[0Th]Ozcoin Pooled Mining |DGM 1%|Stratum+VarDiff port 80|NEW CN mining|

[Pontius]

Looks like there's still a lot to do.

The pool stats on the top are broken, the website loads rather slow (compared to YourBTC.net as you are using their frontend) and after about 70 minutes of mining there I disconnected my miners as I got 8% BTC and 18% NMC rejects (on YourBTC.net is used to be something like 0.2% and 1.2% rejects). I guess I'll retry ozco.in when things have settled in a week or two.

Edit: Oh, I just realized - happy new year! 

[1713921019]

1713921019

[1713921019]

1713921019

[-ck]

I for one welcome our new DGM overlords.

[Graet]

Hi Pontius, thanks for trying us.
Yep still some to do, it was all runing smooth as silk in testing, when we went live wierd errors have cropped up.
Fixing the mining side was given higher priority over site stats etc, got the header stats up last night.
Still working on a lot of the site and need to smooth the stats a bit (maybe lot)

I realise it is no excuse but the festive season has distracted the coders, tho props to gotmilk who went hard coding right up to the minute he had to go out to NYE celebrations and Wayno for a couple of allnighters recently

Thanks to the miners persevering (thank goodness for failover) our goal is to make an excellent pool for all miners
check back regularly

"overlords" hehe
Thanks ckolivas

[Graet]

Did some patching to psj overnight.
Pool isnt crashing like it was, still getting too many comms errors and stales - work continues
small progress

[Graet]

we are winning
au.ozco.in:8332 is back on Australian server
US server has been tweaked only the occasional error around longpolls currently
us.ozco.in:8332 and us.ozco.in:80
eu.ozco.in:8332 coming soon 

Site stats still fluctuate wildly, smoothing is next on todo list

Thanks to miners that have persevered through all the errors

[Graet]

mining on eu.ozco.in:8332 is up for load testing.
us.ozco.in:8332 , us.ozco.in:80 and au.ozco.in:8332 are now on seperate servers and performing well 

[deepceleron]

Looks like I am 1% of the pool with a lowly 5770 thrown your way, let's see how it goes!

[Graet]

one job we must do soon is smooth the stats, the swing between 20and 60Ghash is unrealistic :/
welcome deepceleron

[-ck]

Apart from the stats needing smoothing out, everything's looking really good right now. Good work 

[kano]

...
    Instant payout, pay invalids for 5%+ donors
...

Hmm never noticed that here before.
What exactly do you mean by that?
What are you referring to as "Invalids"?

(I presume you are not referring to blocks that get orphaned or such - coz they should be dealt with by simply calculating shares from the last successful block ...)

[btcbids]

looking nice

[rjk]

170 Ghash really Graet? Why aren't you back in the top 10 list?

[-ck]

...
    Instant payout, pay invalids for 5%+ donors
...

Hmm never noticed that here before.
What exactly do you mean by that?
What are you referring to as "Invalids"?

(I presume you are not referring to blocks that get orphaned or such - coz they should be dealt with by simply calculating shares from the last successful block ...)

Pretty sure that means orphans.

[Graet]

...
    Instant payout, pay invalids for 5%+ donors
...

Hmm never noticed that here before.
What exactly do you mean by that?
What are you referring to as "Invalids"?

(I presume you are not referring to blocks that get orphaned or such - coz they should be dealt with by simply calculating shares from the last successful block ...)

Hi Kanoi,
Hmmm well the pool frontend, backend, payout method and the 1st post all changed with DGM, so yes there will be a lot you "never noticed that here before."
yes invalid/orphan.

This is part of what we inherited with our new front end.
I am well aware that Ozcoin WAS one of the few pools to pay valid block to valid block (effectively ignoring invalid/orphan blocks and paying all valid shares submitted)
The Dillema:
5% donors get instant payout
currently the site software resets to 0 on any block valid or not
we cant tell at 0 confirmations if the block will be valid or not
so if we roll from valid block to valid block and pay donors instantly on invalids do we rest donors scores to 0 and affect their share value on the "rest of the block" or do we only pay valid blocks to non-donors or do we give donors a negative balance until a valid block is found and then adjust thier payouts??

This has been the topic of discussion among staff for days and may well change but currently it is how it is

I would have thought as a coder you would realise "should be dealt with by simply" isnt always simple
thanks for the feedback

170 Ghash really Graet? Why aren't you back in the top 10 list?

Hi rjk,
Closer to 200-220 this morning
A few reasons i guess
1: The top 10 is updated by forum mods when they see fit
2: We only hit 170Ghash yesterday and about 130 of it is one big miner, I will wait a few days for hashrate to settle before hassling mods - hope this guy stays and more join but i tend to err on the side of caution
3: Whist we are counting and paying shares correctly - the displayed hashrates vary wildly both site and user, it is a stats issue that needs to be fixed before I consider us worthy of top 10 I look at the raw figures in psj and update our irc chan regularly on real speeds

Thanks for the kind thoughts - we plan to be back in the sticky list very soon

A huge 48 hours of work has gone on, huge props to gotmilk_, Wayno, shadders and nikade for hours of tireless work and good humor.
Thanks to urstroyer for popping in last night and helping with a few issues
We now have a solid team with staff located in Au Eu and US enabling us to give 24/7 support, we keep an eye on the shoutbox and site forum but the quickest and easiest way to get support currently is via the WebIRC link on the site or if you have an IRC client #ozcoin on chat.freenode.net

Graet

[kano]

...
    Instant payout, pay invalids for 5%+ donors
...

Hmm never noticed that here before.
What exactly do you mean by that?
What are you referring to as "Invalids"?

(I presume you are not referring to blocks that get orphaned or such - coz they should be dealt with by simply calculating shares from the last successful block ...)

Hi Kanoi,
Hmmm well the pool frontend, backend, payout method and the 1st post all changed with DGM, so yes there will be a lot you "never noticed that here before."
yes invalid/orphan.

This is part of what we inherited with our new front end.
I am well aware that Ozcoin WAS one of the few pools to pay valid block to valid block (effectively ignoring invalid/orphan blocks and paying all valid shares submitted)
The Dillema:
5% donors get instant payout
currently the site software resets to 0 on any block valid or not
we cant tell at 0 confirmations if the block will be valid or not
so if we roll from valid block to valid block and pay donors instantly on invalids do we rest donors scores to 0 and affect their share value on the "rest of the block" or do we only pay valid blocks to non-donors or do we give donors a negative balance until a valid block is found and then adjust thier payouts??

This has been the topic of discussion among staff for days and may well change but currently it is how it is

I would have thought as a coder you would realise "should be dealt with by simply" isnt always simple
thanks for the feedback
...
Graet

Well actually it is simple, you complicated it by offering instant payout.
I must admit I also wonder why instant payout is even offered (by any pool)

Also ALL PPS type pools always have paid 'invalid' blocks ...

However, the issue with it is that since this pool hash rate at the moment may mean only a few blocks per day and with variance sometimes only 1 or less blocks per day - this means that anyone who doesn't 'donate' 5% (I donate 0%) can lose an entire day of hashing if you get an orphan block since the pool sent the block out to the network too slowly or the pools bitcoind isn't well enough connected to get it out faster than the block that succeeded and made the orphan.

Paying orphan blocks separately also does another thing:
It takes money from the pool to cover the cost of payouts that are not covered by the following valid block.

If you remove all this crap from the payout process (and remove instant payouts) it does make the coding MUCH simpler and the people not 'donating' 5% are not losing out due to this pool design flaw.

It also makes the term 'donate' incorrect since you must 'donate' 5% if you wish to avoid losing when the next orphan block comes around.
Edit: it is actually a 'fee' since it is insurance - and I'm sure you don't consider the money you pay to any insurance company a 'donation'

I've had this same discussion over at BTCGuild 5 months back (with a much higher block rate thus the variance equated to a much lower hashing time lost and then it was 2.5%)
I'll just post a link here for anyone interested in the discussion:
https://bitcointalk.org/index.php?topic=7760.msg442593#msg442593

There is also the issue of: how is an invalid block any different to a rejected block that is at correct difficulty ...

[Inaba]

Kano,

Just read the other thread you linked to.  You seem to misunderstand what a valid/invalid block really is.  You are right in so far as "invalid" is a misnomer and inaccurate.  However, your reasoning goes off the rails when you state that "all the work is lost before."  It's not (or I dunno, maybe it is on OzCoin, but not most pools, including EMC) - what has actually happened is a strict statistical sense, as well as in the harsh reality, is the "invalid" block is a phatom block as far as mining goes.  While it's hash did in fact satisfy the difficulty criteria, someone else found a lower difficulty/better difficulty and the block that was found really wasn't.  Some/most pools list it as found and then invalidate it later, but purely from a blockchain point of view (which is all that really matters in the end), the block that turns out to be invalid on a given pool should never have been found in the first place, so you/the miners continue mining as if no block had been found (because one hasn't). 

So nothing is lost, you just had your hopes dashed through a false block alert.  A pool could mask a block found for a given amount of time until it's reasonable to assume it's going to be valid and then show miners, but then people would complain about the delay.  In either case, it's impossible to please everyone... however, the general consensus seems to be that immediate reporting and later invalidation is preferable to block announcement delay.  Six of one, half dozen of the other.

[Graet]

Thanks for the input kanoi, as i said this has been the topic of discussion among staff for some days
Currently what is stated on the first post is what is happening - if/when changes are made the main post will be updated
Miners are welcome to mine on the pool or not depending on how they feel about the current conditions at the pool - or any pool for that matter. Actually it would be good if more miners investigated pools closer before jumping on - but thats a whole other subject...

Not sure where you pegged "our current hashrate" from ~30ghash Saturday morn (my tz) to psj reporting over 300Ghash now (2pm monday afternoon)

As a poolop i have noticed many miners do like an instant payout option- some seem desperate for instant BTC, thus the 5% option, maybe "insurance" or "fee" or any number of other words are better name than dontaion...
just one example is non donors that are happy to wait 120 confirmations but would like to be able to manually withdraw at 2% fee, when the auto payout script runs every 10 mins...

I apologise to you and the mining community for launching before the site was perfect...(btw which one is perfect and not still under development?)

We appreciate the miners that have stuck through the instability and other issues, they see the site is developing and are liking what they are seeing. Development is continuing.

kanoi maybe you are one of the miners that should have another look in a week or a fortnight

you will have to excuse me if i have chosen the wrong word here or there or made a misspell hopefully you can understand what i am trying to say, this was typed between picking up and caring for a vomiting child.

best
Graet

[kano]

Thanks for the input kanoi, as i said this has been the topic of discussion among staff for some days
Currently what is stated on the first post is what is happening - if/when changes are made the main post will be updated
Miners are welcome to mine on the pool or not depending on how they feel about the current conditions at the pool - or any pool for that matter. Actually it would be good if more miners investigated pools closer before jumping on - but thats a whole other subject...

Not sure where you pegged "our current hashrate" from ~30ghash Saturday morn (my tz) to psj reporting over 300Ghash now (2pm monday afternoon)

The hash rate was around 200+ when I typed that.
Even at 300 what I have said is still valid.
Even at 400 the expected time for a block is 13430 seconds (3hr 43min 50sec)
So a 5x share variance (6,253,788 shares) would be 18hr 39min 10sec - a lot of hashing time for a person to lose if they don't donate 5% and an orphan occurred.

Also, the 120 confirms is there for a reason.
Unlike a transaction that will simply wait for another block if the block it is in is orphaned, the block itself once orphaned is gone forever.
That logically means to me that it doesn't make sense at all to immediately payout.

As a poolop i have noticed many miners do like an instant payout option- some seem desperate for instant BTC, thus the 5% option, maybe "insurance" or "fee" or any number of other words are better name than dontaion...
just one example is non donors that are happy to wait 120 confirmations but would like to be able to manually withdraw at 2% fee, when the auto payout script runs every 10 mins...

I apologise to you and the mining community for launching before the site was perfect...(btw which one is perfect and not still under development?)

I am not sure why you are apologising or what you are apologising about, Graet ...
That one item is the one I dislike and I have stated why.

We appreciate the miners that have stuck through the instability and other issues, they see the site is developing and are liking what they are seeing. Development is continuing.

kanoi maybe you are one of the miners that should have another look in a week or a fortnight

you will have to excuse me if i have chosen the wrong word here or there or made a misspell hopefully you can understand what i am trying to say, this was typed between picking up and caring for a vomiting child.

best
Graet

Kano,

Just read the other thread you linked to.  You seem to misunderstand what a valid/invalid block really is.  You are right in so far as "invalid" is a misnomer and inaccurate.  However, your reasoning goes off the rails when you state that "all the work is lost before."  It's not (or I dunno, maybe it is on OzCoin, but not most pools, including EMC) - what has actually happened is a strict statistical sense, as well as in the harsh reality, is the "invalid" block is a phatom block as far as mining goes.  While it's hash did in fact satisfy the difficulty criteria, someone else found a lower difficulty/better difficulty and the block that was found really wasn't.  Some/most pools list it as found and then invalidate it later, but purely from a blockchain point of view (which is all that really matters in the end), the block that turns out to be invalid on a given pool should never have been found in the first place, so you/the miners continue mining as if no block had been found (because one hasn't). 

So nothing is lost, you just had your hopes dashed through a false block alert.  A pool could mask a block found for a given amount of time until it's reasonable to assume it's going to be valid and then show miners, but then people would complain about the delay.  In either case, it's impossible to please everyone... however, the general consensus seems to be that immediate reporting and later invalidation is preferable to block announcement delay.  Six of one, half dozen of the other.

Inaba, no I do understand clearly what an orphan block is
The work lost is what I dislike.
BTCGuild prop did that. The new Ozcoin does that.
I'm putting forward my argument why I dislike it.
Hopefully it might change before an orphan block occurs

[Graet]

Thanks for the input kanoi, as i said this has been the topic of discussion among staff for some days

I think this was the most important line in my post - and the line you skipped over?
I do understand your concern and I did raise this with the coding team some days ago
and yes i hope there are no orphans before it is resolved

BUT I also stated in the 1st post how the pool is operating right now - I am not trying to decieve anyone here. Just to state how things are setup at the moment.

[farfiman]

Is there still auto-conversion of namecoin to btc ?