bit pit - (LP, ESMPPS, 8-decimal payout, SSL, API, 0% fee, Almost 0% Stales!)

[1bitc0inplz]

It's been a rather crazy day!

I do apologize for the little bit of down time we had this morning. I do really appreciate all you guys' patience while we sorted out that.

On to block #7! 

[lebuen]

What I'd really like to see is that you put focus on protection from pool hoppers. Either by delaying the stats (tough this can only marginally reduce the effect), or by introducing score/pps-based rewards. The MaxPPS system used by Eligius seems to work well. In the last round we were able to see 10-20 GH leaving in the middle of the round, and this will probably become even worse in the future.

[flower1024]

i don't like pool hoppers too.

but PLEASE don't use score / maxpps or anything else - just stick with pps.

i don't mine 24/7 so for me it's just an additional variance.

if you really think you have to do something about it i suggest "blind mining" and just show stats from previous day.

so the calculation would simple be: (blocksperday*50) / (sharesperday)

[lebuen]

but PLEASE don't use score / maxpps or anything else - just stick with pps.

i don't mine 24/7 so for me it's just an additional variance.

MaxPPS and PPS is actually more or less the same - and the best choice for someone who's not mining 24/7. But right now the pool is using Prop (that's not PPS!)

[flower1024]

sorry my fault.

i wanted to say: please stick with prop.

i just don't like other systems. thats one of the reasons i use this bitpit (just left minecoin for their switch to another payout system).

[lebuen]

You don't "like" other systems... that's ok. But with prop and realtime stats, you actually get ripped off by pool hoppers, so you'd be better off with PPS.

[flower1024]

that's why i suggested a delay?
i really like the idea to go prop on a day basis.

e.g. (blocksperday * 50) / (sharesperday) = your payout for that day
obvious: it has a huge stats delay. but who of you sits in front of your pc and watch estimated rewards the hole day? i cant believe that....

[lebuen]

e.g. (blocksperday * 50) / (sharesperday) = your payout for that day

i don't know how you want to make that work. prop means: you get paid when a block is found. while the pool is still small, this can mean days, weeks.

[flower1024]

there are two possibilites (when no block is found):
1) just take the shares to the next day
2) make it for two, three etc. days

i dont know whats best, i just dont want that every pool switches to pps/maxpps (...) systems. i left btcguild as they are too big, i left minecoin for their switch: i just want to stay with a pool a while. tired of switching.

and (for me) pool selection is an emotional decision (after fees, stales and stableness .

[hugolp]

Was that another round?

That is lucky.

Btw, I support implementing some sort of anti-pool hooping measure. Its tempting for me to do it since its profitable.

[luffy]

this is the best pool right now to stick around :p
even with the pool hopers, they actually help us to find 2 blocks in 2 hours!!!!!!
that was great, the hashrate will reach 100GH soon

[1bitc0inplz]

Wow! I was just about to post congratulations for ending round 7, and lookie there, we're on round 9! Good work everyone.

Just for the record, for our first round we were using Mini's scoring system. It has some major flaws, and due to user feedback we actually left that method.

Due to it's major flaws, I'm not entirely convinced that his scoring system is the right answer to the conversation.

However, I am not against SMPPS or possibly other scoring systems.

SMPPS would be very far to part-time miners, while helping to compensate for long rounds. In general it's not such a bad idea. However, I would appreciate more people chiming in before we'd implement such a change, as I think it merits some type of vote.

I am going to be AFK shortly, but everyone to chime in on what your thoughts are. Ultimately bit pit wants to provide a service that improves the mining experience as a whole, and if changing our reward distribution improves the overal mining experience then it fits without the scope of what we are trying to do.

[flower1024]

what do you think about just print a list of users (maybe without names) which profits most in a "pool hopping"-way?

means: calculate how much they get in fast rounds and how much they get in slow rounds..."free" it of their hashrate (sorry.. dont know how to say that in english correctly).

so all users can see who and how much they gained from hopping
 - then it would be possible to vote them out (or do additional steps if it is just to much)

if you hand me your db scheme (not the data) i may come up with a nice statement (if you use nosql, just give me that schema)

like your pool too! keep going!
just have a small problem with firefox 5: can't change my payout address (switched temporary to ie and it worked)

[1bitc0inplz]

like your pool too! keep going!
just have a small problem with firefox 5: can't change my payout address (switched temporary to ie and it worked)

Keep up the conversations, this is excellent food for thought. I do like hearing all sides of a discussion like this.

Also, thinks for the bug report in Firefox 5. I will look into that shortly.

[roos]

Im all for the system used on the first block. A scoring system with very short alive times of the share really gives everyone that contribute when the block is found a fair share of the lottery win and it doesnt keep new people from joining in on long blocks.

Really, propotional is not propotional at all, its more like redistribution from lucky shares to unlucky shares. That is fine in a sense, but mostly its effect is encouraging hopping and making people who like to optimise their winnings (like myself) restless and likely to change pools when the blocks are slow.

[zoro]

i like only prop in a large pool >100GH. pool jumpers should be no more than 10% i guess, no real influence to the pool's luck.
am i wrong?

[flower1024]

i think all that counts is the percentage of users jumping.

if there are more than > 10% a long block will even take longer, on a VERY long block it shouldnt really matter as your reward get higher over time.

and: pool hoppers do make it more possible for very short rounds - as there is more hashing power

so: at the moment i am unsure if they do any harm at all....

[OldChap]

I hopped in around day 10 of the long round. I thought there would be some benefit to the pool to have what at the time was another ~5% of hashing power. I consider that the hours I spend mining * hash rate are paid the same to all.... I guess I am saying would someone explain the problem with pool hopping...I don't see it http://serve.mysmiley.net/confused/confused0024.gif

[dominatro]

There was a big storm here, flooding, lightning, and water went all over my running hardware. This was just before the long round finished. I had to wait a day to make sure it was dry to even see if it still worked after I moved everything. I'm even surprised it still works after I saw water splash straight into my fans and all over the computer boards. There was water,paint, and dirt on top of capacitors and all the fans. Even the harddrive was on the ground when water poured over it while it ran, and it still works. I was standing in the water on the floor when lightning struck and I felt the current.

But unfortunately that means I missed the good luck streak here, I started mining again now, but in the chaos I forgot my password. Could you add a reset password from e-mail feature?

[hugolp]

I hopped in around day 10 of the long round. I thought there would be some benefit to the pool to have what at the time was another ~5% of hashing power. I consider that the hours I spend mining * hash rate are paid the same to all.... I guess I am saying would someone explain the problem with pool hopping...I don't see it

Hoping in is not a problem, the problem is hoping out.