Dok, thanks for providing an awesome miner.
1. Can you please share you power play table file which you are using with 18.6.1 drivers + Overdrivetool settings for crytonightV7 algo
2. What are your settings for heavy algo in overdrivetool, have you tested yet ?
Thanks
1. powerplay table : https://files.fm/u/s68u78zxOverdrivent: GPU 1450/925mV , MEM 950/910mV Im not looking to get the max hash from the card, just set a value that is stable and works for my miner development. 2. Same settings, as im only using it for testing the miner what i noticed is when connected with AnyDesk or TeamV the hashrate can't get to the max. What i do is restart vega driver, start miner, disconnect from anydesk, connect back in 10 minutes to check hashrate and voila, it's good  |
|
|
|
Allready trying... Please say in 1.6.3 on cn-v7 to got max speed need to wait some time? Or max speed is after start?
it should be in a few minutes max, on my hynix 580 8gb cards i get the same hash as on 1.6.0. Also on 470 8gb. Don't have samsung 580 8g to try. |
|
|
|
V1.6.3
- More speed on Vega56 cards on every algo (probably on Vega64 and VegaFE too, i don't have them to test)
- Added parameter --enableduplicategpuid (used in start.bat), that enables to use the same gpu id in gpu_conf multiple times
- Removed useless kernels 3 and 4
- Removed useless CPU opencl support
+ Vega cards should get even more speed than in previous version (try with adrenaline 18.3.4 or adrenaline 18.6.1)
+ Added parameter --enableduplicategpuid (in .bat) which basically allows you to do something like this :
"gpu_conf" :
[
{ "id" : 0, "intensity" : 120, "worksize" : 8, "threads" : 1},
{ "id" : 0, "intensity" : 120.5, "worksize" : 8, "threads" : 1},
]
So instead of using the same settings for every thread on same GPU, you can define different intensity for example for one thread.
This way maybe you can squeeze out a little more hash from your GPU.
On initialisation this is displayed as DIFFERENT GPU'S, but on hashrate display threads of same gpu are merged, so a per gpu hashrate is still displayed, not a per thread speed.
+ Removed a few useless things like kernels 3&4, and CPU OpenCL support, nobody used these but they just took me more time when implementing something new
What about cn-v7 algo on RX 5xx cards? You wanted to return speed from 1.6.0. On 1.6.1 and 1.6.2 cn-v7 algo on RX 580 is slower... you should try |
|
|
|
V1.6.3
- More speed on Vega56 cards on every algo (probably on Vega64 and VegaFE too, i don't have them to test)
- Added parameter --enableduplicategpuid (used in start.bat), that enables to use the same gpu id in gpu_conf multiple times
- Removed useless kernels 3 and 4
- Removed useless CPU opencl support
+ Vega cards should get even more speed than in previous version (try with adrenaline 18.3.4 or adrenaline 18.6.1)
+ Added parameter --enableduplicategpuid (in .bat) which basically allows you to do something like this :
"gpu_conf" :
[
{ "id" : 0, "intensity" : 120, "worksize" : 8, "threads" : 1},
{ "id" : 0, "intensity" : 120.5, "worksize" : 8, "threads" : 1},
]
So instead of using the same settings for every thread on same GPU, you can define different intensity for example for one thread.
This way maybe you can squeeze out a little more hash from your GPU.
On initialisation this is displayed as DIFFERENT GPU'S, but on hashrate display threads of same gpu are merged, so a per gpu hashrate is still displayed, not a per thread speed.
+ Removed a few useless things like kernels 3&4, and CPU OpenCL support, nobody used these but they just took me more time when implementing something new
Great stuff! I am getting 40H/s extra per Vega 56 with V1.6.3  That is what i measured too, increase on V7 about ~35HS. But listen i just installed 18.6.1 and got even more HS! +30 LOL So nice, this is a huge boost  |
|
|
|
|
V1.6.3
- More speed on Vega56 cards on every algo (probably on Vega64 and VegaFE too, i don't have them to test)
- Added parameter --enableduplicategpuid (used in start.bat), that enables to use the same gpu id in gpu_conf multiple times
- Removed useless kernels 3 and 4
- Removed useless CPU opencl support
+ Vega cards should get even more speed than in previous version (tested with adrenaline 18.3.4)
+ Added parameter --enableduplicategpuid (in .bat) which basically allows you to do something like this :
"gpu_conf" :
[
{ "id" : 0, "intensity" : 120, "worksize" : 8, "threads" : 1},
{ "id" : 0, "intensity" : 120.5, "worksize" : 8, "threads" : 1},
]
So instead of using the same settings for every thread on same GPU, you can define different intensity for example for one thread.
This way maybe you can squeeze out a little more hash from your GPU.
On initialisation this is displayed as DIFFERENT GPU'S, but on hashrate display threads of same gpu are merged, so a per gpu hashrate is still displayed, not a per thread speed.
+ Removed a few useless things like kernels 3&4, and CPU OpenCL support, nobody used these but they just took me more time when implementing something new
Don't copy new version of miner over old one, or if you do make sure you delete .srb files before running it!
|
|
|
|
I can confirm that I am getting about 100 less hash/s on CN7 using the SRB Miner 1.6.2 instead of 1.6.1. thanks. nobody believes me still using 1.6.1 because of that Who said nobody believes you ? Which gpu's are we talking about? Rx550 or Rx580 ? |
|
|
|
vega got more hash on 1.6.2, rx series maybe a little less, but next version should get things back on track.
I have a rig of 5 Vega Frontier Editions. I have not tried on heavy algo but the CN7 is giving me about 100 less. Intensity is set to 120 with double threads. I am very happy with SRB Miner. with XMR STAK, if I set intensity to 2048 for each thread then I would get closer to the hashrate of SRB but still 100 or so less. With SRB 1.6.1 I am getting a stable 10.9 KH/s, if I added another Vega FE card it would be 13 KH/s. You can see photos of my rig and its making here: https://postimg.cc/gallery/ivrtjh24/As you know Vega FE cards have 16 GB of HBM2. Is it possible to run more threads on these GPUs ? It would be awesome if there was a way to utilize this memory. I would appreciate if you could provide an answer to this question, as you are writing the code for SRB Miner and you know better how much memory can be used. Thanks for the awesome SRB Miner I don't have FE vega, but try this for V7 algo: "gpu_conf" : [ { "id" : 0, "intensity" : 126, "worksize" : 8, "threads" : 2} ] |
|
|
|
|
vega got more hash on 1.6.2, rx series maybe a little less, but next version should get things back on track.
|
|
|
|
doktor, this is happening mining bittubev2 in v1.6.2, in other rigs is the same.. after a time i got an hasdropped. Take a printscreen: https://ibb.co/mnEFboThanks you need to crank up the fans to 3000rpm because hbm gets too hot that is why your hash is jumping. |
|
|
|
When i launch the miner under 1.6 thru 1.62 i receive two messages stating "loading kernel for Device bus ID 1 ...." These messages are about 1 second appart. CryptonightV7 mode enabled. ADL initialised successfully and the proper pool name. Then the error appears indicating that the connection was refused.  Something at your side because :  as simple as : { "pools" : [ {"pool" : "xmr.pool.minergate.com:45700", "wallet" : "4A5hJyu2FvuM2azexYssHW2odrNCNWVqLLmzCowrA57xGJLNufXfzVgcMpAy3YWpzZSAPALhVH4Ed7x o6RZYyw2bUtbm12g", "password" : "x"} ] } |
|
|
|
Next version peek, Vega56 bittubev2 algo :  Dok, Have u using the softpowerplaytables? I only get this hashes with my gpu on 1592 Mhz, but consomes 150 watts anyway Thanks yup, powerplay table in registry |
|
|
|
Quite often I stuck on a "Compiling kernel for GPU1 ..." . Just one GPU has this problem. I have to reboot it several times, until it works. Not sure what happens. Does anyone else have this problem?
set only the problematic card in gpu_conf run miner it will nake binary then return all cards in conf. Maybe this will help |
|
|
|
In 1.6.2 on cn-v7 algo intensity higher 100 not give jumps and drops in hashrate like it was on 1.6.1. But speed is low on v7... I tried 54,56,72,90,108,116 but all these intensities gives max 985 h/s on my RX 580 8Gb cards. On 1.6.0 I have 1030-1040 h/s...
Maybe I do something wrong? I tried to wait up to 10 min every try. But speed didn't jumps like in heavy algo. It pretty stable. Couse of it I didn't wait any longer.
Who got speed higher 1000 h/s on v7 with 1.6.2? Share your settings, please...
Same issue: 1.6.0 gets 1080 h/s on RX 588, but 1.6.2 - only 1025 h/s with same config. Looks like that some optimisations that work nicely for vega degrade performance on rx. I will rework my kernel so the speed should be back to v1.6.0 speeds on rx. |
|
|
|
Dok - you need to set your Vega 56 fans to minumum 3000RPM because the card will throttle itself if the HBM2 memory gets too hot.
i know, but this was just a quick test to show off kid was sleeping didnt want to turn on the turbine |
|
|
|
Next version peek, Vega56 bittubev2 algo : |
|
|
|
considering that all closed source miners almost certainly contain open source code or code derived from open source projects and is precisely how crypto assets have come to be.
So if it is like you say, why did you even have to ask me when it looks like to me you already gave the answer before making the question? Also, it's a little strange that you suddenly came from your fpga threads here, with your licensing question. So here is an answer i hope it will satisfy you : I am not using copy pasted code from open source projects, but i am using some good ideas, reworked in my own style. this is likely one of the reasons why you have had to close source yours The reason is obvious, if you follow the open source scene (oh you do, you wouldn be here now if you didn't), then you know that there are always 'good helping people' who remove the devfee, compile, upload and share on forums the 0% fee version. |
|
|
|
I'm not he. No affiliation. Thought it pertinent to ask and for you to answer that question publicly. As a developer, I as sure that you agree that too many miners developers are taking GPL code and using it outside of the license and this is likely one of the reasons why you have had to close source yours and other developers like Imperdin may end up having to close sourcing his.
Thank you for coming here to tell me this.. But im curious, what does he have that he can close source ? |
|
|
|
guys what is your 30 min average speed in statistics 1.6.0 vs 1.6.2?
1080 vs 1024 Strange, really. I have 580 8g (hynix) cards i tested the miner on , and i get the same hashrate on V7 as on previous versions. There are always HR fluctuations, and always will be, that's why i ask for the 30 min average, and not the value you get when pressing 'h'. Same rig, same driver, same everything except miner version? I will check to see if any of the changes i made could cause this HR drop. |
|
|
|
In 1.6.2 on cn-v7 algo intensity higher 100 not give jumps and drops in hashrate like it was on 1.6.1. But speed is low on v7... I tried 54,56,72,90,108,116 but all these intensities gives max 985 h/s on my RX 580 8Gb cards. On 1.6.0 I have 1030-1040 h/s...
Maybe I do something wrong? I tried to wait up to 10 min every try. But speed didn't jumps like in heavy algo. It pretty stable. Couse of it I didn't wait any longer.
Who got speed higher 1000 h/s on v7 with 1.6.2? Share your settings, please...
Same issue: 1.6.0 gets 1080 h/s on RX 588, but 1.6.2 - only 1025 h/s with same config. guys what is your 30 min average speed in statistics 1.6.0 vs 1.6.2? |
|
|
|
V1.6.2
- Added support for Italocoin new algo (from block 55.000)
- Auto intensity for Vega cards improved, also managed to increase hashing speed on Vega cards for about ~1-1.5% (mainly on heavy algos)
- If user uses non existing GPU id in gpu_conf, it will be ignored, no error will be thrown
- Fixed miner crash when using a non AES capable CPU on some algos
- Added Windows version and build in log
- Added video driver version in log
- Added info about CPU AES support in log
- Statistics now shows number of stale shares that were accepted by pool
- Added parameter 'min_rig_speed_duration' that can set period for 'min_rig_speed' parameter (minimum is 30 sec), default is 5 min
+ I finally bought a Vega 56 card, so in the next releases i will try to improve the performance for Vegas. For now a little speedup of about 1-1.5% is achieved, mainly on Heavy based algos (Heavy, Haven, Bittube, Italo)
+ If you use a non existant id in gpu_conf, it will now be ignored, only id's that exist in your system will be used
+ There was a bug in soft AES that caused miner crash on Stellite and Haven algos, this is now fixed
+ Added accepted stale shares number in statistics. It can't know the number of real accepted stale shares on pool side, but the logic is simple, if a new job is received, but the miner sends a result in the mean time that is for the previous job, and this share gets accepted by the pool, miner will treat this like an accepted stale share.
+ min_rig_speed_duration can set the time you want to check the average hashrate for 'min_rig_speed' parameter. Default is 5 min.
Looks like you included Imperdin's heavy optimizations for that small bump. Did you include that GPL3 code in your miner? It just looks like, but dont worry imperdin, as i already told you i dont need to use your crazy insane mega giga optimisations. I already got even more speed increase since we talked, so if you plan releasing another mega giga optimisation please dont do it before i release the next version of miner. |
|
|
|
|
Show Posts
