[1500 TH] p2pool: Decentralized, DoS-resistant, Hop-Proof pool - page 583.

maqifrnswa

sr. member

Activity: 454

Merit: 252

Quote from: mdude77 on March 13, 2013, 08:28:23 PM

p2pool coincidentally went south after stratum support was added. It's been down, quite a bit, since then, with no signs of coming back.

It could be all horrible luck. But the longer it goes on, the less likely it is bad luck.

http://p2pool.info/

M

Thanks - my point is that it doesn't help to keep posting there is a problem without actually finding what the problem is (or if there even is a problem). It's all open source yet no one can find a problem with the code. While the luck charts make it look like something is wrong, other analysis shows that it may be working ok:

Quote from: organofcorti on February 28, 2013, 04:00:57 AM

As a follow up to my last post, here's a better way of judging luck:

My point is to not assume something is wrong just because it feels wrong. It's your money, and can mine how you like - I just wanted to point out that there are no comments because as far as anyone can tell there is no problem.

mdude77

legendary

Activity: 1540

Merit: 1001

Quote from: maqifrnswa on March 13, 2013, 08:22:57 PM

Quote from: PatMan on March 13, 2013, 07:27:43 PM

Quote from: PatMan on January 18, 2013, 02:30:15 PM

Quote from: Subo1977 on January 18, 2013, 02:16:34 PM

No comments?

Indeed. It's also a shame, because if nothing is said or discussed about this problem P2pool will continue to lose miners and cease to exist. That would be a crying shame, because I believe it can work so much better than it is. This was my first and favorite pool, I like the idea of it very much, but at the moment - it's broken.

If it gets fixed, and I hope it does, I will have no problem joining it again.

And, look where we are now.......

Comments about what? What problem needs to be fixed? Is there a bug (besides increased variance combined with increased expected time to solve blocks)?

p2pool coincidentally went south after stratum support was added. It's been down, quite a bit, since then, with no signs of coming back.

It could be all horrible luck. But the longer it goes on, the less likely it is bad luck.

http://p2pool.info/

M

maqifrnswa

sr. member

Activity: 454

Merit: 252

Quote from: PatMan on March 13, 2013, 07:27:43 PM

Quote from: PatMan on January 18, 2013, 02:30:15 PM

Quote from: Subo1977 on January 18, 2013, 02:16:34 PM

No comments?

Indeed. It's also a shame, because if nothing is said or discussed about this problem P2pool will continue to lose miners and cease to exist. That would be a crying shame, because I believe it can work so much better than it is. This was my first and favorite pool, I like the idea of it very much, but at the moment - it's broken.

If it gets fixed, and I hope it does, I will have no problem joining it again.

And, look where we are now.......

Comments about what? What problem needs to be fixed? Is there a bug (besides increased variance combined with increased expected time to solve blocks)?

PatMan

hero member

Activity: 924

Merit: 1000

Watch out for the "Neg-Rep-Dogie-Police".....

Quote from: PatMan on January 18, 2013, 02:30:15 PM

Quote from: Subo1977 on January 18, 2013, 02:16:34 PM

No comments?

Indeed. It's also a shame, because if nothing is said or discussed about this problem P2pool will continue to lose miners and cease to exist. That would be a crying shame, because I believe it can work so much better than it is. This was my first and favorite pool, I like the idea of it very much, but at the moment - it's broken.

If it gets fixed, and I hope it does, I will have no problem joining it again.

And, look where we are now.......

mdude77

legendary

Activity: 1540

Merit: 1001

Quote from: gyverlb on March 13, 2013, 05:32:51 AM

Seems you didn't read me, so I repeat:

Quote from: gyverlb on March 12, 2013, 08:44:57 PM

Past luck doesn't guarantee future luck...

I read it. It's still down 90+ days. It could be luck. Or it could be something wrong. Those 90+ days I've been doing a lot better elsewhere, and I have every reason to believe I'll continue to do better.

It's your money.

M

maqifrnswa

sr. member

Activity: 454

Merit: 252

Quote from: kano on March 12, 2013, 10:49:56 PM

Sorry, you are incorrect, since you have ignored pointing out the 2nd part of the reject issue:

The problem with 10-15% rejects is that means others on p2pool with 4-5% rejects are getting a greater proportion of each block vs their hash rate.

Yes if EVERYONE was mining at 10% rejects, then everyone would get the same proportion of income vs their hash rate.

However, those with better reject rates get a proportionately better payment rate and that extra comes from those with the worse reject rates.

That's true, and needs to be understood. Even so, I frequently am trying to explain p2pool to people that are getting just 4-5% rejects (that is, they are doing well compared to the network), and they still think they are losing money compared to other pools (when they are actually making more compared to other pools for being more efficient than others on p2pool).

All I was trying to emphasize is that reject rate on p2pool does not mean the same thing as reject rate on other pools.

organofcorti

donator

Activity: 2058

Merit: 1007

Poor impulse control.

Quote from: maqifrnswa on March 13, 2013, 04:38:11 PM

Quote from: organofcorti on March 13, 2013, 02:05:45 AM

If the last n blocks are sufficiently unlucky that for 95 runs of n block solvings out of 100 those n blocks will have required fewer shares to solve, then by definition, the next n blocks will have a 95% probability of being "luckier" than the last n blocks.

Ha, yes, if you are unlucky a lot, your next blocks are expected to be luckier than your unlucky blocks. It doesn't mean that your expected value changes, just your return compared to previous blocks. If I buy two lotto tickets, and lose the first one, my expected outcome of the second one is greater than the known result of the first one.

Just so. I thought that's what rav3n_pl meant - when luck has been bad it is more likely to get better than worse. Not better than expected, just better.

maqifrnswa

sr. member

Activity: 454

Merit: 252

Quote from: organofcorti on March 13, 2013, 02:05:45 AM

If the last n blocks are sufficiently unlucky that for 95 runs of n block solvings out of 100 those n blocks will have required fewer shares to solve, then by definition, the next n blocks will have a 95% probability of being "luckier" than the last n blocks.

Ha, yes, if you are unlucky a lot, your next blocks are expected to be luckier than your unlucky blocks. It doesn't mean that your expected value changes, just your return compared to previous blocks. If I buy two lotto tickets, and lose the first one, my expected outcome of the second one is greater than the known result of the first one.

Aseras

hero member

Activity: 658

Merit: 500

Quote from: stevegee58 on March 13, 2013, 10:27:40 AM

We need some of these big new ASIC rigs that have been making the rounds lately on BTCGuild and eligius to show up at p2pool.

Heck, 1 of them would be nice.

I have a batch 1 order ( of several units ) and the same in batch 2. they'll be here as soon as they get to me and I can get them working.

gyverlb

hero member

Activity: 896

Merit: 1000

Quote from: Aseras on March 13, 2013, 10:20:41 AM

Quote from: K1773R on March 13, 2013, 06:33:04 AM

Quote from: gyverlb on March 13, 2013, 05:32:51 AM

Quote from: gyverlb on March 12, 2013, 08:44:57 PM

Past luck doesn't guarantee future luck...

Past luck dosnt affect future's luck at all...

On p2pool, it does. If all the miners leave because they think the pool sucks, and the main page says they are loosing 50% of their income by staying here, then yeah, past luck is going to ruin future luck.

You are mistaking variance with luck. I used p2pool for 6 month, variance over such a period is very small even with p2pool: my payments are currently equivalent to what I would have had with ~99.8% PPS.

rav3n_pl

legendary

Activity: 1361

Merit: 1003

Don`t panic! Organize!

600GH total power and we all be happy.
About 3 blocks/day and every1 will want to mine in P2pool

stevegee58

legendary

Activity: 916

Merit: 1003

We need some of these big new ASIC rigs that have been making the rounds lately on BTCGuild and eligius to show up at p2pool.

Heck, 1 of them would be nice.

Aseras

hero member

Activity: 658

Merit: 500

Quote from: K1773R on March 13, 2013, 06:33:04 AM

Quote from: gyverlb on March 13, 2013, 05:32:51 AM

Quote from: gyverlb on March 12, 2013, 08:44:57 PM

Past luck doesn't guarantee future luck...

Past luck dosnt affect future's luck at all...

On p2pool, it does. If all the miners leave because they think the pool sucks, and the main page says they are loosing 50% of their income by staying here, then yeah, past luck is going to ruin future luck.

Luck in CO-OPs and pools like this is decided almost as much by faith. If p2pool had enough hashrate, everything would be fine.

gyverlb

hero member

Activity: 896

Merit: 1000

Quote from: K1773R on March 13, 2013, 06:33:04 AM

Quote from: gyverlb on March 13, 2013, 05:32:51 AM

Quote from: gyverlb on March 12, 2013, 08:44:57 PM

Past luck doesn't guarantee future luck...

Past luck dosnt affect future's luck at all...

My point exactly (I was just paraphrasing the "past performance doesn't guarantee future results" disclaimer seen in investment contracts).

People using the "current" luck (which in fact is past luck) of a pool to decide if it's a good one to mine at simply don't understand mining.

I could understand using variance to choose a pool and admittedly it is not good on p2pool currently due to low hashrate.

K1773R

legendary

Activity: 1792

Merit: 1008

/dev/null

Quote from: gyverlb on March 13, 2013, 05:32:51 AM

Quote from: gyverlb on March 12, 2013, 08:44:57 PM

Past luck doesn't guarantee future luck...

Past luck dosnt affect future's luck at all...

gyverlb

hero member

Activity: 896

Merit: 1000

Quote from: mdude77 on March 13, 2013, 05:18:52 AM

Quote from: gyverlb on March 12, 2013, 08:44:57 PM

Quote from: mdude77 on March 12, 2013, 07:37:59 PM

Aside from the fact p2pool's 90 day "luck" is at 88.5%, and 30 day "luck" is at 57.1%. Aside from that, yes, it's great.

M

I don't see what luck has to do with greatness. Past luck doesn't guarantee future luck...

Are you here for greatness or coins? If the former, grats. If the latter, you could do better at the most expensive PPS pool.

M

Seems you didn't read me, so I repeat:

Quote from: gyverlb on March 12, 2013, 08:44:57 PM

Past luck doesn't guarantee future luck...

mdude77

legendary

Activity: 1540

Merit: 1001

Quote from: gyverlb on March 12, 2013, 08:44:57 PM

Quote from: mdude77 on March 12, 2013, 07:37:59 PM

Aside from the fact p2pool's 90 day "luck" is at 88.5%, and 30 day "luck" is at 57.1%. Aside from that, yes, it's great.

M

I don't see what luck has to do with greatness. Past luck doesn't guarantee future luck...

Are you here for greatness or coins? If the former, grats. If the latter, you could do better at the most expensive PPS pool.

M

organofcorti

donator

Activity: 2058

Merit: 1007

Poor impulse control.

Quote from: maqifrnswa on March 06, 2013, 12:19:38 PM

Good question, heading out but want to think about it a bit more later. I think it may has to do with conditional probability, which ties into gambler's fallacy (unless I'm reading this too quickly). ......

I just realised that I'd been misunderstanding you. The answer is simple to explain.

If the last n blocks are sufficiently unlucky that for 95 runs of n block solvings out of 100 those n blocks will have required fewer shares to solve, then by definition, the next n blocks will have a 95% probability of being "luckier" than the last n blocks.

Examples:
1. If one block required 3*D of D1 equivalent shares to be solved (CDF = ~ 0.95), then there is a 95% probability that the next block will take less than 3*D D1 equivalent shares to be solved.

2. If the last ten block required 15.7*D of D1 equivalent shares to be solved (CDF = ~ 0.95), then there is a 95% probability that the next ten blocks block will take less than 15.7*D of D1 equivalent shares to be solved.

As for my original example,
I have discovered
a trulymarvelous
proof of this, which
this margin is too
narrow to contain.

kano

legendary

Activity: 4592

Merit: 1851

Linux since 1997 RedHat 4

Quote from: maqifrnswa on March 12, 2013, 04:01:33 PM

Quote from: lenny_ on March 12, 2013, 07:08:29 AM

Can you please share your settings with me? cgminer is able to get similar results?

Many people have said this, but I want to make sure lenny_ understands: The 10-15% "rejects" you see in p2pool have nothing to do with a decrease in income nor with wasted work. 10-15% rejects on p2pool is the same as 2-2.5% rejects on any other pool. The reason is that the p2pool sharechain is different (in behaviour) than the bitcoin blockchain.

I'm periodically fighting FUD of "p2pool sucks because they have a ridiculous reject ratio." I then explain it to people, with examples showing how you make more money with a 10-15% reject rate on p2pool than you would on a 2% reject rate on another pool. A few days later, someone else pops up and says the same thing.
...

Sorry, you are incorrect, since you have ignored pointing out the 2nd part of the reject issue:

The problem with 10-15% rejects is that means others on p2pool with 4-5% rejects are getting a greater proportion of each block vs their hash rate.

Yes if EVERYONE was mining at 10% rejects, then everyone would get the same proportion of income vs their hash rate.

However, those with better reject rates get a proportionately better payment rate and that extra comes from those with the worse reject rates.

gyverlb

hero member

Activity: 896

Merit: 1000

Quote from: mdude77 on March 12, 2013, 07:37:59 PM

Aside from the fact p2pool's 90 day "luck" is at 88.5%, and 30 day "luck" is at 57.1%. Aside from that, yes, it's great.

M

I don't see what luck has to do with greatness. Past luck doesn't guarantee future luck...

Topic: [1500 TH] p2pool: Decentralized, DoS-resistant, Hop-Proof pool - page 583. (Read 2591920 times)