Claymore's CryptoNote AMD GPU Miner v11.3

[Kompik]

Why is the 1070 so bad in CryptoNote? Is there any change of improvement?

[1714470142]

1714470142

[1714470142]

1714470142

[ManeBjorn]

Good to know.
Thanks

That is really good.
What pool?

700 h/s on rx 470
729 h/s on rx 480

Total watt of whole system 340

Just downloaded the miner was only for benchmark so its the stock server and wallet whats standard in the miner

[kirilvvbg]

Guys,

My problem is the following: on some of my rigs with RX470/480 cards' fans are going crazy (100%) on random pick.
It doesn't matter if I use stock setting or I did some overclocking. It even doesn't matter if the cards are with original bios or been flashed.
I've tried with 9.6 and 9.5 - same sh*t.
The only difference is that some rigs manage to restart, while others just shut down. And after they need "cooling time" before they could start again (manually of course).


While on ZEC - no such problem. I read in Wolfo's miner threat about pretty much the same problem.
Any suggestions what it might be?

[mo35]

Guys,

My problem is the following: on some of my rigs with RX470/480 cards' fans are going crazy (100%) on random pick.
It doesn't matter if I use stock setting or I did some overclocking. It even doesn't matter if the cards are with original bios or been flashed.
I've tried with 9.6 and 9.5 - same sh*t.
The only difference is that some rigs manage to restart, while others just shut down. And after they need "cooling time" before they could start again (manually of course).


While on ZEC - no such problem. I read in Wolfo's miner threat about pretty much the same problem.
Any suggestions what it might be?

i struggling with same issue , both on win7 & win10 , and same was happening using wolf0s XMR miner 0.4 & Claymores 9.6. & latest Sgminer-GM. No issues with equihash or dagger hashing. For me its seems related with some AMD driver crap, because i got rid of WHQL-Crimson-16.9.2 with DDU and now testing Non-WHQL-Crimson-16.10.3. So far not seen rig reboots , or fans spinning on 100% randomly ... been stable for 12h stright

[nraltcoin]

kirilvvbg and mo35,

I've been troubleshooting the same issue with a Sapphire Nitro+ RX 480 8gb card. I also use a GTX 970 and a R9 390, but only the 480 is affected. What brand are your cards? Wondering if this is a sapphire issue, or a general rx 400 series issue...

Try googling rx 480 fan issue. It seems like this is maybe a driver issue since non-miners have had this issue as well? Please be sure to report back if you find a solution! I'm banging my head against the wall over here. This doesn't happen with claymore's dual ethereum miner, but I'd love to get into mining XMR.

Are you using any tuning software to control your fans? I've tried several, all with the same issue. The ramp-up and loss of control of the 480 fans is random, though sometimes they stop completely instead of jumping to 100%, and then the pc eventually shuts itself down when temps are exceeded when I am out of the room.

Everyone else, is this an rx 480 issue, or a cryptonote algorithm miner issue with 400 series cards?

I'm on windows 10, crimson 16.12.2, claymore cryptonote 9.6. No issue at all with claymore's 7.4 dual eth miner.

[kirilvvbg]

Yep, all cards are Sapphire - 470/4, 470/8, 480/4...
I am using MSI Afterburner 4.3.0 and the driver is 16.9.2 on win 8.1

Now I will switch to 16.12.1 and will see if this would help...

[mo35]

Yep, all cards are Sapphire - 470/4, 470/8, 480/4...
I am using MSI Afterburner 4.3.0 and the driver is 16.9.2 on win 8.1

Now I will switch to 16.12.1 and will see if this would help...

yeah mine is sapphire 480 nitro+ 8gb too , with latest afterburner

[nraltcoin]

Trying your suggestion now of using non-whql 16.10.3 drivers. Still no fan issues with this driver on your end, mo35?

If this doesn't work out, I might swap this card out for an MSI, since I still have a few days left in the return window. Any reports of this same issue for a 480 from another vendor?

[mo35]

Trying your suggestion now of using non-whql 16.10.3 drivers. Still no fan issues with this driver on your end, mo35?

If this doesn't work out, I might swap this card out for an MSI, since I still have a few days left in the return window. Any reports of this same issue for a 480 from another vendor?

so far runs ok on win10 with 16.10.3 , testing Claymore , not tried Wolf0s or Sgminer yet
p.s. it had 1 short 100% spike , but fan returned to normal state not crashed to 0% speed like with old driver.

[kirilvvbg]

It should be the drivers. I just noticed that one of my rigs has no problems and it is with 16.11.4 drivers instead of 16.9.2
So now I am switching the drivers of the problematic rigs...

[nraltcoin]

I don't know about you, but I've had trouble with windows 10 auto-updating my drivers for my nvidia card, which pulls them out of compatibility (gtx on windows 10 needs drivers that are over a year old to function properly). Could eventually end up being the same issue with using older amd drivers.

If you run into this problem, try using group policy editor (windows 10 pro only). There are guides online to show how to enable a policy that prevents windows from installing a new driver for any hardware that matches the hardware IDs that you specify.

http://www.howtogeek.com/263851/how-to-prevent-windows-from-automatically-updating-specific-drivers/

[mo35]

I don't know about you, but I've had trouble with windows 10 auto-updating my drivers for my nvidia card, which pulls them out of compatibility (gtx on windows 10 needs drivers that are over a year old to function properly). Could eventually end up being the same issue with using older amd drivers.

If you run into this problem, try using group policy editor (windows 10 pro only). There are guides online to show how to enable a policy that prevents windows from installing a new driver for any hardware that matches the hardware IDs that you specify.

http://www.howtogeek.com/263851/how-to-prevent-windows-from-automatically-updating-specific-drivers/

not observed such necessity with amd cards so far on win10 v1607 , stays nicely on drivers im installing , driver update is disabled in CP tho.

[chup]

I could only add that it happens also mining with Claymore ZEC miner, Win 7 x64, Crimson 16.9.1. But pattern is here: Sapphire reference RX470 4GB. I was blaming till now USB riser, ordered and received new riser, but was lazy enough not to troubleshoot (exchanging cable first, than riser) because it happens only once per week or two. Card is mining, only fan is going into 100% madness.

[doktor83]

i can haz 750-760khs on nitro 470/8gb

[bhanu545]

SGMiner-GM, AMDGPU-PRO, Sapphire Nitro+ 470 4GB (Hynix), custom memory timings.

You also might want to mention the chance of him achieving those figures himself is close to zero

That's true.

then whats the trick behind this lol? to achieve your hashrates.

Might be because you are using linux?

I told you already - custom memory timings. Not copied.

I think it must be 1500 memory strap copied in other higher timings right? is that all the changes you made?

[Hotmetal]

SGMiner-GM, AMDGPU-PRO, Sapphire Nitro+ 470 4GB (Hynix), custom memory timings.

You also might want to mention the chance of him achieving those figures himself is close to zero

That's true.

then whats the trick behind this lol? to achieve your hashrates.

Might be because you are using linux?

I told you already - custom memory timings. Not copied.

I think it must be 1500 memory strap copied in other higher timings right? is that all the changes you made?

No. It's custom!

[bhanu545]

SGMiner-GM, AMDGPU-PRO, Sapphire Nitro+ 470 4GB (Hynix), custom memory timings.

You also might want to mention the chance of him achieving those figures himself is close to zero

That's true.

then whats the trick behind this lol? to achieve your hashrates.

Might be because you are using linux?

I told you already - custom memory timings. Not copied.

I think it must be 1500 memory strap copied in other higher timings right? is that all the changes you made?

No. It's custom!

I still do not get what does custom mean here. Sry to ask but is it gonna be a secret? Or can anyone do this?

[doktor83]

SGMiner-GM, AMDGPU-PRO, Sapphire Nitro+ 470 4GB (Hynix), custom memory timings.

You also might want to mention the chance of him achieving those figures himself is close to zero

That's true.

then whats the trick behind this lol? to achieve your hashrates.

Might be because you are using linux?

I told you already - custom memory timings. Not copied.

I think it must be 1500 memory strap copied in other higher timings right? is that all the changes you made?

No. It's custom!

I still do not get what does custom mean here. Sry to ask but is it gonna be a secret? Or can anyone do this?

well probably if wanted to share his knowledge he would already do it.
So you got your answer, he did not only copy lower strap times to higher freq, he modified them.

[Ofardo]

Hi Claymore, very good job w/ all yours miners.. 

Guys Please help with RX 480 o/c in Claymore CryptoNote GPU Miner v9.6 Beta - POOL in Windows 10

I get this error:
12/27/16-11:46:55 - SHARE FOUND (target 10000) - (GPU 0 of 1)
12/27/16-11:46:58 - New job received from xmr-usa.dwarfpool.com:9050
Speed: 792 h/s, TotalHashes: 174K, DevHashes: 3K Mining time: 00:03
12/27/16-11:47:08 - SHARE FOUND (target 10000) - (GPU 0 of 1)
Error in server response
: {"id":1,"jsonrpc":"2.0","error":{"code":-1,"message":"Lowdifficultyshare"}}
Stratum - connecting to 'xmr-usa.dwarfpool.com' <198.50.156.114> port 9050
Stratum - Connected
Pool Diff 10000
12/27/16-11:47:23 - SHARE FOUND (target 10000) - (GPU 0 of 1)
12/27/16-11:47:23 - SHARE FOUND (target 10000) - (GPU 0 of 1)
Error in server response
: {"id":1,"jsonrpc":"2.0","error":{"code":-1,"message":"Lowdifficultyshare"}}

many times..

I tried h 1024 and other, and tried -a 1, -a 4 and nothing
I tried other pool port in dwarf like 8050, 9100, 8800, and nothing
I tried moneropool wich have diff 25000.1 and nothing

ALL specs: Claymore CryptoNote GPU Miner  v9.6 Beta           

Cards available: 1
GPU #0: name: Ellesmere, 8192 MB available, 36 units
Total cards: 1

Catalyst 15.x is REQUIRED for best performance and compatibility
At least 16 GB of Virtual Memory is required for multi-GPU systems ---> IS ON
Make sure you defined GPU_MAX_ALLOC_PERCENT 100 ---> IS ON
Be careful with overclocking, use default clocks for first tests
Initializing...

GPU #0 recognized as RX480
GPU #0: HashCnt = 1024
POOL version
GPU #0: -a 4 mode selected
2 pools specified.
Remote monitoring is enabled on port 3333
Press "m" key for tune mode.
Stratum - connecting to 'xmr-usa.dwarfpool.com' <192.95.39.128> port 9050
Stratum - connecting to 'xmr-usa.dwarfpool.com' <192.95.39.128> port 9050
Stratum - Connected
Stratum - Connected
Pool Diff 10000
DevFee: Pool Diff 10000

Thanks for all

Oswaldo

[HerrDirektor]

Try change port to 8050 or 8100