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

[cracker666]

im more than happy been 25 hrs now running very stable geting 228h/s with my r9 380 set at gpu 1075 mem 1620 with i 7 just want to say thank you claymore

[1714652606]

1714652606

[1714652606]

1714652606

[charles2k]

I am not satisfied with 12.3 version. On my R9 280X 3GB cards I had to decrease intensity to 2 and even with i 2 sometime it freezes (the same with 270X cards).
With Tonga R9 380 cards it works ok.
Also I have problem with all 12.x version on Fury cards (even with i 2 - hashrate 0). Latest stable is 11.1

[majorlee]

To all: you can see low speed at highest possible "-i" value, if so, reduce it. For example, you specify "-i 8" for 280X 3GB card, miner reduces it to "-i 7" automatically (shows "not enough GPU memory" message), but it may work much slower than "-i 6".
So if you see that new version is slower than old version - try to reduce "-i" value a bit.
In some cases you can increase "-i" value, for example, in my tests Nano cards work best with v12.2 using "-i 7", but for v12.3 "-i 8" is faster.

@Claymore Can you make something to improve 380 series? Currently I have 210H. Too low... Thank you in advance

Tonga is really weak for ZEC.

why my nano can't use -i 8?

AMD Cards available: 3
GPU #0: Fiji, 4096 MB available, 56 compute units
GPU #1: Fiji, 4096 MB available, 64 compute units
GPU #1 recognized as Fury X/Nano
GPU #2: Fiji, 4096 MB available, 64 compute units
GPU #2 recognized as Fury X/Nano
POOL version
GPU #0 is going to use too high intensity (, not enough GPU memory, intensity value reduced (7)
GPU #0 algorithm ASM, intensity 7
GPU #1 is going to use too high intensity (, not enough GPU memory, intensity value reduced (7)
GPU #1 algorithm ASM, intensity 7
GPU #2 is going to use too high intensity (, not enough GPU memory, intensity value reduced (7)
GPU #2 algorithm ASM, intensity 7

Make sure you set these env variables:

GPU_MAX_HEAP_SIZE 100
GPU_MAX_ALLOC_PERCENT 100
GPU_SINGLE_ALLOC_PERCENT 100

gunna sound noobish, but where do you set these variables? i never worked it out lol but never seemed to need too

[Iza]

To all: you can see low speed at highest possible "-i" value, if so, reduce it. For example, you specify "-i 8" for 280X 3GB card, miner reduces it to "-i 7" automatically (shows "not enough GPU memory" message), but it may work much slower than "-i 6".
So if you see that new version is slower than old version - try to reduce "-i" value a bit.
In some cases you can increase "-i" value, for example, in my tests Nano cards work best with v12.2 using "-i 7", but for v12.3 "-i 8" is faster.

@Claymore Can you make something to improve 380 series? Currently I have 210H. Too low... Thank you in advance

Tonga is really weak for ZEC.

why my nano can't use -i 8?

AMD Cards available: 3
GPU #0: Fiji, 4096 MB available, 56 compute units
GPU #1: Fiji, 4096 MB available, 64 compute units
GPU #1 recognized as Fury X/Nano
GPU #2: Fiji, 4096 MB available, 64 compute units
GPU #2 recognized as Fury X/Nano
POOL version
GPU #0 is going to use too high intensity (, not enough GPU memory, intensity value reduced (7)
GPU #0 algorithm ASM, intensity 7
GPU #1 is going to use too high intensity (, not enough GPU memory, intensity value reduced (7)
GPU #1 algorithm ASM, intensity 7
GPU #2 is going to use too high intensity (, not enough GPU memory, intensity value reduced (7)
GPU #2 algorithm ASM, intensity 7

Make sure you set these env variables:

GPU_MAX_HEAP_SIZE 100
GPU_MAX_ALLOC_PERCENT 100
GPU_SINGLE_ALLOC_PERCENT 100

gunna sound noobish, but where do you set these variables? i never worked it out lol but never seemed to need too

In your start.bat file is easiest:
SETX GPU_MAX_HEAP_SIZE 100
SETX GPU_MAX_ALLOC_PERCENT 100
SETX GPU_SINGLE_ALLOC_PERCENT 100

This reply makes me curious, should the missing variables:

GPU_FORCE_64BIT_PTR 1
GPU_USE_SYNC_OBJECTS 1

that are recommended in the README!!! file still be set? I keep them just because it says so, but I have newer "higher end" cards.

[DR_TAMFR_DHP]

my R9 380x cards are averaging about 192 H/s.  I haven't set those environment variables yet either.

If the environment variables are recommended in README, why aren't they included by default?

[densets]

my 280x are doing 143 sols , on 12.1 they do 288 sols. not sure whats wrong.

[Tomintx]

Issues with Version 12.3:
- Some rigs will not run all GPUs since v12.0.  With 12.0 and later, some GPUs report 0.00H/s rate.  One of rigs has 7990, only uses one core.  Have not tried 12.3 version on this unit, it runs 12.0 fine. 
- Option -di detect will run very long time on a couple rigs.  Would love to know a procedure to do this manually by switching GPUs on and off.  How to map default GPU order to -di GPU order?  This is an issue because miner will reduce intensity on the wrong card, leaving the hot card to keep getting hotter and eventually stop or crash the rig.  Does the ordering need to be recalculated for different versions?
Linux rigs seem more stable than Wndows rigs.
Thanks for a great miner.

1. Set more virtual memory, use less "-i" value.
2. Use "-di detect" only once to get indexes, then specify them in "-di" option. Use 0...9 keys to enable-disable GPUs if you want to find physical GPU from its index.
GPU indexes are related to hardware/drivers only, the order is the same for all versions of all my miners because they just use OpenCL order.

Thanks - It's just the one rig that will get stuck in -di detect for more than an hour.  I've got indexes set on all of the rigs as provided by -di detect.  It's just the one that won't finish it.  I've been able to mostly guess at doing my own detection manually by switching GPUs off and on with 0..9 command and watching which one cools down/stops running and trying to create my own list.  If you have advice for doing that - it would be great. 

[silentrunner]

my R9 380x cards are averaging about 192 H/s.  I haven't set those environment variables yet either.

If the environment variables are recommended in README, why aren't they included by default?

Seems like you should be getting more than that.  I get 174 H/s from my R9 270 cards using 12.3.

11:58:54:780   3b9c   ZEC: 03/09/17-11:58:54 - New job from zec-us.suprnova.cc:2142
11:58:54:780   3b9c   target: 0x001085cd (diff: 3966H)
11:58:54:780   3b9c   ZEC - Total Speed: 350.740 H/s, Total Shares: 7196, Rejected: 8, Time: 16:58
11:58:54:780   3b9c   ZEC: GPU0 174.680 H/s, GPU1 176.060 H/s

[DR_TAMFR_DHP]

Maybe, I dunno. I saw some others saying about the same amount. 190ish.  These cards are only 4GB.  I need some more power.  Still trying on what additional cards to get - i still have 4 free slots.

[silentrunner]

If that is what others are getting for the 380x then it looks like you are ok.

The 270 cards are only 2GB.

You might try playing with the intensity.

[qberty]

For reference, my AMD 290x's top out at 373 H/s per gpu. 12.3, asm 1, intensity needs to be around 6. Anything else and it diminishes significantly.

http://image.prntscr.com/image/341d9b53d4614ab6af042b136a599d89.png

[DR_TAMFR_DHP]

If that is what others are getting for the 380x then it looks like you are ok.

The 270 cards are only 2GB.

You might try playing with the intensity.

yeah, i did that.  Tried -i 6, 7, 8.  Leaving it on auto and all 3 of my cards being tuned to -i 6 seems to be the best.  I didn't try any lower intensities though.

[Acguy]

Thanks for always updating 

[DR_TAMFR_DHP]

For reference, my AMD 290x's top out at 373 H/s per gpu. 12.3, asm 1, intensity needs to be around 6. Anything else and it diminishes significantly.

http://image.prntscr.com/image/341d9b53d4614ab6af042b136a599d89.png

I wonder why your 290x (373 h/s does so much more than my 380x (193 h/s).....   I need some different vid cards lol.

[Cryptohminer]

I am using 3x Radeon Pro Duo's

- 15.12 is not compatible.
- 16.3.2 Windows will work with 2 cards and miner too. When the 3rd is installed, the pc will BSOD in a loop with error [System Thread Exception Not Handled (atikpmag.sys)]
- 16.10 all cards will show in windows, miner will hash @ 0
- 17.3.1 all cards will show in windows, miner will crash after launching.

Any suggestions ? thanks in advance !

[TheRider]

I am using 3x Radeon Pro Duo's

- 15.12 is not compatible.
- 16.3.2 Windows will work with 2 cards and miner too. When the 3rd is installed, the pc will BSOD in a loop with error [System Thread Exception Not Handled (atikpmag.sys)]
- 16.10 all cards will show in windows, miner will hash @ 0
- 17.3.1 all cards will show in windows, miner will crash after launching.

Any suggestions ? thanks in advance !

What mobo? Can you try 16.11.5?

I'm really interested to know cause I'm thinking of sticking 3 pro duo cards in a tower case in the office, but haven't selected the mobo yet...

[omshree]

I am using 3x Radeon Pro Duo's

- 15.12 is not compatible.
- 16.3.2 Windows will work with 2 cards and miner too. When the 3rd is installed, the pc will BSOD in a loop with error [System Thread Exception Not Handled (atikpmag.sys)]
- 16.10 all cards will show in windows, miner will hash @ 0
- 17.3.1 all cards will show in windows, miner will crash after launching.

Any suggestions ? thanks in advance !

What mobo? Can you try 16.11.5?

I'm really interested to know cause I'm thinking of sticking 3 pro duo cards in a tower case in the office, but haven't selected the mobo yet...

For Pro Duo it's only 16.3.2 Your problem is not driver related I suppose, try proper mobo, or clean win10 install.

[Hamaber]

For reference, my AMD 290x's top out at 373 H/s per gpu. 12.3, asm 1, intensity needs to be around 6. Anything else and it diminishes significantly.

http://image.prntscr.com/image/341d9b53d4614ab6af042b136a599d89.png

I wonder why your 290x (373 h/s does so much more than my 380x (193 h/s).....   I need some different vid cards lol.

290X has more processor cores and higher memory bandwidth.

[majorlee]

Issues with Version 12.3:
- Some rigs will not run all GPUs since v12.0.  With 12.0 and later, some GPUs report 0.00H/s rate.  One of rigs has 7990, only uses one core.  Have not tried 12.3 version on this unit, it runs 12.0 fine. 

- Option -di detect will run very long time on a couple rigs.  Would love to know a procedure to do this manually by switching GPUs on and off.  How to map default GPU order to -di GPU order?  This is an issue because miner will reduce intensity on the wrong card, leaving the hot card to keep getting hotter and eventually stop or crash the rig.  Does the ordering need to be recalculated for different versions?

Linux rigs seem more stable than Wndows rigs.

Thanks for a great miner.

yes i found similar problems

went back to oldest drivers but i also found 1 riser had failed! I also shifted a few cards around just to see what would happen and seemed to fix it

all my 7990's are hashing with new version perfectly for nearly 12hrs+

most of my rigs are mixed cards....one 290x, one HD 7990, and a few 280x's as well on each one, 5-6 GPU's in total

good luck, they DO work and i had similar issues, so there is light for you my 7990 friend

[Fastmartin]

Issues with Version 12.3:
- Some rigs will not run all GPUs since v12.0.  With 12.0 and later, some GPUs report 0.00H/s rate.  One of rigs has 7990, only uses one core.  Have not tried 12.3 version on this unit, it runs 12.0 fine. 

- Option -di detect will run very long time on a couple rigs.  Would love to know a procedure to do this manually by switching GPUs on and off.  How to map default GPU order to -di GPU order?  This is an issue because miner will reduce intensity on the wrong card, leaving the hot card to keep getting hotter and eventually stop or crash the rig.  Does the ordering need to be recalculated for different versions?

Linux rigs seem more stable than Wndows rigs.

Thanks for a great miner.

yes i found similar problems

went back to oldest drivers but i also found 1 riser had failed! I also shifted a few cards around just to see what would happen and seemed to fix it

all my 7990's are hashing with new version perfectly for nearly 12hrs+

most of my rigs are mixed cards....one 290x, one HD 7990, and a few 280x's as well on each one, 5-6 GPU's in total

good luck, they DO work and i had similar issues, so there is light for you my 7990 friend

Did you increase the intensity of the 280X to 6 from 5?