PM

I been mining on 2 rigs for 2 months using:
windows 7 ulti.
h81 bro btc + Ram 8 GB + Intel Pentium Processor G3240
every rig have 5 cards of Sapphire nitro+ 480 8G 11260-01-20G set at 1342/2200 hashing 27.4-27.5
I saw people saying that it can be raises to 28-29 (without any mod or bios flash) only using overclock
I would like to say the rigs are stable and working good with 16.11.5 (pow lim 0 & mv -24)
The cards have samsung rams with asics (75-85) but all same hash 27.4-27.5
any help to raise would be appreciated even 0.5 for 10 cards it will be 5 Mh/s

My batch is:
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:4444 -ewal mywallet.Rig2 -epsw x -ethi 16 -esm 2 -mode 1 -tstop 80,75,75,75,75 -allcoins 1
pause

Hello

1342 it's too high!

1200/2200 or 1150/2150.

and -48mv

With bios mod you can have 30mh

Thank you so much. Ireduced to 1266/2200 same production. noticed on 1266/2250 showing 28 Mh/s but 2250 is the max on the msi afterburner which I dont like as it is the maximum, so returned to 1266/2200. I dont like any mod it might destroy card.

If your gpu blocks the memory but not the coreclock, on these cards you can leave the coreclok at 1200.

But your gpu 4 lowers the memory to 2150.

Thank you it was a new valuable info for me after 10 months of mining. I reduced GPU4 to 2150 and the situation is too much better but I am not sure what is the accepted errors limit. here is the current situation:

GPU0 - errors: 1 - Asic:79.5 @1266/2200 -24mv
GPU1 - errors: 774 - Asic:85.6 @1180/2150 -24mv
GPU2 - errors: 1315 - Asic:73.6 @1266/2200 -24mv
GPU3 - errors: 4 - Asic:75.8 @1266/2200 -24mv
GPU4 - errors: 34 - Asic:75 @1180/2150 -24mv
GPU5 - errors: 295 - Asic:71.1 @1266/2200 -24mv

matmator

sr. member

Activity: 487

Merit: 252

bou !

Quote from: adly3000 on April 03, 2017, 06:00:03 PM

Quote from: matmator on April 03, 2017, 03:57:36 PM

Quote from: adly3000 on April 02, 2017, 10:05:30 PM

I been mining on 2 rigs for 2 months using:
windows 7 ulti.
h81 bro btc + Ram 8 GB + Intel Pentium Processor G3240
every rig have 5 cards of Sapphire nitro+ 480 8G 11260-01-20G set at 1342/2200 hashing 27.4-27.5
I saw people saying that it can be raises to 28-29 (without any mod or bios flash) only using overclock
I would like to say the rigs are stable and working good with 16.11.5 (pow lim 0 & mv -24)
The cards have samsung rams with asics (75-85) but all same hash 27.4-27.5
any help to raise would be appreciated even 0.5 for 10 cards it will be 5 Mh/s

My batch is:
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:4444 -ewal mywallet.Rig2 -epsw x -ethi 16 -esm 2 -mode 1 -tstop 80,75,75,75,75 -allcoins 1
pause

Hello

1342 it's too high!

1200/2200 or 1150/2150.

and -48mv

With bios mod you can have 30mh

Thank you so much. Ireduced to 1266/2200 same production. noticed on 1266/2250 showing 28 Mh/s but 2250 is the max on the msi afterburner which I dont like as it is the maximum, so returned to 1266/2200. I dont like any mod it might destroy card.

If your gpu blocks the memory but not the coreclock, on these cards you can leave the coreclok at 1200.

But your gpu 4 lowers the memory to 2150.

Ossidalm

newbie

Activity: 60

Merit: 0

Quote from: kinsik on April 04, 2017, 02:32:11 AM

Hi everyone.
Please advice if 3-5% of stale shares in ethermine pool is fine.
Please also advice what are the best values for -ethi and -etht.

Thanx

3-5% of stale shares in ethermine pool is fine as the block time of the ETH is just 14s.

kinsik

newbie

Activity: 39

Merit: 0

Hi everyone.
Please advice if 3-5% of stale shares in ethermine pool is fine.
Please also advice what are the best values for -ethi and -etht.

Thanx

adly3000

newbie

Activity: 55

Merit: 0

Quote from: LJLT on April 04, 2017, 12:44:28 AM

Quote from: adly3000 on April 04, 2017, 12:26:38 AM

Quote from: LJLT on April 03, 2017, 08:27:13 PM

Quote from: adly3000 on April 03, 2017, 06:00:03 PM

Thank you so much. Ireduced to 1266/2200 same production. noticed on 1266/2250 showing 28 Mh/s but 2250 is the max on the msi afterburner which I dont like as it is the maximum, so returned to 1266/2200. I dont like any mod it might destroy card.

Are you checking for memory erros with this setting?

how can I? do you mean invalid shares. it is zero.

Download

https://www.hwinfo.com/download.php

There is an indicator for memory errors, they dont reflect in the number of shares but reduce the life of the card.

Thank you so much.. Here is what I found:
GPU0 - errors: 9 - Asic:79.5
GPU1 - errors: 1,794,151 - Asic:85.6
GPU2 - errors: 5123 - Asic:73.6
GPU3 - errors: 752 - Asic:75.8
GPU4 - errors: 88,873,330 - Asic:75
GPU5 - errors: 7811 - Asic:71.1

I reduced GPU1 and GPU4 to 1180/2200 @-24 mv tried 1160/2200 on GPU4 but hangs

LJLT

newbie

Activity: 48

Merit: 0

Quote from: adly3000 on April 04, 2017, 12:26:38 AM

Quote from: LJLT on April 03, 2017, 08:27:13 PM

Quote from: adly3000 on April 03, 2017, 06:00:03 PM

Thank you so much. Ireduced to 1266/2200 same production. noticed on 1266/2250 showing 28 Mh/s but 2250 is the max on the msi afterburner which I dont like as it is the maximum, so returned to 1266/2200. I dont like any mod it might destroy card.

Are you checking for memory erros with this setting?

how can I? do you mean invalid shares. it is zero.

Download

https://www.hwinfo.com/download.php

There is an indicator for memory errors, they dont reflect in the number of shares but reduce the life of the card.

adly3000

newbie

Activity: 55

Merit: 0

Quote from: LJLT on April 03, 2017, 08:27:13 PM

Quote from: adly3000 on April 03, 2017, 06:00:03 PM

Thank you so much. Ireduced to 1266/2200 same production. noticed on 1266/2250 showing 28 Mh/s but 2250 is the max on the msi afterburner which I dont like as it is the maximum, so returned to 1266/2200. I dont like any mod it might destroy card.

Are you checking for memory erros with this setting?

how can I? do you mean invalid shares. it is zero.

Junkey

full member

Activity: 164

Merit: 100

Does the -mclock and -cclock options work in Linux for AMD RX400 cards?

cashen

sr. member

Activity: 434

Merit: 251

Quote from: Truthchanter on April 03, 2017, 06:39:18 PM

Quote from: cashen on April 03, 2017, 06:35:20 PM

Quote from: Truthchanter on April 03, 2017, 06:28:07 PM

This might not be completely related to your issue but i find some of my cards need at least 5% power limit when dual mining or they don't hash as high (maybe even 10-15% for a couple cards). Although other cards seem to dual mine fine at 0% power limit. Another thing that comes to mind is if you have the target temp or max temp set too low and the cards may be throttling/downclocking if the heat gets too high. Otherwise its probably just random pool fluctuations or the normal way mining decred works.. idk

My fans are only running ~30%-40% and I think my target temps are set at 73C. I'm not sure about my max, but i'm not coming close. In Claymore the Hash rate always says 3000+. Its almost like they are not matching. I am wondering if it has to do with my poor wifi connecction.

Poor wifi definitely sounds like a potential culprit. although if you're not fluctuating too much at your eth pool then it just might be the nature of decred pools/mining.. maybe someone else would have a more knowledgeable answer about this

No changes have been made to settings within the last 30 hours. Looks to be correcting slowly.

LJLT

newbie

Activity: 48

Merit: 0

Quote from: adly3000 on April 03, 2017, 06:00:03 PM

Quote from: matmator on April 03, 2017, 03:57:36 PM

Quote from: adly3000 on April 02, 2017, 10:05:30 PM

I been mining on 2 rigs for 2 months using:
windows 7 ulti.
h81 bro btc + Ram 8 GB + Intel Pentium Processor G3240
every rig have 5 cards of Sapphire nitro+ 480 8G 11260-01-20G set at 1342/2200 hashing 27.4-27.5
I saw people saying that it can be raises to 28-29 (without any mod or bios flash) only using overclock
I would like to say the rigs are stable and working good with 16.11.5 (pow lim 0 & mv -24)
The cards have samsung rams with asics (75-85) but all same hash 27.4-27.5
any help to raise would be appreciated even 0.5 for 10 cards it will be 5 Mh/s

My batch is:
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:4444 -ewal mywallet.Rig2 -epsw x -ethi 16 -esm 2 -mode 1 -tstop 80,75,75,75,75 -allcoins 1
pause

Hello

1342 it's too high!

1200/2200 or 1150/2150.

and -48mv

With bios mod you can have 30mh

Thank you so much. Ireduced to 1266/2200 same production. noticed on 1266/2250 showing 28 Mh/s but 2250 is the max on the msi afterburner which I dont like as it is the maximum, so returned to 1266/2200. I dont like any mod it might destroy card.

Are you checking for memory erros with this setting?

Truthchanter

sr. member

Activity: 689

Merit: 253

Quote from: cashen on April 03, 2017, 06:35:20 PM

Quote from: Truthchanter on April 03, 2017, 06:28:07 PM

This might not be completely related to your issue but i find some of my cards need at least 5% power limit when dual mining or they don't hash as high (maybe even 10-15% for a couple cards). Although other cards seem to dual mine fine at 0% power limit. Another thing that comes to mind is if you have the target temp or max temp set too low and the cards may be throttling/downclocking if the heat gets too high. Otherwise its probably just random pool fluctuations or the normal way mining decred works.. idk

My fans are only running ~30%-40% and I think my target temps are set at 73C. I'm not sure about my max, but i'm not coming close. In Claymore the Hash rate always says 3000+. Its almost like they are not matching. I am wondering if it has to do with my poor wifi connecction.

Poor wifi definitely sounds like a potential culprit. although if you're not fluctuating too much at your eth pool then it just might be the nature of decred pools/mining.. maybe someone else would have a more knowledgeable answer about this

cashen

sr. member

Activity: 434

Merit: 251

Quote from: Truthchanter on April 03, 2017, 06:28:07 PM

This might not be completely related to your issue but i find some of my cards need at least 5% power limit when dual mining or they don't hash as high (maybe even 10-15% for a couple cards). Although other cards seem to dual mine fine at 0% power limit. Another thing that comes to mind is if you have the target temp or max temp set too low and the cards may be throttling/downclocking if the heat gets too high. Otherwise its probably just random pool fluctuations or the normal way mining decred works.. idk

My fans are only running ~30%-40% and I think my target temps are set at 73C. I'm not sure about my max, but i'm not coming close. In Claymore the Hash rate always says 3000+. Its almost like they are not matching. I am wondering if it has to do with my poor wifi connecction.

Truthchanter

sr. member

Activity: 689

Merit: 253

Quote from: cashen on April 03, 2017, 06:09:10 PM

I'm duel mining (ETH+DCR) I've noticed some crazy performance. You can see it was solid all night but went wacky around 8AM.

Any Ideas?

This might not be completely related to your issue but i find some of my cards need at least 5% power limit when dual mining or they don't hash as high (maybe even 10-15% for a couple cards). Although other cards seem to dual mine fine at 0% power limit. Another thing that comes to mind is if you have the target temp or max temp set too low and the cards may be throttling/downclocking if the heat gets too high. Otherwise its probably just random pool fluctuations or the normal way mining decred works.. idk

cashen

sr. member

Activity: 434

Merit: 251

I'm duel mining (ETH+DCR) I've noticed some crazy performance. You can see it was solid all night but went wacky around 8AM.

Any Ideas?

adly3000

newbie

Activity: 55

Merit: 0

Quote from: matmator on April 03, 2017, 03:57:36 PM

Quote from: adly3000 on April 02, 2017, 10:05:30 PM

I been mining on 2 rigs for 2 months using:
windows 7 ulti.
h81 bro btc + Ram 8 GB + Intel Pentium Processor G3240
every rig have 5 cards of Sapphire nitro+ 480 8G 11260-01-20G set at 1342/2200 hashing 27.4-27.5
I saw people saying that it can be raises to 28-29 (without any mod or bios flash) only using overclock
I would like to say the rigs are stable and working good with 16.11.5 (pow lim 0 & mv -24)
The cards have samsung rams with asics (75-85) but all same hash 27.4-27.5
any help to raise would be appreciated even 0.5 for 10 cards it will be 5 Mh/s

My batch is:
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:4444 -ewal mywallet.Rig2 -epsw x -ethi 16 -esm 2 -mode 1 -tstop 80,75,75,75,75 -allcoins 1
pause

Hello

1342 it's too high!

1200/2200 or 1150/2150.

and -48mv

With bios mod you can have 30mh

Thank you so much. Ireduced to 1266/2200 same production. noticed on 1266/2250 showing 28 Mh/s but 2250 is the max on the msi afterburner which I dont like as it is the maximum, so returned to 1266/2200. I dont like any mod it might destroy card.

bardacuda

sr. member

Activity: 430

Merit: 254

The devfee always begins x minutes after you start mining (I think x=15...at least it did for the ZEC miner) and then every hour after that. Just don't stop or close the miner and it will only happen once per hour like it's supposed to.

imtav

member

Activity: 79

Merit: 10

DevFee mining happening too frequently!
I have logs to show it. DevFee running at 16:07, then DevFee mining finished and I closed miner. Restarted miner, and again I'm seeing DevFee mining at 16:26. WTF??

Topic: Claymore's Dual Ethereum AMD+NVIDIA GPU Miner v15.0 (Windows/Linux) - page 974. (Read 6590565 times)