after about 24 hours of test, I find 0.6 gives 0S/s for some of my cards more often then 0.5 which is never  , so switch back to 0.5 for now plus, all my GPU usage is around 78% up and down, none of them is reaching 100% like mining other coins |
|
|
|
|
agree, I finally give up
putting amd driver files in the genoil folder doesn't do the trick for me
so I just make sure it's stable mining @ whatever speed I can get out of these old 7950/7970s
|
|
|
|
IT WORKS!!!!!!!!!!!
Copy the entire directory of AMD\Non-WHQL-Win10-64Bit-Radeon-Software-Crimson-16.10.3-Oct27\Packages\Drivers\Display\WT6A_INF\B308236
Into the Genoil directory, will be like 100 files over 250MB long.
I used the Windows 7 AMD 16.10.3 NOT the Windows 10 like shown above.
Interesting I tried to delete all files I just copied in and system prompt one of the files is in use by Genoil.exe...... |
|
|
|
What Sols/s rate are you getting with 290(x)? (Clocks 1030/1250) I'm getting just about 25 Sols/s and 7950 and 280 gives about the same. (Clocks 900/1500) By the way Miner is getting better on every release, thanks  my R9 280 at default get 29 to 37 not the X version and seem to the holdiong at 30 with this version of the miner even my r7 370 seem to run better . Okey, thats much better than me :\ I start my miner (7950, 7950 , 280) with these options, have you changed w or i from Default?
genoil.exe -c POOL:PORT -u USER.WORKER -p z -g 0 1 2 -f 0
(E8400 and 2gigs of ram if these matter) You can't get 30 with a Radeon 280. Not even with a 280x. that is my R9 280 3gb at stock setting water cooled  i use this setting sense everything else appers to be on by default genoil.exe -c us1-zcash.flypool.org:3333 -u ZEC address -p z Nice man. What AMD driver version are you using for your 280x? Did you change intensity, work size or any other options? Stock firmware flashed to card as well? Undervolting? Thanks! Crimson Edition 16.10.3 Hotfix here is the card i have http://www.newegg.com/Product/Product.aspx?Item=N82E16814131570set at defaults including the software miner. I may flash it to see what that does i was very impressed when support for older amd cards came how well this works, it went from using nicehash software hitting at best 10 sol to what you see when 0.42 came out or what ever version gave tahiti cards support . Now i got 30 from my 280 too, 16.10.3 Hotfix did the trick, no change in 7950. Thanks for help and for everyone else-> You dont have to install the driver, just copy these files to same drawer. amdocl64.dll
amdocl12cl64.dll
amdocl12cl.dll
amdocl.dll
amd_opencl64.dll
amd_opencl32.dll
Would be nice to know what is the best driver for 290(x) ' SO we get those files from Windows/SYstem32 correct? And we put them inside the Genoil directory? download 16.10.3 from amd and you will find all files in AMD\Non-WHQL-Win10-64Bit-Radeon-Software-Crimson-16.10.3-Oct27\Packages\Drivers\Display\WT6A_INF\B308236 but this still doesn't work for me, so I stay with 20ish for all my cards for now |
|
|
|
|
if I am already running on 16.10.3 does this help?
|
|
|
|
On my 4 card rig, running 2 instances of Genoil, one for GPU 0+1, the other for GPU 2+3 seems to add to the stability for now.
Running one instance on all 4 cards would simply crash the whole rig. And yes, I've tried just mining on one card before mining on all of them...
Are that trick increase performance at least a little? Or have no impact? I don't think it does, at least not for me, but the good thing is 1 error won't bring down the whole mining rig. I am going to the extreme that open one instance for each card make a batch file to delay each instance for 15 secs and kill them all every 1800 secs see how stable then can get. |
|
|
|
On my 4 card rig, running 2 instances of Genoil, one for GPU 0+1, the other for GPU 2+3 seems to add to the stability for now.
Running one instance on all 4 cards would simply crash the whole rig. And yes, I've tried just mining on one card before mining on all of them...
that sounds a good idea, I can group same cards in my instance |
|
|
|
can someone share the command line for reset if "0" hashrate is found?
simply add -f 0 at the end of genoil.exe line |
|
|
|
is 280 a rebrand 7970? I have two Powercolor 7970 and they gave approx. 22S/s  maybe there might be one setting on i forgot about i used afterburner to turn it on might be why there is a setting called boost in after burner once it on it stys on other then that you think I'm making it up ? . if i turm it off it puts the card at 22 /23 i total forgot i had that on.  nop, I have no doubt in what you are saying, I am just trying to make mine hashing harder  I also use MSI afterburner, and it seems if I adjust clocks in afterburner, mining speed will turn to 5S, 9S or reduced significantly trying to use the boost button, never realize there is a boost :p |
|
|
|
Do you think it's better -f 0 than -f 2?
-f 0 won't guarantee a success restart of your GPU I guess, cos mine just failed one and keep giving me 0S/s for a while I am trying -f 2 and have to find out how to use bat file to check if the miner is running or killed. |
|
|
|
is 280 a rebrand 7970? I have two Powercolor 7970 and they gave approx. 22S/s |
|
|
|
Released 0.6.
- time & thread logging
- fancy console colors
- logs output to file in ./logs directory
- 0 sols/s watchdog
- some minor SilentArmy improvements[/li][/list]
The 0 sols watchdog is enabled by default. Because I don't know yet what is actually causing it, I have added three ways to deal with it. You can specify that with -f flag.
-f 0: attempt to restart the GPU.
-f 1: leave the gpu thread and do nothing (basically what happens now)
-f 2: quit the application. when used in conjunction with a looping bat file, this could be most effective.
Still, loads to do before I can call it 1.0, but it's getting better every release.
Thanks ppl who either accidentally or intentionally leave the default bat file like it is!
thanks for your great work  I am trying it out as I am typing this, I like the fancy console colors  |
|
|
|
|
anyone came across mining speed affected by pcie extension cable?
I just added a asus 7970 to my desktop, I have to use the extension cable to allow the card run outside my case
Asus card doesn't like to work on the pcie-usb 3.0 solution, I have to use a 1x - 16x cable.
I feel that's not long enough so I added another 1x - 1x and connect to the 1x on mobo.
with this config, no matter what I do, the mining speed will be 0.6 - 4.x S/s
so I removed the 1x - 1x and just use the 1x - 16x cable, it back to 20-26 S/s, interesting.
BTW, with a OC AMD Sempron 145 OC to 3.5G, all my 7950/7970 on that rig are doing 20ish S/s, so I guess don't need a fx6300 for now
|
|
|
|
Processor is an AMD Sempron 145. I know it is slow but didn't think anything faster was needed when graphics cards are doing the mining. Was also trying to not use as much electricity with this processor. Anyway, so I need a faster processor? Unlock the other core if you can. One of my all Tahiti rigs uses the same CPU, and it was a real slowpoke with this miner until I realized it was the CPU holding it back and I hadn't unlocked both cores yet. That machine is now mining much faster. Not a problem on my Intel rigs, though the 7 card rig isn't mining to its full potential yet. I suspect that one could use more CPU as well - it's a mixed Tahiti/Hawaii/Pitcairn rig. Incidentally, that one is more stable with 0.4.2 than 0.5. 0.5 will mine for a minute or two and then the cards drop out one by one. I just did a fresh rig, moving all my card from the old E2160 to the "new" 145 cpu rig and I'm going to OC the cpu to 3.5 and see if there is any improvement otherwise I am considering grab a 6300 to support those GPUs and I guess itself can hash about 14S/s |
|
|
|
that looks OLD cpu intensive my E2160 will go to 90% while mining |
|
|
|
|
I use 16.10.2 for my desktop and 16.10.3 for my rig
and I found if it hash @ 0S/s the kernel file is not actually write in the catch folder, I don't know why
try to start with only one card
just came across this prompt
A request to send or receive data was disallowed because the socket is not connected and (when sending on a datagram socket using a sendto call) no address was supplied
and speed back to 0S/s
restart the miner seems solve the problem.
|
|
|
|
my desktop 7950 is mining between 20-28 on stock clock settings, seems -i doesn't affect on this card,
buy all my 7970s are doing no more than 15 per card, maybe my -i is too high for them? (use default -i 20)
or driver is too old? updating the driver now
does ram plays a role in hashing speed? coz my desktop has 32G or ram and only 1 GPU hashing.
what processors do you have in both? desktop I have a fx8320 which is also hashing @ 6 core for 14.5S/s 4 cards rig has a E2160 that doing no hashing it's CPU related? I tried to swap a 7950 out with my desktop card and the on in my desktop is still doing 22-28 but the one back to the rig is doing 16-19..... or does OS matters? My desktop is running windows 10 while my mining rig is running win 7 both are 64bit |
|
|
|
|
my desktop 7950 is mining between 20-28 on stock clock settings, seems -i doesn't affect on this card,
buy all my 7970s are doing no more than 15 per card, maybe my -i is too high for them? (use default -i 20)
or driver is too old? updating the driver now
does ram plays a role in hashing speed? coz my desktop has 32G or ram and only 1 GPU hashing.
|
|
|
|
finally got my rigs running on suprnova my desktop has 8320 @ 6 cores + a 7950 gives me 21.xx-23.xx Sols/s another rig with 4x7970 gives me 37.xxSols/s using 7 cores on my desktop cpu will gain another 3.xx Sols/s but the CPU fan will get louder Still have another 2x7950 need to put the rig up and generate some heat for the winter |
|
|
|
|
Show Posts
