Pages:
Author

Topic: Laurentia Pool - BAD risk for miners - page 7. (Read 2331 times)

legendary
Activity: 3822
Merit: 2703
Evil beware: We have waffles!
May 25, 2020, 11:31:26 AM
#5
We've done testnet and mainnet testing and refinement over several weeks and are 100% confident in our product.
Then Kudos are in order and let's hope your confidence is justified.
I'm curious - who pushed for the testing? I hope the answer is knowledgeable investors...
full member
Activity: 1022
Merit: 221
We are not retail.
May 25, 2020, 11:26:19 AM
#4
We've done testnet and mainnet testing and refinement over several weeks and are 100% confident in our product.
legendary
Activity: 3822
Merit: 2703
Evil beware: We have waffles!
May 25, 2020, 10:31:19 AM
#3
Quote
We are insured, plan to use funds from fee generation as proof of reserves in case of any errors in or out of our control to safety net our users work in lieu of insurance in the future.
That is the one good thing I have yet read about your pool. Too bad a certain other pool op that you have partnered with did not do that to cover losses incurred when they lost yet another block on May 10.

You do realize that you are running a Financial enterprise and in more ways then 1 this is a form of accounting software, right?  Would you run any other business or personal accounting software that has not been properly vetted? I highly doubt it.

Think what you will about Kano but also know that his pool has NEVER lost a block. Part of the reason is that as one of the developers of the code you use, he is well aware of the bugs in it and he addressed them long ago. Before you ask 'then why has he not pushed them to the ckpool git' it is because -ck booted him from it when they had a falling out several years ago.

The other reason is that Kano extensively TESTS his code changes before taking them live. We already know -ck's attitude towards code testing - he believes it to be a waste of time, if it doesn't crash he considers it 'good'...

Oh, huge part of why -ck and Kano parted ways is that issue about testing and having funds set aside to cover any fups. 1 guess who firmly thought it was/is not needed.
full member
Activity: 1022
Merit: 221
We are not retail.
May 25, 2020, 04:32:57 AM
#2
I believe this thread is referred to as defamation and your actions referred to as desperation.

There are zero issues with our code, score, or anything else. If anything laurentia pool is highly optimized for the stratum protocol without need for bloatware that only benefits multi node mining while delivering extremely low latency worldwide.
I would suggest you focus on developing your struggling user base in place of tearing down competition, likely you may receive more reward in heeding this advice and revive your pool.

Though I feel this thread/post should be removed it will serve as great motivation to continue to deliver for the user and for the space where others obviously fail and is the reason the pool exists to begin with.
I do appreciate you giving this opportunity to communicate here as I wouldn't disrupt your pool thread with baseless accusations, mud slinging, and trolling. In that regard if you follow me on twitter, you're way out of your league.

We're more than aware of incompatibility of fw on some asics and communicate to our users as appropriate so they can make the best decisions in their mining venture, and understand that our pool may not be the best choice for their equipment. We also are working on partnerships with fw developers to allow the use of these asics on our pool considering the volume of hashrate we are driving to achieve long term and would like to foster any relationship that would benefit our user base.

Our minimum cadence obviously adjusts in hashrate due to network conditions if you follow our thread or commitment page, it is updated frequently (which we've been getting numerous hits from AUS of late Wink). As a baseline we plan to be well beyond that mark to make relevant argument to your ranting. We are insured, plan to use funds from fee generation as proof of reserves in case of any errors in or out of our control to safety net our users work in lieu of insurance in the future.

Essentially there is zero information you can provide to us that we haven't accounted, and due to your own self destructive nature we're unlikely to take much notice either. Unless it's another desperate vie for attention like this thread. You're more than welcome to keep harassing and producing inflammatory content. If you feel you're doing the community a service here that's a wonderful soap box to be on when we're already 110% transparent.

We invite anyone with interest in our pool service to reach out at anytime. And if you have a productive thoughtful post for our pool thread Kano, it would likely not get removed.

#mineon

PS. Your pool has a high fee.
legendary
Activity: 4634
Merit: 1851
Linux since 1997 RedHat 4
May 25, 2020, 12:18:33 AM
#1
This thread is following the forum rules, Frodo, do not delete it, since that will (yet again) simply be your bias, and not within the rules of the forum.

The original thread owner has deleted both my posts, so following the forum message in that thread:
...
Our minimum average cadence is 5 blockfinds per diffadj.
...


There are issues with paying out to miners in the coinbase.
Some miners wont accept work that has a valid, but larger, coinbase.
The coinbase generated by your chosen pool software is, as is clearly stated, out of date with rewards each time it is sent to a miner.
It does not include the expectation of finding a block.
See the p2pool code about how that should be done correctly.


With the recent improved core bitcoin block distribution code (in the past couple of years), when you find a block, it can be distributed much faster than before.
However, if your coinbase is large due to including payouts in it, it will of course mean that your block distribution data will also be larger, since the coinbase transaction is the one transaction that no other bitcoin will know, and must be distributed to them all with the block.
Be wary of making it too large for miners or too large such that it slows down block distribution.

Lastly, you should understand, and comment about: if you will be using the Fibre Block Relay and how you handle when it isn't working.
The last outage of the Fibre Block Relay was the US-West node, for a number of hours recently, and thus during that time, any blocks you may have found and sent to that specific relay, will not have been distributed quickly around the world and into china, unless you have a secondary solution (like I have)
Ignoring this can easily lead to losing blocks due to stales and orphan races.
Also, not relaying your blocks yourself, directly into china, is an expectation of losing some blocks, since the largest mining pools are in china.

Feel free to ignore all this at the detriment of your miners Sad
Pages:
Jump to: