NastyPoP vs Standard P2Pool

[jonnybravo0311]

Well, p2pool did much better this week with seven day luck of 127.84%.  Also, in the closest race since I've been doing this, p2pool found a block at almost exactly 19:00 UTC.  Looks like it will be included in next week's payout run as it just missed the cutoff by a few seconds.  Because I didn't get to this post until after that block was found, it's influence in the luck is already there... so this week's and next week's luck values might be a bit off because of it.  Like I wrote... it was seconds after the 19:00 cutoff that the block was found.  Both miners beat expected earnings.  NastyPoP was only 101.35%, whereas NastyP2P was 167.63%.

At any rate, here are this week's numbers:

5/1 - 5/8
NastyPoP - 0.03294026BTC
NastyP2P - 0.05447989BTC
Expected - 0.0325BTC
Luck - 127.84%

OP updated

[jonnybravo0311]

Last week p2pool found a block seconds after the payout from NastyPoP and I stated then that it would impact this week's luck numbers because I didn't catch the value before that block was found.  This week, the 7 day luck looks to be 84.97%, so if we average out the two numbers, we get 106.41%.

In terms of the miners, my S3 on the standard NastyP2P payout went on strike apparently, finding very few shares, and hence the payouts were lower than expected.  As usual, the miner on NastyPoP was much less variant.  Here are the numbers for the week:

5/8 - 5/15
NastyPoP - 0.03472755BTC
NastyP2P - 0.02216391BTC
Expected - 0.0325BTC
Luck - 84.97%

OP updated

[jonnybravo0311]

Well, it appears the forums are finally back up after the ordeal of the past few days.  Thankfully I recorded the numbers on Friday .  In other news, my S3 on NastyPoP began misbehaving this weekend.  After nearly a year of operation at a rock steady 440GH/s, it now shows a couple chips with either "x" or "-", resulting in a hash rate of no higher than 400GH/s sustained.  I'm not too happy about it, but since it has long since paid for itself, it is what it is.  It is a batch 1 S3, and as we all know electronics don't last forever.

In an attempt to salvage things, I'm going to down clock both S3s from the standard 218.75 step by step until I find something stable.

So... how about the numbers from the week?  Well, p2pool had the luckiest week we've had since the test started at 182.64% of expectations.  We also see here the drastic effect variance can have on the standard p2pool miner.  With my S3 on the NastyP2P payout, I saw an unfortunately high number of orphaned shares, which resulted in abysmal earnings.  Orphans happen - it's a fact of life mining on p2pool.  This is where mining in the collective NastyPoP benefits you.  Your orphaned share has a far smaller impact on the whole than it does if you're on your own.  The numbers clearly show this, insofar as even with the great pool luck, my S3 mining on NastyP2P didn't even meet expected numbers for the week.

Here are the results:

5/15 - 5/22
NastyPoP - 0.05258228BTC
NastyP2P - 0.03171723BTC
Expected - 0.032BTC
Luck - 182.64%

OP updated

[OgNasty]

Been a good couple of weeks for P2Pool/NastyPoP miners. 

[jonnybravo0311]

I've just gotten back from a much needed vacation where I had virtually zero access to anything online.  Unfortunately, I didn't capture the luck values, but here are the numbers for the past two weeks:

5/22 - 5/29
NastyPoP - 0.04249927BTC
NastyP2P - 0.03263267BTC
Expected - 0.0307BTC

5/29 - 6/5
NastyPoP - 0.03711295BTC
NastyP2P - 0.03414086BTC
Expected - 0.0312BTC

Both NastyPoP and NastyP2P beat expectations over the past two weeks, and the NastyPoP payout method has now beaten the standard p2pool payouts for the past four weeks.

OP updated.

[jonnybravo0311]

Another week gone by.  P2Pool had another good week, and both of the S3s beat expectations.  NastyPoP couldn't pull off a fifth week in a row beating standard payouts, but it was close.  Here are the numbers:

NastyPoP - 0.04351160BTC
NastyP2P - 0.04518917BTC
Expected - 0.03144BTC
Luck - 122.76%

OP updated

[jonnybravo0311]

Just another weekly update... p2pool's good luck streak came crashing to a halt these past 7 days.  Oh, and retrocalc.net needs some loving... the calculations are nowhere near correct (noticed it last week, and it's still going on).  The standard p2pool payout beat NastyPoP, but neither of them met the expected earnings:

NastyPoP - 0.01964088BTC
NastyP2P - 0.02788807BTC
Expected - 0.030107BTC
Luck - 65.12%

OP updated

[jonnybravo0311]

Oops... I guess I forgot to update the thread last week.  I captured the numbers, but never posted.  So, this week I'll give the numbers from both last week and this.  P2Pool had a run of bad luck these past few weeks, which offset the previous few weeks of good luck.  As such, the numbers reflect the luck.

On another side note: if you haven't already done so, and you run your own p2pool node, update it to the latest version.  See the posts in the official p2pool thread upgrade announcement.

So, here are the numbers for the past couple weeks:

6/19 - 6/26
NastyPoP - 0.02159208BTC
NastyP2P - 0.0193531BTC
Expected - 0.030107BTC
Luck - 60.76%

6/26 - 7/3
NastyPoP - 0.02828584BTC
NastyP2P - 0.01623161BTC
Expected - 0.030263BTC
Luck - 81.3%

OP updated

[jonnybravo0311]

And another week of poor luck has passed us by.  A lot of things happened recently... adoption of BIP66, p2pool upgrade to 14.x, "stress tests", and so on.  I'm not trying to insinuate these things caused the bad luck, but we've certainly had a number of things thrown at us all at once.  At least the blocks that p2pool did mine this week contained transactions, unlike our friends over in China who have the vast majority of the BTC network's hash rate.  Tens of thousands of transactions waiting to confirm, and they're sill pumping out empty blocks.

Here are the numbers:

7/3 - 7/10
NastyPoP - 0.02039533BTC
NastyP2P - 0.02521461BTC
Expected - 0.030289BTC
Luck - 71.40%

OP updated.

[jonnybravo0311]

Apparently I've been slacking ... 3 weeks behind.  Sheesh.  Overall, p2pool is on the upward side of a good luck swing.  The week ending 7/17 was a bad week at only 72% luck, but the last 2 weeks p2pool made up for it.  This week shows p2pool at over 130% luck and the past 30 days, p2pool is over 105%.  In the past couple of weeks, with the good luck, NastyPoP has beaten standard p2pool payouts.  Even with luck over 130% this week, standard payouts didn't even meet expected earnings.  Here are the numbers for the past three weeks:

7/10 - 7/17
NastyPoP - 0.02054230BTC
NastyP2P - 0.02884854BTC
Expected - 0.03493028BTC
Luck - 71.40%

7/17 - 7/24
NastyPoP - 0.04771791BTC
NastyP2P - 0.04040042BTC
Expected - 0.03465856BTC
Luck - 162.79%

7/24 - 7/31
NastyPoP: 0.04298881BTC
NastyP2P: 0.03145364BTC
Expected: 0.03401418BTC
Luck - 134.83%

OP updated.

[OgNasty]

It is cool to see the updates.  Thanks for keeping this going so long!  Any chance for a pretty chart to look at?  Have you thought about expanding this to include some other pools as well?

[jonnybravo0311]

I suppose I could make some pretty charts/graphs/whatever... or I could just supply the data and let those who are far more inclined at such tasks take it and run.  As for including other pools, I had considered it.  However, I've not decided which pools I'd compare.  There was a thread on here last year where somebody was comparing p2pool to Eligius and GHash.io (I think), but that thread ran its course and I'm not sure if the author is even still mining - all of his data was based on S1s on those pools.  There's another guy who's doing his own comparison of different pools here: https://bitcointalk.org/index.php?topic=1059040.0.  I'd love to give him my data points so he could include the results of NastyPoP and standard p2pool in his comparisons, but he hasn't really seemed too interested in it.

Are there any pools in particular you'd like to see added?  I'd prefer to stay away from the big ones (AntPool, Discus Fish) because I don't believe in adding any more hashing power (albeit rather minuscule in comparison to their totals) to the two largest pools - both of which use software that has caused a forked chain when BIP66 went into effect and mine empty blocks, even with thousands of transactions waiting to be confirmed.

[OgNasty]

I suppose I could make some pretty charts/graphs/whatever... or I could just supply the data and let those who are far more inclined at such tasks take it and run.  As for including other pools, I had considered it.  However, I've not decided which pools I'd compare.  There was a thread on here last year where somebody was comparing p2pool to Eligius and GHash.io (I think), but that thread ran its course and I'm not sure if the author is even still mining - all of his data was based on S1s on those pools.  There's another guy who's doing his own comparison of different pools here: https://bitcointalk.org/index.php?topic=1059040.0.  I'd love to give him my data points so he could include the results of NastyPoP and standard p2pool in his comparisons, but he hasn't really seemed too interested in it.

Are there any pools in particular you'd like to see added?  I'd prefer to stay away from the big ones (AntPool, Discus Fish) because I don't believe in adding any more hashing power (albeit rather minuscule in comparison to their totals) to the two largest pools - both of which use software that has caused a forked chain when BIP66 went into effect and mine empty blocks, even with thousands of transactions waiting to be confirmed.

I don't have any particular pools in mind.  I thought you have been doing this so long, a chart and some other pools might take it to the next level.  I'm great at giving other people suggestions with stuff to do. 

[jonnybravo0311]

So... here's the chart of the current data set (from 12/26 - 7/31):



I've updated the OP to include it as well.  Also, I have some errors in my data on the OP... math errors mostly because I'm an idiot and did everything by hand previously, but have now dumped it all into an Excel spreadsheet .  I'll fix the OP eventually, but as of now, here's the correct data:



Here's the spreadsheet if anyone wants it:

https://www.dropbox.com/s/vv7r7ejq2pquynl/pop_vs_p2p.xlsx?dl=0

[OgNasty]

So... here's the chart of the current data set (from 12/26 - 7/31):

That is awesome.  Thanks for doing that!

If I were to make a suggestion...  You could have the data listed as a running total on the graph so that you can see the end result over time.

[jonnybravo0311]

Something akin to this?

[OgNasty]

Something akin to this?

Exactly like that.   

[jonnybravo0311]

Well... a lousy week for p2pool - only finding 2 blocks and having 55.38% luck overall.  Neither of my miners made anywhere near expected payouts as a result.  NastyPoP did manage to do better than standard p2pool, though.  Here are the numbers:

7/31 - 8/7
NastyPoP: 0.01487946BTC
NastyP2P: 0.01145337BTC
Expected: 0.02962897BTC
Luck - 55.38%

OP updated.

[jonnybravo0311]

P2Pool's luck certainly turned around this past week.  Both NastyPoP and NastyP2P beat expectations by quite a bit.  NastyPoP would have done better, but it appeared that the nodes were down last weekend.  My miner failed over to a backup pool.  I'm not sure why the pool was down (both OgNasty's and TheAnalogKid's NastyPoP nodes were offline), but the result is that the payout was less than it should have been given the number of blocks p2pool found.

8/7 - 8/14
NastyPoP: 0.03914884BTC
NastyP2P: 0.07990332BTC
Expected: 0.02941349BTC
Luck - 241.06%

OP updated.

[TheAnalogKid]

I'm not sure why the pool was down (both OgNasty's and TheAnalogKid's NastyPoP nodes were offline), but the result is that the payout was less than it should have been given the number of blocks p2pool found.

DNS issues with the domain provider took the nodes offline.