How many shares per block?

[tohwis]

Hey,

I'm new to mining and I have a question. I mine for a pool and currently it has taken quite a bit of time to trying to solve the current block. Is there are a maximum number of shares per block before it gets solved or is it theoretically possible that a block never gets solved?

[Ricochet]

Given infinite time, a block will be found due to the nature of the algorithm.

What's much more likely to happen however is that someone else will find the block rather than the pool you're in.  There really doesn't exist such a thing as a "share" in the real Bitcoin world.  Pools just use them as a way to track how much you're contributing to the effort without micromanaging the exact number of hashes contributed.  For most (all?) pools, a "share" is really a valid block with difficulty 1, as in, a year ago when the difficulty was 1, you'd have solved a block and gotten 50 BTC for it.  Averaged over an extended period of time, counting shares is fairly proportional to counting hashes, but with much less strain on both the server and client.

[tohwis]

Okay, thanks for clearing this up for me.

So I have another question. When pools or individuals (who mine solo) try to solve a block, do they somehow "reserve" a part of the calculation, or are everyone mining just randomly trying to solve the same problem?

Sorry if I'm unclear, I'm still trying to figure out how everything works and my terminology may be way off.

[Dude65535]

The AVERAGE number of shares before a block is solved is equal to the difficulty. It is possible (but VERY unlikely) for a block you are working on to have no solution. However since the block you are working on changes all the time, it is not a problem. Every time a transaction is added to the block or someone else find a block the one you are working on changes.

There is no value to shares but they are a useful thing for pools to track because they can be verified and there is no way to looking for valid shares without looking for a valid block.

There is no duplication of work in mining pools. There is a number in the block that gets changed for every hash, the pool sends out work that start each worker at a different point. This does not give larger pools any noticeable advantage because there are many many possibilities to try and only a tiny fraction are tested before the block changes.

Edit: Also there is no duplication of work between pools or solo miners because the block you are working on contains the address to be paid the block reward.

[tohwis]

Okay, thanks a lot for clearing this up to me!

[kuloch]

Edit: Also there is no duplication of work between pools or solo miners because the block you are working on contains the address to be paid the block reward.

I'd like further clarification on the claim that different pools/soloers have no duplication of work between them.  Perhaps I'm just misreading, but there's no coordination between them.  Within a given pool, different nonce ranges can be doled out to avoid duplication.  And sure, there's a fair likelihood of relatively little duplicated work across different pools/soloers due to the vast problem space.  But relatively little is not the same as none.