[Mining OS] SimpleMining.net - Manage Your GPU farm the easy way! (30 days free) - page 320.

tflux99

member

Activity: 89

Merit: 10

Quote from: adore on October 09, 2017, 02:32:19 PM

Quote from: StyriaTrading on October 09, 2017, 02:28:23 PM

Hi,

I have a problem with undervolting - no setting is applied by my cards, it don't depends if it is 700 or 1000 mV, I have all the time the same wattage draw.

Heard that some other people also experience this problem. Can you give me a hint what can I do that my cards (Sapphire RX 570 Nitro + 4GB) are accepting the undervolting?

Thanks in advance.

have you ever tried to change voltage settings in other OS's or is this the first time you try this with SMOS? some cards just do not support this

all rx series cards should support it if your bios is ok in windows 10 and msi afterburner 4.4.0 beta 17 and latest drivers for me all cards are able to do voltage offset via msi afterburner max -96mv and i have 80 rx gpus from various brands alot of 470 and 480 not so many rx 570 or 580 they are so pricy hehe i had the luck to buy most of my gpus b4 the eth mining hype... so my god old rx 470 are the best crads hehe and even better is i bought most of them for just 160€ here in .de

and the power consumption on most of my 470 is much better than on 570 and they have the same speeds just the rx 570 takes much more energy i think its cause of a lot of my 470 have just one 6 pin connector PCIE power not all but all of my 570 have 8 pin connector so 470 with 6 pin like powercolor rx 470 is best way to go for me in energy usage hehe sry 4 bad english

tflux99

member

Activity: 89

Merit: 10

Quote from: StyriaTrading on October 09, 2017, 02:28:23 PM

Hi,

I have a problem with undervolting - no setting is applied by my cards, it don't depends if it is 700 or 1000 mV, I have all the time the same wattage draw.

Heard that some other people also experience this problem. Can you give me a hint what can I do that my cards (Sapphire RX 570 Nitro + 4GB) are accepting the undervolting?

Thanks in advance.

maybe the voltage controller is incompatible with this undervolting method smos does in mey test it works have without bios tweaks arióund 2200watts with only bios voltage mods mods around 2000 watts and now with a little tests of the undervolting in smos i am down to 1850-1860 watts full 13 gpu rig with different card mix 3 x rx470 10x rx 570 brands powercolor msi and asus and iam mining ethereum + decred dual mining with claymore 10

adore

newbie

Activity: 44

Merit: 0

Quote from: StyriaTrading on October 09, 2017, 02:28:23 PM

Hi,

I have a problem with undervolting - no setting is applied by my cards, it don't depends if it is 700 or 1000 mV, I have all the time the same wattage draw.

Heard that some other people also experience this problem. Can you give me a hint what can I do that my cards (Sapphire RX 570 Nitro + 4GB) are accepting the undervolting?

Thanks in advance.

have you ever tried to change voltage settings in other OS's or is this the first time you try this with SMOS? some cards just do not support this

StyriaTrading

newbie

Activity: 12

Merit: 2

Hi,

I have a problem with undervolting - no setting is applied by my cards, it don't depends if it is 700 or 1000 mV, I have all the time the same wattage draw.

Heard that some other people also experience this problem. Can you give me a hint what can I do that my cards (Sapphire RX 570 Nitro + 4GB) are accepting the undervolting?

Thanks in advance.

tytanick

legendary

Activity: 2660

Merit: 1096

Simplemining.net Admin

Update v1128
- changed start delay from 6seconds to 12seconds in nvidia rigs (temporay fix for slow usb and too slow initialisation). This shoudl fix some OC problems in some cases on Nvidia rigs
- changed verbose logging in xmr-stak-amd miner from 3 to 4 (more speed stats)

citronick

legendary

Activity: 1834

Merit: 1080

---- winter*juvia -----

Try core 100, mem 800, PL 99

Zotac P106-100, Claymore 10, 22-23MHs

Quote from: Fahim021 on October 09, 2017, 09:24:40 AM

It loads DAG file and then hangs. I have removed overlooking. Now let's see.
Also can anyone suggest me a better overclocking setting for p106? I used

Core clock 80
Memory clock 1600
Watt 80

Cards were not stable

This was rather better setting but the rig seem to restart after few hours

Core clock 80
Memory clock 1400
Watt 100

Quote from: MichaelMiner on October 08, 2017, 11:27:32 PM

Quote from: Fahim021 on October 08, 2017, 06:50:57 PM

HI, i am using 13 p106-100 msi card in 1 rig. Since last 2 days I am getting error "binary file (standard input) matches simple mining" very often and rig freezes. It doesn't restart itself. Kindly help !!

You need to provide more information, this has no value, does the OS load? Does the miner start then fail?
Tough to help with out information.
If you have ANY overclocks, take them all out then reboot.

chuck732

member

Activity: 141

Merit: 10

Hi,
May I ask how and where did you put this in SMOS?

/root/utils/rclocal.sh
change 31 line from
sleep 6
to
sleep 20

Quote from: FirstRigger on October 09, 2017, 10:36:42 AM

Quote from: twofreckles on October 09, 2017, 09:51:29 AM

Quote from: tytanick on October 09, 2017, 06:34:27 AM

Quote from: ?? on ??

Hi Guys,
I can confirm that there is still an issue with overclocking Nvidia cards. I'm also getting the the below error.
Invalid MIT-MAGIC-COOKIE-1 keyFailed to connect to Mir: Failed to connect to unable to init server: Could not connect: Connection refused
However the error only seems to happen when I reboot the rig and then it only affects one or two cards per rig.
A fix is to go to overclocking and click save without changing anything and when the miner restarts everything works perfectly - until the next reboot.

I probably know whats causing it and i just need to make some fix in my script.
Anyway if i am thinking right then this issue is only on slow pendrives.

The dummy and temporary fix could be:
/root/utils/rclocal.sh
change 31 line from
sleep 6
to
sleep 20

The Xorg is starting too slow on slow pendrives and thats probably the issue.
I have pretty fast ones so i dont have this issue and this explains why reloading fixes this problem.

I will make some checkups first before miner starts.

I can confirm it, overlocking is working again..

Oh my God, please no more overLocking, we have problem enough with rigs. Wink

FirstRigger

newbie

Activity: 25

Merit: 0

Quote from: twofreckles on October 09, 2017, 09:51:29 AM

Quote from: tytanick on October 09, 2017, 06:34:27 AM

Quote from: ?? on ??

Hi Guys,
I can confirm that there is still an issue with overclocking Nvidia cards. I'm also getting the the below error.
Invalid MIT-MAGIC-COOKIE-1 keyFailed to connect to Mir: Failed to connect to unable to init server: Could not connect: Connection refused
However the error only seems to happen when I reboot the rig and then it only affects one or two cards per rig.
A fix is to go to overclocking and click save without changing anything and when the miner restarts everything works perfectly - until the next reboot.

I probably know whats causing it and i just need to make some fix in my script.
Anyway if i am thinking right then this issue is only on slow pendrives.

The dummy and temporary fix could be:
/root/utils/rclocal.sh
change 31 line from
sleep 6
to
sleep 20

The Xorg is starting too slow on slow pendrives and thats probably the issue.
I have pretty fast ones so i dont have this issue and this explains why reloading fixes this problem.

I will make some checkups first before miner starts.

I can confirm it, overlocking is working again..

Oh my God, please no more overLocking, we have problem enough with rigs. Wink

twofreckles

sr. member

Activity: 366

Merit: 250

Quote from: tytanick on October 09, 2017, 06:34:27 AM

Quote from: ?? on ??

Hi Guys,
I can confirm that there is still an issue with overclocking Nvidia cards. I'm also getting the the below error.
Invalid MIT-MAGIC-COOKIE-1 keyFailed to connect to Mir: Failed to connect to unable to init server: Could not connect: Connection refused
However the error only seems to happen when I reboot the rig and then it only affects one or two cards per rig.
A fix is to go to overclocking and click save without changing anything and when the miner restarts everything works perfectly - until the next reboot.

I probably know whats causing it and i just need to make some fix in my script.
Anyway if i am thinking right then this issue is only on slow pendrives.

The dummy and temporary fix could be:
/root/utils/rclocal.sh
change 31 line from
sleep 6
to
sleep 20

The Xorg is starting too slow on slow pendrives and thats probably the issue.
I have pretty fast ones so i dont have this issue and this explains why reloading fixes this problem.

I will make some checkups first before miner starts.

I can confirm it, overlocking is working again..

Fahim021

newbie

Activity: 7

Merit: 0

It loads DAG file and then hangs. I have removed overlooking. Now let's see.
Also can anyone suggest me a better overclocking setting for p106? I used

Core clock 80
Memory clock 1600
Watt 80

Cards were not stable

This was rather better setting but the rig seem to restart after few hours

Core clock 80
Memory clock 1400
Watt 100

Quote from: MichaelMiner on October 08, 2017, 11:27:32 PM

Quote from: Fahim021 on October 08, 2017, 06:50:57 PM

HI, i am using 13 p106-100 msi card in 1 rig. Since last 2 days I am getting error "binary file (standard input) matches simple mining" very often and rig freezes. It doesn't restart itself. Kindly help !!

You need to provide more information, this has no value, does the OS load? Does the miner start then fail?
Tough to help with out information.
If you have ANY overclocks, take them all out then reboot.

xtoff

full member

Activity: 187

Merit: 100

Tru Reputation Network - Proof of Reputation

Quote from: tytanick on October 08, 2017, 04:47:09 PM

Quote from: xtoff on October 08, 2017, 04:20:08 AM

Hi tytanick,
Can you stop with the time out on the website.
It is impossible now to manage our rigs we are always disconnected. My rigs are alive 24/7 so I need to be connected to the website 24/7 so can you stop the automatic disconnection.

If there is always disconnection there is no more interest in SMOS, we need a gui running with this OS.

Can you make a correction to this issue, thanks.

If you mean the Loggin Out.
Then yes it was a problem and it should be fixed now.
Thx for report.

yes the loggin Out

Allways the same thing log out after 1 hour or 2 hours

no change at all. I must log in each time since you upgrade the server.

:-(

chuck732

member

Activity: 141

Merit: 10

I have tried different risers, all brand new ver 6c from different makers. I have moved the gpu's around to different pcie slots. The motherboard has the newest bios update. The motherboard is a MSI Z170A G45. All the setting you said to do on your website I did. The only thing left is the overclock. Because all the gpu cores are overclocked above 1900 and I cannot lower them. I tried several times and the power usage of these gpu's is averaging 177 watt's and I cannot lower them either.
The only thing left would have to be the USB thumb drive. If I get a faster one then I maybe able to lower the power usage and lower the core clocks.
What USB thumb drives are you talking about?

And thank you for your help!

Quote from: tytanick on October 09, 2017, 07:43:14 AM

Quote from: chuck732 on October 09, 2017, 07:04:10 AM

Tytanick,
So you saying that if I get a faster USB stick it will solve my problems with overclocks and strange GPU3 stop's that move to different pcie slots?

Only OC problems.
The GPU3 stopped could be becasue of:
- too much OC (most common)
- bad riser (most common)
- some incompatibility in linux or incompatible motherboard bios (sometimes this is becasue if this yes - worth to try after rulling out above problems)
- bad gpu (slight chance)

tytanick

legendary

Activity: 2660

Merit: 1096

Simplemining.net Admin

Quote from: chuck732 on October 09, 2017, 07:04:10 AM

Tytanick,
So you saying that if I get a faster USB stick it will solve my problems with overclocks and strange GPU3 stop's that move to different pcie slots?

Only OC problems.
The GPU3 stopped could be becasue of:
- too much OC (most common)
- bad riser (most common)
- some incompatibility in linux or incompatible motherboard bios (sometimes this is becasue if this yes - worth to try after rulling out above problems)
- bad gpu (slight chance)

chuck732

member

Activity: 141

Merit: 10

Tytanick,
So you saying that if I get a faster USB stick it will solve my problems with overclocks and strange GPU3 stop's that move to different pcie slots?

Quote from: tytanick on October 09, 2017, 06:34:27 AM

Quote from: ?? on ??

Hi Guys,
I can confirm that there is still an issue with overclocking Nvidia cards. I'm also getting the the below error.
Invalid MIT-MAGIC-COOKIE-1 keyFailed to connect to Mir: Failed to connect to unable to init server: Could not connect: Connection refused
However the error only seems to happen when I reboot the rig and then it only affects one or two cards per rig.
A fix is to go to overclocking and click save without changing anything and when the miner restarts everything works perfectly - until the next reboot.

I probably know whats causing it and i just need to make some fix in my script.
Anyway if i am thinking right then this issue is only on slow pendrives.

The dummy and temporary fix could be:
/root/utils/rclocal.sh
change 31 line from
sleep 6
to
sleep 20

The Xorg is starting too slow on slow pendrives and thats probably the issue.
I have pretty fast ones so i dont have this issue and this explains why reloading fixes this problem.

I will make some checkups first before miner starts.

tytanick

legendary

Activity: 2660

Merit: 1096

Simplemining.net Admin

Quote from: ?? on ??

Hi Guys,
I can confirm that there is still an issue with overclocking Nvidia cards. I'm also getting the the below error.
Invalid MIT-MAGIC-COOKIE-1 keyFailed to connect to Mir: Failed to connect to unable to init server: Could not connect: Connection refused
However the error only seems to happen when I reboot the rig and then it only affects one or two cards per rig.
A fix is to go to overclocking and click save without changing anything and when the miner restarts everything works perfectly - until the next reboot.

I probably know whats causing it and i just need to make some fix in my script.
Anyway if i am thinking right then this issue is only on slow pendrives.

The dummy and temporary fix could be:
/root/utils/rclocal.sh
change 31 line from
sleep 6
to
sleep 20

The Xorg is starting too slow on slow pendrives and thats probably the issue.
I have pretty fast ones so i dont have this issue and this explains why reloading fixes this problem.

I will make some checkups first before miner starts.

chuck732

member

Activity: 141

Merit: 10

I spent that day in a half trying to figure this out. I set the core overclock to 0 and the save reboot. Then I look at the dash and all the cores are overclocked to above 1900. I set the overclock to 25%, 50%, 75%, 100%, 125% and 150%. I even did -25% etc. and still 1900 overclock! Does nothing. Started about 6 days ago or so. And power usage for my 1070's was 115 watts with the power limit set to 70%. Now for that last 3 days my power usage for these cards is 177 watts. I even set the power limit to 115 watts. It does nothing. Now for the last 3 days I keep getting a gpu has stopped and it restarts it if it can. I reboot works for a short time! So I change risers does nothing. I move cards, it last for maybe an hour then it does it again. I leave that pcie slot empty move the card to a different one. Does it again gpu stop's so i switch cards but not the same one as last time. And it still does it. GPU3 is driving me frekin nuts.

Sorry but I pay for a service and I expect it to at lease work.

Quote from: ?? on ??

Hi Guys,

I can confirm that there is still an issue with overclocking Nvidia cards. I'm also getting the the below error.
Invalid MIT-MAGIC-COOKIE-1 keyFailed to connect to Mir: Failed to connect to unable to init server: Could not connect: Connection refused

However the error only seems to happen when I reboot the rig and then it only affects one or two cards per rig.

A fix is to go to overclocking and click save without changing anything and when the miner restarts everything works perfectly - until the next reboot.

chuck732

member

Activity: 141

Merit: 10

I have it running but it likes to reboot it self alot. so far to day 18 times. The overclocking is messed up doesn't do crap. Dam it just rebooted while I'm typing this.

Here this might help you but not me. It's a guide.

https://docs.google.com/document/d/1SvtkI_ZCHAxqbIsaWnuA81hRbI3Tp2Zf3iyEQlORLTU/edit

Quote from: itongy on October 09, 2017, 05:16:53 AM

Help me please... i cannot overclocking

Invalid MIT-MAGIC-COOKIE-1 keyFailed to connect to Mir: Failed to connect
to unable to init server: Could not connect: Connection refused

ERROR: The control display is undefined; please run 'ncidia-setting --help'

itongy

newbie

Activity: 2

Merit: 0

Help me please... i cannot overclocking

Invalid MIT-MAGIC-COOKIE-1 keyFailed to connect to Mir: Failed to connect
to unable to init server: Could not connect: Connection refused

ERROR: The control display is undefined; please run 'ncidia-setting --help'

https://i.imgur.com/BHhBnCL.jpg

FirstRigger

newbie

Activity: 25

Merit: 0

I would like to know too what about wattage?
I dont receive any reply. Nothing here and nothing direct from admin (from email).

I dont understand:
I mine dual (eth+lbc) with smOS, 11xRX580: 2150W from Wall WITH smOS undervolting/overclocking (1130@2175@850, stable)
I mine with the SAME RIG with other linux mining software with SAME undervolting/overclocking, I receive 1850W from Wall.

Same rig, everything same, only other linux program =300W diff.

Really nobody knows the answer? How can I/WE right setup the smOS undervolting?

chuck732

member

Activity: 141

Merit: 10

How do you set the clocks back to stock? I tried removing the overclock on the core and it does nothing. On the dash it shows the same clocks overclock and stock are the same. It's like the overclock feature in Simplemining is broken. Does nothing, but I did notice the the power limit won't let you drop it down. My cards at first 5 days ago where averaging 115watts now they are averaging 177 watts. I'm lost here. It's like the cards are running at 100% power. Can someone PLEASE help me???

Quote from: MichaelMiner on October 08, 2017, 11:19:33 PM

Chuck732, don't overclock anything and see if it runs for a few days before playing with it again.

Topic: [Mining OS] SimpleMining.net - Manage Your GPU farm the easy way! (30 days free) - page 320. (Read 835509 times)