Topic: [1500 TH] p2pool: Decentralized, DoS-resistant, Hop-Proof pool - page 475. (Read 2591928 times)

Wow.

The newest firmware from KnC (0.97) makes their miners work FLAWLESSLY with p2pool!  The difference is staggering.  Before p2pool was literally unusable with my Saturn - but now I am getting close to max hashrate

Hello all,

I'm new to p2pool and one thing I miss from previous pools is an email notification when I get a payout.

Anyone know if this will work?



Edit: nevermind...answered my own question just now! For those interested it does work.

2 x 60GH/s miners here. Seems pretty good to me so far. What's the problem?

2013-10-22 23:08:48.262337 P2Pool: 17382 shares in chain (17386 verified/17386 total) Peers: 9 (3 incoming)
2013-10-22 23:08:48.262503  Local: 126GH/s in last 10.0 minutes Local dead on arrival: ~2.8% (1-5%) Expected time to share: 59.7 minutes
2013-10-22 23:08:48.262634  Shares: 32 (3 orphan, 1 dead) Stale rate: ~12.5% (4-29%) Efficiency: ~107.9% (88-118%) Current payout: 0.1353 BTC
2013-10-22 23:08:48.262764  Pool: 22105GH/s Stale rate: 18.9% Expected time to block: 14.5 hours

I'm referring to the KFC 10s LP delay
(Yeah ain't that scary that they did that!)

It's 30s now, but your point is still valid.

Th 10s LP delay makes p2pool mining with KFC a bad move until they work around that.

Yeah, I agree that it'd be nice to see Jupiters able to hash properly on P2Pool.

At this point in the process I have no concerns about the design.

Encouraging? Or, does it at least sound as if any shortcomings in the driver implementation can be fixed by you guys subsequent to you receiving a Minion/Prospero? (I understand the chips are called Minions and the units are called Prosperos). Assuming they send you a sample 100 GH/s at the least...  

If you mean the minions?
He replied in my cgminer hardware design thread:
https://bitcointalksearch.org/topic/m.3159041
(and yes I have PMed back and forth with him about the design)

Graciously accepted. Humility is a priceless quality, and exceptionally rare on these forums.

Now back to the important business of mining...

Moving forward, we've been trying to get manufacturers to take the idea of work restart a step beyond the current notion of throw all work out and start again and have at least 2 different levels of restart. One for "updated work" to throw out queued work that's not been started on, and one for "block change" which stops all current work and throws out all queued work. As ASIC manufacturers so far have only been interested in getting hardware out ASAP, it's clear these issues have not remotely concerned them, but we see encouraging signs that the next generation of ASICs will be far more professional and raise the bar for what the standard should be.

@ Kano & ckolivas

Just got a reply in the Black Arrow thread that their new units have got you guys collaborating on the miner driver... I asked them (simply) whether they would be p2pool ready or not, but what with all the long poll and work flushing talk in this thread, would you care to shed any light on how p2pool capable the prospero will be from your perspective? (Black Arrow just answered the question by saying that they will be using Stratum protocol; a fat lot of good these things would be without it! 2 TH/s on getwork, um, yeah)

Just a quick update on my 10GH processing via P2pool with my BE. I guess Ive been lucky as everything has always worked perfect for me on P2Pool, until this last difficulty rise. I used to get payouts in virtually every new block found. Ive been told by respected members of the forum I will still get my fair payout, and so I am continuing to processes via P2Pool even though I have not had a payout in around four solid days. I trust the members that have given me this advise. I will ofc feel happy when I see a payment again.......  ty all again.

avalon with 100G, but p2pool show 80G
why?  how to fix it?
thanks a lot to your reply.

Got a couple of ASICMiner Erupter blades running smoothly on p2pool for quite sometime now.
No need for proxy. Just configure yourusername+1 on the blade soft (note the +1 at the end to maintain the correct p2pool difficulty).
Results are actually better than obtained at other pools. The high DOA rate for these blades does not affect final results which are inline with the expected total payout.

A feature that will be more than welcome on p2pool will be a way to mitigate DDOS attacks that happen every couple of weeks or so. I had to block quite a few IPs on the firewall -many from Indonesia- . These IPs come and visit you node hundreds of times per second until they take it down. This forces to restart p2pool and block the attacking IPs on the firewall. Some kind of protection inside p2pool will be very welcome.

Agreed, sorry, I went in a 'big kid who lost his lolly' crusade to OffTopicLand. 

I also apologise for being a TAUC towards kano and ck - there's no need for it, even though I disagree with their choice of USB drivers, they're still producing one of the top-two mining programs, without which BTC wouldn't be where it is now.

I hope you accept my apologies.

Lulz. Funny how I seemed to list a number of things in my reply that you ignored

This thread is about p2pool.
But as I said in my reply to you that you completely ignored most of,
you're arguing in here about p2pool fail - yet you don't even understand the design of p2pool or why it works that way.

The hardware issue is simply hardware designed badly - problems that can easily be avoided.

These companies have ignored input about what should be done and gone ahead and made blunders.
To help with this, forrestv has changed p2ool to help deal with some of these more blatant hardware design flaws.

Seriously, the KFC miner effectively stops mining for 10s every LP.
Who but an idiot or BTC ignorant hardware designer would do something like that.
You can't get around that with p2pool, but you can avoid making your miner go idle in the hardware design ... as anyone, not ignorant of how BTC in general works, would do.

The basic design of p2pool is a sub chain of BTC.
That design is actually what will be required in the future for BTC to continue or it will die.
It won't be p2pool, but the concept is indeed what BTC will require to grow beyond it's rather blatant, current limitations.
... and it's nothing new about how to design the correct hardware, the problem is we have BTC noobs, who think they are know it alls, who then go and design hardware with totally fail mistakes, that are unnecessary and obvious to avoid.

+1

Seems to be a bit of a shitstorm going on here.

Guys, whatever your feelings are towards other forum members - personal attacks are uncalled for & just not cool.

Can we keep it clean & on-topic por favor?  

Edit: For the record, I'm a cgm man all the way - always have been & always will be 

And again, you pick up on the least important part of my post.  I've bolded the important parts for the hard of thinking.   

But, yeah, about the hot plug - who the hell unplugs their miners?  Mine are plugged in 24/7, mining 24/7.  Only time mine aren't mining is on the very rare occasion I'm rebooting the host PC.  Going by the "post your rig" threads, I think I'm in the majority group who mine the hell out of their machines, and aren't unplugging their miners willy-nilly.

I'd suggest you learn English.

So in your opinion hotplug is pointless - well I'm glad we got that clear.
Your opinion of hotplug is indeed just that and no more.

As for your master you praise - you forgot to mention that he extorts payments in IRC ... yes acolyte - continue your blind faith.
You think there might be a reason he was voted LAST at the Bitcoin Foundation election and got less than half the votes of 2nd last?
https://electionbuddy.com/elections/9787/results/pqdyb3n39
Though the difference between 1st and 2nd last wasn't very large.

Also learn about the design of p2pool and a whole new term for you to read up about ... LP ... and what the actual problem is with crap hardware design like the KFC.
The KFC has a 10s idle on LP
If that is beyond your understanding of what the problem is then I guess you've made it clear your argument is worth jack shit.