Hello
please pay attention
something went wrong
I bring a log
[2019-07-11 16:55:18] Startup monitor attached
[2019-07-11 16:55:18] Miner version: 1.9.1
[2019-07-11 16:55:18] Windows version: 10.0 build 17763
[2019-07-11 16:55:18] Video driver version: 26.20.11015.5009
[2019-07-11 16:55:18] Please run miner as administrator to enable tweaking
[2019-07-11 16:55:23] AMD Platform ID: 0
[2019-07-11 16:55:23] AMD platform FOUND
[2019-07-11 16:55:23] Found 7 AMD devices
[2019-07-11 16:55:23] CPU AES-NI: FALSE
[2019-07-11 16:55:23] PREPARE DATA: FALSE
[2019-07-11 16:55:23] GPU0: Radeon(TM) RX 460 Graphics [baffin] [2048 MB][CU: 14][I: 20.0][W: 8][T: 2][F: 8][BUS: 1]
[2019-07-11 16:55:23] GPU1: Radeon(TM) RX 460 Graphics [baffin] [2048 MB][CU: 14][I: 27.0][W: 8][T: 2][F: 8][BUS: 3]
[2019-07-11 16:55:23] GPU2: AMD Radeon R9 200 Series [tonga] [2048 MB][CU: 28][I: 28.0][W: 8][T: 2][F: 8][BUS: 4]
[2019-07-11 16:55:23] GPU3: Radeon RX 560 Series [baffin] [4096 MB][CU: 16][I: 27.0][W: 8][T: 2][F: 8][BUS: 8]
[2019-07-11 16:55:23] GPU4: AMD Radeon (TM) R7 370 Series [pitcairn] [2048 MB][CU: 16][I: 28.0][W: 8][T: 2][F: 8][BUS: 9]
[2019-07-11 16:55:23] GPU5: AMD Radeon R7 200 Series [pitcairn] [2048 MB][CU: 16][I: 28.0][W: 8][T: 2][F: 8][BUS: 10]
[2019-07-11 16:55:23] GPU6: AMD Radeon (TM) R7 370 Series [pitcairn] [2048 MB][CU: 16][I: 28.0][W: 8][T: 2][F: 8][BUS: 11]
[2019-07-11 16:55:23] ADL is enabled
[2019-07-11 16:55:23] Algorithm: Cryptonight Conceal
[2019-07-11 16:55:23] Algo switching: disabled
[2019-07-11 16:55:23] Coin forking: disabled
[2019-07-11 16:55:23] Gpu watchdog: enabled [gpu disable mode]
[2019-07-11 16:55:23] Startup monitor: enabled
[2019-07-11 16:55:23] Tweaking: disabled
[2019-07-11 16:55:23] Socket initialiasation done
[2019-07-11 16:55:23] Starting init of gpu threads
[2019-07-11 16:55:23] Created OCL context
[2019-07-11 16:55:23] Created OCL input buffer for DeviceID 0 (Thread 0)
[2019-07-11 16:55:23] load_binary: Couldn't load default binary for DeviceID 0 (Thread 0)
[2019-07-11 16:55:23] Loading [conceal] kernel for DEVICE BUS_ID[1]
[2019-07-11 16:55:23] ctx->Program for DeviceID 0 (Thread 0) loaded
[2019-07-11 16:55:23] Created OCL input buffer for DeviceID 0 (Thread 1)
[2019-07-11 16:55:23] load_binary: Couldn't load default binary for DeviceID 0 (Thread 1)
[2019-07-11 16:55:24] Loading [conceal] kernel for DEVICE BUS_ID[1]
[2019-07-11 16:55:24] ctx->Program for DeviceID 0 (Thread 1) loaded
[2019-07-11 16:55:24] Created OCL input buffer for DeviceID 1 (Thread 2)
[2019-07-11 16:55:24] load_binary: Couldn't load default binary for DeviceID 1 (Thread 2)
[2019-07-11 16:55:24] Loading [conceal] kernel for DEVICE BUS_ID[3]
[2019-07-11 16:55:24] ctx->Program for DeviceID 1 (Thread 2) loaded
[2019-07-11 16:55:24] Created OCL input buffer for DeviceID 1 (Thread 3)
[2019-07-11 16:55:24] load_binary: Couldn't load default binary for DeviceID 1 (Thread 3)
[2019-07-11 16:55:24] Loading [conceal] kernel for DEVICE BUS_ID[3]
[2019-07-11 16:55:24] ctx->Program for DeviceID 1 (Thread 3) loaded
[2019-07-11 16:55:25] Created OCL input buffer for DeviceID 2 (Thread 4)
[2019-07-11 16:55:25] load_binary: Couldn't load default binary for DeviceID 2 (Thread 4)
[2019-07-11 16:55:25] ctx->headers[0] created for DeviceID 2 (Thread 4)
[2019-07-11 16:55:25] Error CL_COMPILE_PROGRAM_FAILURE when calling clCompileProgram [ctx->Program] for DeviceID 2 (Thread 4)
[2019-07-11 16:55:25] Error while creating GPU threads.. Stopping miner process.
[2019-07-11 16:55:25] Socket cleanup done
[2019-07-11 16:55:25] Shutting down miner...
previous version works fine
should be fixed. try it, re - download 1.9.1 |
|
|
|
V1.9.1- Added support for Navi10 (gfx1010) - Arto, conceal, normal, normalv7, marketcash, b2n, fast, hycon, hospital algos have lower power consumption on Vegas with auto settings - Hopefully fixed bug with invalid job id / same job id - If more than 5 'invalid share' messages received from pool within same connection, auto miner restart - Fixed cn-gpu hashrate (it was lower from v.1.8.8 -> my mistake) - CN/R program precompilation process is now only logged, not displayed on screen + Added support for gfx1010 card. Well at least i hope, because i don't own a rx 5700 card, so i couldn't test  It would be good if someone could test it and report back. + On Vega, intensity 0 (auto setting) now uses a little less power on next algorithms : Arto, conceal, normal, normalv7, marketcash, b2n, fast, hycon, hospital + I tried to fix the 'invalid job', 'invalid share' issue, unfortunately i couldn't reproduce it neither earlier neither now, that's why i say i tried ..  Anyways if the problem still persists, i also added a 'protection' against this, if within the same pool connection more than 5 invalid share messages received from the pool, miner will auto restart. + I accidentally messed up CN/GPU hashrate from version 1.8.8, now it is fixed and hashing again at full speed tested on 5700xt Algo: normalv4 1 min avg: 231 H/s 5 min avg: 206 H/s Max Speed: 866 H/s Thanks for trying it out. You probably tried the auto setup, guess didnt play with manual config yet.. i hope more can be squeezed out from those cards and the right intensity/worksize/fragments. |
|
|
|
V1.9.1- Added support for Navi10 (gfx1010) - Arto, conceal, normal, normalv7, marketcash, b2n, fast, hycon, hospital algos have lower power consumption on Vegas with auto settings - Hopefully fixed bug with invalid job id / same job id - If more than 5 'invalid share' messages received from pool within same connection, auto miner restart - Fixed cn-gpu hashrate (it was lower from v.1.8.8 -> my mistake) - CN/R program precompilation process is now only logged, not displayed on screen + Added support for gfx1010 card. Well at least i hope, because i don't own a rx 5700 card, so i couldn't test It would be good if someone could test it and report back. + On Vega, intensity 0 (auto setting) now uses a little less power on next algorithms : Arto, conceal, normal, normalv7, marketcash, b2n, fast, hycon, hospital + I tried to fix the 'invalid job', 'invalid share' issue, unfortunately i couldn't reproduce it neither earlier neither now, that's why i say i tried .. Anyways if the problem still persists, i also added a 'protection' against this, if within the same pool connection more than 5 invalid share messages received from the pool, miner will auto restart. + I accidentally messed up CN/GPU hashrate from version 1.8.8, now it is fixed and hashing again at full speed
|
|
|
|
Iridium Coin
It is time for VERSION 5! New release is out containing new algorithm (Cn Pico) !
When new version srb?
CN Pico = turtle, so you can already mine it |
|
|
|
Letter by letter! It worked!
Thanks a lot, @Topcd, @doktor83!
Now you just need to find the best settings for your VII |
|
|
|
You are using the wrong algorithm, you should use cryptonight_r : config.txt{
"cryptonight_type" : "cryptonight_r",
"intensity" : 0,
"double_threads" : true
}{
"pools" :
[
{
"pool" : "xmr.pool.minergate.com:45700",
"wallet" : "myemail",
"password" : "x",
}
]
}SRBMiner-CN.exe --config config.txt --pools pools.txt Also looks like pool rejects your connection, maybe you got banned because trying with the wrong algo? |
|
|
|
upd
changed mobo(to ga 970a ds3p) and now it works fine, ~4k
could it be PCI bus thing?
upd2
if one of vc is in PCI ex1 slot both cards shows low hashrate and if both in PCI ex4-16 then both works fine.
Great, im glad you found the solution  |
|
|
|
Thanks for help, but did both, and still have low hashrate with 2 vc. normal with one.
Also tryed Adrenalin 18.6.1 instead of newest, updating mobo bios, putting vc in diff slots (both plugged via razers).
Can you paste your config file here ? btw. that's almost the same hashrate i get on my Vega56 Hynix sometimes when the --resetvega isn't applied.. |
|
|
|
Miner works fine with one vc (vega56 2000 H/s).
But with 2 plugged it shows only half hashrate.
8GB ram, 60GB ssd swap, intel g4400, ASUS H110M-K D3
What could it be?
--setcomputemode as admin once, and add --resetvega to start.bat |
|
|
|
Hi all,
I tried started mining with SRB, the last version 1.9, configured all but I canīt connect to the pool that Iīm used to connect, xmr.pool.minergate.com.
Is there some incompatibility with the pool or similar? I reviewed N tutorials from the videos that are available and is correctly configured.
By the way, is advised to use Radeon Adrenalin Edition 18.5.1, Radeon Adrenalin Edition 18.5.2 or Radeon Adrenalin Edition 18.6.1 but the Radeon VII is not detected by them, I had to use the last ones.
If someone help me I will appreciate it!
Thanks!
Just tried and it works with Minergate : {
"pools" :
[
{
"pool" : "xmr.pool.minergate.com:45700",
"wallet" : "your-email-address",
"password" : "x",
}
]
}
Show your configs, so i can help you. |
|
|
|
Interesting. So is this only on 1.9.0 or 188, 189 too ?
Can some of you upload a log file somewhere for me to inspect?
Also which pool are you guys using?
same shit on 1.8.0 - work perfect on 1.9.0 -random "invalid/bad shares", but +150 h/s, using https://mining.bit.tube/i think it's not the same issue, because those guys have problem with CN/R and precompilation, not Tube. https://bitcointalk.org/index.php?topic=3167363.msg50571884#msg50571884heavy_mode default value is now 3, before it was 1. Mode 3 is faster but can produce more compute errors. also it can happen that after a pool disconnect/reconnect shares get sent to the pool , those will automatically be flagged as invalid shares. I will make some changes regarding this, so it shouldnt happen. |
|
|
|
start
about 1-10 hours all are ok
and then
[2019-07-03 18:02:41] JOB error: Last and current jobs have same job_id
[2019-07-03 18:02:41] Reconnecting to pool.hashvault.pro:3333 in 15 seconds
[2019-07-03 18:02:56] Connected to pool.hashvault.pro:3333
[2019-07-03 18:02:56] Pool difficulty: 20000
[2019-07-03 18:02:56] Job received, block height 1870497 [ID: b89226cb-e47f-4285-80ca-bbb357dc1454]
[2019-07-03 18:02:58] JOB error: Last and current jobs have same job_id
and again, and again...
What's that?
A nasty bug i should fix Could you please send me the log file in PM so i can get more info on what's happening? Thanks for reporting |
|
|
|
Hello all I've started using SRBMiner exclusively, mining Monero, I get flawless performance on my Vega 56 at tweak profile 9 approx. 2020 H/S for approx. 36-48 Hrs with no invalid/bad shares, somewhere between that time it starts spitting all invalid/bad shares, any recommendations or ideas as to what may be causing this. Is there a way to configure the miner to automatically restart when it does start submitting invalid shares?
exactly the same on 1.9 with lower gpus and even WITH NO TWEAKS. 1.87 is also not perfect but works at least. restart on invalid shares is good idea as they go after few hours --gpuerrorrestart needed I have a similar issue. Everything works fine, and then after 16-20 hours I only get rejected shares. Interesting. So is this only on 1.9.0 or 188, 189 too ? Can some of you upload a log file somewhere for me to inspect? Also which pool are you guys using? |
|
|
|
I'm searching for a "summer algo" a light one for my old amd rx cards what is the lighter algo to mine with SRBMiner, the one that will not overheat my cards too much? Maybe litev7 algo in merged mining, like AEON + IRD |
|
|
|
Doesn't seem to be helping, unless I did something wrong.
Hard to tell what's causing it.. Something you could always try is to delete video drivers with DDU and then reinstall them. Also 18.6.1 are recommended. |
|
|
|
Yo, I'm back. A couple of days ago I finished mining another monero with my two 64s. I was on the 1.8.0 but had to update because I didn't mine anything afterwards. So, I'm at 1.9.0 now, and my main GPU never goes to its full potential. Always starts around 1000h/s and just drops and drops over time. Sure, that happened before If I was using anything besides the miner, but now its the case even after fresh restart, and nothing besides the miner is active. Can someone help, or if its a bug could you look at it?
Try adding --nobinaries parameter to start.bat , maybe your cards have a problem with the built in binaries. |
|
|
|
Hello doktor83,
what is the characteristic of the ramp-up curve, when using --enablegpurampup ? Is it linear, and, how long does it take unil it reaches the maximum?
Thank you for your reply in forward,
RBM
Every thread is delayed 2 seconds. So if you have 6 gpu, every gpu is running 2 threads, that is 12 threads total so ~ 24 seconds until every gpu starts working. Except on CN/R, because of the precompilation process. Until precompilation is finished, the gpu is idling. |
|
|
|
Hello Doctor83,
With 1.8.9 the rigs stops after 10-12 hours, and the 1.9.0 make to many compute errors any algorithm with vega 64 cards. The 1.8.8 was works fine. Please check what was the problem.
Cheers.
From 1.8.9 to 1.9.0 nothing changed in kernels, so i don't see a reason for more compute errors. Can you give more info, a log would be the most helpful i can see almost everything from it. Maybe you have problems with the embedded binaries (but i dont see any reason why would it cause problems). You can add --nobinaries in start.bat so they won't be used, try it. Thanks doctor, now works fine! Great, but nevertheless, can you send me a log ? |
|
|
|
Hello Doctor83,
With 1.8.9 the rigs stops after 10-12 hours, and the 1.9.0 make to many compute errors any algorithm with vega 64 cards. The 1.8.8 was works fine. Please check what was the problem.
Cheers.
From 1.8.9 to 1.9.0 nothing changed in kernels, so i don't see a reason for more compute errors. Can you give more info, a log would be the most helpful i can see almost everything from it. Maybe you have problems with the embedded binaries (but i dont see any reason why would it cause problems). You can add --nobinaries in start.bat so they won't be used, try it. |
|
|
|
|
V1.9.0
- Added gpu ramp-up (slow start) possibility, so your PSU won't get suddenly overloaded (--enablegpurampup)
- Added low mode tweak profiles
- Added parameter --precompileblocks, which can set the number of blocks that get precompiled on CN/R algo
- Miner will try to restart if CN/R precompilation process takes more than 10 minutes for a gpu
- Minor bug fixes
+ Added gpu ramp-up (slow start) so your PSU won't get suddenly overloaded. It is disabled by default, to enable it add --enablegpurampup to start.bat
+ If your GPU can't handle a tweak profile, try using the low (weaker) profile. Just add 'L' after the tweak number. If using config files change the tweak profile to be between quotation marks. I updated the examples on the first page so you can see how to set it up there.
+ CN/R needs to precompile blocks from time to time, the default number for blocks to be precompiled is 15. This can now be changed by using --precompileblocks value.
If you set a bigger number precompilation process will be longer, but less often.
|
|
|
|
|
Show Posts
