Exact same as 0.32b but bad shares fixed (at least it runs fine on my Polaris)
edit: nope
Topic: [XMR] JCE Miner Cryptonight/forks, now with GPU! - page 72. (Read 90842 times)
Exact same as 0.32b but bad shares fixed (at least it runs fine on my Polaris)
edit: nope
My opencl is generated per-gpu (even per card) it's very possible it works on Vega but not Polaris. it's still a broken version i cannot decently put online back.
at least the perf are encoureaging
at least the perf are encoureaging
Really, every time i rush a release, i break something, it's getting annoying…
i remove the 0.32b i really need to test it more…
excepted for the bad shares, did you notice more speed on vega?
i remove the 0.32b i really need to test it more…
excepted for the bad shares, did you notice more speed on vega?
yes vega got more hash about +10 h/s, so still not good enough I guess.
I more amaze with increase on RX 580 8GB.
one more thing on vega rig I think there is no rejected shares, is it possible?
Ok thanks for the reply, and sorry for the bad release, who lived something like 30 min.
i've so many features to add that i tested just two shares, ang go go go shipping. which is very lame
i've so many features to add that i tested just two shares, ang go go go shipping. which is very lame
Really, every time i rush a release, i break something, it's getting annoying…
i remove the 0.32b i really need to test it more…
excepted for the bad shares, did you notice more speed on vega?
i remove the 0.32b i really need to test it more…
excepted for the bad shares, did you notice more speed on vega?
Didn't test on a Vega but the polaris card were on fire (most likely not correct)
. I saw 1180 on my 8gb card.
Really, every time i rush a release, i break something, it's getting annoying…
i remove the 0.32b i really need to test it more…
excepted for the bad shares, did you notice more speed on vega?
i remove the 0.32b i really need to test it more…
excepted for the bad shares, did you notice more speed on vega?
minor version 0.32b
Now i stop OpenCL changes and I focus on GPU Fan/Temp
If I success to improve Heavy, I expect Tube to be faster too, right.
Code:
Reworked a little of CN-Heavy
Not sure it's better on Vega...Now i stop OpenCL changes and I focus on GPU Fan/Temp
If I success to improve Heavy, I expect Tube to be faster too, right.
Hi JCE,
I tried mine on Haven and Bittube got a lot of rejected share.
Please help.
Thanks.
lower your overclocking
you should try your self then you will know it's not because of to much overclocking.
minor version 0.32b
Now i stop OpenCL changes and I focus on GPU Fan/Temp
If I success to improve Heavy, I expect Tube to be faster too, right.
Code:
Reworked a little of CN-Heavy
Not sure it's better on Vega...Now i stop OpenCL changes and I focus on GPU Fan/Temp
If I success to improve Heavy, I expect Tube to be faster too, right.
Hi JCE,
I tried mine on Haven and Bittube got a lot of rejected share.
Please help.
Thanks.
lower your overclocking
Tried the new version too. Get a bunch of rejects + accepts. Seems broken. No problems in OC here since it had been mining ipbc with 0 rejects for days.
15:47:25 | GPU 4 Thread 5 Lane 480 finds a Share, value 15000
15:47:25 | Accepted by the pool.
15:47:26 | GPU 1 Thread 1 Lane 414 finds a Share, value 15000
15:47:26 | Accepted by the pool.
15:47:28 | GPU 0 Thread 0 Lane 906 finds a Share, value 15000
15:47:28 | Rejected by the pool.
15:47:28 | Message from the pool: Low difficulty share
15:47:30 | GPU 5 Thread 6 Lane 312 finds a Share, value 15000
minor version 0.32b
Now i stop OpenCL changes and I focus on GPU Fan/Temp
If I success to improve Heavy, I expect Tube to be faster too, right.
Code:
Reworked a little of CN-Heavy
Not sure it's better on Vega...Now i stop OpenCL changes and I focus on GPU Fan/Temp
If I success to improve Heavy, I expect Tube to be faster too, right.
Hi JCE,
I tried mine on Haven and Bittube got a lot of rejected share.
Please help.
Thanks.
lower your overclocking
minor version 0.32b
Now i stop OpenCL changes and I focus on GPU Fan/Temp
If I success to improve Heavy, I expect Tube to be faster too, right.
Code:
Reworked a little of CN-Heavy
Not sure it's better on Vega...Now i stop OpenCL changes and I focus on GPU Fan/Temp
If I success to improve Heavy, I expect Tube to be faster too, right.
Hi JCE,
I tried mine on Haven and Bittube got a lot of rejected share.
Please help.
https://imgur.com/a/dmJFLbt
Thanks.
Code:
Reworked a little of CN-Heavy
Not sure it's better on Vega...Now i stop OpenCL changes and I focus on GPU Fan/Temp
If I success to improve Heavy, I expect Tube to be faster too, right.
That's what i call multipool/multicoin, it's planned.
JCE has a good Cn-ipbc and a bad CN-heavy. since bittube is CN-ipbc plus CN-heavy, both compensate and it has good overall performance. i know my Heavy is bad on Vega but i'd need a real vega to bench on.
JCE has a good Cn-ipbc and a bad CN-heavy. since bittube is CN-ipbc plus CN-heavy, both compensate and it has good overall performance. i know my Heavy is bad on Vega but i'd need a real vega to bench on.
I am actually talking about Polaris numbers.
That's what i call multipool/multicoin, it's planned.
JCE has a good Cn-ipbc and a bad CN-heavy. since bittube is CN-ipbc plus CN-heavy, both compensate and it has good overall performance. i know my Heavy is bad on Vega but i'd need a real vega to bench on.
JCE has a good Cn-ipbc and a bad CN-heavy. since bittube is CN-ipbc plus CN-heavy, both compensate and it has good overall performance. i know my Heavy is bad on Vega but i'd need a real vega to bench on.
Hi JCE,
just curious when new version will release? and what new feature will be implemented?
is that mean if you optimized CN-Heavy later, hash rate for CN-Saber (Bittube) will improve also?
if after sometimes (few hours) the hash rate become zero, should I lower cclock or mclock? because I try set lower intensity from 944 to 864 the hash rate still become zero after sometimes (few hours).
I try combination of setting for RX Vega 56 in CN-Saber (Bittube) so far the best I can achieve with JCE is 1550 h/s while SRB can reach more than 1600 h/s with the same cclock and mclock.
setting cclock 1480/900 and mclock 950/900.
JCE with driver 18.6.1 1550 h/s while SRB with driver 18.3.4 1600 h/s.
JCE with driver 18.3.4 cannot reach more than 1500 h/s.
Thanks.
That's what i call multipool/multicoin, it's planned.
JCE has a good Cn-ipbc and a bad CN-heavy. since bittube is CN-ipbc plus CN-heavy, both compensate and it has good overall performance. i know my Heavy is bad on Vega but i'd need a real vega to bench on.
JCE has a good Cn-ipbc and a bad CN-heavy. since bittube is CN-ipbc plus CN-heavy, both compensate and it has good overall performance. i know my Heavy is bad on Vega but i'd need a real vega to bench on.
brah, if you can allow multi algo in the same window (without launching two seprate instances for heavy / normal etc. thats my issue with rigs, running multiple variants that are faster on a specific gpu.
@JCE-Miner.
The CN-Saber (bittube) algo is touted to be heavier than CN-heavy. Yet the hashrates on the miner are same for both the algos.
Is there some optimization that was done for cn-saber over cn-heavy?
Always noticed so when compared to loki. The CN-Saber (bittube) algo is touted to be heavier than CN-heavy. Yet the hashrates on the miner are same for both the algos.
Is there some optimization that was done for cn-saber over cn-heavy?
Never heard that windows is running for years? Linux is not the holy grail of operating systems, as some here are talking about.
Our Windows Server is running for years without a reboot. my pc at home run for several months without reboot and i work with it every day.
Our Windows Server is running for years without a reboot. my pc at home run for several months without reboot and i work with it every day.
MS has greatly improved reliability in the last 5 years. For me it lacks customization.
@JCE-Miner.
The CN-Saber (bittube) algo is touted to be heavier than CN-heavy. Yet the hashrates on the miner are same for both the algos.
Is there some optimization that was done for cn-saber over cn-heavy?
The CN-Saber (bittube) algo is touted to be heavier than CN-heavy. Yet the hashrates on the miner are same for both the algos.
Is there some optimization that was done for cn-saber over cn-heavy?
Never heard that windows is running for years? Linux is not the holy grail of operating systems, as some here are talking about.
Our Windows Server is running for years without a reboot. my pc at home run for several months without reboot and i work with it every day.
Our Windows Server is running for years without a reboot. my pc at home run for several months without reboot and i work with it every day.
It's practically impossible...
It's not. I have a couple of systems, too, which were rebootet last time several years ago. When the system is stable and auto updates are off, why should a system not run for years? 
Jump to: