Before about DAG 142 or so and the AMD drivers with the memory paging fix, I was able to get > 28Mh/s on all 4 of my Polaris cards without too much tweaking (using Claymore/Win7). Neither of them have Samsung RAM either...that was with AJR, MJR, and BABG.

With the fix now they all do 29+.

I'm hoping I can get Debian 9 running with a more recent kernel and amdgpu drivers soon so I can try out sgminer/linux to see if they'll go faster.

btw I guess I'll just throw this out here...anyone have a fix for this error in the Windows version of sgminer?

^THIS!

btw Claymore has posted several times in the last few days.

https://bitcointalk.org/index.php?action=profile;u=306958;sa=showPosts


I've already tried with ethminer 0.11.0 and it doesn't work either. Crashes instantly. I think the problem is related to the driver not the miner.
For Win7 the only solution I've found is to use the early leaked drivers. They are signed by AMD so I don't think they can have any malware...and it's not the crimson .exe type either, just straight-up display drivers with the .dlls and .infs.

http://1stminingrig.com/unofficial-amd-driver-fix-for-ethereum-dag-epoch-released/

lol

@talboris2

Try actually reading the post in that link. Literally the first sentence.

Those work great if you just want to undervolt...but I assume the 'ROP shredding' is not so straightforward.

What you want can be found here:

https://drive.google.com/drive/folders/0B2XVfIhkHZlfMllHSHp5RmFvYlU

The explanation for what is what you can read about here:

https://archive.litecointalk.org/index.php?topic=12830.msg121827#msg121827

Apparently it only works for reference cards.

Yeah apparently it just doesn't work on Win7 

Mining with a laptop is a bad idea. It will overheat and kill it. If you want to mine you should have a PC.

I just tried them and the signature check fix doesn't seem to work anymore.

An idiot that knows where the 's' key is and how to click a link. 

And I'm simply telling you that cgminer is connected to coinotron pool with a share target of 320. It is not solo mining, has found no solo blocks, and has no relevance to anything.

@notbatman

You don't seem to be getting it. When you solo mine, the difficulty is the total network difficulty. (~84T for ETC right now). For a rig that gets 180MH/s, it would take on average (84 * 10^12) / (180 * 10^6) = ~467 000 seconds (or 130 hours or 5.5 days). That is an average and it depends a lot on luck...so you might find nothing for 11 days, and then all of a sudden find 2 in a row.

You will not find any shares unless you find a share that finds a block. Therefore you will not see an average share difficulty or best share found until you find at least 1 block. If you do find a share, you can be assured the difficulty of that share was equal to network difficulty (84T). You can see network difficulty here: http://etcstats.net/

No need to see share difficulty in the miner when you can just check it with a browser any time you want.

Here you go:

http://etcstats.net/

86 Trillion. No sweat.

@flamefaker

Did that solve your issue?

Try running atikmdag-patcher after you install the AMD drivers. They have a BIOS signature check that needs to be removed when using a modded BIOS.

https://www.monitortests.com/forum/Thread-AMD-ATI-Pixel-Clock-Patcher

I keep getting this error when trying to mine ETC:


I searched around but can't seem to find any info on a fix.

I am using Win7x64Pro w/ driver 17.4.4. Card is a 580 8GB.

I have tried removing the standard envars:

(
GPU_FORCE_64BIT_PTR 0
GPU_MAX_HEAP_SIZE 100
GPU_USE_SYNC_OBJECTS 1
GPU_MAX_ALLOC_PERCENT 100
GPU_SINGLE_ALLOC_PERCENT 100
)

It complained about GPU_USE_SYNC_OBJECTS and GPU_MAX_ALLOC_PERCENT not being set and gave the same error (-4)...so I re-set those two but still no change.
I tried differing/lowering worksize (from 192) and xintensity (from 1024)...no change.
I also tried setting gpu-platform 0 (from -1) and changing gpu-map 1:0 and device # (I also have a 2GB 270 in the system as primary card) but still same issue.

Interestingly, when I start it on the other device (the 270), the program starts fine, and I actually get a few hundred khash, even though that card doesn't have enough memory for the DAG. I don't understand how that's even possible.

Any ideas on cause/solution for this error?

btw the miner works fine for mining zcash albeit with a low sol rate.

Ahh okay I'm getting it. And with some searching it seems you already answered my 20 follow-up questions a couple months ago

https://forum.ethereum.org/discussion/comment/75007/#Comment_75007

So if I understand correctly...you have to change that if you are going to insert extra commands, but the structure of the commands could still be different depending on the VRM controller type?

Does the byte at offset 06 always point to the last data byte of the controller programming section, or is that specific to the IR controller? And also are the commands always in the format of <command code> 00 <data> 00, or is that controller-specific as well?

In FLT's orginal VOI that byte is 0E....but if the structure was the same then wouldn't that mean it's trying to send command code = 00 and data = 00 (the values from 0C - 0F)

Hmm I never actually used this one so dunno what to tell you other than try one of the other 2 methods.

On another note I finally narrowed down the problem with this Armor card after being half tempted to throw it out the window, and it was tFAW. Apparently 10 is too low. I have no clue how nerdralph was setting that to zero and getting away with it. Maybe the 500 series works differently than the 400 series? Anyone else have issues setting tFAW too low?

You can use a hex editor if you want to but you can do it in PBE as well. I wouldn't touch the 600 strap because it might actually get used during idle or some 2d stuff. In my case I had a 1425 MHz strap, so copied over it with the 1500 strap. Then I copied the 1625 strap over the old 1500 strap, etc. Then ended up with 2 2000 straps, and modified the second one to be a 2250 strap.

Just make sure you also modify the frequencies. Like where it said 2:1425, I had to change it to 2:1500, etc.