As i was typing this, LOL stopped hashing. No error displayed, and system did not crash / reboot. It simply stopped hashing. Fired right back up and hashing at about the the same rates as above.
Hu, thats odd ... nothing displayed ... could it be that the text output was just over-scrolled? This some times happening on Windows. Overall I found out it is likely more stable mining in two Command Lines open, one with --devices 0,1,2 and the second with --devices 3,4,5 I hope to get rid of this instabilities with many GPUs soon. By the way you still can adjust OC settings maybe a bit. Afaik ETH reacts only to memory, while equihash has a certain clock sensitive part in it. Therefore keeping clocks high - but then with under-volt & lower power target may be worth a try. On my other, 390 rig, 0.5 lolminer, when start mining says gpu0 stucked, miner is restarting.. Driver 18.3.4.
Any help?
It seems the 380 does not like the new Minexcoin code ... its always a bit difficult to tune, what makes the one GPU fly can push the other down. Maybe run it for short to see if its also lower on pool. The 0.42 was rather unstable afaik. But I will try to investigate what may have happened there. Regarding the 2nd rig: here I would trs to run two windows. But I will check back again the code if I missed some memory check somewhere there - usually with the new code gen this stucking stuff should have been eliminated. *Search* I just deleted all the old and downloaded v5 again and same result. Is there a new download I didn't see? Thank you for your hard work!
Thats odd ... Linux or Windows? I was just mining BTCZ with it ... (this are the current links) https://www.dropbox.com/s/nu0oe4tv2ncu50c/lolMiner_v05_Win64.zip?dl=1https://www.dropbox.com/s/p3f2g5w08hs2jlu/lolMiner_v05_Lin64.tar.gz?dl=1-------------- Ps: I just silently updated the archive again. I added two fields about the accepted shares in the API and created an option for moving the log file (--logfile /some/path/to/file.txt) so you now can move it. Note you need --logs 1 and the new param to see a file created and printed. Also you need write access to the directory you want to have it. |
|
|
|
|
I just discovered the bug that may cause problems when you try to mine BTCZ.
Update:
The bug is fixed. I updated the download links. If you want to mine BTCZ load the newer version to do so.
|
|
|
|
Ok then, first of all thanks to all testing this new stuff so early after release and commenting here. I was thinking a lot about consequences from what I read and I think this will be the way it goes: 1) I will stop advertising this miner to be usable for NVidias (at least for the moment) It seems it is ok for AMD. Some times more then 4 or 5 GPUs in one instance make it non-stable but thats an issue one can deal with. On Nvidia it is completely different somehow. For MNX and AION it is an easy pick: use ewbf, its just much quicker. For 144/5 and 192/7 the picture is very mixed. Performance for 1070 is ok, for 1080 is good, but below it does not scale. And this in a way I can not explain because well... Ideally the 1060 has 60% of bandwidth of a 1080 and exactly 50% of the shader clusters, so one expect the performance to be between 50 and 60% ... but here its at only 40% which just does not make sense. I expect that this is due to some strange things the OpenCL compiler does on this cards... 2) The current code / situation seems to be a trap. The host side mining code I have is still similar to the one I used for MNX back when I did 0.21. Not bad, but I am hitting the limitations of it now. Still supporting Nvidia ends up that my scheduling for AMD is non optimal (I could use way less CPU load and also get rid of problems with more then 4 GPUs when I would not have to support NV). Also supporting Nvidias causes that I can not use more modern OpenCL 2.0 or 2.1 language features. Other way around sticking to OpenCL makes the results on Nvidia unpredictable. So I decided that I will change some things for future versions. First I will design a more flexible mining back end that schedules work better then now. This means that 0.51 will likely be AMD only. Then I will see if I finally learn some CUDA to add support for Nvidia back in, algorithm by algorithm. Seems like a lot of work? It is! But I got the feeling that the algorithms are not bad in itself. In fact on AMD they work pretty well on on Nvidia you can see here and there its potential. So I think it is worth doing the effort. Stay tuned. Ps: will later reply to the concrete support requests here. Right now busy with typing a new front page entry. Keep mining  |
|
|
|
miner not working
lolMiner is configured to mine Zelcash(ZEL), Equihash 144/5
And closing without errors
At that point is interesting. I will check code if we have a bug here. Can you paste me the full error message as private msg? It would be also more informative to add time stamp.
And those ~20sol/s on GPU0 against all other is also seen on 1080 (less difference but still exist). GPU0 is faster in your miner. But GPUs are the same.
You can add timestemps (see manual ^^) It seems we have a bug here with multi GPU. Does this difference also exist when you use olets say only 2 GPU in one instance (--devices 0,1) Any ideas?
Thanks
james
As mentioned above it seems that the miner does not lile too much cards... hmmm can you start two instances, first with --devices 0,1,2 and second with --devices 3,4,5? Lets see if that works. I definately have to restructure the miners main loop it seems. |
|
|
|
Got a 8*1060 rig ewbf makes 210-240sol/s depending on PL(60-90)
Your miner makes only 116-130sol/s also depending on PL(60-90).
But with 5*1080 it is closer:
ewbf - 240sol/s
lol ~ 200sol/s
Ah, thats interesting. I have seen that with MNX in the past that the 1060 series reacts different to code then 1080. Good to know that this is the case here. Can you do me a favor and delete the 144_5_NV.bin file from.the kernel dir and see again what happens? The Nvidias should then load the AMD kernel and I would like to know if that works better for the 1060 cards. Thx for reporting! Memo to myself: need a 1060 for testing... Great Big News, Lolliedieb!
can't make this work though? can someone paste a BTCZ config that works? anyone?
Thank you for all of the hard work. Anyone? Thank you, again!
If you need a one liner look to the post before yours and replace --coin MNX by --coin BTCZ next pool, port and so on... should be self explaining. Else 0.43 had a working config file for btcz if remember right. |
|
|
|
One note: I am a bit short on time, so I will update the front page later. Who ever now already tests, I would appreciate some benchmarks For all wondering how to start the miner with new command line structure it is Linux: ./lolMiner --coin <shortCut> --pool <poolAddr> --port <portNumber> --user <Wallet or user name> --pass <userPassword>
Windows: lolMiner.exe --coin <shortCut> --pool <poolAddr> --port <portNumber> --user <Wallet or user name> --pass <userPassword> So for example for MinexCoin: lolMiner.exe --coin MNX --pool stratum.minexpool.eu --port 9999 --user XQZoBeGPZJoLhTdkcF8bScsNH5Wyesd9g4.lolRig --pass x Have fun! |
|
|
|
Hey folks, I have worked for a while now and although there might be still headroom here and there I think it is time to present lolMiner 0.5  For those who can not wait (plz read the manual ^^) here are the download links: [Lin 64] Click me (0.5 Linux) [Win 64] Click me (0.5 Windows) So lets start with the performance first. I did benchmarks as well with 0.43 as 0.5 and ewbf 0.5 on Nvidia. All benchmarks lasted at least an hour to get stable results and let the miner heat up. Note that for 96/5 I had to correct the 0.43 results down due to defect shares. Here are the resultsAMD RX 580 4G (Sapphire Pulse, stock):| | lolMiner 0.43 | lolMiner 0.5 | | Equihash 96/5 | 12096.7 sol/s | 13951.6 sol/s | | Equihash 144/5 | 21.2 sol/s | 25.1 sol/s | | Equihash 192/7 | 12.5 sol/s | 15.2 sol/s | | Equihash 210/9 | - | 107,5 sol/s |
NVidia GTX 1080 FE (stock):| | lolMiner 0.43 | lolMiner 0.5 | ewbf 0.5 | | Equihash 96/5 | 17200 sol/s* | 19984.1 sol/s | 22141.4 sol/s | | Equihash 144/5 | 34,4 sol/s | 45.5 sol/s | 43.7 sol/s | | Equihash 192/7 | 19,5 sol7s | 23.1 sol/s | 19.8 sol/s | | Equihash 210/9 | - | 178.4 sol/s | 209.0 sol/s |
Note: I observed that newer AMD drivers are a little quicker then 2017 ones. Maybe worth a try. * Corrected value due to invalid shares. So well – you see some Nvidia algos may be better (I wonder what black magic he used for 96/5 xD) but overall I think improvements are nice compared to 0.43. So what is new:- Complete rework of all GPU algorithms for both stability and speed Note that this also changed the memory footprint of the algorithms. 144/5 is now fitting to 2.2G and thus works on 3G cards when GPU is attached and 192/7 is now close to 3G and should work on any 4G card fine. 210/9 (AION) by the way uses only ~800M. - Added Equihash 210_9 (Aion) - Added Coins: AION, ANON and BTH - Renamed the AUTO144 coin profile to AUTO144_5. Also added AUTO192_7 (see manual where it is possible to use this) - Changed the command line style.Before it was -parameter=value and now it is --parameter value This is more Linux style and will help integrating the miner into mining OS solutions - Rewrote the complete manual - Added a function to write the text output to a log (see manual) - Added support for custom personalization strings for Equihash 144/5 and 192/7 (see manual) - The disabling of the memory check is no longer required on 3G & 4G Nvidia GPUS (disabled by default) - Reduced the maximum work batch for 96/5 to 24[/b] because higher values caused crashed some times. - Changed the standard short / long statistic intervals to 30 seconds and 5 minutes. - Changed the run_miner scripts not to loop any more (unless you use the _with_restart version), see known issues. Migration Guide:Kernels and command line interface have changed, but most config files from 0.43 will continue to work. So if you want to migrate from 0.43 best create a new folder for 0.5, unzip the archives there and then copy over your config file from old to new folder. Do not forget to update the run_miner scripts with the profiles you like to use! Know issues:The new codes have all been stable so far with except for 192/7 on Nvidia. I observed approx a miner close in 4-5 hours for this. For this GPUs I recommend using the _with_restart starting scripts that will restart the miner directly. |
|
|
|
Before answering I just wanted to note that I got 2 mails recently regarding mining on R7 370. The problem there is that GCN1 does not allow full memory for compute, so this cards fail compatibility check even when 4G are there. I am currently tuning the memory footprint of 0.5 - hopefully I can get below the limits of this cards so they work then. Lolliedieb pleeease implement that sh#t temp control.
...
Eg.: if gpuX temp >= Ytemp then stop mining for Z time period then Restart mining.
Hmm - indeed that idea sounds easy to implement. I plan temperature sensor reading for 0.6, so when I got the temps it should be easy to do so. I will give it a try the worst miner for the reds, it's silly to use your rx580 which is worse than 1060))) the developer sucks, the miner is shit !!!
For a moment I considered to report to mod and delete it, but no, will not do that. I never claimed of doing an better job then the other miner devs, especially I really respect ewbf, because he did an insanely performant and stable miner for Nvidia GPUs that is a really tough competition (But in fact my 0.5 144/5 implement is hashing 4% faster on my 1080 test sys  ) Anyways. If you think this one is not profitable then just do not use it. I think mining something else and then buy the coin you like is also a perfect way to support your favorite coin, because it keeps price up. And that I say as a miner dev - If you do so I earn nothing, but you do the better deal. But why so bold language here? Without this miner there would be no 144/5 miner for AMD yet, the 192/7 miner would be extremely slow for 4G cards, for Minex the miner would have only 70% of current performance and for AION there would be no new miner next week, so that would be almost (except for Linux Rocm) Nvidia only, too. So it may be that the miner is not a top profitable option for AMD - but I still think it adds value to the community and thus has a right to exist. Regarding performance ... I could now try to argue why 580 is slower then 1060 on Equihash (cache controller structure and amount of local memory per shader cluster), but well... Equi was never best on AMD, Nvidia usually take the sols / watts crown here. But one question: why so hard personal words against me? Did I do something against you? Try to find someone the is more responsive on this thread among the mining software devs... Good luck. is there any help to run lol miner as a custom miner in hive os?
I will contact hive team one or two days before release (when I write manuals and so on) to add 0.5. Hope they can do that. Any update on the 0.5 release?
I currently plan for Sep 25th or 26th - if my brain does not melt before. (Re-) writing 8 mining codes (4 algos in 2 tuning variants each) plus a new stratum (for aion) in only 4 weeks is / was a hell of work xD |
|
|
|
Thank you. That worked!
However, it doesn't seem to be doing any hashing..
I can see that it connected to my pool (on my local pool server) but it doesn't do anything past the output above. It sits there at "Connected to...".
If I read your output correctly the miner is waiting for the pool to authorize the worker. But there where the authorization message should appear I only see "Warning: value is type " what indicates that the pool sends out a defect stratum message. Please check that and also that the pool supports the extended mining.notify message with the personalization string given by pool. When you plan to release 0.50?
Thanks for helping me, seems for Minex it makes no difference with HIGH or VERYHIGH, can you check it?
Maybe something to do with drivers as using older drivers than latest ATI
Planning to do so beginning of next week  Good question... hmm ... I do not know much about hive. But I will contact the team so they integrate 0.5 asap. |
|
|
|
I'm trying to mine ANON with AUTO144 and i'm getting errors.. anyone know what I'm doing wrong?
The configuration is looking good so far. The error message is a bit vague I have to admit. I will try to improve them soon. There are two potential reasons: 1) The miner fails the memory check. That could be on 4G cards, when some driver environment variables are not set. For Windows use setx GPU_FORCE_64BIT_PTR 1
setx GPU_MAX_HEAP_SIZE 100
setx GPU_USE_SYNC_OBJECTS 1
setx GPU_MAX_ALLOC_PERCENT 100
setx GPU_SINGLE_ALLOC_PERCENT 100
before starting the software, on Linux names of the variables are the same, they are set with "export NAME=value". 2) The miner does not find the kernel / the kernel directory. Ensure it is in the folder where you start the miner. ------------- Update of the 0.5 development status: | Algorithm | AMD | Nvidia | | 96/5 | Ready | In Tuning | | 144/5 | Ready | Ready | | 192/7 | Ready | In Testing | | Aion | In Tuning | In Tuning | | (210/9) | (Stratum ready) | (Stratum ready) |
|
|
|
|
Thanks I added this line "PASS" : "x"
"WORKBATCH" : "HIGH"}
To my user config file, but soon as I run the miner goes crazy, doesnt work windows keep popping up.
In next version can you check it, also why the need for two windows, can we keep all in one like other miners.
You missed a "," because every line needs a comma at its end except for the last line in the block - I think its the json parser complaining here. --------- So I'm seeing an odd behavior with 144.5 mining (zelcash in this case, 2 Vega64's, 3 Vega56's). If/when I lose connection to the pool, the moment the worker is reauthorized, my linux box crashes and reboots:
...
and then the immediate reboot. I don't see the kernel panic in the log, but I see ubuntu 18 uses syslog caching by default. I've removed caching and we'll see if it catches something on the next crash.
This is more of an FYI unless you see something in your reconnect code that might cause a panic. However, I'm not even sure whether the lost connection is a symptom of an imminent crash or the actual cause. Who knows, it could simply be instability with one of my risers...
I'll see if I can track down more info in the logs next time.
edit: I also notice that if I ctrl-c out of the miner, it's a 50/50 toss up that restating the miner causes a reboot on reconnect. So this is likely related. Perhaps something about the state that the card is left in during an abnormal stop? Reuse that card and cause the panic? I find that I usually have to power off the clear the cards before the system is stable again. Just speculating...
I'm undervolting and using the latest amd 18.30 drivers for ubuntu.
Hmm have not seen this till now, but may be. ctrl+c is a bit difficult to handle because it stops program immediately, so I even can not catch this to wait until cards have stopped what they just did to exit gracefully. Maybe I must think of an other way to stop the miner. |
|
|
|
192/5  I just wanted to check if you guys read carefully Of cause its 192/7 By the way, current 0.5 development status:
Please, dont forget fix password for yiimp pools: c=LTC,mc=INN/GUN/GBX No.worry, there are more changes then just new mining codes. This is also on the List |
|
|
|
I'm bit confused where to set: Setting the WorkBatch (MNX, Equihash 96.5 only)
Can you provide me a example line and in where to place it.
In your guide it does not mention this cheers.
All option are used either as "PARAMETER" : VALUE in config file or -parameter=value in command line. So here it is "WORKBATCH" : "HIGH" (Or for example "WORKBATCH":12 for a fixed number) or -workbatch=HIGH in cmd. The parameter controls how many instances of 96/5 are allowed to run at same time / how much of the available GPU memory the miner is allowed to use. ------------- By the way, current 0.5 development status: | Algorithm | AMD | Nvidia | | 96/5 | In testing | In Tuning | | 144/5 | Ready | In testing | | 192/7 | Ready | In testing | | Aion | In dev | In dev | | (210/9) | (Stratum ready) | (Stratum ready) |
|
|
|
|
its not working yet for anonimous coin right?
Just read the 3 posts above yours. Anon is not supported officially yet, but you can mine on suprnova with -coin=AUTO144 (cmd) or "COIN":"AUTO144" (config file). |
|
|
|
Hi dev please add "ANON" Equihash 144,5 Thanks I will do with 0.5 that we will see within next 10 days. But honestly: there are at the moment only 3 or 4 mining softwares out there supporting Equihash 144/5 and you do not start or fork a coin within 5 days. Why I hear of this project now and no dev tried before if all mining software is compatible? I would not trust such a team... When version 0.5, I wish for better Equihash 96.5 (Minexcoin) hashrate, I know you can do it.
Maybe look at other miners and how they done the tweaks.
Please keep project alive and well, don't wish for another miner gone bust
Yes, all algorithms for all cards will get a speedup. AMD cards on MNX about 10%, Nvidia slightly more, but there the 0.4x series was pretty bad. I hope to catch up to EWBF For the 144/5 coins that already works, just use coin=AUTO144 ... the format suprnova uses for this is same as for zergpool. I will add this option for 192/7 next version. |
|
|
|
Hello ,
I have a question . Does auto144 option means it can be used with the new support on MRR for auto peers ?
As for the API .. can you/is it possible to make it to support ethman ? The Remote Control/Monitor program Claymore uses ? I`m guessing its just syntax of response since the program receive mostly GPU temps/speed only .
Regards
Yes, thats true, some pools send the personalization string as 10th entry in the mining.notify method even if they do not auto exchange (e.g. have seen that on suprnova). This can alliw to mine a coin not officially supported yet. Regarding the API: will look into that - most likely will not be able to do it before 0.5, but maybe one of the next minor versions. Maybe u can lookup sgminer, claymore or ccminer that software is compatible for awesome miner..
Will do so. When all new mining codes are done API compatiblity for existing monitoring tools is quite high on priority list. Running lolminer 0.43 for BTG 144 on 4xRX580, works great. Another old rig has 4xRX460 2 gb,
claymore supported this before the fork, but lolminer rejects these cards.
Could I use disable_memcheck for this, or should I wait for lolminer 5 ?
regards
snofte
At the moment the 144/5 algo needs more then 2G of memory, thats why it does not want to start. lolMiner 0.43 needs about 2.8G of memory, 0.5 will be approx 2.2G. So with 0.5 you will be able to mine MNX (96/5) and Aion (210/9) but not 144/5 and 192/7 that need more memory. But I will pick this as a vote for a memory reduced form of 144/5 - sub 2G, but slightly slower then the 2.2G code is possible with some dev effort. |
|
|
|
Still doesn't work on linux ubuntu 16.04 with ROCm amd driver
Hi there, I just wonder if this affects only 144/5 and 192/7 or are all algorithms not working? Because the kernel code compiles for Rocm (tested with offline compiler). I currently wait for a Ubuntu 18.04 compatible rocm driver, then I will do some tests, bit usually it should work... hmm I had never used lolMiner and never mine Equihash algo.
Tell me please, when i start to mine one of equihash coin, i see Kh/s, not KSol. And i don't see my worker on pool statistic page.
You are sure you use lolMiner? Because currently the string "sol/s" is hardcoded and it should never be able to display Kh/s. Something s wrong there. hai..
can u add API to your software? so i can use this with awesome
Thanks
Currently the miner has a simple read-only api. Do you know a good source for the awesome miner api (what it needs for fields to read), because if I can get such thing it would be easy to make lolMiner fully compatible to it. |
|
|
|
Just tried this Miner.
...
Is the Dev going to lower the Dev fee?
...
Well this won't be used as it is the slowest one.
Hi there, thanks for trying out and reporting the results Well as for your last point: I know and I immediately see that in my own wallet - when a miner is often used cash is flowing well on devs side, but if the software is to bad - well he will get no bugs. Kind of turbo capitalism here. That said: I suspect I have many Nvidia users atm - or rather I am sure for 192/7 and 144/5 I have none. Point is that 0.4x trunk was mainly build to make a software for 144/5 on AMD. I am currently working on new codes that will also speed up for both - AMD and NVIDIA. I already can say that new 192/7 code is faster then current EWBF iteration - but well he knows what he is doing and I am sure he will also bring an optimized version soon. Regarding fee height: Developing this stuff is fun but also a bunch of working hours. Therefore I have a simple rule of thumb: when I earned about 500$ with a new code then I lower the fee by 0.5%. This is not yet happened for 144/5, but was already for 96/5 some time ago. There I even made more money in the past, so there the fee is down to 1%. I hope when new code kicks in we will see fee drop to 1.5% soon (obviously because I gained some users then ^^) Lolliedieb
Please, integrate your kernels to executable file.
Had that some time ago, but problem is that this makes it much harder for me to quickly change some stuff, test, change again. What is problem with current solution? With the parameters "KERNELSDIR" and -usercfg you even can put the config file and the kernel directory wherever you like, so does not have to be at same position as the software itself. So why you need it integrated? |
|
|
|
Hi all, I just created a file containing only executables (Both, Windows & Linux in one .zip) that will update lolMiner to 0.43b. Changes: - rebranded SafeCash (SCASH) to Genesis (GENX). Use "COIN" : "GENX" or -coin=GENX to mine it. - Added Bitcoin Candy. Use "COIN" : "CDY" or -coin=CDY to mine it. - Added Bitcoin RM. Use "COIN" : "BCRM" or -coin=BCRM to mine it. - lolMiner will now print a warning, when it does not find its kernel files Here is the link to the update-pack: Klick MeMore updates to come next 2 weeks - I am doing good progress on the new 0.5 mining codes ----------------------------------------------- One RX 580 4Gb isnt worked.
In system: Intel HD + RX580
Hi there, can you check with new executable if the miner finds its kernel codes? Because from the memory output of clinfo the memory check should be passed and that two are the only two active checks that may cause the message (the one for compiling error looks different). ------------ I set up a friend with LolMiner a couple of days ago. Running fine and smiling except his miner has colors and mine is black and white. Was there an upgrade or am I missing something? Thanks.
What OS is your friend on? On Windows that can happen on older Windows versions, because the coloring uses WinAPI instructions that are in Win 10 ... hmm ... at least since the October 17 update. Here at home, Win 10 April update I see color, but on an older Win 7 system not, since the API calls are not supported there. On the other hand: Who is watching his rig 24/7 just because of the colors xD As long as its mining, all is good ------------ Im wtite pm with log of amd util.
Replied in private MSG. The 560 is detected as 2G, so 96/5 (MinexCoin) should work, the other two currently not. ------------ Hi, I am a developer of Bitcoin Candy (official website is cdy.one), can you add support for this coin? we are using Equihash 144_5 with pers CandyPoW.
Done ------------ Glad to see that the team continue their development to improve this miner, I used this recently to mine MNX, but seems the difficulty of MNX is climbed so high and not profitable, I hope that this version may add additional algo and coin to mine. Great job to lolMiner Developer!
Atm mining indeed is hard, but next pump will come *hope for that, too*. By the way: I unfortunately do not have a team - this is a one man show |
|
|
|
Feature request for nex version: Please allow for different config files, e.g. -CONFIGFILE=[json file] This will allow for easy inclusion in programs like www.MultiPoolMiner.ioYou do not need to wait, the parameter for this is -usercfg=<pathToFile> as described bottom of page 2 of the manual |
|
|
|
|
Show Posts
