PhoenixMiner 6.2c: fastest Ethereum/Ethash miner with lowest devfee (Win/Linux)

[B00gieman]

yeah i've noticed that the voltage commands seem to not work at all. and applying drastically low values (like 700mV) do nothing and power consumption is not reduced.

That is amd drivers, not the miner per se, amd drivers used to be like that, you set and it did, now you have to mod the bios with the voltage you want and you need to write -cvddc xx and you must pray the driver will accept the voltage cause if it does not then it will use another voltage like 950 or even 1150 and you will see your card overheating after few minutes. Most noob miners think is using the correct set voltage but most of the time if you did not do it right then no.

Did you read my previous post?
The command line works on Claymore With no problem.
Im using blockchain drivers and already testes With last amd drivers.

People With AMD, can you please test only AMD With Claymore and PM? PLEASE report config (Core / mem - OC and voltage), drivers used and Consumption in Watt.

[inv1s1bl3]

I have tried mining some akroma with the 2.9e on my 3card rig and I'm getting almost 96mh/s with the settings 1200/2200! Running stable for the past week or so without any hangs. Great miner!

[B00gieman]

That is amd drivers, not the miner per se, amd drivers used to be like that, you set and it did, now you have to mod the bios with the voltage you want and you need to write -cvddc xx and you must pray the driver will accept the voltage cause if it does not then it will use another voltage like 950 or even 1150 and you will see your card overheating after few minutes. Most noob miners think is using the correct set voltage but most of the time if you did not do it right then no.

Did you read my previous post?
The command line works on Claymore With no problem.
Im using blockchain drivers and already testes With last amd drivers.

People With AMD, can you please test only AMD With Claymore and PM? PLEASE report config (Core / mem - OC and voltage), drivers used and Consumption in Watt.

[pinamalina]

Here we go again:

Miner:
7*1070 + 5*580

So on Claymore v. 11.6 i use This bat file:

Code:

setx GPU_FORCE_64BIT_PTR 0
setx GPU_MAX_HEAP_SIZE 100
setx GPU_USE_SYNC_OBJECTS 1
setx GPU_MAX_ALLOC_PERCENT 100
setx GPU_SINGLE_ALLOC_PERCENT 100
Miners\Claymore_Dual_Ethereum_v10.2\EthDcrMiner64.exe -epool eu1.ethermine.org:4444 -ewal walletaddress.rigcozinha -epsw x -allpools 1 -mport 4444 -cclock 1150 -mclock 2225 -cvddc 850 -mvddc 850

MSI Afterburnner 4.5.0:
Core: 0
Mem: +715
PL: 65%
Power Consumption: 1750w

PhoniexMiner config bat file:

Code:

 REM
REM Example bat file for starting PhoenixMiner.exe to mine ETH
REM

setx GPU_FORCE_64BIT_PTR 0
setx GPU_MAX_HEAP_SIZE 100
setx GPU_USE_SYNC_OBJECTS 1
setx GPU_MAX_ALLOC_PERCENT 100
setx GPU_SINGLE_ALLOC_PERCENT 100
PhoenixMiner.exe -pool eu1.ethermine.org:4444 -pool2 us1.ethermine.org:4444 -wal walletaddress.RIGcozinha -proto 3 -cclock 1150 -mclock 2225 -cvddc 850 -mvddc 850 -log 0

Afterburnner config are the same

Power Consumption: 2150w

Whats the problem?
I like verter PM, but With This difference i cannot keep using.

After some tests i realized that the extra consumption is on AMD cards. So for now... the solution i'm using is Claymore for AMD cards, and PM for Nvidia cards.

I dont know why but PM is not executing the "in BOLD" command line:
PhoenixMiner.exe -pool eu1.ethermine.org:4444 -pool2 us1.ethermine.org:4444 -wal walletaddress.RIGcozinha -proto 3 -cclock 1150 -mclock 2225 -cvddc 850 -mvddc 850 -log 0

Use hwinfo64 to check actual values the gpus are running at. For PhoenixMiner specify the voltage for each card so - cvddc 850,850,850,... -mvdcc 850,850,850... Same for frequencies.

I have multiple rigs configured this way on -18.x.x drivers perform as expected.

[B00gieman]

Will try that...
********
 Tested as sugested. But was not the solution. Thanks anyway for the help.
Its strange... I know.
.......

Uninstaling everything. Start from zero.
If it works for you, it should work for me.

[crypper]

I've been using claymore for my RX X80s for over a year now. Decided to give phoenix a go...and I'm damn happy I did: slight improvement in hashrate at the same clocks and power and significant reduction in stale shares! I've been suffering from 5-15% stales (ethermine.org) since Byzantium fork in 2017. With phoenix I have <1% stales.
Great job, man.

[B00gieman]

Will try that...
********
 Tested as sugested. But was not the solution. Thanks anyway for the help.
Its strange... I know.
.......

Uninstaling everything. Start from zero.
If it works for you, it should work for me.

So.... start over again...

The drivers:
So. I didnt use DDU because de program only gives me a white screen... (everything appens to me).
Downloaded latest Adrenalin Drivers.
Install drivers... instalation program says it needs to remove older ones... yes...reboot...and the new adrenalin drivers start pumping... wait...
Install finished.
Reboot
No need to patch using ATimdagpatch.
Execute SM... and click "y" to activate compute mode. (if dont do this you will get only 19hashes). The computer needs to reboot.
Execute again SM and now the ComputeMode is activated and you should get 31/32 hashes.

The best... finally... the consumption is low... as on Claymore.
Funny... i've already tried this in the past...something possible went wrong...

Thanks to you all that showed me the way.

[sexontoast]

This a common issue ?
Would love to move away from Claymore for my 3 rigs.

Eth: Generating light cache for epoch #186
GPU1: 26C 60%
build: Failed to GPU1 program: clBuildProgram (-11)
GPU1: Failed to load kernels: clCreateKernel (-46)
Thread(s) not responding. Restarting.
Eth speed: 0.000 MH/s, shares: 0/0/0, time: 0:00



Eth: New job #107ff4c3 from eth-au.dwarfpool.com:8008; diff: 2000MH
GPU1: Starting up... (0)
build: Failed to GPU1 program: clBuildProgram (-11)
GPU1: Failed to load kernels: clCreateKernel (-46)

[4play]

For now i have never problem with PhoenixMiner. Now i have a problem with this:
I using PM 2.9e

Code:

2018.05.13:10:26:26.668: GPU2 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)
2018.05.13:10:26:26.668: GPU2 GPU2 search error: unspecified launch failure
2018.05.13:10:26:26.678: GPU5 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)
2018.05.13:10:26:26.678: GPU5 GPU5 search error: unspecified launch failure
2018.05.13:10:26:26.694: GPU3 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)
2018.05.13:10:26:26.694: GPU3 GPU3 search error: unspecified launch failure
2018.05.13:10:26:26.709: wdog Thread(s) not responding. Restarting.
2018.05.13:10:26:26.725: GPU6 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)
2018.05.13:10:26:26.725: GPU6 GPU6 search error: unspecified launch failure
2018.05.13:10:26:26.725: GPU1 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)
2018.05.13:10:26:26.725: GPU1 GPU1 search error: unspecified launch failure
2018.05.13:10:26:26.756: GPU4 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)

My settings are same all time. I have gtx 1070 (8gb) gpus. How to solve this problem any idea?

[rboer]

For now i have never problem with PhoenixMiner. Now i have a problem with this:
I using PM 2.9e

Code:

2018.05.13:10:26:26.668: GPU2 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)
2018.05.13:10:26:26.668: GPU2 GPU2 search error: unspecified launch failure
2018.05.13:10:26:26.678: GPU5 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)
2018.05.13:10:26:26.678: GPU5 GPU5 search error: unspecified launch failure
2018.05.13:10:26:26.694: GPU3 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)
2018.05.13:10:26:26.694: GPU3 GPU3 search error: unspecified launch failure
2018.05.13:10:26:26.709: wdog Thread(s) not responding. Restarting.
2018.05.13:10:26:26.725: GPU6 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)
2018.05.13:10:26:26.725: GPU6 GPU6 search error: unspecified launch failure
2018.05.13:10:26:26.725: GPU1 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)
2018.05.13:10:26:26.725: GPU1 GPU1 search error: unspecified launch failure
2018.05.13:10:26:26.756: GPU4 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)

My settings are same all time. I have gtx 1070 (8gb) gpus. How to solve this problem any idea?

try lowering the GPU2 overclock settings a bit, looks like that is the card that is causing the issue

[4play]

For now i have never problem with PhoenixMiner. Now i have a problem with this:
I using PM 2.9e

Code:

2018.05.13:10:26:26.668: GPU2 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)
2018.05.13:10:26:26.668: GPU2 GPU2 search error: unspecified launch failure
2018.05.13:10:26:26.678: GPU5 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)
2018.05.13:10:26:26.678: GPU5 GPU5 search error: unspecified launch failure
2018.05.13:10:26:26.694: GPU3 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)
2018.05.13:10:26:26.694: GPU3 GPU3 search error: unspecified launch failure
2018.05.13:10:26:26.709: wdog Thread(s) not responding. Restarting.
2018.05.13:10:26:26.725: GPU6 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)
2018.05.13:10:26:26.725: GPU6 GPU6 search error: unspecified launch failure
2018.05.13:10:26:26.725: GPU1 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)
2018.05.13:10:26:26.725: GPU1 GPU1 search error: unspecified launch failure
2018.05.13:10:26:26.756: GPU4 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)

My settings are same all time. I have gtx 1070 (8gb) gpus. How to solve this problem any idea?

try lowering the GPU2 overclock settings a bit, looks like that is the card that is causing the issue

 Ok, tnx for advice. I’ll try this

[Digital_Seytan]

I have unspecified launch failure 719 for some days now, what does this mean, and what can you do about it? phoenixminer 2.9.e nvdia 388.31 win 10/64

[jonfrakes]

DAG generation freezes/crashes the whole system with the latest PhoenixMiner. I'm using 8 GTX 1070s (6 MSI Gaming Xs and 2 Zotac Minis). I tried running the miner using -lidag 3, but to no avail.

Windows 10 Pro
8 GB RAM
73728 MB Virtual Memory / Pagefile Size

[sidaliroy]

Still no info on 3.0?

[PhoenixMiner]

I would love me some nice juicy and fat linux version

    Yes, but hopefully not that fat There is some progress on this but no ETA yet, sorry.

Hi Phoenix, can you explain how -nvf option works (and, of course, for -clf) more detailed and can it be used with new kernels or with old ones only? Also I would like to have more details about -mi and -gt options. If you can explain it in understandable way for regular users, it will be so good. May be it will be better also to include this information in readme file. Thank you for very good work!

   It's hard to this do in non-technical terms (and frankly, it's not that understandable even in technical terms) but here it goes:

   -clf/-nvf

   Theoretically, it would be best to leave it to the driver to decide how to synchronize the individual kernel dispatches (calls) - this should provide the best speed and utilization of the GPU. In practice however, the driver often doesn't do very good job, and here is the role of the -clf and -nvf options (first one is for AMD cards, second one - for Nvidia cards). When these options are at their default values (1), PhoenixMiner will "force" the synchronization periodically to avoid driver issues. This still leaves some room for the driver to do optimizations between these "force" points. If the options are at 2, the driver has no say at all. The other extreme is when these options are at 0: no forced synchronization will be performed and the driver will be fully in control. Possible problems when using -clf/-nvf at 0 are unstable hashrate or semi-permanent drops of the hashrate. -clf/-nvf 2 should provide much more stable hashrate but sometimes at the price of slightly lower absolute speed.
   These options are used for both the new kernels and the old ones, and can be specified per-GPU.

 -mi

   This effectively sets the size of job performed by each kernel dispatch (call). The bigger -mi values correspond to larger kernel sizes. This usually increases the hashrate because less time is lost in overhead when starting/stopping the kernel. However the bigger job sizes can also cause crashes and unresponsive video drivers, especially if one of the GPUs is also used as main GPU (connected to a display or a plug). In most cases AMD cards have less problems with high -mi values than Nvidia cards, which are almost unusable at high -mi values and sometimes just crash with high job sizes.

 -gt

   Again, theoretically, the GPU should "know" the best how to combine the individual threads in each kernel dispatch in order to maximize the throughout. It does this by switching very fast from one thread to another whenever the first thread is waiting for a memory read/write to finish. This is something like hyper-threading on steroids. Again, the theory doesn't correspond to practice too well, so we are doing "manual" reordering of the memory accesses but this should be fine-tuned for each GPU depending on the core clocks, memory clocks, and memory access latency. This is done by the -gt parameter. Note that there may be more than one "best" or "equally best" -gt value and the difference between these will be almost 0. We are going to release an auto-tune option in PhoenixMiner 3.0, which will automatically find the best -gt values for each AMD card individually (note that -gt only works for AMD cards).

I cannot use

Code:

-coin clo 

param. Please, update the miner to support CLO.

   Coming in PhonixMiner 3.0 along with a lot of other ethash coins.

Here we go again:

Miner:
7*1070 + 5*580

So on Claymore v. 11.6 i use This bat file:

Code:

setx GPU_FORCE_64BIT_PTR 0
setx GPU_MAX_HEAP_SIZE 100
setx GPU_USE_SYNC_OBJECTS 1
setx GPU_MAX_ALLOC_PERCENT 100
setx GPU_SINGLE_ALLOC_PERCENT 100
Miners\Claymore_Dual_Ethereum_v10.2\EthDcrMiner64.exe -epool eu1.ethermine.org:4444 -ewal walletaddress.rigcozinha -epsw x -allpools 1 -mport 4444 -cclock 1150 -mclock 2225 -cvddc 850 -mvddc 850

MSI Afterburnner 4.5.0:
Core: 0
Mem: +715
PL: 65%
Power Consumption: 1750w

PhoniexMiner config bat file:

Code:

 REM
REM Example bat file for starting PhoenixMiner.exe to mine ETH
REM

setx GPU_FORCE_64BIT_PTR 0
setx GPU_MAX_HEAP_SIZE 100
setx GPU_USE_SYNC_OBJECTS 1
setx GPU_MAX_ALLOC_PERCENT 100
setx GPU_SINGLE_ALLOC_PERCENT 100
PhoenixMiner.exe -pool eu1.ethermine.org:4444 -pool2 us1.ethermine.org:4444 -wal walletaddress.RIGcozinha -proto 3 -cclock 1150 -mclock 2225 -cvddc 850 -mvddc 850 -log 0

Afterburnner config are the same

Power Consumption: 2150w

Whats the problem?
I like verter PM, but With This difference i cannot keep using.

    Most probably you are using the blockchain beta drivers, and the undervolt settings are not applied properly by PhoenixMiner. Check with GPU-Z or something like this to see the actual voltage of the GPUs. The solution is either to use third-party program to control the clocks and voltages, or use newer drivers (18.x.x are working good for us).

Hello,

i am running 2 different instances of the miner, 1 for nvidia and 1 for amd, its more comfortable for me to control the settings this way,

for now i used -cdm and 2 diferent ports, so i see them in my control panel as 2 miners,

is there an option to see them as 1 miner  ? somehow make the cdm know that there are 2 windows running ?


thanks

   No, unfortunately there is no way to see them as single miner as each instance needs a different port to listen to.

Still no info on 3.0?

   Around the end of the month. We are adding a lot of new features, and we don't want to release new versions with too little changes between them to avoid upgrade fatigue.

   

[j2james]

   It's hard to this do in non-technical terms (and frankly, it's not that understandable even in technical terms) but here it goes:

   -clf/-nvf

   Theoretically, it would be best to leave it to the driver to decide how to synchronize the individual kernel dispatches (calls) - this should provide the best speed and utilization of the GPU. In practice however, the driver often doesn't do very good job, and here is the role of the -clf and -nvf options (first one is for AMD cards, second one - for Nvidia cards). When these options are at their default values (1), PhoenixMiner will "force" the synchronization periodically to avoid driver issues. This still leaves some room for the driver to do optimizations between these "force" points. If the options are at 2, the driver has no say at all. The other extreme is when these options are at 0: no forced synchronization will be performed and the driver will be fully in control. Possible problems when using -clf/-nvf at 0 are unstable hashrate or semi-permanent drops of the hashrate. -clf/-nvf 2 should provide much more stable hashrate but sometimes at the price of slightly lower absolute speed.
   These options are used for both the new kernels and the old ones, and can be specified per-GPU.

 -mi

   This effectively sets the size of job performed by each kernel dispatch (call). The bigger -mi values correspond to larger kernel sizes. This usually increases the hashrate because less time is lost in overhead when starting/stopping the kernel. However the bigger job sizes can also cause crashes and unresponsive video drivers, especially if one of the GPUs is also used as main GPU (connected to a display or a plug). In most cases AMD cards have less problems with high -mi values than Nvidia cards, which are almost unusable at high -mi values and sometimes just crash with high job sizes.

 -gt

   Again, theoretically, the GPU should "know" the best how to combine the individual threads in each kernel dispatch in order to maximize the throughout. It does this by switching very fast from one thread to another whenever the first thread is waiting for a memory read/write to finish. This is something like hyper-threading on steroids. Again, the theory doesn't correspond to practice too well, so we are doing "manual" reordering of the memory accesses but this should be fine-tuned for each GPU depending on the core clocks, memory clocks, and memory access latency. This is done by the -gt parameter. Note that there may be more than one "best" or "equally best" -gt value and the difference between these will be almost 0. We are going to release an auto-tune option in PhoenixMiner 3.0, which will automatically find the best -gt values for each AMD card individually (note that -gt only works for AMD cards).

The very thank you Phoenix!

Your description has added some clarity. Once again, I'm grateful to you for not ignoring your users' questions and trying to answer all of them. This is one more advantage of your team, in addition to the fact that your product, in my opinion, has already surpassed Claymore.

[Digital_Seytan]

For now i have never problem with PhoenixMiner. Now i have a problem with this:
I using PM 2.9e

Code:

2018.05.13:10:26:26.668: GPU2 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)
2018.05.13:10:26:26.668: GPU2 GPU2 search error: unspecified launch failure
2018.05.13:10:26:26.678: GPU5 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)
2018.05.13:10:26:26.678: GPU5 GPU5 search error: unspecified launch failure
2018.05.13:10:26:26.694: GPU3 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)
2018.05.13:10:26:26.694: GPU3 GPU3 search error: unspecified launch failure
2018.05.13:10:26:26.709: wdog Thread(s) not responding. Restarting.
2018.05.13:10:26:26.725: GPU6 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)
2018.05.13:10:26:26.725: GPU6 GPU6 search error: unspecified launch failure
2018.05.13:10:26:26.725: GPU1 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)
2018.05.13:10:26:26.725: GPU1 GPU1 search error: unspecified launch failure
2018.05.13:10:26:26.756: GPU4 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)

My settings are same all time. I have gtx 1070 (8gb) gpus. How to solve this problem any idea?

try lowering the GPU2 overclock settings a bit, looks like that is the card that is causing the issue

I have also had the same problem for a couple of days, have you solved it and what did you like to report?

[Digital_Seytan]

For now i have never problem with PhoenixMiner. Now i have a problem with this:
I using PM 2.9e

Code:

2018.05.13:10:26:26.668: GPU2 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)
2018.05.13:10:26:26.668: GPU2 GPU2 search error: unspecified launch failure
2018.05.13:10:26:26.678: GPU5 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)
2018.05.13:10:26:26.678: GPU5 GPU5 search error: unspecified launch failure
2018.05.13:10:26:26.694: GPU3 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)
2018.05.13:10:26:26.694: GPU3 GPU3 search error: unspecified launch failure
2018.05.13:10:26:26.709: wdog Thread(s) not responding. Restarting.
2018.05.13:10:26:26.725: GPU6 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)
2018.05.13:10:26:26.725: GPU6 GPU6 search error: unspecified launch failure
2018.05.13:10:26:26.725: GPU1 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)
2018.05.13:10:26:26.725: GPU1 GPU1 search error: unspecified launch failure
2018.05.13:10:26:26.756: GPU4 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)

My settings are same all time. I have gtx 1070 (8gb) gpus. How to solve this problem any idea?

try lowering the GPU2 overclock settings a bit, looks like that is the card that is causing the issue

 Ok, tnx for advice. I’ll try this

I have also had the same problem for a couple of days, have you solved it and what did you like to report?

[j2james]

For now i have never problem with PhoenixMiner. Now i have a problem with this:
I using PM 2.9e

Code:

2018.05.13:10:26:26.668: GPU2 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)
2018.05.13:10:26:26.668: GPU2 GPU2 search error: unspecified launch failure
2018.05.13:10:26:26.678: GPU5 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)
2018.05.13:10:26:26.678: GPU5 GPU5 search error: unspecified launch failure
2018.05.13:10:26:26.694: GPU3 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)
2018.05.13:10:26:26.694: GPU3 GPU3 search error: unspecified launch failure
2018.05.13:10:26:26.709: wdog Thread(s) not responding. Restarting.
2018.05.13:10:26:26.725: GPU6 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)
2018.05.13:10:26:26.725: GPU6 GPU6 search error: unspecified launch failure
2018.05.13:10:26:26.725: GPU1 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)
2018.05.13:10:26:26.725: GPU1 GPU1 search error: unspecified launch failure
2018.05.13:10:26:26.756: GPU4 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)

My settings are same all time. I have gtx 1070 (8gb) gpus. How to solve this problem any idea?

try lowering the GPU2 overclock settings a bit, looks like that is the card that is causing the issue

I have also had the same problem for a couple of days, have you solved it and what did you like to report?

I have this error regularly beginning from 2018.04.22:18:14. I do not remove logs, so I have full history. Sometime I have other errors. My error history is the following:

error 719 - 5 times (the 1st of them was 2018.04.22:18:14)
error 999 - 1 time
error 719 - 1 time
error 719 and error 999 after 719 - 1 time
error 719 - 1 time
error 4 - 1 time
error 719 - 7 times
error 999 - 1 time
error 719 - 1 time (this one was 2018.05.11:15:24)

Also I had some restarts during this period w\o errors in log files.

All these errors in PM 2.9e. Before that I used 2.7c and 2.8c. And only one time I had only one error 999 in 2.7c. So looks like something wrong in 2.9e.

If PM developers want I can provide all logs.

[pinamalina]

For now i have never problem with PhoenixMiner. Now i have a problem with this:
I using PM 2.9e

Code:

2018.05.13:10:26:26.668: GPU2 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)
2018.05.13:10:26:26.668: GPU2 GPU2 search error: unspecified launch failure
2018.05.13:10:26:26.678: GPU5 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)
2018.05.13:10:26:26.678: GPU5 GPU5 search error: unspecified launch failure
2018.05.13:10:26:26.694: GPU3 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)
2018.05.13:10:26:26.694: GPU3 GPU3 search error: unspecified launch failure
2018.05.13:10:26:26.709: wdog Thread(s) not responding. Restarting.
2018.05.13:10:26:26.725: GPU6 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)
2018.05.13:10:26:26.725: GPU6 GPU6 search error: unspecified launch failure
2018.05.13:10:26:26.725: GPU1 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)
2018.05.13:10:26:26.725: GPU1 GPU1 search error: unspecified launch failure
2018.05.13:10:26:26.756: GPU4 CUDA error in CudaProgram.cu:264 : unspecified launch failure (719)

My settings are same all time. I have gtx 1070 (8gb) gpus. How to solve this problem any idea?

try lowering the GPU2 overclock settings a bit, looks like that is the card that is causing the issue

 Ok, tnx for advice. I’ll try this

I have also had the same problem for a couple of days, have you solved it and what did you like to report?

I got these same errors when OC-ing my 1070 to the limits.
Follow next steps to resolve this problem - worked 100% in my case:

• Reduce the OC settings for the nvidia cards
• Reinstall nvidia drivers
• Run Phoenix ... it will for sure work again

I got this problem twice; In both cases when searching for maximum stable mem freq I can push the 1070 gpus.
Currently running stable at:
  Core: -200 (minus 200)
  Mem: 765
  Power Limit: 59
  Hashing: 32.770 MH/s
To tune, I am using nvidia Inspector as it allows me to easily tune individual gpus.
I have this line "call nvidia-tune.bat" in start.bat before the PhoenixMiner.exe line:

Code:

::nvidia-tune.bat file content: 7 nvidia cards 0..7 - 
:: Note1: card 3 - it is stable at max 625 mem clock - any atampt to go higher will produce errors in minutes after the mining starts
:: Note2: The fans are set to run at constant speed - I have found these values (different for each fan) that keep my cards at around 50-55oC, depending on the ambient temperature
start C:\dev\Tools\Guru3D.com\nvidiaInspector.exe -setBaseClockOffset:0,0,-200 -setMemoryClockOffset:0,0,765 -setPowerTarget:0,59 -setTempTarget:0,1,65 -setFanSpeed:0,45
start C:\dev\Tools\Guru3D.com\nvidiaInspector.exe -setBaseClockOffset:1,0,-200 -setMemoryClockOffset:1,0,765 -setPowerTarget:1,59 -setTempTarget:1,1,65 -setFanSpeed:1,60
start C:\dev\Tools\Guru3D.com\nvidiaInspector.exe -setBaseClockOffset:2,0,-200 -setMemoryClockOffset:2,0,765 -setPowerTarget:2,59 -setTempTarget:2,1,65 -setFanSpeed:2,45
start C:\dev\Tools\Guru3D.com\nvidiaInspector.exe -setBaseClockOffset:3,0,-200 -setMemoryClockOffset:3,0,625 -setPowerTarget:3,55 -setTempTarget:3,1,65 -setFanSpeed:3,40
start C:\dev\Tools\Guru3D.com\nvidiaInspector.exe -setBaseClockOffset:4,0,-200 -setMemoryClockOffset:4,0,765 -setPowerTarget:4,59 -setTempTarget:4,1,65 -setFanSpeed:4,45
start C:\dev\Tools\Guru3D.com\nvidiaInspector.exe -setBaseClockOffset:5,0,-200 -setMemoryClockOffset:5,0,765 -setPowerTarget:5,59 -setTempTarget:5,1,65 -setFanSpeed:5,60
start C:\dev\Tools\Guru3D.com\nvidiaInspector.exe -setBaseClockOffset:6,0,-200 -setMemoryClockOffset:6,0,765 -setPowerTarget:6,59 -setTempTarget:6,1,65 -setFanSpeed:6,45

In my experience the nvidia is not gracefully handling errors when it comes to errors from OC. In best scenario (when slightly overclocked), PhoenixMiner will restart as expected. In worst case, when OC is too high, nvidia driver/configuration gets corrupted and the error as above ends any attempt to mine with the card. The reboot, restart nor cold restart will not make it go away. Only the driver reinstall fixes the problem.