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

[Masked_Immortal]

hmm..
I have problem with 4.1 version
I have 4 cards on one of my rigs

r9 280x
r9 280x
7970
r9 390x

i adjust core clock for each card in msi afterburner, but when i start miner 390x downclocks to around 950 MHz, this is not throttling because temp is blow 70C for all cards!

[Trimegistus]

hmm..
I have problem with 4.1 version
I have 4 cards on one of my rigs

r9 280x
r9 280x
7970
r9 390x

i adjust core clock for each card in msi afterburner, but when i start miner 390x downclocks to around 950 MHz, this is not throttling because temp is blow 70C for all cards!

In my rigs, using 270x and 7950, every time I use Afterburner to change some settings I always restart the system before starting the miner.

This way, I make sure the settings are permanent because I double check they are in effect before firing up the miner.

It works for me 

[malefactor]

Took first crack at using this (linux).

Works fine for a bit, then crashes one of the GPUs.  Interestingly enough, the watchdog doesn't.  :-)  YOU HAD ONE JOB, WATCHDOG!

...excerpt from debug log...
WATCHDOG: GPU 2 hangs in OpenCL call, exit
<hang>


can't fall back & try older version (no linux build for pre-4.1).  If you can fix this, I'd prefer to deploy this, even with a fee.  The problem here is that it's constantly crashing at variable rates.

[arielbit]

i have this rig with 2x r9 390 and 1x r9 390x

using gpu-z monitoring (sensors tab), i OC all the cards 1120 clock and 1600 ram ...

looking at sensors tab only one card is at 1120 clock the rest are ~ 930 clock (i'm not sure which card is 390x coz gpuz labels them all 390)

therefore only one GPU is OC'ed...

all ram are at 1600mhz

graphics card tab at gpu-z shows uniform 1120 clock for all cards...sensors do not

any tips?

my 280x rig is uniform, all clocks matched.

edit: these issue occurs only in dual mining mode, mining in ETH only (-mode 1) shows uniform clocks in sensors tab of GPU-Z.

claymore..any suggestions?

Sounds like down throttling, of course.  If cards are getting too hot or drawing too much power, the firmware downclocks them.

Why do you overclock RAM?  It doesn't seem to offer any hashrate increase from my testing.  In fact, on my R9 290 cards, I downclock VRAM to 1125mhz for increased power savings (with no hashrate decrease).

I also find downvolting usually results in more stable hashrates, as the GPU is less stressed and is more likely not to throttle.

tried cclock 1100 vram 1125

dual mining core clocks (gpu-z sensors tab) are 970, 995, 1100 ...all vram are 1500

gpu-z (gpu-z graphics card tab) core clock are all 1100 and vram are 1125..

in ETH only mode (gpu-z sensors tab and graphics card tab) all core clock 1100 and vram 1125 (sensors tab 1500)

i could say in dual mining mode, two out of three of my cards are under-performing.

my cards are ~77 temp, only msi r9 390 can reach up to 81 temp ..... i could say it is not down throttling due to heat, besides i run gpu-z right after i started the miner...i don't know if it is a power draw issue.

i haven't done downvolting yet..

sensors tab says..you cannot downlclock below 1500 or VRAM..

@restless

i OC thru cclock and mclock

[restless]

Check with Hwinfo, temperatures of VRMs and voltages (12V line)
With 80 on core, VRMs can easily go above 90-95.
Dual mining needs more power..., ether alone barely heats VRMs to 60

[Q_R_V]

sensors tab says..you cannot downlclock below 1500 or VRAM..

In that case you would have to flash gpu with modified bios and set memclocks in it (vbe7 and atiflash would be the easiest way).

[Claymore]

Within 24 hours I will release version with Ethereum solo mining support (and with some new options/features).

[Dum3]

For some reason, version 4 crashes my gfx card drivers on one of my rigs.  I've moved back to 3.3, and everything is working fine again.  Not sure what is going on...

Any solution to this,i encountered same problem with 4.1, driver crashing,windows 10 x64 , 16 gb virtual memory 8 gb ram, 4 x 390x ,catalyst 15.12,all cards are stable on 1.0 or 3.2 on 1100/1500 -100 mv for days! on 4.1 they crash after a few hours even default freq/voltage not working....

[Vaslime]

For some reason, version 4 crashes my gfx card drivers on one of my rigs.  I've moved back to 3.3, and everything is working fine again.  Not sure what is going on...

Any solution to this,i encountered same problem with 4.1, driver crashing,windows 10 x64 , 16 gb virtual memory 8 gb ram, 4 x 390x ,catalyst 15.12,all cards are stable on 1.0 or 3.2 on 1100/1500 -100 mv for days! on 4.1 they crash after a few hours even default freq/voltage not working....

I had similar problem. I reduce the clock frequency and also reduce the intensity. It is working well now.

[arielbit]

Check with Hwinfo, temperatures of VRMs and voltages (12V line)
With 80 on core, VRMs can easily go above 90-95.
Dual mining needs more power..., ether alone barely heats VRMs to 60

i OC clock to 1160 so if there is down throttling it would become more obvious due to heat..but i think VRM temp is no issue here..

hwinfo

VRM temps and voltage (maximun temp and volt only)..gpu 1 is 390x the rest are 390

eth mode only

gpu 0

81c 83c 11.531v

gpu 1

90c 75c 11.563v

gpu 2

61c 77c 11.781v

dual mode..

gpu 0 - down to 1000+ mhz

90c 81c 11.531v

gpu 1 - down to 1000+ mhz

97c 76c 11.531v

gpu 2 still at 1160 mhz

87c 61c 11.781v

sensors tab says..you cannot downlclock below 1500 or VRAM..

In that case you would have to flash gpu with modified bios and set memclocks in it (vbe7 and atiflash would be the easiest way).

maybe i'll tinker with the 280x's but not with the 390's because it is still under warranty.

[citronick]

Within 24 hours I will release version with Ethereum solo mining support (and with some new options/features).

fantastic - looking forward to get my rigs to test

[adamvp]

what should I expect with my r9 280x (how much Decred hashrate without big ethhashrt drop.. let's say I'll get not less than 90% of Ethhashrte?) and if how much dual mine mode increase power consumption?
And what with: r7 370
and r9 380x
i will aprecciate your help

[phzi]

Claymore - do you think further optimizations for fiji chips are possible? Memory controller usage seems to be sitting pretty low, and I get the feeling after playing with intensities that there's more room left for optimization.

It seems a higher ethi allows for increased decred hashrate without ethash loss. I am now pushing ~30MH/s ethash and 1650MH/s decred with ethi 20 and dcri 110.  With ethi 16, dcri can't go that high without losing a chunk of ethash.  Compare that to 30MH and under 1000MH/s decred max with lower ethi.

[phzi]

sensors tab says..you cannot downlclock below 1500 or VRAM..

In that case you would have to flash gpu with modified bios and set memclocks in it (vbe7 and atiflash would be the easiest way).

Or you can use the 15.7.1 drivers, where downclocking still works.  Downclocking is broken (intentionally by amd?) in later drivers.  From my testing, there didn't seem to be an advantage between 15.7.1 or 15.12 in terms of performance, but 15.7.1 lets me downclock ram and save on power for free.

[ExcelThroughKnowledge]

Claymore - do you think further optimizations for fiji chips are possible? Memory controller usage seems to be sitting pretty low, and I get the feeling after playing with intensities that there's more room left for optimization.

It seems a higher ethi allows for increased decred hashrate without ethash loss. I am now pushing ~30MH/s ethash and 1650MH/s decred with ethi 20 and dcri 110.  With ethi 16, dcri can't go that high without losing a chunk of ethash.  Compare that to 30MH and under 1000MH/s decred max with lower ethi.

Wasn't ethi 16 the max? For my case the ethi isn't doing anything though, setting it to 8 or 16 doesn't affect a single thing.

[ExcelThroughKnowledge]

@Claymore

Could you make the ethi intensity change easily adjustable like dcri? So that I could test the speed out on different ethi without closing the miner. Currently ethi 8 and ethi 16 has no difference on my rate, which I don't know why.. You have any idea about that?

[arielbit]

sensors tab says..you cannot downlclock below 1500 or VRAM..

In that case you would have to flash gpu with modified bios and set memclocks in it (vbe7 and atiflash would be the easiest way).

Or you can use the 15.7.1 drivers, where downclocking still works.  Downclocking is broken (intentionally by amd?) in later drivers.  From my testing, there didn't seem to be an advantage between 15.7.1 or 15.12 in terms of performance, but 15.7.1 lets me downclock ram and save on power for free.

what do you use for downclocking? 15.7.1 is my driver here.

you may see your downclocked VRAM in info but "sensors" do not show downclocked speed.

[phzi]

Claymore - do you think further optimizations for fiji chips are possible? Memory controller usage seems to be sitting pretty low, and I get the feeling after playing with intensities that there's more room left for optimization.

It seems a higher ethi allows for increased decred hashrate without ethash loss. I am now pushing ~30MH/s ethash and 1650MH/s decred with ethi 20 and dcri 110.  With ethi 16, dcri can't go that high without losing a chunk of ethash.  Compare that to 30MH and under 1000MH/s decred max with lower ethi.

Wasn't ethi 16 the max? For my case the ethi isn't doing anything though, setting it to 8 or 16 doesn't affect a single thing.

I thought claymore said 32 was max, but values higher then 16 cause very long kernel runtimes.

[malandante]

Nice work. With v3.0 i have 100% eth hashrate + 33% dcr hashrate per card and near 0%/2% rejects respectively.
Are there any plans to add vcash (Blake algo) mining in dual mode?

Yes I'd like vcash support too

[mmitech]

I need a proxy for group management, so in other words I want to point a bunch of rigs using this miner to a proxy so I don't have to mess with each rig individually but I would be able to just edit/change the proxy config.

The problem with dwarfpool stratum-proxy is that it works only one way Stratum=>getwork, and I need a proxy that support Stratum<=>Stratum, any Ideas ?