OFFICIAL CGMINER mining software thread for linux/win/osx/mips/arm/r-pi 4.11.0 - page 622.

-ck

legendary

Activity: 4088

Merit: 1631

Ruu \o/

Quote from: Hawkix on March 08, 2012, 03:55:09 PM

I have not checked the source code, but it seems that CGMINER keeps getting work from all defined pools, regardless if it mines on them or not. I am on very slow line (GPRS), so this hogs it the line a lot. Can it be run in "no poll" mode, just being switched on failure of primary pool?

BTW, i already use --failover-only, but it has no effect on amount of work being downloaded.

This is a gross exaggeration. It asks for precisely one work item (about 100 bytes) from pools that are not in use with --failover-only when starting up to see if those pools are alive and then never communicates with them again unless your primary pool fails. If you are having network trouble then almost certainly the problem is the rate the requests are sent out rather than the total amount of requests because often multiple requests are sent at exactly the same time. Try --net-delay to cope with that.

DeathAndTaxes

donator

Activity: 1218

Merit: 1079

Gerald Davis

Quote from: coblee on March 09, 2012, 02:10:55 PM

When I run cgminer with 6 GPUs, 3-4 of them end up in the DEAD state after a while. Anyone know why that happens or how I can debug it? Quitting cgminer and restarting it fixes things.

Reduce your clocks. Dead = GPU crashed. cgminer miner tends to drive GPU harder at a given Mhz than other miners. Try backing your clocks off by 5Mhz or so.

coblee

donator

Activity: 1654

Merit: 1351

Creator of Litecoin. Cryptocurrency enthusiast.

When I run cgminer with 6 GPUs, 3-4 of them end up in the DEAD state after a while. Anyone know why that happens or how I can debug it? Quitting cgminer and restarting it fixes things.

The00Dustin

hero member

Activity: 807

Merit: 500

Quote from: localhost on March 09, 2012, 09:55:21 AM

Quote from: The00Dustin on March 09, 2012, 09:47:19 AM

EDIT: Also, with Cypress and phatk, I am under the impression that SDK 2.1 is better when the GPU memory is underclocked. You might want to dowload phatk120223Cypressv2w128l4.bin from http://ck.kolivas.org/apps/cgminer/bins/2.1/ and try out the SDK 2.1 binary that is actually compiled in 2.3.1-2.

Just tried it, doesn't work (error 42, then recompiles the slower new version over it)

Weird! I guess he did do some major backtracking by 2.3.1-2, and that kernel must have been compiled in 2.3.0-1. I do wonder why mine is working, though, pretty sure I didn't delete the kernel after I upgraded to 2.3.1-2 from 2.3.0-1. Oh well, you've got your system working well enough to make you happy, right?

localhost

sr. member

Activity: 392

Merit: 250

Quote from: sharky112065 on March 09, 2012, 11:09:03 AM

I get 442 Mh/s on my 5870's clocked at 970/170 on 2.5 (793.1). So 2.5 does just fine at lower mem clock settings (yes verified by GPUz).

I'm getting 396MH/s @890/300 currently - 333MH/s @750/300 on the other one (underclocked for silence), about 2% slower clock for clock. Tried setting the RAM @170, lost about 10% hashrate. I guess I'll have to play with the drivers a bit some time then (next time I upgrade cgminer I suppose Cheesy

). Not that urgent though, given the small speed difference.

sharky112065

sr. member

Activity: 383

Merit: 250

Quote from: localhost on March 09, 2012, 09:34:59 AM

Hm, alright, thanks for all this information. I grabbed the old phatk110817.cl and phatk110817Cypressbitalignv2w128long4.bin from my cgminer 2.0.6, renamed them to phatk120223.cl and phatk120223Cypressv2w128l4.bin and this did restore my hashrate.

Judging from their date, they were built either under Catalyst 11.9 or 11.10. Should I expect any additional benefit from messing with my drivers as you describe? The changelog doesn't seem to mention performance improvements, "just" tons of features, bugfixes, and tweaks for newest GPUs.

Well you need to get rid of that AMD-APP Run-time (2.6) and go to 2.5 or you will always have to be copying and renaming bin files every time you switch to a newer version of cgminer. Follow step 2.

I get 442 Mh/s on my 5870's clocked at 970/170 on 2.5 (793.1). So 2.5 does just fine at lower mem clock settings (yes verified by GPUz).

localhost

sr. member

Activity: 392

Merit: 250

Quote from: The00Dustin on March 09, 2012, 09:47:19 AM

I'm surprised that worked, actually.

Yeah, I wasn't really expecting it to work either, but I thought this was worth trying since I've got some stuff running that I can't really interrupt right now. Working and looking good, although I'm not using "advanced" features such as overclocking/fan control/etc - the only thing I configure from cgminer, apart from the proxy now, is the intensity.

Quote from: The00Dustin on March 09, 2012, 09:47:19 AM

EDIT2: The .cl file isn't used when you have a .bin file, you probably sohuldn't have overwritten it with the older one, but if you never change settings, it won't matter.

Kept a backup anyway

Quote from: The00Dustin on March 09, 2012, 09:47:19 AM

EDIT: Also, with Cypress and phatk, I am under the impression that SDK 2.1 is better when the GPU memory is underclocked. You might want to dowload phatk120223Cypressv2w128l4.bin from http://ck.kolivas.org/apps/cgminer/bins/2.1/ and try out the SDK 2.1 binary that is actually compiled in 2.3.1-2.

Just tried it, doesn't work (error 42, then recompiles the slower new version over it)

The00Dustin

hero member

Activity: 807

Merit: 500

Quote from: localhost on March 09, 2012, 09:34:59 AM

Hm, alright, thanks for all this information. I grabbed the old phatk110817.cl and phatk110817Cypressbitalignv2w128long4.bin from my cgminer 2.0.6, renamed them to phatk120223.cl and phatk120223Cypressv2w128l4.bin and this did restore my hashrate.

Judging from their date, they were built either under Catalyst 11.9 or 11.10. Should I expect any additional benefit from messing with my drivers as you describe? The changelog doesn't seem to mention performance improvements, "just" tons of features, bugfixes, and tweaks for newest GPUs.

I'm surprised that worked, actually. At some point on 2.3.0 or 2.3.1, someone tried that and it didn't work at all. I thought ck subsequentsy posted that the old binaries weren't compatible with 2.3 and that was why he did the minor vesion change from 2.2.x. I guess somewhere along the way to 2.3.1-2 he did some major reverting.

EDIT: Also, with Cypress and phatk, I am under the impression that SDK 2.1 is better when the GPU memory is underclocked. You might want to dowload phatk120223Cypressv2w128l4.bin from http://ck.kolivas.org/apps/cgminer/bins/2.1/ and try out the SDK 2.1 binary that is actually compiled in 2.3.1-2.

EDIT2: The .cl file isn't used when you have a .bin file, you probably sohuldn't have overwritten it with the older one, but if you never change settings, it won't matter.

localhost

sr. member

Activity: 392

Merit: 250

Hm, alright, thanks for all this information. I grabbed the old phatk110817.cl and phatk110817Cypressbitalignv2w128long4.bin from my cgminer 2.0.6, renamed them to phatk120223.cl and phatk120223Cypressv2w128l4.bin and this did restore my hashrate.

Judging from their date, they were built either under Catalyst 11.9 or 11.10. Should I expect any additional benefit from messing with my drivers as you describe? The changelog doesn't seem to mention performance improvements, "just" tons of features, bugfixes, and tweaks for newest GPUs.

sharky112065

sr. member

Activity: 383

Merit: 250

Quote from: localhost on March 09, 2012, 08:02:42 AM

Quote from: kano on March 09, 2012, 07:45:46 AM

I should say quite a lot more than that.
Post it all.
Also in the cgminer directory in a command prompt: dir/a *.bin

Woops, I ran it on version 2.0.6 instead of 2.3.1, that's why. There you go:

Code:

>cgminer -n
[2012-03-09 12:59:20] CL Platform 0 vendor: Advanced Micro Devices, Inc.
[2012-03-09 12:59:20] CL Platform 0 name: AMD Accelerated Parallel Processing
[2012-03-09 12:59:20] CL Platform 0 version: OpenCL 1.1 AMD-APP (898.1)
[2012-03-09 12:59:20] Platform 0 devices: 2
[2012-03-09 12:59:20] GPU 0 ATI Radeon HD 5800 Series hardware monitoring enable
d
[2012-03-09 12:59:20] GPU 1 ATI Radeon HD 5800 Series hardware monitoring enable
d
[2012-03-09 12:59:20] 2 GPU devices max detected

Code:

>dir/a *.bin
 Volume in drive E is xxxx
 Volume Serial Number is xxxxx

 Directory of E:\xxx\cgminer-2.3.1-2

2012-03-07  16:53           751,168 phatk120223Cypressv2w128l4.bin
               1 File(s)        751,168 bytes
               0 Dir(s)  198,869,528,576 bytes free

Edit: actually, on cgminer 2.0.6 I'm even ~730MH/s right now

I believe you may be on 2.6 AMD APP (898.1). I think you want 2.5 (793.1). The reason 2.0.6 is still probably working at the correct hash rate is because you still have the bins that were created prior to upgrading your driver (which sneaks the newer APP in). If you were to delete your .bin files in your 2.0.6 directory I'm betting that your hash rate will drop as well.

The following was posted in an earlier post...

1) Run Catalyst 12.1 installer -> Custom install -> Unselect All button -> Select GPU Driver only -> Next... Untill it's installed
2) Run Catalyst 11.11 installer -> Custom install -> Unselect All button -> Select SDK Runtime only -> Next... Untill it's installed (2.5.793.1)
3) Done!

That should give you a driver without the 100% CPU bug and the last version of SDK 2.5 runtime.

Note: I had to also install catalyst when I did step 1 to get the temp and fan monitoring in cgminer.

Don't forget to remove every thing first.

In the future when upgrading your driver, uncheck all check boxes and recheck the driver and catalyst. (custom install)

localhost

sr. member

Activity: 392

Merit: 250

Quote from: kano on March 09, 2012, 07:45:46 AM

I should say quite a lot more than that.
Post it all.
Also in the cgminer directory in a command prompt: dir/a *.bin

Woops, I ran it on version 2.0.6 instead of 2.3.1, that's why. There you go:

Code:

>cgminer -n
[2012-03-09 12:59:20] CL Platform 0 vendor: Advanced Micro Devices, Inc.
[2012-03-09 12:59:20] CL Platform 0 name: AMD Accelerated Parallel Processing
[2012-03-09 12:59:20] CL Platform 0 version: OpenCL 1.1 AMD-APP (898.1)
[2012-03-09 12:59:20] Platform 0 devices: 2
[2012-03-09 12:59:20] GPU 0 ATI Radeon HD 5800 Series hardware monitoring enable
d
[2012-03-09 12:59:20] GPU 1 ATI Radeon HD 5800 Series hardware monitoring enable
d
[2012-03-09 12:59:20] 2 GPU devices max detected

Code:

>dir/a *.bin
 Volume in drive E is xxxx
 Volume Serial Number is xxxxx

 Directory of E:\xxx\cgminer-2.3.1-2

2012-03-07  16:53           751,168 phatk120223Cypressv2w128l4.bin
               1 File(s)        751,168 bytes
               0 Dir(s)  198,869,528,576 bytes free

Edit: actually, on cgminer 2.0.6 I'm even ~730MH/s right now

kano

legendary

Activity: 4592

Merit: 1851

Linux since 1997 RedHat 4

Quote from: localhost on March 09, 2012, 07:43:53 AM

Quote from: Vbs on March 09, 2012, 05:48:48 AM

Have you updated drivers/sdk on that machine since you first started 2.0.6? If you did, you are probably using SDK 2.6 now...

I've kept my drivers updated, but I haven't upgraded since I switched from 2.0.6 to 2.3.1. To be sure (Catalyst 12.2 was indeed released not long before I switched from 2.0.6 to 2.3.1), I just relaunched 2.0.6 right now: I'm still getting around 710MH/s

@kano: was "cgminer -n" for me? (it just tells me I have 2 GPUs - that's correct I have 2 5870s)

I should say quite a lot more than that.
Post it all.
Also in the cgminer directory in a command prompt: dir/a *.bin

localhost

sr. member

Activity: 392

Merit: 250

Quote from: Vbs on March 09, 2012, 05:48:48 AM

Have you updated drivers/sdk on that machine since you first started 2.0.6? If you did, you are probably using SDK 2.6 now...

I've kept my drivers updated, but I haven't upgraded since I switched from 2.0.6 to 2.3.1. To be sure (Catalyst 12.2 was indeed released not long before I switched from 2.0.6 to 2.3.1), I just relaunched 2.0.6 right now: I'm still getting around 710MH/s

@kano: was "cgminer -n" for me? (it just tells me I have 2 GPUs - that's correct I have 2 5870s)

kano

legendary

Activity: 4592

Merit: 1851

Linux since 1997 RedHat 4

cgminer -n
should tell you.

Vbs

hero member

Activity: 504

Merit: 500

Quote from: localhost on March 09, 2012, 04:55:21 AM

I don't really upgrade this one often, because I tend to avoid fixing things that already work, but I recently upgraded from 2.0.6 to 2.3.1 because I wanted to use a SOCKS proxy. It turns out it's quite slower: from about 710-720 MH/s I'm now around 650 MH/s with the same very settings (same GPU clock, same intensity). This doesn't seem to be due to a change in hash speed measurement on the client because I get the same figures from my pool's speed estimations. Anyone else noticed such a speed drop at some point between v 2.0.6 and 2.3.1?

Have you updated drivers/sdk on that machine since you first started 2.0.6? If you did, you are probably using SDK 2.6 now...

localhost

sr. member

Activity: 392

Merit: 250

I don't really upgrade this one often, because I tend to avoid fixing things that already work, but I recently upgraded from 2.0.6 to 2.3.1 because I wanted to use a SOCKS proxy. It turns out it's quite slower: from about 710-720 MH/s I'm now around 650 MH/s with the same very settings (same GPU clock, same intensity). This doesn't seem to be due to a change in hash speed measurement on the client because I get the same figures from my pool's speed estimations. Anyone else noticed such a speed drop at some point between v 2.0.6 and 2.3.1?

kano

legendary

Activity: 4592

Merit: 1851

Linux since 1997 RedHat 4

Quote from: O_Shovah on March 09, 2012, 03:15:38 AM

...
Thank you fred0,

I tried but it seems to run / not run on a random basis.
One problem is also that i have to try getting it run in paralell to ther MPBMiner ( there are also a x6500 and an Icarus)
Maybe they argue about a module.
I will further dig into this this evening.

If there is an Icarus plugged into a serial port and cgminer does a Bitforce check on it - cgminer will hang.

Firstly be sure which is which.
On linux: ls -o /dev/serial/by-id/
Icarus says usb-Prolific_Technology_Inc.... blah blah
BFL says something with "Bitforce" in the name.

Of course if you have 2 or more of any of any device (e.g. 2xBFL or 2xIcarus) it will only show one of each.

Bottom line is you don't want cgminer doing an auto scan with an Icarus plugged in.

It does an auto scan if:
1) you don't specify any "--scan-serial /dev/ttyUSBn"
2) it doesn't find any BFL devices in the list of "--scan-serial /dev/ttyUSBn" you supply
3) if you specify "--scan-serial auto"

Also if you specify "--scan-serial /dev/ttyUSBn" that points to an Icarus, it will hang.

O_Shovah

sr. member

Activity: 410

Merit: 252

Watercooling the world of mining

Quote from: fred0 on March 08, 2012, 05:57:08 PM

Quote from: O_Shovah on March 08, 2012, 05:29:29 PM

Hello

I have tried to compile and run cgminer for my BFL box.
While the compialtion runs fine with diabled open CL i get the following error when i start cgminer.

'All devices disabled, cannot mine!'

The bitforce box is connected and running.

Anybody has an idea ?

Your configure command should end with ./configure --enable-bitforce

Also, when starting, use
   --scan-serial /dev/ttyUSB0 \
   --scan-serial /dev/ttyUSB1 \
or under windows
   -S com5 -S com6

if you have more than one device.

Thank you fred0,

I tried but it seems to run / not run on a random basis.
One problem is also that i have to try getting it run in paralell to ther MPBMiner ( there are also a x6500 and an Icarus)
Maybe they argue about a module.
I will further dig into this this evening.

Red Emerald

hero member

Activity: 742

Merit: 500

Quote from: kano on March 08, 2012, 07:57:23 PM

Quote from: coblee on March 08, 2012, 07:30:52 PM

any plans to support the x6500 fpgas?

Send me one and I'll do it

(Edit: after the issues with ZTEX not working the same as the others and lacking any documentation on the mining, I'll not guess I can do it without the device ever again)

kudos for trying.

kano

legendary

Activity: 4592

Merit: 1851

Linux since 1997 RedHat 4

Quote from: coblee on March 08, 2012, 07:30:52 PM

any plans to support the x6500 fpgas?

Send me one and I'll do it

(Edit: after the issues with ZTEX not working the same as the others and lacking any documentation on the mining, I'll not guess I can do it without the device ever again)

Topic: OFFICIAL CGMINER mining software thread for linux/win/osx/mips/arm/r-pi 4.11.0 - page 622. (Read 5805546 times)