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

[vmozara]

I am totally cool with devfee. In fact, when I switched to JCE i thought i am paying 1.5% and later JCE told me it is actually 0.9% for gpu mining :*

JCE, since you are so cool, i have 2 more suggestions:

1. Do not have separate blue and violet colors for GPUs. Just insert the hashrates into violet text and add one extra line with total rig statistics.

18:02:16 | GPU 6: Temp: 55C - Fan: 86% -- Shares: Good: 1 Bad: 0 - Thread 02 1050h/s Thread 03 1030h/s Total 2080h/s

Unless you are limited by line length or something but you can always remove unnecessary letters

2. If you press R, it would be cool to see highest difficulty share your miner found.

PS. I was always thinking that 0.32x versions with fast compile have somewhat lower hashrate in vega cn7, than 0.31 versions. (14280 vs 14400) But now i just randomly entered multihash 2000 instead of 1936 and bam! 14400h/s.  

[JCE-Miner]

All on the same line is too long, I already tried it.

Also the purple log is special for GPU, that's the very physical status. Fan, Temp and shares that were physically good or bad.
Not related to pool side results.

The blue hashrate is the instant physical hashrate for all devices
Not related to pool side results neither.

Yellow is about the pool-side info: coin, wallet, difficulty, effective net hashrate...

I want to avoid to mix those notions. However the yellow part is incomplete, i admit, i must add the triplet of shares found/accepted/rejected

The memory model of 0.32+ is very different of 0.31 and yes, it tends to be faster with more vram (which is not that good, but i tested on my low-mem card HD7790 1G and i still get the peak of 268h/s, on par with claymore.
So i can expect the 0.32 to be as fast as the 0.31, with a few more mem, but starting a lot faster.

highest difficulty share your miner found.

For what? For fun? it's a useless info since mining is all about random values and only the pool difficulty counts, a 1G diff hash with pool difficulty of 10000 worths 10000.

[Lermite]

The too long line can be easyly shortened:

18:02:16 | GPU 6: 55°C 86%  Shares: 1 | 0   Avg 1030 h/s  Max 1083 h/s

"Temperature" and "Fan speed" are obvious because a temperature and a percentage can only come from them.

The hashrate per thread is not as useful as the total one per device (GPU or CPU) so the global hashrate per device would be enough in the automatically displayed values, while the detailed hashrates would be still available through the key H.

The hashrate is not a direct measure from hardware, but its value depends directly on the hardware state: If a hashrate drops significantly, the GPU has some issue or is already crashed.
That's why I believe it could be included to the violet line about each CPU, and those lines could have one about the CPU too.

[WebTosha]

good afternoon.
On the first versions when I first started using this miner, I could install on the cards Sapphire RX552 "multi_hash": 464 and the cards were given out at 520 h/s, but now I can not install more "multi_hash": 432 everything above the card is not keep the memory frequency and reset again and as a result the speed is from 0 to 60 h/s. The system did not change anything, tried to reduce the frequency to indecent 1800 MHz, the result was not given. Tell me what it can depend on, because it's on the forum guys use 464 on similar cards and I like everything hangs on the line

     ^{{ "mode" : "GPU", "worksize" : 8, "alpha" : 128, "beta" : 16, "gamma" : 4, "delta" : 4, "epsilon" : 4, "zeta" : 4, "index" : ..., "multi_hash":432 },
     { "mode" : "GPU", "worksize" : 8, "alpha" : 128, "beta" : 16, "gamma" : 4, "delta" : 4, "epsilon" : 4, "zeta" : 4, "index" : ..., "multi_hash":432 },}

[QuirkSilver]

good afternoon.
On the first versions when I first started using this miner, I could install on the cards Sapphire RX552 "multi_hash": 464 and the cards were given out at 520 h/s, but now I can not install more "multi_hash": 432 everything above the card is not keep the memory frequency and reset again and as a result the speed is from 0 to 60 h/s. The system did not change anything, tried to reduce the frequency to indecent 1800 MHz, the result was not given. Tell me what it can depend on, because it's on the forum guys use 464 on similar cards and I like everything hangs on the line

     ^{{ "mode" : "GPU", "worksize" : 8, "alpha" : 128, "beta" : 16, "gamma" : 4, "delta" : 4, "epsilon" : 4, "zeta" : 4, "index" : ..., "multi_hash":432 },
     { "mode" : "GPU", "worksize" : 8, "alpha" : 128, "beta" : 16, "gamma" : 4, "delta" : 4, "epsilon" : 4, "zeta" : 4, "index" : ..., "multi_hash":432 },}

{ "mode" : "GPU", "worksize" : 8, "alpha" : 128, "beta" : 16, "gamma" : 4, "delta" : 4, "epsilon" : 4, "zeta" : 4, "index" : ..., "multi_hash":448 },
     { "mode" : "GPU", "worksize" : 8, "alpha" : 128, "beta" : 16, "gamma" : 4, "delta" : 4, "epsilon" : 4, "zeta" : 4, "index" : ..., "multi_hash":416 },
try worksize, 16, alpha 64 as well.

[zbucks_project]

    Monero (XMR)
    Monero-V (XMV)
    Electroneum (ETN)
    Karbowanec (KRB)
    Bytecoin (BCN)
    Sumokoin (SUMO)
    Bitcoal (COAL)
    Bitcedi (BXC)
    Dinastycoin (DCY)
    Leviarcoin (XLC)
    Fonero (FNO)
    Turtlecoin (TRTL)
    Graft (GRFT)
    Dero (DERO)
    Stellite (XTL)
    UltraNote (XUN)
    Intense (INTS)
    Crepcoin (CREP)
    Pluracoin (PLURA)
    Haven (XHV)
    FreelaBit (FBF)
    BlueberriesCoin (BBC)
    B2BCoin (B2B)
    Bitsum (BSM)
    Masari (MSR)
    SuperiorCoin (SUP)
    EDollar (EDL)
    Interplanetary Broadcast (IPBC)
    Alloy (XAO)
    BBSCoin (BBS)
    BitcoiNote (BTCN)
    Elya (ELYA)
    Iridium (IRD)
    Italo (ITA)
    Lines (LNS)
    Niobio (NBR)
    Ombre (OMB)
    Solace (SOL)
    Triton (TRIT)
    Truckcoin (TRKC)
    Qwertycoin (QWC)
    Loki (LOK)
    Gadcoin (GAD)
    MarketCash (MKT)
    ArtoCash (RTO)
    Bloc (BLOC)
    Wownero (WOW)
    PrivatePay (XPP)
    Nicehash Cryptonight
    Minergate Cryptonight
    MiningPoolHub Cryptonight
    MiningRigRentals Cryptonight
    Suprnova Cryptonight

Hi there,

Could you add, zBucks to your list?

https://zbucks.org

[Lermite]

You can use the option --any to mine any coin without wallet address syntax checking.

[JCE-Miner]

Hello all,

Sorry i've very little time to do the support here, i'm polishing the GPU version g
I'll skip some new features for now to save dev and test time, and just focus on the Warmup fix and CN-Heavy perf boost.

No problem to add a new coin, and as it has been said, it can already be mined with parameters
-- variation 5 --any

any allow to mine anything, 5 is CN-Heavy fork in JCE, complete list is:

Code:

 1 = Original Cryptonight
 2 = Original Cryptolight
 3 = Cryptonight V7 fork of April-2018
 4 = Cryptolight V7 fork of April-2018
 5 = Cryptonight-Heavy
 6 = Cryptolight-IPBC
 7 = Cryptonight-XTL fork of May-2018
 8 = Cryptonight-Alloy
 9 = Cryptonight-MKT
10 = Cryptonight-Arto
11 = Cryptonight-Fast MSR fork of June-2018
12 = Cryptonight-Haven fork of June-2018
13 = Cryptonight-BitTube v2 of July-2018

Online is the
0.32g CPU Linux 32/64

with netcode fixes, no other change

edit:
My new warm-up sequence gives very good result on my Pitcairn rig, the one where the bug was the most obvious.
I think it fixes, at the cost of a very instable hashrate during the first 5 min (on purpose)

[QuirkSilver]

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

Well, out of experience, if a gpu crashes as in memory errors, nothing can reboot it except a power loss (forced shutdown) windows is so stupid.

[heavyarms1912]

Well, out of experience, if a gpu crashes as in memory errors, nothing can reboot it except a power loss (forced shutdown) windows is so stupid.

even linux won't be able to recover from that state.  Typically those kind of crashes are the ones stuck in interrupt handling.

[JCE-Miner]

I plan to put a Claymore-like watchdog: in case a GPU falls under a certain hashrate (default: 0) the miner quits. Then it's up to you to have a script (.bat or .sh) whose next line is a reboot command.

The watchdog wouldn't apply during the warmup session, since then the hashrate is very unstable and inconsistent, on purpose. It may crate both false positive (hashrate higher than real hardware max) or false negative (hashrate very low). That's because the VRAM memory accesses are managed to ensure 100% hashrate after the warmup. Something that was broken in the 0.32+
And not applied to CPU, since a slower CPU is caused by a legitimate other app that need some compute (the system, a driver, whatever...).

0.32g released soon. I skip the minor updates and will release them later in a minor upgrade. 0.32g will only have:

Code:

Faster CN-Heavy/Haven/Tube
Fixed warmup
Netcode fixes
Removal of gamma/delta/epsilon/zeta greeks

Then next:

Code:

Cryptolight-Red fork
Reworked log about shares
zBucks coin

[frozen80]

Hello JCE,

Can you add support for https://webchain.network/?
Their algo it's a modified version of CryptoNightV7.
They have official miner for CPU here: https://github.com/webchain-network and unofficial for GPU here: https://github.com/monkins1010.

Tnx and keep up the good work!

[vmozara]

Can you still keep the greek letters optional? I  was playing yesterday a little bit and managed to squeeze extra 60h/s from entire rig by playing with these letters 

[WebTosha]

I plan to put a Claymore-like watchdog: in case a GPU falls under a certain hashrate (default: 0) the miner quits. Then it's up to you to have a script (.bat or .sh) whose next line is a reboot command.

it seems to me that this does not always help those who do not have a hardware watchdog. I have one or two cards falling and when I close the miner it just hangs for a long time (20 minutes the system may hang in this state), so the next line will not be executed for a long time and the reboot will not happen. It may be better to call a pre-defined script, even if it's empty, but there everyone will write what he needs in it.

[samvicky]

Can you still keep the greek letters optional? I  was playing yesterday a little bit and managed to squeeze extra 60h/s from entire rig by playing with these letters 

Can you elaborate, please..?

[lebuawu2]

I plan to put a Claymore-like watchdog: in case a GPU falls under a certain hashrate (default: 0) the miner quits. Then it's up to you to have a script (.bat or .sh) whose next line is a reboot command.

The watchdog wouldn't apply during the warmup session, since then the hashrate is very unstable and inconsistent, on purpose. It may crate both false positive (hashrate higher than real hardware max) or false negative (hashrate very low). That's because the VRAM memory accesses are managed to ensure 100% hashrate after the warmup. Something that was broken in the 0.32+
And not applied to CPU, since a slower CPU is caused by a legitimate other app that need some compute (the system, a driver, whatever...).

0.32g released soon. I skip the minor updates and will release them later in a minor upgrade. 0.32g will only have:

Code:

Faster CN-Heavy/Haven/Tube
Fixed warmup
Netcode fixes
Removal of gamma/delta/epsilon/zeta greeks

Then next:

Code:

Cryptolight-Red fork
Reworked log about shares
zBucks coin

can you please add keepalive option in the next release plan?

[Beruang123]

hi

can you add RYO?

[JCE-Miner]

you can keep the greeks in the config, but the four less important won't be logged anymore at startup.

RYO wallet are the same as Sumo so i. cannot deduce if a sumo wallet mines ryo. but you can mine ryo with
--any --variation 5
5 is fork number of CN-Heavy

i get perfect results of warmup with my Pitcairn but mixed with the rx560 where sometime the gpu with the display attached stays at 90% speed. i'll probably release as-is so you can give feedback. i cannot just restore behavior of 0.31 because everything has changed inbetween

keepalive: possible but low priority to dev


webchain: nope they changed too many things in the netcode, not just in the algo. i won't double my netcode volume for one obscure coin.

watchdog: JCE currently fails to quit because it tries to clean every memory at exit. in case of watchdog fail, it won't clean and just successfully quit. so the line to be called will be the next in the caller script, i prefer this way rather than forwarding a command line.

[JCE-Miner]

Here comes

0.32g GPU

Code:

Smaller release package (~4MB)
Netcode fixes
CN-Heavy/Tube/HVX optimization
Automated last four greeks (now deprecated and lo longer logged)
New warm-up phase

During the warmup (about 5min) the hashrate can be crazy, your vega can display 50 or 5000 h/s, that's normal (as least, expected). That's because all timings are tweaked during this time, so two hashes can be mined in a row, leading to a crazy number. That's cosmetical, during this time you mine normally and can find shares. If a bad share is found, that's a bug or GPU problem, the warum phase is not expected to produce bad shares.

I tested it on my RX560 and after 5mn of crazy hashrates, i get my stable 527 h/s i got on 0.31, same for all my rigs.

[Iamtutut]

Here comes

0.32g GPU

Code:

Smaller release package (~4MB)
Netcode fixes
CN-Heavy/Tube/HVX optimization
Automated last four greeks (now deprecated and lo longer logged)
New warm-up phase

During the warmup (about 5min) the hashrate can be crazy, your vega can display 50 or 5000 h/s, that's normal (as least, expected). That's because all timings are tweaked during this time, so two hashes can be mined in a row, leading to a crazy number. That's cosmetical, during this time you mine normally and can find shares. If a bad share is found, that's a bug or GPU problem, the warum phase is not expected to produce bad shares.

I tested it on my RX560 and after 5mn of crazy hashrates, i get my stable 527 h/s i got on 0.31, same for all my rigs.

Your file name is wrong, it's called "cpu" instead of "gpu".