Wolf's XMR GPU Miner

[cryptomined]

does this work in windows 10 with drivers 16.10.#? or 16.9.#?

I keep trying to run it but my command windows just closes right after trying to run it..
not sure if I fixed the config file proper, but I think I did
is there some instrucitons for setting up the config file somwhere that I missed?

Thanks

[mirny]

@dallase

Hi, where can I find that coinpicker application?
cheers

[pacers101]

7770 Does good ~162h/s, Claymore does ~170h/s

I get this a lot:
"Share rejected (Low difficulty share)"

http://imgur.com/a/UQdRs

5830 Throws error: "Error -54 when calling clEnqueueNDRangeKernel for kernel 3."

Config:

Code:

{

 "Algorithms":
[{
      "name": "CryptoNight",
      "devices":[{
            "index": 0,
                                "rawintensity": 216,
            "worksize": 8
                },
{
            "index": 1,
                                "rawintensity": 128,
            "worksize": 8
                }
          ],

      "pools":[{
            "url": "stratum+tcp://mine.moneropool.com:8080",
            "user": "user",
            "pass": "x"
         }]
}]
}

It is neat you are working on an open source miner.

I get a similar error, mine is error 63 and kernel 6 but it also says "when calling clEnqueueNDRangeKernel for kernel" . have you fixed this at all, I'd like to know how if you have. thanks

[oldDIN]

to take 7770 version 3,
4 falls on them due to lack of memory or reject

[Koiner247]

+1

I know it's a wee bit off topic, but does anyone have WORKING links to Wolf's CPU miner for Cryptonight?

[Ned Kelly]

+1

I know it's a wee bit off topic, but does anyone have WORKING links to Wolf's CPU miner for Cryptonight?

Just download file from 1st post and use -1 as device index.

[Trimegistus]

Guys,

Trying to make my way into XMR mining but I'm getting the old error on my rigs:

Code:

Error -1 when calling clGetDeviceIDs for number of devices

I'm using miner version 0.4 on a rig with 6x480 4GB under Win10 x64, Crimson 16.9.2.

This is my conf file:

Code:

{
	
"Algorithms":

[
		
	{

	"name": "CryptoNight",
	"devices":
	[
	{
	"index": 0,
	"threads": 1,
	"rawintensity": 640,
	"worksize": 16
	},
	{
	"index": 1,
	"threads": 1,
	"rawintensity": 640,
	"worksize": 16
	},
	{
	"index": 2,
	"threads": 1,
	"rawintensity": 640,
	"worksize": 16
	},
	{
	"index": 3,
	"threads": 1,
	"rawintensity": 640,
	"worksize": 16
	},
	{
	"index": 4,
	"threads": 1,
	"rawintensity": 640,
	"worksize": 16
	},
	{
	"index": 5,
	"threads": 1,
	"rawintensity": 640,
	"worksize": 16
	}
	],
	"pools":
	[
	{
	"url": "stratum+tcp://xmr-eu.dwarfpool.com:8005",
	"user": "47sghzufGhJJDQEbScMCwVBimTuq6L5JiRixD8VeGbpjCTA12noXmi4ZyBZLc99e66NtnKff34fHsGRoyZk3ES1s1V4QVcB.a34e4b35838dde3cf5e68bca39ed1f64376b71eaee082407d50d4ca4d11101ef.02",
	"pass": "x"

Any ideas?

Thanks!

[Trimegistus]

C'mon guys, anyone using this miner on a multi AMD GPU rig?

Can you please help me?

[QuintLeo]

I suspect most of us that have multi-AMD rigs are mining more profitable stuff like ETH/ETC/ZEC over mining XMR with those cards.

[Trimegistus]

I suspect most of us that have multi-AMD rigs are mining more profitable stuff like ETH/ETC/ZEC over mining XMR with those cards.

Yeah, but XMR profitability is now close to ETH/ZEC and I wanted to try this miner just in case...

[Termie]

Guys,

Trying to make my way into XMR mining but I'm getting the old error on my rigs:

Code:

Error -1 when calling clGetDeviceIDs for number of devices

I gave it up with this miner. Never get it to run with my 1x290x / 3x Nano rig.
I tried v0.3 and v0.4 with Crimson 15.12 and 16.7.3. No chance, always getting this -1 error.
With Claymore's it runs immediately.

[ku4eto]

Okay guys.

R9 290 Sapphire Trix Here, Hynix Memory.

Running at 16.12.1 drivers, using Wattman global OC at 1000/1375.

With conf of:

rawintensity: 800
threads: 2

I am getting ~3.4GB of VRAM used and around 700 H/s.

Any more rawintensity and i either run out of VRAM and BSOD the PC, or it turns into negative scaling.

How do Worksize affect the hash rate?

Temps are ~75 core, <70 on VRMs.

Fan speed is default profile, ramps up to 40%.

Also, why my 1605T gets 22H/s at default conf, and the work VM with E5 2660v2 gets 25H/s with AVX1 cpu miner? I thougth the difference would be bigger...

Also, for some reason, when running the miner, i get occasional hitches (mouse, scrolling). The DPC latency seems normal, ideas on that?

[antantti]

How do Worksize affect the hash rate?

With 2 threads I would try to lower that worksize. 32, 16, 8 and so on.

[ku4eto]

How do Worksize affect the hash rate?

With 2 threads I would try to lower that worksize. 32, 16, 8 and so on.

The default one is 16, i have tried changing between 24 and 8, and saw no difference.

I couldn't find any specific data about this. And didn't found much about the clock speed relations. What does the Wolf miner benefit more from? Core or memory overclock? I guess if i lower the core clocks, i can increase the memory, without needing to increase the voltage.

[antantti]

How do Worksize affect the hash rate?

With 2 threads I would try to lower that worksize. 32, 16, 8 and so on.

The default one is 16, i have tried changing between 24 and 8, and saw no difference.

I couldn't find any specific data about this. And didn't found much about the clock speed relations. What does the Wolf miner benefit more from? Core or memory overclock? I guess if i lower the core clocks, i can increase the memory, without needing to increase the voltage.

First find those best settings, then start to fine tune with core & mem. Lower than 8 could work.

[QuintLeo]

Okay guys.

R9 290 Sapphire Trix Here, Hynix Memory.

Running at 16.12.1 drivers, using Wattman global OC at 1000/1375.

 1) No reason to run the 16.12.1 drivers for ANY reason, they're demonstrated to be junk for mining - 5-10% LOWER hashrate than 16.9.2 or 15.12
     Bloated junk with very poor performance AND no tolerance for modded BIOS on many cards that IMO should never have been released.

 2) No reason to run anything past 15.12 on a card prior to the RX series, as NONE of the 16.x series drivers offer ANY improvement and tend to be LESS stable on the older cards.
     16.9.2 is a tossup on hashrate though, if you feel the need to be compatable with newer cards for future upgrades to a rig.

 3) Drop Wattman. It's total junk. PITA interface, very ANTI-intuitive controls that don't WORK very well if they work at all, UI is totally FUBAR.
     MSI Afterburner blows it away TOTALLY on configurability and ease of use, especially on fan control which is CRITICAL for 290/29x/390/390x cards.



 As far as profitability - I have yet to see XMR even get CLOSE to ETH/ETC/ZEC on profitability on AMD cards in the ballpark year I've been watching them all.
 Nvidia, yes - XMR is somewhat more profitable on a regular basis on my GTX 750 Ti, but that card seems to be THE sweet spot for XMR but doesn't do ETH/ETC very well at all (and soon not at all) and doesn't do well on ZEC even with the NiceHash eqm miner.


 Your number may vary of course.



 I don't know if TheStilt ever did any BIOS that would work with non-Reference R9 290s, but they made a HUGE difference on my Sapphire ref design R9 290s - 20 or so degrees cooler even when bumping up the core clock from the "can barely run at 947 before thermal limiting hits" pre-bios to "runs comfortably at 1100 core clock" post-bios, and the memory timings are a TON tighter which gave a major hashrate jump even before I started moving the clocks from the stock 947/1250.

[ku4eto]

Okay guys.

R9 290 Sapphire Trix Here, Hynix Memory.

Running at 16.12.1 drivers, using Wattman global OC at 1000/1375.

 1) No reason to run the 16.12.1 drivers for ANY reason, they're demonstrated to be junk for mining - 5-10% LOWER hashrate than 16.9.2 or 15.12
     Bloated junk with very poor performance AND no tolerance for modded BIOS on many cards that IMO should never have been released.

 2) No reason to run anything past 15.12 on a card prior to the RX series, as NONE of the 16.x series drivers offer ANY improvement and tend to be LESS stable on the older cards.
     16.9.2 is a tossup on hashrate though, if you feel the need to be compatable with newer cards for future upgrades to a rig.

 3) Drop Wattman. It's total junk. PITA interface, very ANTI-intuitive controls that don't WORK very well if they work at all, UI is totally FUBAR.
     MSI Afterburner blows it away TOTALLY on configurability and ease of use, especially on fan control which is CRITICAL for 290/29x/390/390x cards.



 As far as profitability - I have yet to see XMR even get CLOSE to ETH/ETC/ZEC on profitability on AMD cards in the ballpark year I've been watching them all.
 Nvidia, yes - XMR is somewhat more profitable on a regular basis on my GTX 750 Ti, but that card seems to be THE sweet spot for XMR but doesn't do ETH/ETC very well at all (and soon not at all) and doesn't do well on ZEC even with the NiceHash eqm miner.


 Your number may vary of course.



 I don't know if TheStilt ever did any BIOS that would work with non-Reference R9 290s, but they made a HUGE difference on my Sapphire ref design R9 290s - 20 or so degrees cooler even when bumping up the core clock from the "can barely run at 947 before thermal limiting hits" pre-bios to "runs comfortably at 1100 core clock" post-bios, and the memory timings are a TON tighter which gave a major hashrate jump even before I started moving the clocks from the stock 947/1250.

1) Wrong, i am using this rig to game as well, not a dedicated zombie machine.

2) ^

3) Wattman offers exactly what i need, profile based OC. MSI AB or TriXX voltage controls and overclock is global and screws up all other Windows profiles as well.  Wattman offers fan control as well, but i am using the default fan curve, which is actually enough.


Anyway, i figured out the mouse hitching - it was due to GPU-Z.
When i stopped it (and all other monitoring programs), the hashrate went from ~700h/s to 800h/s.
And the pool reports that i am actually doing 0.9 - 1.2 kh/s right now (PC left on, while i am on work).

Anyway, tried with workunit: 2 and 4, no differennce again O_o ?

[QuintLeo]

Stock fan profile will likely get your card DEAD, or at best will severely limit it for mining usage due to frequent if not constant thermal limiting.

 As a gaming machine that only mines secondarily though, TheStilt bios mods are probably a bad option for you.


 It's odd that GPU-Z would impact your hashrate - I've NEVER seen it do that in a lot of usage on a lot of varied hardware platforms.

 I suspect something underlying as the actual issue, not GPU-Z itself.

 I also suspect you'll see a lot fewer issues OVERALL if you run the 16.11.5 drivers vs the 16.12.1, even if you game.
 I find it VERY interesting that nobody is comparing performance of 16.12.1 vs 16.11.5 or vs 16.9.2 - makes me wonder if the mining performance isn't the ONLY performance hit the new version offers.

[moonmaths]

This is me killin it on Ubuntu 16.04 w/ 6x MSI rx470 4GB

[14:31:16] Thread 3, GPU ID 3, GPU Type: Ellesmere: 635.87H/s
[14:31:16] Thread 2, GPU ID 2, GPU Type: Ellesmere: 634.97H/s
[14:31:17] Thread 5, GPU ID 5, GPU Type: Ellesmere: 641.74H/s
[14:31:18] Thread 1, GPU ID 1, GPU Type: Ellesmere: 644.15H/s
[14:31:18] Thread 4, GPU ID 4, GPU Type: Ellesmere: 641.23H/s
[14:31:18] Thread 0, GPU ID 0, GPU Type: Ellesmere: 641.01H/s
[14:31:18] Share accepted: 72/72 (100.00%)
[14:31:18] Total Hashrate: 3838.97H/s

BONUS --> Only 640w total draw at the wall... (custom bios 1160core/1900mem/1500straps)

How are you modifying your bios?  You just going into UEFI boot and changing the options?  I'm new to this obviously. 

Core 1160 = gpu core clock rate?
Mem = memclock rate?
What is straps? 
Also, have you tried undervolting the gpu's or is it possible on ubuntu? 

I have 2 unmodded sapphire rx480's running at 577H/s
"threads": 1,
"rawintensity": 1024,
"worksize": 8

[bkelley0804]

Guys,

Trying to make my way into XMR mining but I'm getting the old error on my rigs:

Code:

Error -1 when calling clGetDeviceIDs for number of devices

I'm using miner version 0.4 on a rig with 6x480 4GB under Win10 x64, Crimson 16.9.2.

This is my conf file:

Code:

{
	
"Algorithms":

[
		
	{

	"name": "CryptoNight",
	"devices":
	[
	{
	"index": 0,
	"threads": 1,
	"rawintensity": 640,
	"worksize": 16
	},
	{
	"index": 1,
	"threads": 1,
	"rawintensity": 640,
	"worksize": 16
	},
	{
	"index": 2,
	"threads": 1,
	"rawintensity": 640,
	"worksize": 16
	},
	{
	"index": 3,
	"threads": 1,
	"rawintensity": 640,
	"worksize": 16
	},
	{
	"index": 4,
	"threads": 1,
	"rawintensity": 640,
	"worksize": 16
	},
	{
	"index": 5,
	"threads": 1,
	"rawintensity": 640,
	"worksize": 16
	}
	],
	"pools":
	[
	{
	"url": "stratum+tcp://xmr-eu.dwarfpool.com:8005",
	"user": "47sghzufGhJJDQEbScMCwVBimTuq6L5JiRixD8VeGbpjCTA12noXmi4ZyBZLc99e66NtnKff34fHsGRoyZk3ES1s1V4QVcB.a34e4b35838dde3cf5e68bca39ed1f64376b71eaee082407d50d4ca4d11101ef.02",
	"pass": "x"

Any ideas?

Thanks!

This happened to me when i went from 1 --> 2 GPUs. My fix was to dump OS and re-install everything. If anyone knows an easier fix, please share!