NebuTech / NBMiner

GPU Miner for ETH, RVN, BEAM, CFX, ZIL, AE, ERGO
https://nbminer.com
3.18k stars 518 forks source link

Miner reports way less accepted shares than on pool #522

Open webcmrj opened 3 years ago

webcmrj commented 3 years ago

Hi everyone.

Have anyone seen this happens before? On HiveOS NBMiner says 50 accepted total but on pool (tried more than one) its showing 250 accepted shares.

This is causing a real hashrate and real average hashrate to be a third of the reported hashrates on pool.

FDB4084E-A96E-4911-BCF2-E7C21790A9AE

Any guesses? Thanks in advance people!

webcmrj commented 3 years ago

Here a visual explanation:

52F06107-984E-4191-ACEC-26BCC4EE906C 994C4305-49A3-4150-BC3E-6BBDFA590053

webcmrj commented 3 years ago

And the consequences are(different pool from original post but same results):

6C0DC74D-0DA8-48E4-BD67-192073610066

pdieppa commented 3 years ago

I'm not sure if this is the case but every time the miner restarts, the stats on HiveOs will reset. In your last graph, it shows two dips in reported hash rate making me wonder if the miner restated at some point.

webcmrj commented 3 years ago

You are right @pdieppa are those miner’s resets the reason for real x reported hashrate discrepancy here?

057A81B8-4EF8-4409-B43C-A48A5127794F

![Uploading 831081EA-12C1-49E8-9FB1-CB1025DA3547.jpeg…]()

pdieppa commented 3 years ago

There is always a small variation in the reported hash rate, but in your case the hash rate is dropping down to 0 which makes the think the miner or even the whole rig is restating. Check the stats for that rig in HiveOs and look at the power. This is one of my rigs and you can see the dips when I restated it.

image

webcmrj commented 3 years ago

@pdieppa the rig itself is not restarting...status on HiveOS is good...could the problem be something on the communication between miner and pool? Check this:

44332EF5-4469-4C18-B4A9-18A78AD30C0C

webcmrj commented 3 years ago

And lots of times on pool goes to ZERO as you could see on graphs

pdieppa commented 3 years ago

I'm not sure if the problem is the pool since Flexpool.io is pretty large. One sure way to check it is to point your rig to another pool and let it run for while to see that the hash rate is at the pool and at nbminer.

GDMalyska commented 3 years ago

Much Love );G

webcmrj commented 3 years ago

It was the LHR thing....I accidentally released 100% of power but it didn't reflect on the pool results. I was wrongly using lhr code on the latest drives of nvidia on Hive.

lhr-mode: 1 lhr: 100

This released full power for my 3060 LHR but unfortunatelly it decreases the real power to less than 20% on the pool. Maybe this can end pointing a new way to release full power for 3060 cards in a near future 🙏🏼

GDMalyska commented 3 years ago

Awesome great work tho all around guys I swear the best in the biz I care not high five no touch mad props to all

On Wed, Oct 13, 2021, 5:40 PM webcmrj @.***> wrote:

It was the LHR thing....I accidentally released 100% of power but it didn't reflect on the pool results. I was wrongly using lhr code on the latest drives of nvidia on Hive.

lhr-mode: 1 lhr: 100

This released full power for my 3060 LHR but unfortunatelly it decreases the real power to less than 20% on the pool. Maybe this can end pointing a new way to release full power for 3060 cards in a near future 🙏🏼

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/NebuTech/NBMiner/issues/522#issuecomment-942741682, or unsubscribe https://github.com/notifications/unsubscribe-auth/AUGEAC7LIJ6M5EW5VPHJFJDUGX4DDANCNFSM5F45DHBA .