KanoPool since 2014 🐈 - PPLNS and Solo 0.5% fee - Worldwide - 2435 blocks

[NotFuzzyWarm]

So as I'm reading this topic, BrainsOS+ cannot find block on solo mining? I'm currently testing it on s9 but I'm confused after your (kano) post that BrainsOS+ is not suitable to find a block

Um just *where* did you see that?
Braiins works here just fine. In fact in the early days of Braiins, it was only allowed on solo and not allowed on the main pool. Why - because the devs refused to offer any proof of it being able to find a block. Only once it was proved that Braiins could find a block it was it allowed on the main pool.

[1714626175]

1714626175

[Demanu]

So as I'm reading this topic, BrainsOS+ cannot find block on solo mining? I'm currently testing it on s9 but I'm confused after your (kano) post that BrainsOS+ is not suitable to find a block

Um just *where* did you see that?
Braiins works here just fine. In fact in the early days of Braiins, it was only allowed on solo and not allowed on the main pool. Why - because the devs refused to offer any proof of it being able to find a block. Only once it was proved that Braiins could find a block it was it allowed on the main pool.

here is kano post about BrainsOS+:

You can run that miner on solo mining here.
The non+ version is also free to run.
It's restricted on PPLNS since no one seems to ever report if the code actually finds blocks, and like all the hack firmware out there, the developers aren't interested in testing it either, they leave that as a problem for the people who use it.

[NotFuzzyWarm]

AFAIK that changed in 2022 when a solo miner running it produced a high enough share (>32bit) to verify that Braiins can indeed work at the diffs needed.

[kano]

Yes - as NotFuzzyWarm stated it is now allowed for that exact reason.

The main issue with most if not all miner developers, is they are hackers without any tertiary computer education, and haven't worked in the computing field doing any sort of critical software development.

They wont test their software properly and expect problems to be the problem of those paying exorbitant fees for the software.

So here, with hack firmware like the apollo and braiins, I required them to find a >32 bit share before risking them running on pplns - since mining on pplns is forcing sharing that risk with everyone else on pplns.

Also, for the same reason, since I'm the largest miner on the pool, I don't want to be sharing my rewards with people running untested software that might throw away blocks.

Bitcoin mining is abound with known license violations and hack software.
To be blunt, I don't know of anyone who doesn't fall in that category (except me)

[Demanu]

So let's summarize: I can use BrainsOS+ on solo pool and there was a proof that this software can find block. Am I correct?

I'm using different software for my antminers because of custom cooling and I want to protect the device if that cooling will fail (also autotuning option convinced me). I cannot do this on Bitmains software as it has disabled any access into it (tried to ssh and telnet but without success) and without fans it won't start. I know that with access to shell with root I can "cheat" any sensor to make it run  (I'm linux daily user and DevOps engineer so I believe I have enough knowledge to do this).

I heard about Vnish which is using open source software under the hood but I can be wrong.

Do you have any suggestions how to solo mine "the right way"?

[kano]

So let's summarize: I can use BrainsOS+ on solo pool and there was a proof that this software can find block. Am I correct?

No idea if it can find a block, but proof it can hash >32bit shares - which is one well known historical example of losing blocks ... that happened on slush

I heard about Vnish which is using open source software under the hood but I can be wrong.

Vnish is a full on lying bunch of scumbags who also violate my cgminer license.

Do you have any suggestions how to solo mine "the right way"?

Mine here solo for the ridiculous low 0.5% fee

If you want to do it yourself, then be prepared to spend $15k a year on fast bitcoin nodes to distribute you blocks around the world.
Of course it will still be slower than here.

[Demanu]

Mine here solo for the ridiculous low 0.5% fee

I am and I will My question was more about firmware for antminers you suggest to use. As I mentioned before: Bitmain makes a lot of limitations for its firmware.

[kano]

My miners use the original firmware

[Demanu]

My miners use the original firmware

I will try to buy fan emulators as I'm running immersion cooling so I need to "cheat" original firmware that fans are still there. Thank you for your help and make it clear for me. I will still run it on your pool.

Or maybe you have any solution to ssh into original firmware of antminers? I will do shell scripting on my own but I need access.

[n0nce]

~

Since I'm not sure exactly how this works on a technical level right now (might have slipped my memory): is the mining software / firmware used communicated through Stratum v2 and / or could it be spoofed? How do you make sure that someone pointing hashpower to your pool is running a tested / 'known working' software stack?

My miners use the original firmware

I will try to buy fan emulators as I'm running immersion cooling so I need to "cheat" original firmware that fans are still there.

Should be fairly easy to hack together if you can't find anything ready-made, right? I guess the original firmware expects a fan RPM reading on the yellow cable of a 3-pin fan connector?



The cable usually gives you 2 pulses per revolution so if you want to emulate 500RPM you just need to provide a 1kHz square wave. You can get that from a cheap Arduino clone and a few lines of code such as:

Code:

void setup() {
pinMode(0, OUTPUT);
}
void loop() {
digitalWrite(0, HIGH);
delay(1); // wait 1ms 
digitalWrite(0, LOW);
delay(1); // wait 1ms 
}

If it should be cheaper, you could also get such a square wave from a 555 timer IC, although I doubt that the time invested to build and test it is worth it over an affordable microcontroller devkit.

Lastly, easiest might be to use fan cable extensions and have actual fans running outside of the immersion cooling tank.

[kano]

No comment

[NotFuzzyWarm]

~

Since I'm not sure exactly how this works on a technical level right now (might have slipped my memory): is the mining software / firmware used communicated through Stratum v2 and / or could it be spoofed? How do you make sure that someone pointing hashpower to your pool is running a tested / 'known working' software stack?
...

Kano has made his opinion of Stratum v2 well known and safe to say it will be a mighty cold day in hell before he uses it. As for how he determines if the miner is running a hacked version of cgminer (which are quickly kicked off) vs accepted OEM, Braiins, BFGminer or Bitfury firmware - trade secret...  

[Demanu]

I've just ordered fan emulators for antminers, there is plenty of them on the market.

I have one more question about worker diff in solo mining. Is setting this parameter have any sense on solo mining? At the moment, its calculated by pool and its 6500, is that number for solo mining should be changed depend on hardware I'm using?

[NotFuzzyWarm]

Unless you are running small USB stick miner just leave suggested diff alone. The pool and miners work together to quickly assign the best value. The diff target is aimed at the pool & miner working at 18 shares per-minute.

[kano]

The work diff has no effect on your chance of finding a solo block.

When solo mining, the pool's work diff ensures it provides regular feedback, so that the hash rate the pool shows has low variance.
If you set it too high, the hash rate on the pool will jump all over the place and effectively be meaningless.

The option to set it high on a solo account, is in case you do a rental that requires a high diff before it will work.
The only other use for setting the diff on solo (or pplns) is to set a higher starting diff than the pool default of 8190,
if your miners normally run much higher i.e. so they don't send a bunch of 'hi' ignored shares when they connect.

In all cases, once the pool corrects the diff after you connect, the pool's diff is the best diff to use.

[JessieJames13]

My asic shows 81T, but on kano it reports 72T.
What happened to the other 9 hash?

[NotFuzzyWarm]

My asic shows 81T, but on kano it reports 72T.
What happened to the other 9 hash?

Folks here are not mind readers, some basic info would be nice. Things like:
What make/model of miner?
How long has it been running?
Does it use modified firmware or Braiins?

[os2sam]

Or, are you overriding the pools difficulty setting?

[kano]

My asic shows 81T, but on kano it reports 72T.
What happened to the other 9 hash?

Well most miners take an hour to actually get an average close to what they are really doing.
If you are running any 3rd party firmware, and the pool isn't blocking it, they are also taking a % of your hash rate.
The workers page on the pool gives lots of details about each miner and the invalids,
and Help->Stats explains a lot of that page if you are logged in.

[JessieJames13]

What make/model of miner?
How long has it been running?
Does it use modified firmware or Braiins?

1166 pro
stock os, no custom firmware
2 days running

Or, are you overriding the pools difficulty setting?

no

Well most miners take an hour to actually get an average close to what they are really doing.
If you are running any 3rd party firmware, and the pool isn't blocking it, they are also taking a % of your hash rate.
The workers page on the pool gives lots of details about each miner and the invalids,
and Help->Stats explains a lot of that page if you are logged in.

no 3rd party f/w
invalids between 0.04% and 0.07%