I'll give it a try!
Topic: [XMR] JCE Miner Cryptonight/forks, now with GPU! - page 63. (Read 90814 times)
Thank you vmozara and kapitulieren
I'll give it a try!
I'll give it a try!
Could you tell us which amd driver, config, algo, and hashrate you have?
I'm very curious because last time I tested JCE-miner, it was very unstable! So I'm still running SRBminer with 4 vega 56, and it's very stable so far!
I'm very curious because last time I tested JCE-miner, it was very unstable! So I'm still running SRBminer with 4 vega 56, and it's very stable so far!
Adrenalin driver 18.6.1 or similar depending on the rig
Windows 10 1803
Rig 7x Vega 56 1408/800 1100/80, multihash 1968 on JCE miner
Hashrate 14400H/s effective
Consumption 1250W from the wall
Rig 6x Vega 64
same config
Hashrate 12700 effective at 1130W
All rigs using octominer riserless motherboards and server power supplies
Experienced no hashrate drop or any other issue associated with other miners
Discovered JCE since first GPU version and fell in love
The worst part about JCE-miner for Vega rigs is that it is so stable - that I feel so bad to shut down 600 hours uptime rig just to update the miner version. JCE, we need hot swap updates! LOL! 
Could you tell us which amd driver, config, algo, and hashrate you have? I'm very curious because last time I tested JCE-miner, it was very unstable! So I'm still running SRBminer with 4 vega 56, and it's very stable so far!
I'm also using JCE miner with great results on my 6x vega 56 rig.
6x Sapphire pulse vega 56 - 1408 core 905 mV. / 3 with 950 memory 900 mV & 3 with 900 memory
Radeon adrenalin 18.5.1 drivers
win 10 1803
Monero algo cryptonightv1
https://imgur.com/awMAjgt current hashrate around 11950 h/s ( pool is showing 11.7-11.9k per 24h)
Gpu config is
{ "mode" : "GPU", "worksize" : 8, "alpha" : 64, "beta" : 8, "gamma" : 4, "delta" : 4, "epsilon" : 4, "zeta" : 4, "index" : 0, "multi_hash":1936 },
The worst part about JCE-miner for Vega rigs is that it is so stable - that I feel so bad to shut down 600 hours uptime rig just to update the miner version. JCE, we need hot swap updates! LOL!
Could you tell us which amd driver, config, algo, and hashrate you have? I'm very curious because last time I tested JCE-miner, it was very unstable! So I'm still running SRBminer with 4 vega 56, and it's very stable so far!
The worst part about JCE-miner for Vega rigs is that it is so stable - that I feel so bad to shut down 600 hours uptime rig just to update the miner version. JCE, we need hot swap updates! LOL!
--any --variation 3
for v7
--help to get the list
for v7
--help to get the list
what is the Variation number for crptonight v7 algo?. im trying to mine an unknown coin which has cryptonight v7 algo.
Hi JCE,
Why my rig show temp and fan always 0c and 0%? Did i miss something?
Why my rig show temp and fan always 0c and 0%? Did i miss something?
maybe you should describe more which OS (Windows 10 build 1709), driver (18.3.4), GPU (RX 580 8GB) so dev can find the root cause more easily.
for me with the details above it work fine.
Windows 10 LTSB and 1709, Driver 18.4.1 and 18.6.1, RX Vega 56
Code:
"gpu_status":
[
{ "index": 0, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 29, "bad_shares": 0 },
{ "index": 1, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 39, "bad_shares": 0 },
{ "index": 2, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 43, "bad_shares": 0 },
{ "index": 3, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 38, "bad_shares": 0 },
{ "index": 4, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 35, "bad_shares": 0 },
{ "index": 5, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 30, "bad_shares": 0 },
{ "index": 6, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 27, "bad_shares": 0 },
{ "index": 7, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 42, "bad_shares": 0 }
[
{ "index": 0, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 29, "bad_shares": 0 },
{ "index": 1, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 39, "bad_shares": 0 },
{ "index": 2, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 43, "bad_shares": 0 },
{ "index": 3, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 38, "bad_shares": 0 },
{ "index": 4, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 35, "bad_shares": 0 },
{ "index": 5, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 30, "bad_shares": 0 },
{ "index": 6, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 27, "bad_shares": 0 },
{ "index": 7, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 42, "bad_shares": 0 }
Double check to make sure that your AMD drivers are in COMPUTE MODE, they should display gfx901, NOT gfx900.
Yes, I ues "switch-radeon-gpu" to COMPUTE MODE, gfx900 is my Vega64, It does't work
Manual enable HBM
I did no fix for the GPU fan/temp in the h version.
Xeons: sure, it was a blind fix, i've no dual xeon to test on, can you try again with --probe to see if the numbers have changed (still numbered 0-15 + 64-79 or 0-31). Thanks for report.
I tried drivers 18.8.1 on one rig and i needed to lower a lot the multi_hash (from 464 to 432) on my RX560 not to make it crash. I still prefer 18.6.1
Xeons: sure, it was a blind fix, i've no dual xeon to test on, can you try again with --probe to see if the numbers have changed (still numbered 0-15 + 64-79 or 0-31). Thanks for report.
I tried drivers 18.8.1 on one rig and i needed to lower a lot the multi_hash (from 464 to 432) on my RX560 not to make it crash. I still prefer 18.6.1
I tried 32h with dual xeon, same error, CPU ID cannot exceed 16
0.32h online
For windows, 32/64 and GPU
For windows, 32/64 and GPU
API doesn't change
Hi JCE,
Why my rig show temp and fan always 0c and 0%? Did i miss something?
Why my rig show temp and fan always 0c and 0%? Did i miss something?
maybe you should describe more which OS (Windows 10 build 1709), driver (18.3.4), GPU (RX 580 8GB) so dev can find the root cause more easily.
for me with the details above it work fine.
Windows 10 LTSB and 1709, Driver 18.4.1 and 18.6.1, RX Vega 56
Code:
"gpu_status":
[
{ "index": 0, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 29, "bad_shares": 0 },
{ "index": 1, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 39, "bad_shares": 0 },
{ "index": 2, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 43, "bad_shares": 0 },
{ "index": 3, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 38, "bad_shares": 0 },
{ "index": 4, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 35, "bad_shares": 0 },
{ "index": 5, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 30, "bad_shares": 0 },
{ "index": 6, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 27, "bad_shares": 0 },
{ "index": 7, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 42, "bad_shares": 0 }
Double check to make sure that your AMD drivers are in COMPUTE MODE, they should display gfx901, NOT gfx900.
Yes, I ues "switch-radeon-gpu" to COMPUTE MODE, gfx900 is my Vega64, It does't work
Hi JCE,
Why my rig show temp and fan always 0c and 0%? Did i miss something?
Why my rig show temp and fan always 0c and 0%? Did i miss something?
maybe you should describe more which OS (Windows 10 build 1709), driver (18.3.4), GPU (RX 580 8GB) so dev can find the root cause more easily.
for me with the details above it work fine.
Windows 10 LTSB and 1709, Driver 18.4.1 and 18.6.1, RX Vega 56
Code:
"gpu_status":
[
{ "index": 0, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 29, "bad_shares": 0 },
{ "index": 1, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 39, "bad_shares": 0 },
{ "index": 2, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 43, "bad_shares": 0 },
{ "index": 3, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 38, "bad_shares": 0 },
{ "index": 4, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 35, "bad_shares": 0 },
{ "index": 5, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 30, "bad_shares": 0 },
{ "index": 6, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 27, "bad_shares": 0 },
{ "index": 7, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 42, "bad_shares": 0 }
Double check to make sure that your AMD drivers are in COMPUTE MODE, they should display gfx901, NOT gfx900.
0.32h online
For windows, 32/64 and GPU
For windows, 32/64 and GPU
Hello,
monero benchmarks : thanks for having added JCE I love your site and used a lot when I started mining, a lot time before i deved my own miner.
CPU counter bug: that's the first time i see that. Do you use linux version?
The CPU numbers should be in a row, but here there are gaps, I can fix it, that's a high priority bug.
Other high priority one is the Warmup sequence.
I note the bug about temp/fan but that's a lesser important bug, i'll look at this one when the other are fixed, and when i do the watchdog.
edit: i think I found the CPU counter bug, but of course i've nothing like a dual Xeon machine to test on, so it will be, once again, a blind fix.
I'm preparing a minor release with:
The warmup should be a bit longer (closer to 6-7 min) but more reliable. It works well on my own rigs.
The GPU log will be added to the yellow log, and specific to the GPU version, of course. You'll have info about shares and overall hashrate. I keep it separate from the purple log, because they have different meaning:
The purple is all about hardware (fan, temp, correct or bad computation)
The yellow is all about pool-side efficiency (accepted/reject shares, hashrate...)
of course a correct share tends to be accepted, and a broken one will always be rejected, but that's two different things that happens at two different places (purple: in your GPU, yellow: in your pool).
edit:
a preview
* all CPUs are grouped, if any
* Accepted means: explicitely accepted by the pool
* Rejected means: explicitely rejected by the pool, or if --doublecheck is enabled, a broken share we didn't even try to send.
* Ignored are shares either not sent yet because of lag, or sent but the pool never answered, or we got disconnected from the pool, or the share was too stale, etc.
the hashrate is the per-device effective hashrate: only explicitely accepted shares are counted, stale/bad/outdated ones are ignored.
monero benchmarks : thanks for having added JCE
I love your site and used a lot when I started mining, a lot time before i deved my own miner.CPU counter bug: that's the first time i see that. Do you use linux version?
The CPU numbers should be in a row, but here there are gaps, I can fix it, that's a high priority bug.
Other high priority one is the Warmup sequence.
I note the bug about temp/fan but that's a lesser important bug, i'll look at this one when the other are fixed, and when i do the watchdog.
edit: i think I found the CPU counter bug, but of course i've nothing like a dual Xeon machine to test on, so it will be, once again, a blind fix.
I'm preparing a minor release with:
Code:
new coin XCash
new coin zBucks
Fixed CPU counter for dual socket CPU
Reworked warmup
More complete log about GPU
The warmup should be a bit longer (closer to 6-7 min) but more reliable. It works well on my own rigs.
The GPU log will be added to the yellow log, and specific to the GPU version, of course. You'll have info about shares and overall hashrate. I keep it separate from the purple log, because they have different meaning:
The purple is all about hardware (fan, temp, correct or bad computation)
The yellow is all about pool-side efficiency (accepted/reject shares, hashrate...)
of course a correct share tends to be accepted, and a broken one will always be rejected, but that's two different things that happens at two different places (purple: in your GPU, yellow: in your pool).
edit:
a preview
Code:
12:03:58 | Devices results - Shares Accepted/Ignored/Rejected - Average Hashrate
12:03:58 | * CPUs - 1/0/0 - 120.48 h/s
12:03:58 | * GPU 0 - 4/0/0 - 481.93 h/s
12:03:58 | * GPU 1 - 1/0/0 - 120.48 h/s
* all CPUs are grouped, if any
* Accepted means: explicitely accepted by the pool
* Rejected means: explicitely rejected by the pool, or if --doublecheck is enabled, a broken share we didn't even try to send.
* Ignored are shares either not sent yet because of lag, or sent but the pool never answered, or we got disconnected from the pool, or the share was too stale, etc.
the hashrate is the per-device effective hashrate: only explicitely accepted shares are counted, stale/bad/outdated ones are ignored.
Hi JCE,
Why my rig show temp and fan always 0c and 0%? Did i miss something?
Why my rig show temp and fan always 0c and 0%? Did i miss something?
maybe you should describe more which OS (Windows 10 build 1709), driver (18.3.4), GPU (RX 580 8GB) so dev can find the root cause more easily.
for me with the details above it work fine.
Windows 10 LTSB and 1709, Driver 18.4.1 and 18.6.1, RX Vega 56
Code:
"gpu_status":
[
{ "index": 0, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 29, "bad_shares": 0 },
{ "index": 1, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 39, "bad_shares": 0 },
{ "index": 2, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 43, "bad_shares": 0 },
{ "index": 3, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 38, "bad_shares": 0 },
{ "index": 4, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 35, "bad_shares": 0 },
{ "index": 5, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 30, "bad_shares": 0 },
{ "index": 6, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 27, "bad_shares": 0 },
{ "index": 7, "temperature": 0, "fan": 0, "processor": "gfx900", "memory": 8176, "good_shares": 42, "bad_shares": 0 }
Hi JCE,
Why my rig show temp and fan always 0c and 0%? Did i miss something?
Why my rig show temp and fan always 0c and 0%? Did i miss something?
maybe you should describe more which OS (Windows 10 build 1709), driver (18.3.4), GPU (RX 580 8GB) so dev can find the root cause more easily.
for me with the details above it work fine.
Hi JCE,
Why my rig show temp and fan always 0c and 0%? Did i miss something?
Why my rig show temp and fan always 0c and 0%? Did i miss something?
I believe this miner software is fast, because i have use this in PC mode. But never try it in GPU mode, but i believe it is also fast. Thank you dev for this software, and thank you for the update also.
I don´t know, it is either 18.6.1 or 18.7.1, i didn´t try newest driver. 2000 worked on vegas 56 but didn´t work on vegas 64 when i tried, they were hashing 0. I will be back home in 10 days (i am onboard the ship) then i need to standardize my miners, drivers and settings. now it is all over the place, my rigs have miners from JCE from 31a all the way to 32g, so can´t really give any reliable information now
Thanks for your information, what is your hashrate for RX Vega 56 ?
I got around 2100 h/s for RX Vega 56 samsung chipset with cclock 1480/900, mclock 950/900, and multi_hash 1968. hynix chipset with the same setting I got around 2000 h/s.
windows 10 build 1709 amd driver 18.8.1.
Jump to: