Author

Topic: [1500 TH] p2pool: Decentralized, DoS-resistant, Hop-Proof pool - page 250. (Read 2591928 times)

legendary
Activity: 1260
Merit: 1008
Thank you for the reply! I understand more now. And thanks for the calculation for intro hash power.

unfortunately, with my "the new R box" kickin out 110 Gh/s, I guess I'll have to stick to other pools. Maybe someday I'll dump the 750 for one of them bitcranes and really get things going.

Good luck finding that solution to the pool hash rate variance problem! When I get my bitcrane, i'll plugin into a p2p.

Anyone wanna buy me a bitcrane? I'll pay the electricity!
hero member
Activity: 686
Merit: 500
WANTED: Active dev to fix & re-write p2pool in C
''local'' and public, 300km away. The node is now down.
The strange part was that i had 0shares all the time Huh

You should have got some shares with that hash rate. What miner settings & hardware are you using? Location?
hero member
Activity: 742
Merit: 500
''local'' and public, 300km away. The node is now down.
The strange part was that i had 0shares all the time Huh
hero member
Activity: 686
Merit: 500
WANTED: Active dev to fix & re-write p2pool in C
nice as fuck! 4th/s in the pool for 2days and 0btc! back to my regular pools Smiley

It sounds like you're doing something wrong there - are you using your own local node or a public one? Post some info & we'll see what the problem is  Wink
hero member
Activity: 742
Merit: 500
Super NOOB question. Flame if you must.
if "x" miner have no shares = no btc even after mining for "x" amount of time after p2pool network hits "x" blocks

Right or wrong? Just want to confirm and clarify.

No active shares in the chain = no payout when a block is found

nice as fuck! 4th/s in the pool for 2days and 0btc! back to my regular pools Smiley
hero member
Activity: 686
Merit: 500
WANTED: Active dev to fix & re-write p2pool in C
+1

this

Anyone is able to develop that?

@forrestv: can this work? B-)

Hi lenny, not seen you here in a while,

I don't recall seeing a reply from forrestv regarding this - in fact I can't remember the last time forrestv commented here at all  Tongue

From what I can gather from various well informed devs, p2pool has outgrown python, it's just too slow (& single threaded), plus finding a dev who's understanding of p2pool & python has proven impossible to date. I believe a complete rewrite of p2pool is required in C/++ as there are far more coders out there who are familiar with it - plus it is faster & less resource hungry.

My 2 cents  Wink
legendary
Activity: 1500
Merit: 1002
Mine Mine Mine
Super NOOB question. Flame if you must.
if "x" miner have no shares = no btc even after mining for "x" amount of time after p2pool network hits "x" blocks

Right or wrong? Just want to confirm and clarify.

No active shares in the chain = no payout when a block is found

thx Smiley understand more about p2pool daily ....
legendary
Activity: 1258
Merit: 1027
Super NOOB question. Flame if you must.
if "x" miner have no shares = no btc even after mining for "x" amount of time after p2pool network hits "x" blocks

Right or wrong? Just want to confirm and clarify.

No active shares in the chain = no payout when a block is found
legendary
Activity: 1500
Merit: 1002
Mine Mine Mine
Super NOOB question. Flame if you must.
if "x" miner have no shares = no btc even after mining for "x" amount of time after p2pool network hits "x" blocks

Right or wrong? Just want to confirm and clarify.
full member
Activity: 312
Merit: 100
Bcnex - The Ultimate Blockchain Trading Platform
I really like it to.  Just wish the blocks were more consistent then every other day or 2, but every time i do the math it usually does close or better then Eligius.
legendary
Activity: 1344
Merit: 1024
Mine at Jonny's Pool
I've mined in p2pool since the end of March, with forays into other pools every now and then to see what they were about.  Always ended up back here.  I like running my own node and the concept of p2pool.
full member
Activity: 312
Merit: 100
Bcnex - The Ultimate Blockchain Trading Platform
I just cjecked the link you posted.  I am doing great.  Est is almost 3.6Th and i am averaging about 3TH according to my local pool hash rate. 

Do you mine in P2pool only or are you balanced between morn then one pool.
legendary
Activity: 1344
Merit: 1024
Mine at Jonny's Pool
In the last post are you saying that each miner should be a S3+ or higher per miner or a total of each user hash rates of all miners assigned the one payout address. I have 7 s3+ mining on my node and there all using the same Pay out address.  this seems to work ok, or would it be better to assign each miner there own payout address.

I also have 5 S2 that are pointing to a other pool. that seem to average 8-9 blocks a day so they make about double what my 7 S3+ do on p2pool node.


I was saying that minimal entry would be about an S3+ worth of hashing power.  You're doing the same thing I do - all my S3 use the same address to mine, so my node thinks I've got about 4TH/s (well, less now because I'm doing some testing).  As for earnings, there are a number of threads showing how p2pool has done compared to other pools.  Over the long run (since January of this year when then experiment began for BTCGuild and Eligius, and since July when BAN opened its door), p2pool has out-earned Eligius, BTCGuild and BAN.

You're doing just fine if you've found 8 shares in 24 hours.  A quick and easy way to check and see what the p2pool network thinks your hashing power is, go to http://minefast.coincadence.com/p2pool-stats.php and click on the Active Miners tab.  Find your address in that list.  If it shows higher than your actual hash rate, you've found more shares than average.  If it's lower, you've found fewer shares than you should have.
full member
Activity: 312
Merit: 100
Bcnex - The Ultimate Blockchain Trading Platform
In right about 24 hours my 7 S3+ have found 8 shares 1  orphaned. I did not think that was to bad.
full member
Activity: 312
Merit: 100
Bcnex - The Ultimate Blockchain Trading Platform
In the last post are you saying that each miner should be a S3+ or higher per miner or a total of each user hash rates of all miners assigned the one payout address. I have 7 s3+ mining on my node and there all using the same Pay out address.  this seems to work ok, or would it be better to assign each miner there own payout address.

I also have 5 S2 that are pointing to a other pool. that seem to average 8-9 blocks a day so they make about double what my 7 S3+ do on p2pool node.

legendary
Activity: 1344
Merit: 1024
Mine at Jonny's Pool
Okay, this is probably an idiot question, but here it goes. I really support the whole concept behind P2p, so I tried to mine in this pool.

I had low ping with coincadence.com, so I pointed my miner their...

now, all I did was point my miner their via cgminer.... if I'm pointing to a node, do I need to run an instance of p2p and bitcoind local?

On the stats page I was getting a hashrate, but on my cgminer output I was getting "share accepted" once every blue moon, and instead I was getting a lot of "stop work requests".

I found a small non-p2p pool with low ping and pointed their, and saw cgminer start spitting out accepted shares constantly.

NOw, I assume this is because the p2p pool has a a bajillion users, whereas the non-p2p pool I found has ~200, so my miner was really only getting a smaller amount of work on the p2p pool (Huh)

But the more I look into p2p, I'm thinking I may have done something wrong.
Not stupid questions at all...

If you point your miners to a remote node (like windpath's), you do not need to locally run an instance of p2pool or even bitcoind.  All you need is a valid bitcoin wallet address to which your portion of the block reward can be paid when a block is found.

The reason you see "share accepted" only once in a blue moon on a p2pool node vs all the time on a regular node has to do with the share difficulty.  If you connect to a p2pool node using only your bitcoin wallet address, you will be assigned a share difficulty by the pool.  The pool determines share difficulty based upon its total hashing power.  You can override this if you want by using +XXXX at the end of your wallet address.  For example, if you put 1DevLdogN52pHdjZnsgi4HzreFDB4ZHVre+500 as your username, you're telling the node that you wish it to consider all shares from your miner of difficulty 500 or more.  The lower that number, the more "share accepted" messages you'll see.

Remember, however, that those low difficultly shares mean absolutely squat.  Their only purpose is to give you visual reassurance that your miners are actually submitting work to the node.

A typical traditional pool will assign you a variable difficulty.  For example, if you were to mine on ckolivas' solo pool, that pool dynamically adjusts the share difficulty based upon your hash rate.  Again, and especially on his solo pool, those low difficulty shares mean nothing.  There are other payout methodologies and share difficulty settings in use, so it can get confusing.

Back to p2pool, the only share you submit that matters is one that at least matches the target difficulty of the share chain.  If it does, then your share is added to the chain and if that share happens to be in the payout list when the pool finds a block, you get your coin.  Right now, that difficulty is just about 10 million.  P2Pool constantly adjusts the share difficulty to try and maintain an average of 30 seconds per share added to the chain (just like Bitcoin adjusts its difficulty every 2016 blocks to try and keep the average block time of 10 minutes).  

People have different views on how much hash rate you should have to be mining on p2pool.  A good rule of thumb I use is "do I have a high enough hash rate to expect to find 1 share in the average amount of time it would take p2pool to solve 3 blocks".  If yes, then you would expect to have at least 1 share on the chain for every block p2pool finds.  If we use this rule, we can figure out how much hashing power constitutes the entry point:
Code:
Difficulty * 2^32 / hash rate = expected time to find a share
10436491.83 * 2^32 / hash rate = 179100
hash rate = ~250GH/s
Keep in mind that the difficulty is constantly changing and the p2pool total hash rate is constantly changing.  For example, while I was typing this reply, the expected time to block for p2pool went from over 16 hours to under 15.  This has an effect on the above calculations, which in turn changes the baseline for how much hashing power you should have.

So, how much hashing power should you really have on p2pool?  An S3+ typically does it... an S2 is better... an SP20 even better still.  If you've got anything under an S3 (like an old S1), then expect to see a ton of variance where you'll go for a number of blocks with no payout at all because you haven't found any shares.  This is one of the primary issues p2pool has - the more hashing power the pool has, the greater the variance experienced by individual miners.  As you can see from the first post on this page, people have been trying to find a solution for a long time.

Sorry about the long-winded post.  I tried to give you as much information as possible to help you better make a decision on whether p2pool is the right choice for you.
legendary
Activity: 1258
Merit: 1027
Okay, this is probably an idiot question, but here it goes. I really support the whole concept behind P2p, so I tried to mine in this pool.

I had low ping with coincadence.com, so I pointed my miner their...

now, all I did was point my miner their via cgminer.... if I'm pointing to a node, do I need to run an instance of p2p and bitcoind local?

On the stats page I was getting a hashrate, but on my cgminer output I was getting "share accepted" once every blue moon, and instead I was getting a lot of "stop work requests".

I found a small non-p2p pool with low ping and pointed their, and saw cgminer start spitting out accepted shares constantly.

NOw, I assume this is because the p2p pool has a a bajillion users, whereas the non-p2p pool I found has ~200, so my miner was really only getting a smaller amount of work on the p2p pool (Huh)

But the more I look into p2p, I'm thinking I may have done something wrong.

This sounds like expected behavior.

Because the p2pool share chain is much faster then the bitcoin blockchain the stop work requests happen much more frequently.

You do not need to run bitcoind or p2pool locally to mine on someone elses node, however if you have the resources to set it up it is the most efficient way to mine on p2pool.

Shares in p2pool have a much higher difficulty target then with a traditional pool, you will still earn the about the same (and in many cases a little more) compared to a traditional pool. Most nodes will give you an expected time to share (give it a while for your hashrate to report correctly), this expected time is subject to luck/variance so you will occasionally find share much faster, and occasionally much slower....

Welcome to p2pool!
legendary
Activity: 1260
Merit: 1008
Okay, this is probably an idiot question, but here it goes. I really support the whole concept behind P2p, so I tried to mine in this pool.

I had low ping with coincadence.com, so I pointed my miner their...

now, all I did was point my miner their via cgminer.... if I'm pointing to a node, do I need to run an instance of p2p and bitcoind local?

On the stats page I was getting a hashrate, but on my cgminer output I was getting "share accepted" once every blue moon, and instead I was getting a lot of "stop work requests".

I found a small non-p2p pool with low ping and pointed their, and saw cgminer start spitting out accepted shares constantly.

NOw, I assume this is because the p2p pool has a a bajillion users, whereas the non-p2p pool I found has ~200, so my miner was really only getting a smaller amount of work on the p2p pool (Huh)

But the more I look into p2p, I'm thinking I may have done something wrong.
legendary
Activity: 1036
Merit: 1000
DARKNETMARKETS.COM
+1

this

Anyone is able to develop that?

I was thinking, how to "pull  back" small miners (all miners?) to P2Pool and this is some assumptions for new version of share chain:
- we have 3-5 separate share chains
- each chain has own min and max share power/diff (let say sc1 is from 1k to 10k, sc2 11k-100k, 101k-1M etc)
- each share have flag, that tells it is already used/paid or not
- when user find a share in one of lower chains, earlier share is marked as used and power of new share and old share is summarized (we can`t pay from lowest chain directly because of dust threshold) 
- once power or his share reach chain threshold he need to find one stronger share to bump it to higher chain till it reach one that can be paid
- each miner (payout address) starts in highest chain for 1-10 mins that node can recognize its hash rate and select proper chain/diff for him
- goal is, that every miner found share every 1/10 -1/2 block ETA time

Pros:
- share chain length can be reduced because of "summing" thing, every miner can/need have 2 active shares in each chain, no more need
- no more "wasted work" when block time exceeds share chain length - new share sum power of last share and current, oldest share can be removed from payout computations
- virtually any miner can mine
- big miners are in high power chains that small miners can easily participate in mining

Cons:
- more P2P data overhead (more chains to transmit)
- more CPU overhead: more data to analyze to create payout tx, more job to be done when share found


Sadly, I`m too small in Python tot try implement that.

@forrestv: can this work? B-)
hero member
Activity: 686
Merit: 500
WANTED: Active dev to fix & re-write p2pool in C
Just wasted 4 hours of my life updating all my S3's to the latest firmware that was supposed to fix the last broken firmware not being able to do a factory reset, only to watch them all slowly drop off into oblivion..... Roll Eyes Not only that, but I was unable to get ckolivas p2pool binary installed on them either. Just rolled back to the previous broken firmware & everything is fine again. Just hope I don't actually need to reset them.......

I give up on Bitmain, really. Once again they have released another firmware that beaks their miners, as they did with the S2 & S4. They seem to just cobble something together & throw it out there in the hope that it works, instead of testing it first to ensure that it actually does work.

Thank the cyber gods for Spondoolies  Smiley
Jump to: