which merchants do that? Because for me that's obviously scamming. Can't they even save a few bytes for more than 15min :O

I didn't understand how they want to track the carts and if it works with such a high accuracy i wouldn't use one.

I like the idea of a p2p text message network, the devices don't need a high range if many ppl use it and it could decrease fees by a lot i guess.

I'm sorry about that and have to admit that i didn't read your post too closely. But I think my post explains why it's useless to have a bounty on oprhaned blocks which still makes it of some use to read for you.

There's no advantage. bitcoind should be able to serve far enough getworks

are you using testnet in a box? Then what's your CPU?

A bounty on orphaned blocks doesn't make sense. As already mentioned noone has to and noone probably would include a transaction to the miner who mined the orphaned block.
And another point is that it wouldn't increase the "fairness" because normally everyone should have the same rate of orphans in average. The miners who are much above the average rate are doing something wrong they are wasting hashing power for which we shouldn't give a bounty since it doesn't increase the network security.

I like the lottery analogy. Let's assume every block starts a lottery round then normally the only one who wins is the one who is the first that brings his lottery ticket to the lottery company. With your proposal the one who is a bit late would also get a small amount of money. In the long run it's nothing else than giving the first one more money. For Bitcoin giving a bounty for orphaned blocks is the same as increasing the block reward to 51BTC because everyone has the same chance to find a block that gets orphaned. If we would still want to limit the total BTC amount we would have to change the block reward to 49BTC and the bounty for an orphaned block to 1BTC.
But then we didn't change anything we only decreased the variance for miners and we didn't increase the fairness for miners. And to decrease variance which is the only thing your proposal could do there are much better methods like pools or even decreasing the average time to find a block.

In the end your proposal doesn't help anyone and I have no idea how to implement it and that's enough to simply drop it.

The number of connections really doesn't matter. I've downloaded the chain at pretty much the same speed with 120 connections.
If you don't believe us forward the bitcoin port to your pc in your router. It doesn't matter if the additional connections are incoming or outgoing but it's better for the network health to have incoming connections and you wont have to recompile your client.

The limiting factor is almost always the harddisk and it has been improved in the new beta release AFAIK. Again if you don't believe us create a ramdisk and download the blockchain there. You should see a significant speed up since the RAM is much faster than your harddisk.

I can confirm that. For the bitcoind 256MB is just enough better go with 512MB. CPU doesn't matter.

As already mentioned you can relay your block through a proxy but that's actually not necessary because every node is a proxy already. Let's say I create lots of blocks. They probably don't go directly to you because it's pretty unlikely that we have a connection to each other. I send the blocks to the nodes im connected to and they send them further until they reach you. So you will only know the IP of the node you got the block from and not the IP of the creator of the block (me).

Did it now the interesting code was easier to find than expected. It's in main.py starting from line 574 if someone's interested to do that too.

all you need is bitcoind and this tool: https://bitcointalk.org/index.php?topic=18313.0

 
I've put it here because it's a question about the p2pool software and not about the pool itself. But if it is the wrong place a mod can move it

All my miners are individual persons and therefore all are using unique payout adresses and I understood the reward distribution now and it makes sense because one p2pool share can only contain a single payout address and everyone receives generated coins directly.


Imho they are not totally ignored because p2pool calculates the local hash rate and the graph based on them. And true there doesn't seem to be a way to export that data which is why I was asking here if someone knows one. My friends could mine on there own but since they are only mining sometimes they would have to wait for their blockchain to download and some of them don't even have a blockchain locally. That's why i invited them to my p2pool node which is running 24/7 anyways to get them away from the (too) big pools. And they are actually quite happy because no downtime so far but i wanted to offer them PPS.

So I guess I'll have to change the p2pool source code somewhere after the difficulty 1 shares are checked. If there's an easier way to export that data please let me know.

Thanks for the help

I couldn't think about a more specific title for this thread.

My first question is about the reward distribution of p2pool if I'm using multiple clients. Let's say there's a Miner A and a Miner B both pointing their hashing power to one p2pool node and they use their address as their username when connecting. Miner A finds a real p2pool share now and then p2pool finds a block. Does Miner B get something of that reward too based on his submitted difficulty 1 shares or does the entire reward go to Miner A?

My second question is about counting the difficulty 1 shares each miner submitted. Is there an easy way to export that count so I could use my own application to create statistics or do i have to change the code of p2pool?

I'm asking that because i've set up a p2pool node about 1month ago and more and more of my friends started to hash there instead of their traditional pools and the built in graphs of p2pool are not readable anymore. And i also wanted to offer them PPS payout to reduce their variance although they are pretty satisfied with it until now.

WTF?

start your own thread and try to use google or search the forum...
https://bitcointalk.org/index.php?topic=62842.0

Bitcoin mining is like playing the lottery and every !!hash!! is a ticket.
Let's say i buy 10 lottery tickets and in the household next to me live 10 people and everyone of them buys one ticket. The household of my neighbor now has obviously the same chance to win. That's pretty much your example with the 1GH/s card and the 10 100MH/s cards.
Others have already said whats wrong with your math.

But there actually is this kind of effect with distributed hashing power:

Let's say i have 2000 CPUs and each of them needs 2hours for one hash.
You have 1CPU that needs 0.002h=7.2s for each hash.

That means that we have the same hashing power but I will never find a hash and therefore no block because i have to start all over again after ~10min when a new block comes out.

But that problem doesn't exist, because noone is hashing with such a CPU :/
And it's not related to your calculations which are just wrong.

From my point of view the reason why ppl can do this without losing anything are the small transaction fees. But in the long term no processed transactions will cause higher transaction fees and that will cause more miners to include transactions.

I agree that they are harming the network because they could include transactions with little effort. But higher tx fees are anyways necessary for the future and this seems like a small step to it for me.

The transactions are currently not worth beeing included and that needs to change.

Are you running bitcoin-qt only? Then your mining reward will go directly to your wallet you can see with your bitcoin client. The only thing you should take a look at is the Local hashing rate, p2pool says is around 50MH/s but GuiMiner says its 250MH/s. Did you make that screen shot right after you launched p2pool?

Check http://localhost:9332/graphs if your Local hashing power is around 250MH/s

It will take a while until you get your first payment from p2pool but in total you will get way more than from deepbit if run it for a few weeks.

I don't think it's CGMiner clocking them down it's probably the driver that's why ppl said you should take an other program to check the GPU load.

First post outside the newbie section WOW

about your suggestion to abolish miners:

True without miners things might be better but the miners are needed. The Blockchain is the consens of the network. For every block there needs to be work involved because if there wasn't everyone could just create as much blocks as he wanted and there wouldn't be any consens. There also needs to be a block reward atleast until transaction fees become higher because why would someone solve blocks if he doesn't get anything for it. And how do you want to put the first money in circulation?

//edit just read that you want to hold the first coin.

So you decide who gets them :/

But if you do that you have to trust the one who gave you his priv key, because he could have a copy of it enabling him to spend the coins too. So it's not a very safe way and i doubt many people are using this method.