Claymore's ZCash/BTG AMD GPU Miner v12.6 (Windows/Linux)

[marcos5872]

Make sure your AMD Gaming Evolve app didn't automatically update your drivers, I had to use DDU and reinstall 4 times until I figured out it does that EVEN IF you uncheck the box during installation. You have to manually open AMD Gaming Evolved, click on the wheel besides deriver update and uncheck auto updates.

That's not it. And this is a 100% clean and updated Windows 10 x64 install with 4GB ram, 20 GB virtual mem, and NOTHING but AMD 16.3 drivers and claymore installed. Blows my damn mind.

General behavior is:

1. A card (random) will hash at 0% while all others are fine
2. OpenCL call hangs.
3. Crash

Sometimes all cards will be hashing just dandy for a minute or two, then one drops, then OpenCL call fail on 1 card.

Hello, I has the same problem wit my R9 280x and R 390, always one GPU shows at 0% and OpenCL call hangs. Did you found a solution?

[1714633930]

1714633930

[1714633930]

1714633930

[1714633930]

1714633930

[1714633930]

1714633930

[1714633930]

1714633930

[IOTUSA]

Make sure your AMD Gaming Evolve app didn't automatically update your drivers, I had to use DDU and reinstall 4 times until I figured out it does that EVEN IF you uncheck the box during installation. You have to manually open AMD Gaming Evolved, click on the wheel besides deriver update and uncheck auto updates.

That's not it. And this is a 100% clean and updated Windows 10 x64 install with 4GB ram, 20 GB virtual mem, and NOTHING but AMD 16.3 drivers and claymore installed. Blows my damn mind.

General behavior is:

1. A card (random) will hash at 0% while all others are fine
2. OpenCL call hangs.
3. Crash

Sometimes all cards will be hashing just dandy for a minute or two, then one drops, then OpenCL call fail on 1 card.

Experience has shown me that the OpenCL hang for any of my rigs (I am small time with 16 rigs) is that it is one of three things:

1. Video drivers need to be re-installed.  I have been using 16.11.5 for all my rigs for several months now without issue, or at least with the least number of issues.  Install it with only one card connected to the motherboard, and then add the other cards and let them auto-populate/auto-install.  This is probably 50% of my issues.
2.  I run a mod'd bios on all my cards that I make that uses the #5 voltage strap across #6 and #7 straps and runs the mem speed at 1750 Mhz.  Sometimes this causes the voltage to be too low on the #7 strap and I have to raise that voltage up in 20 mV steps until I get it stable, and then I will flash the bios with that change, or you can use WattTool or whatever you have to make that change.  None of the cards are exactly the same due to variations in manufacturing even though they are all the same make and model, so getting the voltage right is key.  This is probably 40% of my issues.
3.  Bad riser cards or bad riser cables, or a bad connection to the motherboard.  This has only happened twice in 6 months, but it is a possible contributor on a less frequent case.

A few times I have had to do a complete Windows 10 clean install, but I have come to realize that most of the issues are really video drivers and undervolting the GPU core clock.  Hope this helps.

My cards are BIOS modded slightly for power draw. Maybe I'll start with dropping the mem/core speed some and see if that solves the issue. Thank you.

[IOTUSA]

Make sure your AMD Gaming Evolve app didn't automatically update your drivers, I had to use DDU and reinstall 4 times until I figured out it does that EVEN IF you uncheck the box during installation. You have to manually open AMD Gaming Evolved, click on the wheel besides deriver update and uncheck auto updates.

That's not it. And this is a 100% clean and updated Windows 10 x64 install with 4GB ram, 20 GB virtual mem, and NOTHING but AMD 16.3 drivers and claymore installed. Blows my damn mind.

General behavior is:

1. A card (random) will hash at 0% while all others are fine
2. OpenCL call hangs.
3. Crash

Sometimes all cards will be hashing just dandy for a minute or two, then one drops, then OpenCL call fail on 1 card.

Hello, I has the same problem wit my R9 280x and R 390, always one GPU shows at 0% and OpenCL call hangs. Did you found a solution?

Nothing solid yet. I am wildly speculating it might be power draw. What PSU do you use? Setup?

[br2459]

Anyone have a BIOS mod for a XFX r9 280x Black Edition with Elpida Memory?

[marcos5872]

Make sure your AMD Gaming Evolve app didn't automatically update your drivers, I had to use DDU and reinstall 4 times until I figured out it does that EVEN IF you uncheck the box during installation. You have to manually open AMD Gaming Evolved, click on the wheel besides deriver update and uncheck auto updates.

That's not it. And this is a 100% clean and updated Windows 10 x64 install with 4GB ram, 20 GB virtual mem, and NOTHING but AMD 16.3 drivers and claymore installed. Blows my damn mind.

General behavior is:

1. A card (random) will hash at 0% while all others are fine
2. OpenCL call hangs.
3. Crash

Sometimes all cards will be hashing just dandy for a minute or two, then one drops, then OpenCL call fail on 1 card.

Hello, I has the same problem wit my R9 280x and R 390, always one GPU shows at 0% and OpenCL call hangs. Did you found a solution?

Nothing solid yet. I am wildly speculating it might be power draw. What PSU do you use? Setup?

Actually I have only 2 R9 280x with PSU EVGA 1300w tow both on stock settings, One r9 390 move to another rig and works fine, I'm thinking the problem is the Mobo.

With Otiminer one R9 280x Drops hash but not crash.

[IOTUSA]

Make sure your AMD Gaming Evolve app didn't automatically update your drivers, I had to use DDU and reinstall 4 times until I figured out it does that EVEN IF you uncheck the box during installation. You have to manually open AMD Gaming Evolved, click on the wheel besides deriver update and uncheck auto updates.

That's not it. And this is a 100% clean and updated Windows 10 x64 install with 4GB ram, 20 GB virtual mem, and NOTHING but AMD 16.3 drivers and claymore installed. Blows my damn mind.

General behavior is:

1. A card (random) will hash at 0% while all others are fine
2. OpenCL call hangs.
3. Crash

Sometimes all cards will be hashing just dandy for a minute or two, then one drops, then OpenCL call fail on 1 card.

Hello, I has the same problem wit my R9 280x and R 390, always one GPU shows at 0% and OpenCL call hangs. Did you found a solution?

Nothing solid yet. I am wildly speculating it might be power draw. What PSU do you use? Setup?

Actually I have only 2 R9 280x with PSU EVGA 1300w tow both on stock settings, One r9 390 move to another rig and works fine, I'm thinking the problem is the Mobo.

With Otiminer one R9 280x Drops hash but not crash.

I might try a new MB as well. I can't for the damn life of me figure it out. I run 6 x 480 on a single 1300w but they are Bios modded (Eco mod) so that shouldn't be an issue.

I'll swap MB and report back.

[ocminer]

dwarwpool is so fast, lets make it great pool, it needs more workers. It does not need ssl/tlc connection btw...you dont need to worry about 2.5 dev fee

dwarfpool is a thief.

flypool is doing good job now a days

Maybe the variation in Dwarfpool is too high as it has fewer users.

Supernova has issues paying and are not replying to emails

Never had any issues paying and my inbox is empty

[rpg]

dwarwpool is so fast, lets make it great pool, it needs more workers. It does not need ssl/tlc connection btw...you dont need to worry about 2.5 dev fee

dwarfpool is a thief.

flypool is doing good job now a days

Maybe the variation in Dwarfpool is too high as it has fewer users.

Supernova has issues paying and are not replying to emails

Never had any issues paying and my inbox is empty

Sent 2 messages via the web site and have .0535 unpaid for 3 days

[ocminer]

dwarwpool is so fast, lets make it great pool, it needs more workers. It does not need ssl/tlc connection btw...you dont need to worry about 2.5 dev fee

dwarfpool is a thief.

flypool is doing good job now a days

Maybe the variation in Dwarfpool is too high as it has fewer users.

Supernova has issues paying and are not replying to emails

Never had any issues paying and my inbox is empty

Sent 2 messages via the web site and have .0535 unpaid for 3 days

Sorry I didn't receive any message but I just received your PM, it looks like you've used the same address for an registered account and also for an anonymous acount - this is not possible, the system doesn't allow an anonymous account to "grab" the address of an registered account.

I've fixed that and you should receive the payment within a few seconds.

[rpg]

dwarwpool is so fast, lets make it great pool, it needs more workers. It does not need ssl/tlc connection btw...you dont need to worry about 2.5 dev fee

dwarfpool is a thief.

flypool is doing good job now a days

Maybe the variation in Dwarfpool is too high as it has fewer users.

Supernova has issues paying and are not replying to emails

Never had any issues paying and my inbox is empty

Sent 2 messages via the web site and have .0535 unpaid for 3 days

Sorry I didn't receive any message but I just received your PM, it looks like you've used the same address for an registered account and also for an anonymous acount - this is not possible, the system doesn't allow an anonymous account to "grab" the address of an registered account.

I've fixed that and you should receive the payment within a few seconds.

Ty, and look into those web messages that may not be working. In this crypto world suspicion is a must. For the community ocminer fixed the issue

[MilkTea]

I have to remarks on linux version 12.2 (will test 12.3 quit soon)
1. when you display the fan speed, you should read first, set it then, wait half a second and read it again : sometimes a card does not want you to set the fan-speed. 
   eg : cat /sys/class/drm/ ..blahblahblah/pwm1 reads 50%  - even though the program seays "100%" because you set it, and then you read it (too short - it takes about 100 msec to "settle" :-) )
2.  when you have the WATCHDOG detecting an OPENCL HANG .... The program quits.  and immediately start again.  Bad bad bad.  You should sleep some time before restarting. SYSCALL HANG (impossible to recover from in Linux, bec  I handle it starting zcash64miner in an infinite while-do loop (while true; do; zec64miner;sleep 15; done.  (with r 1 option) and it prevents an indefinite sys-call trap (no reboot possible under linux.   Lift you *ss, press the on-off switch for 10 secs.   Reboot (and hope your filesystem is still there). 
What I see often is : your miner detects that the card has limited memory and therefore lessens the intensity to 1 (rather than 6).   When you wait 15 secs: no problem.

mvg
M.

[haxiboy]

I have a question regarding to monitoring hashrate. Is there a way to read out the current hashrate reported by the miner via 3rd party apps on the remote port?
Is there some documentation how it's possible?

I have an arduino project and would be cool reading out realtime stats of a miner instead of waiting for console output to text.

Thanks!

Peter

[beeboop]

Just an update about my rig and how I temporarily fixed my hang issue.

I have a rig with 6 rx470s and ever since version 12.0 i have had issues with one of my GPUs hanging. I would have to restart the entire system every time this happened which was quite often. The last stable version for me was version 11. So I compared the differences from the two versions and the only thing I noticed was the different mining algorithm. All 12 versions defaulted my 470s to the faster ASM algorithm which worked fine on all my GPUs except one. the one that kept hanging. so i changed the algorithm for the GPU that was hanging back to the original one and I haven't had any issues since on version 12.3. My config file looks like this.

ZecMiner64.exe -zpool ssl://eu1-zcash.flypool.org:3443 -zwal zcashaddress.YourWorkerName -zpsw x -asm 1,1,1,1,1,0

0=card 5 (the card that would always hang)

Hope this helps!

[forzendiablo]

Just an update about my rig and how I temporarily fixed my hang issue.

I have a rig with 6 rx470s and ever since version 12.0 i have had issues with one of my GPUs hanging. I would have to restart the entire system every time this happened which was quite often. The last stable version for me was version 11. So I compared the differences from the two versions and the only thing I noticed was the different mining algorithm. All 12 versions defaulted my 470s to the faster ASM algorithm which worked fine on all my GPUs except one. the one that kept hanging. so i changed the algorithm for the GPU that was hanging back to the original one and I haven't had any issues since on version 12.3. My config file looks like this.

ZecMiner64.exe -zpool ssl://eu1-zcash.flypool.org:3443 -zwal zcashaddress.YourWorkerName -zpsw x -asm 1,1,1,1,1,0

0=card 5 (the card that would always hang)

Hope this helps!

Good idea, thanks i also have a card that fucks up. I will try this now

[MilkTea]

Just an update about my rig and how I temporarily fixed my hang issue.

I have a rig with 6 rx470s and ever since version 12.0 i have had issues with one of my GPUs hanging. I would have to restart the entire system every time this happened which was quite often. The last stable version for me was version 11. So I compared the differences from the two versions and the only thing I noticed was the different mining algorithm. All 12 versions defaulted my 470s to the faster ASM algorithm which worked fine on all my GPUs except one. the one that kept hanging. so i changed the algorithm for the GPU that was hanging back to the original one and I haven't had any issues since on version 12.3. My config file looks like this.

ZecMiner64.exe -zpool ssl://eu1-zcash.flypool.org:3443 -zwal zcashaddress.YourWorkerName -zpsw x -asm 1,1,1,1,1,0

0=card 5 (the card that would always hang)

Hope this helps!

Good idea, thanks i also have a card that fucks up. I will try this now

Same here. It only happens in multi-rig systems. The card in a slot on a one-rig system works fine (no riser, I mean). And I do not try to avoid it any more, because I found a remedy.  Zec64miner restarts too quickly.  You have to wait between iterations  Set -r 1 option (quit on hang). and put Zec64Miner in an enless loop (while true; do; ... done) and WAIT at least 15 sec between  iterations. Timewait (windows) , sleep 15s (linux). 
Every 2-3 hours the hang occurs, But I never need to do a "hard reset" (push the on-off button and hope my filesystem is not gone)
 And my rig(s) continues to produce sweet ZEC  ;.. whilst sleeping 

[ivomm]

My Sapphire 280x tri-x hang often with 12.3 (with stock clock and voltage). With 12.0 they work well, with 12.1 and 12.2 one of them keeps hanging, and with 12.3 all of them hang  Tried lowering clocks, intensity and lowering/raising power control with no effect.

[bitcoinsay]

The stability is nice for 24 hours. But there seems to be no increase in speed.

[HunterFX]

That's not it. And this is a 100% clean and updated Windows 10 x64 install with 4GB ram, 20 GB virtual mem, and NOTHING but AMD 16.3 drivers and claymore installed. Blows my damn mind.

General behavior is:

1. A card (random) will hash at 0% while all others are fine
2. OpenCL call hangs.
3. Crash

Sometimes all cards will be hashing just dandy for a minute or two, then one drops, then OpenCL call fail on 1 card.

Parameter "-old 1" will help you

[thesaltonsea]

Hello,

how do you get 300 from 280x stock ? With anything more than -i 2 is freezing , and on 2 is doing 270-275sols. Can i use the new -old option and bigger -i without crash ?

[kinsik]

ZEC: GPU0 455.346 H/s, GPU1 481.857 H/s, GPU2 460.153 H/s
v12.3 -i 7
gpu0 is fury
gpu1 is fury x
gpu2 is nano
for v12.2 can't use -i 7 , for v.12.3 i can use -i 7 but can't -i 8

Do you use "asm" algo for Fury or "2" ?