[JCE]Fast & stable CN/v8/Heavy/Tube/XHV miner, CPU+GPU, Vega56 1800+ RX580 1200+

[heavyarms1912]

P.S. Looks like at start and during several minutes in heavy algo speed every time differs...

I don't get this behavior while mining cn-saber.  I haven't tried cn-heavy.  Will report back.

EDIT: Yup I get fluctuating hash rates.  Kind of swings between 50-75 H/s range from avg total.

[1713495791]

1713495791

[1713495791]

1713495791

[geck]

Just tried out 32a with the auto function and im getting CL_MEM_ALLOCATION_ERROR on one thread. The thread number that fails is random. Reducing the multihash value to 1840 solves it but it seems like there is a lot of speed left on the table

I have tried reducing the memory clocks all the way down but it still fails. This is on an 8x vega56 rig.

[ZCXND]

0.32a no more bad shers  
550/460 no difference compare to 0.31f
0.32a on vega 64 CC ~1407 MC 1099 i got 2153 h/s v7 wich this config :
     { "mode" : "GPU", "worksize" : 8, "alpha" : 64, "beta" : 4, "gamma" : 4, "delta" : 4, "epsilon" : 4, "zeta" : 4, "index" : x, "multi_hash":2000 },
     { "mode" : "GPU", "worksize" : 8, "alpha" : 64, "beta" : 4, "gamma" : 4, "delta" : 4, "epsilon" : 4, "zeta" : 4, "index" : x, "multi_hash":2000 },
on 0.31f i can get 2135 h/s wich :
     { "mode" : "GPU", "worksize" : 8, "alpha" : 64, "beta" : 4, "gamma" : 4, "delta" : 4, "epsilon" : 4, "zeta" : 4, "index" : x, "multi_hash":1968 },
     { "mode" : "GPU", "worksize" : 8, "alpha" : 64, "beta" : 4, "gamma" : 4, "delta" : 4, "epsilon" : 4, "zeta" : 4, "index" : x, "multi_hash":1968 },
when i increase multi_hash above 1968 on 0.31f hashrate get lower
drive 18.5.1

[JCE-Miner]

Both memory model and openCL code of version 0.32+ are completely new.
It's somehow like Claymore 9 and 10+ : maybe better, maybe worse, but completely different. I cannot go back to the long-compiling versions, now i must still improve the new code.

Good news !

Code:

Detecting OpenCL-capable GPUs...
Calling clGetPlatformIDs
Returned CL_UNKNOWN_ERROR -1001
Found 0 OpenCL PlatformIDs
0█└☻

I can reproduce the no-gpu bug on my Win7+betablockchain rig i just installed, and on which i'm writing from right now. I didn't find the cause yet, but i can reproduce!

edit: Bad news this time
I found the reason it failed on Win7 is that my protection technique of injecting my opencl code directly into the drivers, bypassing the normal way of calling OpenCL compiler, to prevent hackers to dump my code on the fly, is just not compatible with Win7. I've the choice to give up security, or to giveup Win7 support. I couldn't find any way to keep both, so I drop Win7 support on the GPU part (cpu mining still works down to Vista 32).

Sorry JuanHungLo for the bad ending, but at least i fully understood the problem. Other miners work because they use a clear openCL api, I use a hacky one, on purpose, for security.
I'll update my documentation to state Win7 is not supported. I tested on Win8.1 and it works fine.

[UnclWish]

Both memory model and openCL code of version 0.32+ are completely new.
It's somehow like Claymore 9 and 10+ : maybe better, maybe worse, but completely different. I cannot go back to the long-compiling versions, now i must still improve the new code.

Good news !

Code:

Detecting OpenCL-capable GPUs...
Calling clGetPlatformIDs
Returned CL_UNKNOWN_ERROR -1001
Found 0 OpenCL PlatformIDs
0█└☻

I can reproduce the no-gpu bug on my Win7+betablockchain rig i just installed, and on which i'm writing from right now. I didn't find the cause yet, but i can reproduce!

edit: Bad news this time
I found the reason it failed on Win7 is that my protection technique of injecting my opencl code directly into the drivers, bypassing the normal way of calling OpenCL compiler, to prevent hackers to dump my code on the fly, is just not compatible with Win7. I've the choice to give up security, or to giveup Win7 support. I couldn't find any way to keep both, so I drop Win7 support on the GPU part (cpu mining still works down to Vista 32).

Sorry JuanHungLo for the bad ending, but at least i fully understood the problem. Other miners work because they use a clear openCL api, I use a hacky one, on purpose, for security.
I'll update my documentation to state Win7 is not supported. I tested on Win8.1 and it works fine.

Maybe it's even better. Windows 10 often gives more mining speed than Win7. Windows 10 with at least 4Gb vmemory works with the same speed as Win7, and often quicker. I not understand why use Windows 7 at all.
For these, who affraid Win10 spying and many unuseful options and apps, exists LTSB version of Windows 10 wich have minimal size and don't have many apps... Just google Windows 10 LTSB.

[UnclWish]

JCE, i can't understand, why miner can't allocate all of 4Gb vmemory of my 270X. Max it can use is 3,5Gb. If I set multihash to use higher than 3Gb, miner gives CL_memory allocate error...
In info about devices at start miner shows that card have 4Gb vmemory...

[JuanHungLo]

Both memory model and openCL code of version 0.32+ are completely new.
It's somehow like Claymore 9 and 10+ : maybe better, maybe worse, but completely different. I cannot go back to the long-compiling versions, now i must still improve the new code.

Good news !

Code:

Detecting OpenCL-capable GPUs...
Calling clGetPlatformIDs
Returned CL_UNKNOWN_ERROR -1001
Found 0 OpenCL PlatformIDs
0█└☻

I can reproduce the no-gpu bug on my Win7+betablockchain rig i just installed, and on which i'm writing from right now. I didn't find the cause yet, but i can reproduce!

edit: Bad news this time
I found the reason it failed on Win7 is that my protection technique of injecting my opencl code directly into the drivers, bypassing the normal way of calling OpenCL compiler, to prevent hackers to dump my code on the fly, is just not compatible with Win7. I've the choice to give up security, or to giveup Win7 support. I couldn't find any way to keep both, so I drop Win7 support on the GPU part (cpu mining still works down to Vista 32).

Sorry JuanHungLo for the bad ending, but at least i fully understood the problem. Other miners work because they use a clear openCL api, I use a hacky one, on purpose, for security.
I'll update my documentation to state Win7 is not supported. I tested on Win8.1 and it works fine.

Thank you for your extra time and research.  I still continue to use your miner on all of my Win10-Vega56 rigs.  Awesome work.  Now I need to consider the upgrade to LTSB version of Windows 10 on a dozen or so rigs.  Hmmm.

[dingdongtobias]

Both memory model and openCL code of version 0.32+ are completely new.
It's somehow like Claymore 9 and 10+ : maybe better, maybe worse, but completely different. I cannot go back to the long-compiling versions, now i must still improve the new code.

Good news !

Code:

Detecting OpenCL-capable GPUs...
Calling clGetPlatformIDs
Returned CL_UNKNOWN_ERROR -1001
Found 0 OpenCL PlatformIDs
0█└☻

I can reproduce the no-gpu bug on my Win7+betablockchain rig i just installed, and on which i'm writing from right now. I didn't find the cause yet, but i can reproduce!

edit: Bad news this time
I found the reason it failed on Win7 is that my protection technique of injecting my opencl code directly into the drivers, bypassing the normal way of calling OpenCL compiler, to prevent hackers to dump my code on the fly, is just not compatible with Win7. I've the choice to give up security, or to giveup Win7 support. I couldn't find any way to keep both, so I drop Win7 support on the GPU part (cpu mining still works down to Vista 32).

Sorry JuanHungLo for the bad ending, but at least i fully understood the problem. Other miners work because they use a clear openCL api, I use a hacky one, on purpose, for security.
I'll update my documentation to state Win7 is not supported. I tested on Win8.1 and it works fine.

Maybe it's even better. Windows 10 often gives more mining speed than Win7. Windows 10 with at least 4Gb vmemory works with the same speed as Win7, and often quicker. I not understand why use Windows 7 at all.
For these, who affraid Win10 spying and many unuseful options and apps, exists LTSB version of Windows 10 wich have minimal size and don't have many apps... Just google Windows 10 LTSB.

all this protection would make sense if your miner was 20,30,50% faster than the others. But.its not the case 

[geck]

wanted to add, awesomeminer is reporting a much lower hashrate with 0.32a compared to 0.31f. the miner shows 15k but awesomeminer reports around 11k.

[Iamtutut]

all this protection would make sense if your miner was 20,30,50% faster than the others. But.its not the case 

That's his code, his intellectual property. He does whatever he wants with and his miner is awsome. Other CN miner progammers are not idiots, being even 1-2% consistently faster is huge.

[JCE-Miner]

Thanks,

i did more test on my A8 apu i just installed, and… i don't get what amd did.
I had to go down to old 15.7 drivers to find a version with OpenCL support. later version supports the APU but not its OpenCL (just directx).
of course with such old drivers, no real GPU works, RX500 need decently recent drivers.

but, with those old drivers… it runs, the APU is detected as Devastator and it mines. i didn't try to tune, i got a valid share and stopped.

i still need to make the special APU version.

[Iamtutut]

Thanks,

i did more test on my A8 apu i just installed, and… i don't get what amd did.
I had to go down to old 15.7 drivers to find a version with OpenCL support. later version supports the APU but not its OpenCL (just directx).
of course with such old drivers, no real GPU works, RX500 need decently recent drivers.

but, with those old drivers… it runs, the APU is detected as Devastator and it mines. i didn't try to tune, i got a valid share and stopped.

i still need to make the special APU version.

The Q2 drivers are supposed to work with the APUs, that's what a senior AMD engineer wrote on his twitter, he claimed they are the first drivers unified for RXs and APUs.

[JCE-Miner]

it looks it supports the Ryzen G apu, i need support of the A-series APU (the HD7000D), which are no longer supported by amd opencl since 2015

to answer above about 4G memory : not the full 4G can be allocated, the mining program and the drivers take some room. GPUZ gives info about max opencl allocation. on 4G with a screen (or dummy plug) on, it may be limited to ~3.5G
don't forget to setup the environment variable GPU_MAX...


geck: i didn't even know Awesome supports jce. look at your pool, you'll see i don' cheat 35% of your shares…

i don't know what Awesome get its numbers from. the Effective hashrate is avaible in the native json output.

[Iamtutut]

it looks it supports the Ryzen G apu, i need support of the A-series APU (the HD7000D), which are no longer supported by amd opencl since 2015


i don't know what Awesome get its numbers from. the Effective hashrate is avaible in the native json output.

I does (accord to the AMD engineer), you can also try the 18.5.1 that officially supports older APUs + Ryzen G + RX

[Uaciuganadu]

Hey JCE,

Can you please add more data to the R command and the API? It lacks statistics like the number of invalid shares or the errors. This is vital for testing stability on a rig.

The more detail the better and we should be able to see every possible detail related to the functionality and stability so we can analyze how our rigs perform on the long run and set them up accordingly.

This in my opinion this should be one of the biggest priorities, not the automatic config, as most miners already know how to test the rigs. We need to find rock stable versions from the mining code standpoint and to do that we need to report back invalid shares or strange errors and see if they are related to the miner code or to an unstable OC/config.

Please also if possible create an option that can log ONLY the errors/invalid shares and strange behavior in a pre-determined TXT file for later analysis.

Love the new and fast start but it can still be improved a lot.


Thank you for the hard work.

[geck]

it looks it supports the Ryzen G apu, i need support of the A-series APU (the HD7000D), which are no longer supported by amd opencl since 2015

to answer above about 4G memory : not the full 4G can be allocated, the mining program and the drivers take some room. GPUZ gives info about max opencl allocation. on 4G with a screen (or dummy plug) on, it may be limited to ~3.5G
don't forget to setup the environment variable GPU_MAX...


geck: i didn't even know Awesome supports jce. look at your pool, you'll see i don' cheat 35% of your shares…

i don't know what Awesome get its numbers from. the Effective hashrate is avaible in the native json output.

I'm not saying you cheat the shares. I'm thinking there maybe something wrong with the hash rate reporting via api to awesomeminer?

awesomeminer picks up the right hashrate with 0.31f but something must have changed with 0.32a. It isn't a big deal, i only need the hash rate reading as i set up rules to restart the miner

[lebuawu2]

Hi JCE-Miner,

Please help to add temp and fan speed for each GPU, that information really useful to make sure my GPU not too hot.

Thanks.

[UnclWish]

And it's need hw controls to overclock, undervolt, fan control, etc...

[Lermite]

You still can use some monitoring tool such HWiNFO.

https://puissanceled.com/vrac/HWiNFO_RX.png

[justoiurii]

Hello guys! Could someone help with right .bat and .config? I'd like to test using rig with six RX480 8G, blockch. driver. Thanks in advance)