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

[dingdongtobias]

Can You tell us JCE how is the fee mined on GPU ?
All we knou is 0.9% and not in first minut.

How often and how long is the fee mining?

[JCE-Miner]

It's randomized, average of ~54s every 100min, like SRB, on the long term. Since it's randomized, it may be slightly more or less every time, but in a small extent. The random is to avoid raisonnance with the user pool, in case it itself sends a new session every N minutes (like the 10mn of Nicehash).

You can check all this with you pool, the real pool side hashrate is supposed to be about 2% lower than the average hashrate of your rig (~1% for fees and ~1% for stale shares, with Nicehash losing more stale shares than other pools).

edit: the GPU hashrate @0.00 is a bug i already fixed in the 0.32g i'm testing for now. It's a display bug and does not block your GPU.

[dingdongtobias]

It's randomized, average of ~54s every 100min, like SRB, on the long term. Since it's randomized, it may be slightly more or less every time, but in a small extent. The random is to avoid raisonnance with the user pool, in case it itself sends a new session every N minutes (like the 10mn of Nicehash).

You can check all this with you pool, the real pool side hashrate is supposed to be about 2% lower than the average hashrate of your rig (~1% for fees and ~1% for stale shares, with Nicehash losing more stale shares than other pools).

I ask this because there are mad people at SRB thread, because the way the fee is mine is changed, now they cant kill process every hour and skip fee, because fee 1 is mined at 4 min.
And if they start the miner 24 time a days, then its 24 fee mining.

You got also the same problem probably, just people still not mad 

[JCE-Miner]

I mixed the Claymore fee system (randomized timed fees) and the SRB overall rate (~54s/100mn). If you want to mine fee-free with JCE, kill it after 1mn, or even use the parameter -q 1 so it closes by itself after 1mn. It's the documented and official no-fee mode of JCE.

JCE never takes any fee during the first minute, so if you run it, and kill it after one minute, and repeat again and again, then you'll never pay any fee

- github doc

I prefered it compared to the stupid -nofee mode of Claymore that dropped 5% of your shares (10% measured on Claymore 10+) as a vengeance, keeping the then wrong 100% hashrate display and 100% power consumption.

edit: I took a look at SRB topic, and yes people are crazy. Doktor speech is right: he takes 0.85% on the long term, period. on 24h that's 0.85% of 24h, period. JCE does that since the very first version. As I said, that's the Claymore-style. I'm surprised Doktor still used fixed timed fees up to his 1.6.4...

My argumument for JCE is simple : look at your pool. Forget the stale shares, the fees, the average rates... Your pool, focus on your pool. Mine with miner X, Y... and JCE. Keep the one which gives more shares to your pool, period.

[Iamtutut]

Hello all,
* Cryptonight-Red fork if i've time enough to do it. It's a mix of XTL and CN-Light

This fork is for which coin ?

[dingdongtobias]

I agree with you, alsow what is the point of mining a coin 15 minute, than switch to other
People are crazy.

[JCE-Miner]

CN-Red is for MOX coin

Some miners dont show you even this.

I think Doktor talks about JCE here, but that's pointless, since i've control over my netcode, i can mine devfee for 10mn and report on screen i mine fee for 1mn. A close source miner internal report worth nothing. Like the Claymore -nofee hashrates that were tweaked. Only one info is trustfull : your pool report. Period.

[heavyarms1912]

I agree with you, alsow what is the point of mining a coin 15 minute, than switch to other
People are crazy.

It seems like they were cheating the dev fee.  It's not worth switching a coin in an hour or two.  You introduce additional latencies: drop current work, connect to new pool, switch kernels, get new work.  Am I missing something.

@JCE-Miner.  I mine CN-saber and pretty good results with it (more than any other miner).  
CN-Heavy is supposed to be lighter than CN-Saber and so potential we should get higher hashrates on that algo.

[vmozara]

When I get back home, I will give a 7 day test to srb miner again to count the hashes.

I can tell you this, that the hashes reported by JCE-miner are spot on with what the pool reports. That is why I stick with JCE from the very moment first GPU version was out.

Anyway, how many of you guys are mining using Vegas? What version you use and what you mine? Did anybody tinker with settings to pull out some extra hashes?

[Uaciuganadu]

Anyway, how many of you guys are mining using Vegas? What version you use and what you mine? Did anybody tinker with settings to pull out some extra hashes?

Here are my 2 Vega 64 mining V7:

Code:

{ "mode" : "GPU", "worksize" : 8, "alpha" : 64, "beta" : 8, "gamma" : 4, "delta" : 4, "epsilon" : 4, "zeta":4, "index" : 1, "multi_hash":2000 },
{ "mode" : "GPU", "worksize" : 8, "alpha" : 64, "beta" : 8, "gamma" : 4, "delta" : 4, "epsilon" : 4, "zeta":4, "index" : 1, "multi_hash":2000 },

2000 is under testing as i get a few more H

Code:

| Hashrate GPU Thread 1: 1139.96 h/s
| Hashrate GPU Thread 2: 1139.90 h/s - Total GPU 1: 2279.86 h/s ( Sapphire ) ~ 1500GPU - 1170MEM @960mv
| Hashrate GPU Thread 3: 1044.88 h/s
| Hashrate GPU Thread 4: 1045.47 h/s - Total GPU 2: 2090.34 h/s ( Asus ) ~1400GPU - 1080MEM @960mv

The sapphire is a beast and can take more but with few errors.
The Asus is kinda crap in thermals so i keep it a bit lower.

[JCE-Miner]

online is the
0.32g CPU - Windows

Linux version will come next.

Code:

Netcode fixes

[obm_dan]

From the JSON that's returned via the API, how do you differentiate between the hash rate from CPU mining vs the hash rate from GPU mining?

I'm newer to this miner.  The thread rates for _gpu seem to be the thread_all rates, but paired (_all[0] + _all[1] = _gpu[0], _all[2] + _all[3] = _gpu[1], etc).  Which of these is the cpu rate if you're mining with both, assuming that's possible.

[lebuawu2]

When I get back home, I will give a 7 day test to srb miner again to count the hashes.

I can tell you this, that the hashes reported by JCE-miner are spot on with what the pool reports. That is why I stick with JCE from the very moment first GPU version was out.

Anyway, how many of you guys are mining using Vegas? What version you use and what you mine? Did anybody tinker with settings to pull out some extra hashes?

when srb introduce option --sendallstales and you use it in .bat file you should get close to what miner reported.
actually I also notice that maybe since 1.5.x or 1.6.x CN-V7 is far away from miner reported. because when I tried 1.4.1 pool hashrate close to miner reported, in that version the miner also counted stale shares. but somehow that information lost in 1.5.x or 1.6.x.
I didn't report because I know their respond sometimes is bad or simply ignored it. so I try to stop posting on their thread and find another miner.

the best dev attitude so far at least in my opinion is Phoenix Miner. even if someone said bad to them, they will answer politely.

[lebuawu2]

CN-Red is for MOX coin

Some miners dont show you even this.

I think Doktor talks about JCE here, but that's pointless, since i've control over my netcode, i can mine devfee for 10mn and report on screen i mine fee for 1mn. A close source miner internal report worth nothing. Like the Claymore -nofee hashrates that were tweaked. Only one info is trustfull : your pool report. Period.

I understand that all miner start dev fee at first few minutes to avoid people cheating by close miner right before dev fee kick in and start miner again.
The problem when doktor83 change the way dev fee work from 2 hours at first kick in to 15 minutes at first kick in, he didn't state in 1.6.3 release  https://bitcointalk.org/index.php?topic=3167363.msg41828317#msg41828317 even though at first page I saw he wrote that changes, but not sure when he wrote it.

so for me the problem not because of dev fee 15 minutes start kick in, but because of he didn't inform boldly in his thread like when he asked miner to download new version because of his cn-heavy dev fee mining in sumo coin while sumo back to CN ASIC friendly.

[lebuawu2]

Hello all,

I do have an experimental version that fixes the warm-up phase stuck @80% forever, but I want to test it well, no more disasters like the 0.32 c-d-t... that produced bad shares. No more rushed release, i really need time to test.

If i can achieve it in time, i will release a CPU version with backports of the GPU netcode fixes. It's mostly targetted to the Linux miners since that's the version I update the less.

Next planned GPU version will have :

* Warm-up phase fixed
* CN-Heavy optim (i expect to come closer to SRB, but probably not above yet)
* Shares sent to the void bug on exotic pools (Nicehash, MoneroOcean...) fixed
* Auto blue hashrate back
* Correct display of very low GPU hashrates (for HD5000...)
* Auto yellow report (period: about 10mn)
* Cryptonight-Red fork if i've time enough to do it. It's a mix of XTL and CN-Light

can't wait for this release, especially point 1 and 2. any estimation when it will be release?

[dingdongtobias]

CN-Red is for MOX coin

Some miners dont show you even this.

I think Doktor talks about JCE here, but that's pointless, since i've control over my netcode, i can mine devfee for 10mn and report on screen i mine fee for 1mn. A close source miner internal report worth nothing. Like the Claymore -nofee hashrates that were tweaked. Only one info is trustfull : your pool report. Period.

I understand that all miner start dev fee at first few minutes to avoid people cheating by close miner right before dev fee kick in and start miner again.
The problem when doktor83 change the way dev fee work from 2 hours at first kick in to 15 minutes at first kick in, he didn't state in 1.6.3 release  https://bitcointalk.org/index.php?topic=3167363.msg41828317#msg41828317 even though at first page I saw he wrote that changes, but not sure when he wrote it.

so for me the problem not because of dev fee 15 minutes start kick in, but because of he didn't inform boldly in his thread like when he asked miner to download new version because of his cn-heavy dev fee mining in sumo coin while sumo back to CN ASIC friendly.

If you use program properly, you dont  need to worry is fee at 1 min or 100min, if You get on pool the hashrate you shuld.
JCE is smarter dont show when fee is mining, so the restart process killing users dont know how much fee they mine

but will probably the same users who use both miners be now mad on JCE too if they using it improperly and killing process 50 time a day 

[JCE-Miner]

what happens to SRB is crazy, people who cheated his fees complain they cannot cheat any longer. i never had any complain about fee on jce and i won't start just because another miner dev has some. the fee netcode of jce is the exact same since very first version, and is the same as Claymore, except there's no revenge mode in case a fee fails.
an important note is that jce use the same algo for fees as the user, to ensure no power consumption change.

imagine: if your card is fine tuned for v7, it mines, then the devfee of miner X switch to cn-heavy, your card eats 5W more and crashes. i don't like this way, again i use Claymore principle: devfee mined in the same algo than the user.


estimate for 0.32g cpu linux : one day
0.32g GPU : a few days

[lebuawu2]

CN-Red is for MOX coin

Some miners dont show you even this.

I think Doktor talks about JCE here, but that's pointless, since i've control over my netcode, i can mine devfee for 10mn and report on screen i mine fee for 1mn. A close source miner internal report worth nothing. Like the Claymore -nofee hashrates that were tweaked. Only one info is trustfull : your pool report. Period.

I understand that all miner start dev fee at first few minutes to avoid people cheating by close miner right before dev fee kick in and start miner again.
The problem when doktor83 change the way dev fee work from 2 hours at first kick in to 15 minutes at first kick in, he didn't state in 1.6.3 release  https://bitcointalk.org/index.php?topic=3167363.msg41828317#msg41828317 even though at first page I saw he wrote that changes, but not sure when he wrote it.

so for me the problem not because of dev fee 15 minutes start kick in, but because of he didn't inform boldly in his thread like when he asked miner to download new version because of his cn-heavy dev fee mining in sumo coin while sumo back to CN ASIC friendly.

If you use program properly, you dont  need to worry is fee at 1 min or 100min, if You get on pool the hashrate you shuld.
JCE is smarter dont show when fee is mining, so the restart process killing users dont know how much fee they mine

but will probably the same users who use both miners be now mad on JCE too if they using it improperly and killing process 50 time a day  

I used program properly, that's why I said I understand all miner start dev fee at first few minutes. I used claymore and phoenix miner a lot, their dev fee is same with srb. cast xmr and jce also I think the same mechanism. I just pointing why people angry, people angry because they know it changed by finding it not because of doktor show them.

sorry for my bad English if you can't understand for what I want to say.

[lebuawu2]

what happens to SRB is crazy, people who cheated his fees complain they cannot cheat any longer. i never had any complain about fee on jce and i won't start just because another miner dev has some. the fee netcode of jce is the exact same since very first version, and is the same as Claymore, except there's no revenge mode in case a fee fails.
an important note is that jce use the same algo for fees as the user, to ensure no power consumption change.

imagine: if your card is fine tuned for v7, it mines, then the devfee of miner X switch to cn-heavy, your card eats 5W more and crashes. i don't like this way, again i use Claymore principle: devfee mined in the same algo than the user.


estimate for 0.32g cpu linux : one day
0.32g GPU : a few days

can't wait for GPU release.
My understanding for Claymore Ethash, dev fee mining currently only have option eth or etc, so if you mine musicoin or other ethash coin it will switch dag even though power consumption same sometimes it cause crash. Phoenix Miner solve this problem by introduce a lot coin dev fee mining in order to avoid switching dag.
maybe my understanding also wrong, since I just used it.

[samvicky]

Your miner is simply awesome .All I need is auto reboot feature that auto reboot the rig if any gpu crashes