Quote from: jgarzik on March 12, 2011, 07:36:50 AM

Quote from: geebus on March 12, 2011, 06:42:41 AM

Quote from: jgarzik on March 12, 2011, 03:08:08 AM

Did I ask you to change anything? No. Simply stating the facts.

I didn't say you did. I just asked you to allow us to run our pool the way we choose to.

You seem to be implying that I am attempting to interfere with "allow[ing] us to run our pool the way we choose to." That implication is false, just like the assertion that changing the ask rate from 5 to 20 or more seconds "would find more shares" (direct FairUser quote).

You are and have always been free to run your server however you choose. I don't have the power to interfere with that, nor do I want to.

On the other hand, I and others are free to hang out in this forum thread and point out things like the above quote.

In any case, long pooling in slush's and Tycho's pool looks interesting. If it takes off, maybe that will eliminate the need for policies like this, in your pool.

To be fair, you're not interfering, just trying to publicly discount our comments, in our thread, about our pool, in reference to our users, and their participation in our pool.

Likewise, view the above datasets. Longer askrates yield a higher probability of shares in a lower amount of getworks when compared to incredibly low askrates and hoping an answer is discovered in less than a second of hashing, with a slow hashrate.

It's not a theory, it's not guessing, it's not some equation we're basing an assumption off of. It's data we've collected, and analyzed, and derived a reasonable principal from. That principal just so happens to be that a higher askrate, up to and including the number of seconds required to process an entire 2^32 getwork, does yield the same amount of results, with less getworks... and that the likelihood of finding an answer in milliseconds, or single seconds, is incredibly low.

Therefore: More answers for the same number of requests.