[DEAD] DeepBit.net PPS+Prop,instant payouts, we pay for INVALID BLOCKS too

[DeepBit]

Why do you think that coinbase payment would differ from normal sendmany TX ?

200 sends (not sendmany) is roughly 7.5 times the size of 200 coinbase payments in a single block.

i.e. a single send txn is around 258 bytes and a coinbase payment is around 34.5 bytes (if the coinbase txn pays around 200 addresses)

My single TXes are 233-234 bytes each, about 9% less than normal ones.

We may consider using sendmany in the future, but of course "coinbase" payments are not suitable in our case.

[cmg5461]

Any update on the time between blocks issue?

[kano]

Why do you think that coinbase payment would differ from normal sendmany TX ?

200 sends (not sendmany) is roughly 7.5 times the size of 200 coinbase payments in a single block.

i.e. a single send txn is around 258 bytes and a coinbase payment is around 34.5 bytes (if the coinbase txn pays around 200 addresses)

My single TXes are 233-234 bytes each, about 9% less than normal ones.

We may consider using sendmany in the future, but of course "coinbase" payments are not suitable in our case.

They are if you look at it a different way:
You could say: Your payment will appear in the next DeepBit block after 0, 1, 2 etc. DeepBit blocks
(once payments reach 50 it then goes in the next block after that ... etc)
Then use coinbase payments to do those payments ...
Since DeepBit blocks are 'usually' pretty regular - I'm sure most people would be happy with that
(and do send/sendmany for the users who aren't)
... and knowing they are reducing the block-chain requirements of their payments would help encourage them also.
... and of course coinbase payments do not need fees

Oh and just an FYI Deepbit uses about 1/6 of the block-chain space to pay their miners, per miner, compared to P2Pool - even with single sends.
Using only coinbase payments would reduce it by up to approx another 1/7 of it's current size
i.e. total block-chain usage ~1/42 of the P2Pool size

[DeepBit]

My single TXes are 233-234 bytes each, about 9% less than normal ones.

We may consider using sendmany in the future, but of course "coinbase" payments are not suitable in our case.

They are if you look at it a different way:
You could say: Your payment will appear in the next DeepBit block after 0, 1, 2 etc. DeepBit blocks
(once payments reach 50 it then goes in the next block after that ... etc)
Then use coinbase payments to do those payments ...

Why do you think that coinbase payments are better than sendmany ? They are much worse and I hope we will never use them for payments.

Do you know that users can't spend coinbase payments for ~20 hours after receiving ? I prefer to pay with ready-to-use/mature bitcoins :)

[kano]

My single TXes are 233-234 bytes each, about 9% less than normal ones.

We may consider using sendmany in the future, but of course "coinbase" payments are not suitable in our case.

They are if you look at it a different way:
You could say: Your payment will appear in the next DeepBit block after 0, 1, 2 etc. DeepBit blocks
(once payments reach 50 it then goes in the next block after that ... etc)
Then use coinbase payments to do those payments ...

Why do you think that coinbase payments are better than sendmany ? They are much worse and I hope we will never use them for payments.

Do you know that users can't spend coinbase payments for ~20 hours after receiving ? I prefer to pay with ready-to-use/mature bitcoins

Yep - and yep I left that point out - forgot to mention it.
Yes coinbase payments have to wait for the block to mature
(currently 120 blocks - and that is the same time for all pools of course)

Anyway, as I said in the rest of it, it's a payment method that is better for the block chain than send/sendmany and I'm sure if people had it as an option and bothered to look at the option to choose, some would certainly choose it (I would on all pools I mine on )

Being the biggest pool and advertising it as an option, I'm sure at least some of the other pools would also implement it for the sake of comparison - ok time for me to suggest it to another pool

[DeepBit]

Why do you think that coinbase payments are better than sendmany ? They are much worse and I hope we will never use them for payments.

Do you know that users can't spend coinbase payments for ~20 hours after receiving ? I prefer to pay with ready-to-use/mature bitcoins :)

Yep - and yep I left that point out - forgot to mention it.
Yes coinbase payments have to wait for the block to mature
(currently 120 blocks - and that is the same time for all pools of course)

Anyway, as I said in the rest of it, it's a payment method that is better for the block chain than send/sendmany

WHY do you think that coinbase payment is better than sendmany ? No way it can be true.
It may be a couple of percents smaller because no previous TX is needed, but it's not worth the 20-hours delay.

P.S.: And no, maturing time is not equal to 120 blocks for all the pools.

[kano]

Why do you think that coinbase payments are better than sendmany ? They are much worse and I hope we will never use them for payments.

Do you know that users can't spend coinbase payments for ~20 hours after receiving ? I prefer to pay with ready-to-use/mature bitcoins

Yep - and yep I left that point out - forgot to mention it.
Yes coinbase payments have to wait for the block to mature
(currently 120 blocks - and that is the same time for all pools of course)

Anyway, as I said in the rest of it, it's a payment method that is better for the block chain than send/sendmany

WHY do you think that coinbase payment is better than sendmany ? No way it can be true.
It may be a couple of percents smaller because no previous TX is needed, but it's not worth the 20-hours delay.

P.S.: And no, maturing time is not equal to 120 blocks for all the pools.

I'm referring to the coinbase payment of course, where the maturing time is the same 120 blocks after it goes in the block, no matter what.
You can't spend the BTC until 120 network blocks - and you certainly have the BTC at 120 network blocks.

[DeepBit]

Why do you think that coinbase payments are better than sendmany ? They are much worse and I hope we will never use them for payments.

Yep - and yep I left that point out - forgot to mention it.
Yes coinbase payments have to wait for the block to mature
(currently 120 blocks - and that is the same time for all pools of course)

Anyway, as I said in the rest of it, it's a payment method that is better for the block chain than send/sendmany

WHY do you think that coinbase payment is better than sendmany ? No way it can be true.

P.S.: And no, maturing time is not equal to 120 blocks for all the pools.

I'm referring to the coinbase payment of course, where the maturing time is the same 120 blocks after it goes in the block, no matter what.
You can't spend the BTC until 120 network blocks - and you certainly have the BTC at 120 network blocks.

1. 120-blocks limit is not a network-enforced rule, only the "official" client works that way. Network-enforced period is somewhere around 101 blocks.
And no, you won't "certainly" have the BTC at 120 blocks because the block can be orphaned. This should be checked by a special routine and re-sent if something happens.

2. I'll repeat my question for the THIRD time: Why do you think that coinbase payments are better than sendmany ?

[kano]

Why do you think that coinbase payments are better than sendmany ? They are much worse and I hope we will never use them for payments.

Yep - and yep I left that point out - forgot to mention it.
Yes coinbase payments have to wait for the block to mature
(currently 120 blocks - and that is the same time for all pools of course)

Anyway, as I said in the rest of it, it's a payment method that is better for the block chain than send/sendmany

WHY do you think that coinbase payment is better than sendmany ? No way it can be true.

P.S.: And no, maturing time is not equal to 120 blocks for all the pools.

I'm referring to the coinbase payment of course, where the maturing time is the same 120 blocks after it goes in the block, no matter what.
You can't spend the BTC until 120 network blocks - and you certainly have the BTC at 120 network blocks.

1. 120-blocks limit is not a network-enforced rule, only the "official" client works that way. Network-enforced period is somewhere around 101 blocks.

I find that rather disturbing that you of all people would make that statement.

The code in the current client indeed does say:

Code:

static const int COINBASE_MATURITY = 100;
& in GetBlocksToMaturity():
return max(0, (COINBASE_MATURITY+20) - GetDepthInMainChain());

and the earliest src I have on my computer at the moment: bitcoin-0.3.19, says that also.

I'm not sure what you mean by "network-enforced rule" vs "official client" unless you have changed that code in your pool to some other value and thus are trying to change the bitcoin rules yourself?

Of course I can write my own client to create a transaction worth 100 times my BTC balance, but no one's clients would accept it because that breaks the "Bitcoin rules"

And no, you won't "certainly" have the BTC at 120 blocks because the block can be orphaned. This should be checked by a special routine and re-sent if something happens.

Yeah OK, of course orphaned blocks can mean delays, however that is the whole point of the 100+20 code to ensure that the probability of a coinbase generation ever being orphaned once it gets to 120 confirms is so unlikely that it doesn't matter.

2. I'll repeat my question for the THIRD time: Why do you think that coinbase payments are better than sendmany ?

I thought I had already answered that

...

... and knowing they are reducing the block-chain requirements of their payments would help encourage them also.
... and of course coinbase payments do not need fees

Oh and just an FYI Deepbit uses about 1/6 of the block-chain space to pay their miners, per miner, compared to P2Pool - even with single sends.
Using only coinbase payments would reduce it by up to approx another 1/7 of it's current size
i.e. total block-chain usage ~1/42 of the P2Pool size

Edit: even when compared to a sendmany, a coinbase payment is smaller per payment.

[ancow]

I'm not sure what you mean by "network-enforced rule" vs "official client"

Geez, seriously? All compatible clients (even the official one) accept payments after 100 blocks maturity, the official client just won't generate them until after 120 blocks. 

[DeepBit]

1. 120-blocks limit is not a network-enforced rule, only the "official" client works that way. Network-enforced period is somewhere around 101 blocks.

I find that rather disturbing that you of all people would make that statement.

Well, it's expected that I know how mining works :)

The code in the current client indeed does say:

Code:

static const int COINBASE_MATURITY = 100;
& in GetBlocksToMaturity():
return max(0, (COINBASE_MATURITY+20) - GetDepthInMainChain());

and the earliest src I have on my computer at the moment: bitcoin-0.3.19, says that also.
Of course I can write my own client to create a transaction worth 100 times my BTC balance, but no one's clients would accept it because that breaks the "Bitcoin rules"

Why do you think it says "COINBASE_MATURITY = 100" instead of 120 ? :)
Official client won't create TXes with coins younger than 120, but will happily accept at 101.

Code:

            // If prev is coinbase, check that it's matured
            if (txPrev.IsCoinBase())
                for (const CBlockIndex* pindex = pindexBlock; pindex && pindexBlock->nHeight - pindex->nHeight < COINBASE_MATURITY; pindex = pindex->pprev)
                    if (pindex->nBlockPos == txindex.pos.nBlockPos && pindex->nFile == txindex.pos.nFile)
                        return error("ConnectInputs() : tried to spend coinbase at depth %d", pindexBlock->nHeight - pindex->nHeight);

And no, you won't "certainly" have the BTC at 120 blocks because the block can be orphaned. This should be checked by a special routine and re-sent if something happens.

Yeah OK, of course orphaned blocks can mean delays, however that is the whole point of the 100+20 code to ensure that the probability of a coinbase generation ever being orphaned once it gets to 120 confirms is so unlikely that it doesn't matter.

No, it's not how it works.
If I send normal TX and the block with it gets orphaned, this TX will be automatically re-mined again.
But if block with coinbase TX is orphaned, this TX won't be re-mined automatically and I'll need to check it after 100 blocks and if it's lost then there is another 20 hours period to wait.

2. I'll repeat my question for the THIRD time: Why do you think that coinbase payments are better than sendmany ?

I thought I had already answered that :P
Edit: even when compared to a sendmany, a coinbase payment is smaller per payment.

No, it's not. Why do you think it is ?

[rjk]

No, it's not how it works.
If I send normal TX and the block with it gets orphaned, this TX will be automatically re-mined again.
But if block with coinbase TX is orphaned, this TX won't be re-mined automatically and I'll need to check it after 100 blocks and if it's lost then there is another 20 hours period to wait.

This is the main point. Since Deepbit pays for invalid blocks, coinbase payments will not work properly without a lot of additional hassle.

[bittenbob]

Is it just me or is Deepbit finding a lot less blocks than it used to? I went from average rewards of .35BTC per day to .20-.25 with no change in hashing power. I originally thought my ISP might causing issues because I would see random server down-times but that hasn't happened in at least week. My average reward per block hasn't changed much but my total rewards haven't gone back up to their previous levels, making me suspect that the pool is finding less blocks than it used to.

Can anyone confirm this?>

[rjk]

Is it just me or is Deepbit finding a lot less blocks than it used to? I went from average rewards of .35BTC per day to .20-.25 with no change in hashing power. I originally thought my ISP might causing issues because I would see random server down-times but that hasn't happened in at least week. My average reward per block hasn't changed much but my total rewards haven't gone back up to their previous levels, making me suspect that the pool is finding less blocks than it used to.

Can anyone confirm this?>

The difficulty just went up.

[bittenbob]

The difficulty just went up.

Not by the same percentage as the drop in rewards though. It looks like difficulty increased by about 10% which should give me about .32BTC per day not .20-.25

[Edit]

After some research this appears to be a direct result of the so called mystery miner. Time will tell if this can be reverse or if it is going to be a permanent problem for Bitcoin.

[rjk]

The difficulty just went up.

Not by the same percentage as the drop in rewards though. It looks like difficulty increased by about 10% which should give me about .32BTC per day not .20-.25

[Edit]

After some research this appears to be a direct result of the so called mystery miner. Time will tell if this can be reverse or if it is going to be a permanent problem for Bitcoin.

Well besides the difficulty difference, you might also be losing to hoppers. How would the mystery miner affect it, unless he is somehow stealing your hashes (not possible)?

[bittenbob]

The difficulty just went up.

Not by the same percentage as the drop in rewards though. It looks like difficulty increased by about 10% which should give me about .32BTC per day not .20-.25

[Edit]

After some research this appears to be a direct result of the so called mystery miner. Time will tell if this can be reverse or if it is going to be a permanent problem for Bitcoin.

Well besides the difficulty difference, you might also be losing to hoppers. How would the mystery miner affect it, unless he is somehow stealing your hashes (not possible)?

I was a bit short sighted on that one. You are right that it should only affect the difficulty.

The only way it could be affecting my rewards were if it were confusing Deepbit every time it orphaned a block but I don't see that as a possibility.

[DeepBit]

I was a bit short sighted on that one. You are right that it should only affect the difficulty.
The only way it could be affecting my rewards were if it were confusing Deepbit every time it orphaned a block but I don't see that as a possibility.

No, MM is not affecting our mining in any way.
Only luck and difficulty are important.

If you aren't loosing any shares then your reward should be correct.

[Domchi]

I found a funny bug in Deepbit JSON API. When I pay out the exact amount on the balance, leaving balance at 0.0, I get this (incorrect) value of confirmed_reward:

"confirmed_reward":1.0e-8

Can anyone fix this?

[DeepBit]

I found a funny bug in Deepbit JSON API. When I pay out the exact amount on the balance, leaving balance at 0.0, I get this (incorrect) value of confirmed_reward:

"confirmed_reward":1.0e-8

Can anyone fix this?

Thanks for your report :)
I'll fix it.