Hi Ciuciu,

Chlorine notified me of the situation, and the fee compensation for april through october has just been transferred for your account. All other affected accounts have also just received their compensation for these months.

Sorry for the delay. I've been a bit demotivated by the fact that getting the new pool software ready is taking so much longer to than I had hoped, and had not checked the forum and my PM's for quite some time as a result.

Rest assured the compensation agreement still stands until your stolen BTC are fully compensated.

Regards,
MC

Thanks for sharing this observation MacT. Let me explain what you're seeing.

At particularly busy moments, shares from the pool server cannot be written to the database right away. While they are cached, the stats page does not know about them yet and displays a lower hashrate. Once the load returns to lower levels (load spikes are usually due to payouts and statistics aggregation being processed), the cached shares are written to the database and they are added to your balance.

The shares are however timestamped at their original arrival time. Since the hashrate calculation only takes the last x seconds of submissions, it will not overestimate your hashrate to 'make up' for the earlier underestimation. Instead it just ignores the old entries and displays the correct hashrate.

I had seen a sporadic rise in the number of cached shares before, but did not realize its effects would show on the stats pages.

We're still working diligently on the new version of the pool software, and I'm glad to be able to announce it's almost ready. This new version will be able to better support the new generations of mining hardware while keeping our system load lower. This will help in keeping the number of shares in the caches lower.

Hope this helps,
MC

Sorry to hear that ShadowAlexy. Can you PM us the details of the problem you are having with statistics?
 
As for GPUMax, in reference to the thread linked below: We don't block GPUMax. It is our opinion that their users should decide for themselves wether to trust them. Rest assured that any GPUMax presence does not affect ABCPool miners, since ABCPool pays PPS.

If you are experiencing technical issues in combining GPUMax with ABCPool, please elaborate so we can look into it. You may also PM about that if you like.

Regards,
MC

https://bitcointalk.org/index.php?topic=109197.80

Many a pool op must be scrambling right now to deal with the quick increases in hashpower: Network speed touched 20TH recently. Accross the board we're seeing varying symptoms of capacity problems at upstream pools, and those are never good for the miners. And we're having our own growing pains as well. The next release of our pool software (due in a few weeks) will address these, and furthermore will offer both time-rolling and higher difficulty shares for those interested.

Your address was probably perfectly valid, but we were restarting the payment bitcoind. It's up again. I've also just improved that error message

Dear miners,

We're having some serious issues with high load, causing our upstream share submissions to fail significantly more than usual. Since we pay you guys for a share no matter if we succeed in getting paid for it ourselves, this is costing us a lot, and the situation is financially unsustainable for us.

We're working like crazy to remedy this. The solution we are working on is quite involved though, and it looks like it's going to be around three weeks before we have a handle on the problem.

In the meantime I regret to inform you that we need to raise the fee to 4.8% temporarily. Current fee is 2.8%, so it will be 2.0% higher. Once the issue is under control we can return to a lower fee.

This measure will cover part of our losses and may also lead to somewhat lower load in the short term. The new fee will go into effect 24 hours from now, 30-8-2012 12:00:00 UTC.

Sorry to have to do this now and on such short notice. I hope you can understand.

MC & Chlorine

420,
Unrounded PPS amounts are accumulated as double precision floating point values every minute, then added to your balance with a precision of eleven decimals. The site displays necessarily displays a rounded amount, because some fractions would be infinite.

I can tell you there's a lot more than 2 pools not using roll-n-time. It's in our short-term plans to start supporting it, for reasons you are very familiar with

We are not really 'gone' from ozco.in. We're still mining there, albeit a little less, but I get what you mean. Our backend chooses where to get work based on a myriad of factors. Our own fee is not one of those factors, so the upcoming fee change will (barring other changes) not have any effect on work selection.

Please see our FAQ for an example calculation using the current difficulty.

That's unknown at the moment because of the difficulty change coming up soon. You can calculate it if you know the difficulty: PPS = 50/difficulty*0.972

NB: Our top priority for now remains fixing the current idling-on-new-block issue a number of you are experiencing.

For over a year now ABCPool has operated with the now proven PPS recipe. In this time we have developed into one of the best Bitcoin mining pools out there, even more so if you're looking for the steady and transparant rewards our PPS mechanism provides.

The Bitcoin mining scene has changed considerably in the time ABCPool has been operating. With the next generation of mining equipment on the horizon we see more changes and development will be needed by all pools that want to continue to offer mining services.

ABCPools approach to redistributing risk over a multitude of pools is what has made our low fees possible, and we intend to stick with this strategy because it benefits our miners (low fees), us (low risk) and the upstream pools (less variance for their miners). Ozco.in is one of our main partners for this approach. They recently raised their fees by 3%. Of course this decision had an impact our costs as well. We cannot help but react to this by reconsidering our own fee.

After long deliberation we came to the conclusion that to ensure the stability, long term viability and continued development of ABCPool as a Premium Bitcoin Mining pool we need to raise our fee for service by 1.3%. Naturally we will keep paying for your stale shares, and all other benefits we're offering stay in place just like you're used to.

The new policy will come into effect 16:00:00 UTC, August 24 2012.

We would like to take this opportunity to thank all miners, donors and developers that have supported the site so far, without you we couldn't have come this far!

MC & Chlorine

My estimate is we'll have it corrected within the next 5 days.

I can confirm the getwork request failures from earlier today, I experienced them on my own miners too. I've tuned the server configuration since then; how's it been for the past hour?

Good news,
The deployment seems to have been successful. The new version has been running without any problems for a few hours now, and we've verified a significant drop in resource usage. You may want to return your miners to ABCPool.

We're aiming to process the announced manual balance correction before the weekend.

Also, registration has been opened up again. If you were unable to join ABCPool last week, now's your chance!

Happy hashing,
MC

We'll be doing another production trial of yesterday's patch shortly.

You'll be credited correctly. We need to make a manual correction for the period earlier today, which we will do when things are stable again. You may keep mining at ABCPool with an occasional hickup, or switch for the next 36 hours.

It's quite a comprehensive patch we're rolling out today, and there were some issues with it. The pool was unavailable for a while, and some shares have not been accounted for yet. We will adjust your balances for that in the coming days.

Things have stabilized now it seems, but I can't promise that we're out of the woods for today. We'll continue to monitor the pool closely.

I'm sorry we didn't provide you with more advanced warning of this.

MC

Usually, it's not necessary. In some transitions the LP connection is not severed neatly, which consequently causes buggy miners to fail doing LP from that point on. They never recover from that one broken LP. So the essence is: If you are missing log messages about LP and are suddenly experiencing an elevated amount of invalid shares, try restarting your miner.

This can actually happen at any pool, even during normal operation, when the connection has a hickup. Therefore it's wise to check for it now and then anyway.

I remember seeing this problem with Phoenix v1, but only because I was using it myself. I have not tested other mining software, or more recent versions of Phoenix.

When we see elevated invalid shares pool-wide after a patch, we always post a message here on the board asking you to restart your miners.