[XMR] JCE Miner Cryptonight/forks, now with GPU! - page 33.

TigTex

jr. member

Activity: 55

Merit: 5

Quote from: JCE-Miner on November 09, 2018, 06:32:12 PM

Please read the full doc about how to setup a different config for each algo:
https://github.com/jceminer/cn_cpu_miner#pool-managed-autoswitch

Hmm, I was reading the cn_gpu_miner github page and that info is missing from there. My bad, didn't noticed that there are 2 projects
Thank you for your fast and kind reply. I'll look into it

JCE-Miner

member

Activity: 350

Merit: 22

Quote from: Lermite on November 09, 2018, 06:39:16 PM

I still have hashrate droops with the 0.33b6.
One has just occurs on my RX 570 4G after two hours of mining CN V8: from 1100 to 965 h/s.
It's the same hashrate droop I had with the b4 and b5 versions.

I fixed drops on CN-Heavy (and related) only. If i have good feedbacks about it, so i'll put it on V8 too.

JCE-Miner

member

Activity: 350

Merit: 22

What you need is already available to script this:

* Do short tests: parameter --autoclose N to close after N minutes (the 1st minute has no fee)
* Generate a different config for each test: just generate the JSON config, it's a piece of standard JSON
* Read the hashrate: the JSON output is perfect, or use --log then parse the file

and repeat repeat until you get the optimum.

Lermite

newbie

Activity: 54

Merit: 0

I still have hashrate droops with the 0.33b6.
One has just occurs on my RX 570 4G after two hours of mining CN V8: from 1100 to 965 h/s.
It's the same hashrate droop I had with the b4 and b5 versions.

Uaciuganadu

newbie

Activity: 39

Merit: 0

Hey JCE,

Would you be able to create a benchmark flag? For example we can use it for testing all the compatible variations of intensity in order to find the best one for a specific algo. It can run for 2min at a time for eatch variation and export them to TXT at the end. It can take a while but it can find unexpected variations that no one ever atempted to try and will take the guessing out of the configuration.

JCE-Miner

member

Activity: 350

Merit: 22

Github page: that's on first page of this topic
direct: https://github.com/jceminer/cn_gpu_miner

Quote

Again I need the config for a ryzen 1600 for cnv7

I answered just above

Quote

And that's with usage of FX4330(4-core, 8M, 4GHz)

CN-Heavy consumes 4M of cache, better use two threads with multi-hash 1

Quote

pool managed autoswitching

Please read the full doc about how to setup a different config for each algo:
https://github.com/jceminer/cn_cpu_miner#pool-managed-autoswitch

Three config arrays for three algo class (Light, normal, Heavy)

Quote

multiple algos on config.txt

That's exactly how it works. Example from the doc:

Code:

/* This is the configuration when Cryptolight class algo is mined */
"cryptolight_cpu_threads_conf" :
[
{ "cpu_architecture" : "auto", "affine_to_cpu" : 0, "use_cache" : true, "multi_hash":1 },
{ "cpu_architecture" : "auto", "affine_to_cpu" : 1, "use_cache" : true, "multi_hash":1 },
{ "cpu_architecture" : "auto", "affine_to_cpu" : 2, "use_cache" : true, "multi_hash":1 },
{ "cpu_architecture" : "auto", "affine_to_cpu" : 3, "use_cache" : true, "multi_hash":1 },
],

/* This is the configuration when Cryptonight class algo is mined */
"cryptonight_cpu_threads_conf" :
[
{ "cpu_architecture" : "auto", "affine_to_cpu" : 0, "use_cache" : true, "multi_hash":1 },
{ "cpu_architecture" : "auto", "affine_to_cpu" : 2, "use_cache" : true, "multi_hash":1 },
],

/* This is the configuration when CN-Heavy class algo is mined */
"cn_heavy_cpu_threads_conf" :
[
{ "cpu_architecture" : "auto", "affine_to_cpu" : 0, "use_cache" : true, "multi_hash":1 },
],

And adding a hint for the pool to select your most profitable algo is planned, it has been requested in a Github ticket

TigTex

jr. member

Activity: 55

Merit: 5

Hello

I'm testing your miner on 5 rx580 4gb on moneroocean.
Since 0.33c you support "pool managed autoswitching" but it doesn't seems to work like the other miners
For example, I would like to set each algo performance so that the pool always mines the most profitable coin. Also, if I use the same config for the different algorithms, the miner will run out of memory (for example, cn8 intensity cant be used on a cn-heavy algo). It needs to support multiple configuration files or multiple algos on config.txt
Is this going to be implemented?

1033ruben

newbie

Activity: 22

Merit: 0

Quote from: JCE-Miner on November 09, 2018, 02:36:17 PM

Thanks all for report and sharing your optimal config. Don't forget to tell what algo you're mining Wink

so i can put those config in the Github doc.

well maybe someone can send me a link to this.
THANKS
P.S. Again I need the config for a ryzen 1600 for cnv7. the reasin why i ask is because as of right now am only getting like high 200's with this miner and am getting like high 400s with xmr stak.

BS0D

newbie

Activity: 29

Merit: 0

The last point I've reached is quite stable heavy 1920-1930h/s, single Vega 64 (1478/920 1100/910). But still I don't understand what it depends on.
{ "mode" : "GPU", "worksize" : 8, "alpha" : 68, "beta" : 64, "gamma" : 4, "delta" : 4, "epsilon" : 4, "zeta":4, "index" : 1, "multi_hash":960 },
{ "mode" : "GPU", "worksize" : 8, "alpha" : 68, "beta" : 16, "gamma" : 4, "delta" : 4, "epsilon" : 4, "zeta":4, "index" : 1, "multi_hash":1008 }
About 80-100h/s appears when playing with alpha 64-68 on threads.
It can start with rising to 1950-1970 but then it falls to 1830-1840 (which is very good too). But it can also start with straight 1920 and work with stable 1920-1930h/s with no issues.
And that's with usage of FX4330(4-core, 8M, 4GHz) in that same config:
{ "cpu_architecture": "auto", "affine_to_cpu": 0, "use_cache": true, "multi_hash": 1 },
{ "cpu_architecture": "auto", "affine_to_cpu": 3, "use_cache": true, "multi_hash": 2 }
which does 120h/s.
I think it's all about some megabytes of memory usage.

vmozara

member

Activity: 190

Merit: 59

Quote from: Iamtutut on November 09, 2018, 05:28:48 PM

My GPUs could get through the warmup, still with swings (mostly RAM frequency, sometimes core freq). Declared speed in itself is highly improved compared to my reference JCE 0.32G version, +15% for same power draw (at wall). If my computer stay stable during the night, I'll check the reported HR at the pool.

We don´t have same cards, but I did 3 things at same time that equalized and stabilized hashrate over my rigs:
1; Changed multihash from 960 to 896 (vega cards)
2. Changed miner to low priority ( --low). My CPU is very slow (celeron 3855U) and probaby has issues when any load is applied to it
3. Updated from b5 to b6

try fiddling with any of these settings, maybe it gets stable

vmozara

member

Activity: 190

Merit: 59

b6 version:
vega rig i think all 7 cards are 56@64, platinum PSU
custom 880 ppt, 1450/1100 forgot the driver version, multihash 896

total hashrate 12120 pulling 1120W from the wall. This are great figures that put heavy algos highly on the profitability list. This is my first epxerience with heavy algos, i don´t know any coins (mining bit tube now for testing) but if this proves good and stable for my vega machines, i will probably switch all rigs to reduce the temperatures and strain on my rigs.

not sure if I want to leave it running overnight, as i have no access to internet once i leave the workshop.

Please share your full setup when writing about hashrates

JCE thanks for awesome miner, please keep developing, you think you can squeeze more hash out from heavy?

Iamtutut

full member

Activity: 1120

Merit: 131

Quote from: JCE-Miner on November 09, 2018, 04:31:08 PM

It's online 0.33b6 GPU

I mined Haven for 48 hours on 3 capricious RX and had a very stable hashrate, varying +/- 2% not more, I hope it will be better for everybody.

My GPUs could get through the warmup, still with swings (mostly RAM frequency, sometimes core freq). Declared speed in itself is highly improved compared to my reference JCE 0.32G version, +15% for same power draw (at wall). If my computer stay stable during the night, I'll check the reported HR at the pool.

salamor

newbie

Activity: 1

Merit: 0

V0.33.B4:
rx 564 - ~580 h
rx 584 - ~760 h
rx 588 - ~1030 h
нecтaбильный зaпycк, мaкc xeш co 2-4 пoпытки

V0.33b5:
rx 564 - ~450 h
rx 584 - ~930 h
rx 588 - ~900 h
нecтaбильный зaпycк, мaкc xeш co 2-4 пoпытки

i3 - 8350k вceгдa - ~143h

измeнeниe config нe пoмoгaeт

JCE-Miner

member

Activity: 350

Merit: 22

It's online 0.33b6 GPU

I mined Haven for 48 hours on 3 capricious RX and had a very stable hashrate, varying +/- 2% not more, I hope it will be better for everybody.

Code:

Fixed the OpenCL bug 0-2.10
Stabilized hashrate
Updated doc
Updated autoconfig for older cards
Little perf boost
Temperature watchdog (default: disabled)

Quote from: 1033ruben on November 09, 2018, 04:28:57 PM

Will any one be so kind as to share their config file for CNV7 for a ryzen 5 1600 cpu.
THANKS IN ADVANCE

Code:

"cpu_threads_conf" :
[
   { "cpu_architecture" : "auto", "affine_to_cpu" : 0, "use_cache" : true, "multi_hash":1 },
   { "cpu_architecture" : "auto", "affine_to_cpu" : 2, "use_cache" : true, "multi_hash":1 },
   { "cpu_architecture" : "auto", "affine_to_cpu" : 4, "use_cache" : true, "multi_hash":1 },
   { "cpu_architecture" : "auto", "affine_to_cpu" : 6, "use_cache" : true, "multi_hash":1 },

   { "cpu_architecture" : "auto", "affine_to_cpu" : 8, "use_cache" : true, "multi_hash":1 },
   { "cpu_architecture" : "auto", "affine_to_cpu" : 10, "use_cache" : true, "multi_hash":1 },
   { "cpu_architecture" : "auto", "affine_to_cpu" : 11, "use_cache" : true, "multi_hash":1 },
   { "cpu_architecture" : "auto", "affine_to_cpu" : 1, "use_cache" : true, "multi_hash":1 },
],

1033ruben

newbie

Activity: 22

Merit: 0

Will any one be so kind as to share their config file for CNV7 for a ryzen 5 1600 cpu.
THANKS IN ADVANCE

dnkn

newbie

Activity: 11

Merit: 0

Unfortunately it has happened again. Hashrate drop. Cruel hashrate drop. From 26-27kh to 20.5-22. This was at the first day I tried 0.33b5 but then I tried different settings and everything seemed fine until several hours ago. I noticed that my hashrate on pool decreased and I noticed that effective hashrate reported by jce has decreased too. I tried to restart miner/reboot rigs, nothing helps. I'll wait for the next version. I'm sure it'll be fixed.

JCE-Miner

member

Activity: 350

Merit: 22

Quote from: UnclWish on November 09, 2018, 02:35:14 PM

I don't think that on HD6990 or 390 mining is profitable at this moment... Power costs more than these cards can mine...

If my hybrid miner works, it may resurect the old card mining, like I resurected Core2 and Opteron mining with JCE CPU on V7. I admit they lost a lot of efficiency on v8. Cry

JCE-Miner

member

Activity: 350

Merit: 22

I'm claiming to be the fastest on old GPU and I don't even provide a decent autoconfig for old cards, what a shame...
I'll update some autoconf for those cards like Cayman, Tahiti, Bonaire or Pitcairn.

Perf on HD6990:
https://github.com/jceminer/cn_gpu_miner/issues/26

This guy shared his config that goes to 270 h/s per GPU (HD6990 is a dual GPU) with no memory strip hack (which is lame, i will advise him to patch his BIOS).

Quote

Settings used : 32 worksize and 864 intensity.

I guess that's on one thread per GPU, the dual threading is for GCN cards.
It translates:

Code:

{ "mode" : "GPU", "worksize" : 32, "alpha" : 64, "beta" : 8, "index" : ...., "multi_hash":864 }

Thanks all for report and sharing your optimal config. Don't forget to tell what algo you're mining Wink

so i can put those config in the Github doc.

Quote

Tuning is just a bit of a hobby of mine

I usualy get complains about the parameters being too numerous (there are four). Not to few.
Greeks above beta are deprecated, worksize is very like in XMRstak, multi_hash is like XMRstak intensity (but unlike SRB intensity) and alpha/beta are internal data packing parameters, no general rule, sometimes doing x2 or /2 on the defaults gives a little boost. Depends on the card and algo you mine. My autoconfig always defaults to 64 and 8.

UnclWish

sr. member

Activity: 1484

Merit: 253

I don't think that on HD6990 or 390 mining is profitable at this moment... Power costs more than these cards can mine...

varusisog

member

Activity: 123

Merit: 10

I have a few R9 390/390X. What is the optimum configuration?

Topic: [XMR] JCE Miner Cryptonight/forks, now with GPU! - page 33. (Read 90841 times)