vane91
Member


Activity: 133
Merit: 26
|
 |
July 04, 2015, 10:25:52 PM |
|
yes unless someone starts developing a ASIC for block verification... which will be actually good news! 
|
|
|
|
|
Peter R
Legendary

Activity: 1162
Merit: 1034
|
 |
July 04, 2015, 10:43:30 PM Last edit: July 04, 2015, 11:11:20 PM by Peter R |
|
|
|
|
|
|
|
solex
Legendary

Activity: 1078
Merit: 1007
100 satoshis -> ISO code
|
 |
July 05, 2015, 12:02:59 AM |
|
Some numbers:
Assume it takes on average 30 seconds to verify 1 MB of typical transactional data (k =0.5 min / MB). Since T = 10 min, this means the maximum average blocksize (network capacity) is limited to:
Seffective = T / (4 k) = (10 min) / (4 x 0.5 min / MB) = 5 MB.
QED. We've shown that there exists a limit on the maximum value of the average blocksize, due to the time it takes to verify a block, irrespective of any protocol enforced limits.
Great work Peter, but do we have any empirical evidence for the 30 seconds? Seems surprisingly high and I would have guessed just a few seconds.
|
|
|
|
Peter R
Legendary

Activity: 1162
Merit: 1034
|
 |
July 05, 2015, 12:07:18 AM |
|
Some numbers:
Assume it takes on average 30 seconds to verify 1 MB of typical transactional data (k =0.5 min / MB). Since T = 10 min, this means the maximum average blocksize (network capacity) is limited to:
Seffective = T / (4 k) = (10 min) / (4 x 0.5 min / MB) = 5 MB.
QED. We've shown that there exists a limit on the maximum value of the average blocksize, due to the time it takes to verify a block, irrespective of any protocol enforced limits.
Great work Peter, but do we have any empirical evidence for the 30 seconds? Seems surprisingly high and I would have guessed just a few seconds. No, I just made it up. I think I'll change it to 15 second, as I agree it's probably too high. I guess we could estimate it by looking at the ratio of empty blocks to non-empty blocks produced by F2Pool. If someone wants to tabulate that data, I'll update my post.
|
|
|
|
|
TheRealSteve
|
 |
July 05, 2015, 12:23:37 AM |
|
I guess we could estimate it by looking at the ratio of empty blocks to non-empty blocks produced by F2Pool.
If someone wants to tabulate that data, I'll update my post.
If you by empty you mean coinbase-only, then in the last 27,027 blocks (basically since jan 1st 2015), f2pool-attributed blocks: 5241, of which coinbase-only: 139 For antpool, this is 4506 / 246. Not sure if that's all the info you'd need, though. See also: Empty blocks [bitcointalk.org]
|
|
|
|
thezerg
Legendary

Activity: 1246
Merit: 1018
|
 |
July 05, 2015, 12:31:33 AM |
|
Your modification to require the inputs to state which block it comes from is a clever way to reduce the addr does not exist proof. But I dont understand your subsequent complexity. If the txn input states that block B is the UTXO then the invalid proof is simply to supply B, right?
|
|
|
|
|
justusranvier
Legendary

Activity: 1400
Merit: 1069
|
 |
July 05, 2015, 12:41:24 AM |
|
If the txn input states that block B is the UTXO then the invalid proof is simply to supply B, right? That's one way to do it, however even this can be shortened. Right now with all the blocks < 1 MB it's not really a big deal to supply the entire block to prove that the referenced transaction doesn't exist, but it'd be nice to not require the entire block especially for when blocks are larger. By adding a rule to new blocks that require all the transactions to be ordered by their hash, you don't need to supply the entire block to prove that the transaction doesn't exist. It would be good to have that ordering requirement in place before blocks are allowed to grow to make sure that fraud proof size is bounded.
|
|
|
|
|
Peter R
Legendary

Activity: 1162
Merit: 1034
|
 |
July 05, 2015, 12:41:29 AM Last edit: July 05, 2015, 01:16:29 AM by Peter R |
|
I guess we could estimate it by looking at the ratio of empty blocks to non-empty blocks produced by F2Pool.
If someone wants to tabulate that data, I'll update my post.
If you by empty you mean coinbase-only, then in the last 27,027 blocks (basically since jan 1st 2015), f2pool-attributed blocks: 5241, of which coinbase-only: 139 For antpool, this is 4506 / 246. Not sure if that's all the info you'd need, though. See also: Empty blocks [bitcointalk.org]Awesome! Thanks!! We can estimate the average effective time it takes to process the blocks, then, as τ ~= T (N empty / N notempty) ~= T (N empty / (N total - N empty)) F2Pool: ~= (10 min) x [139 / (5241 - 139)] = 16.3 seconds AntPool: ~= (10 min) x [246 / (4506 - 246)] = 34.6 seconds
|
|
|
|
cypherdoc (OP)
Legendary

Activity: 1764
Merit: 1009
|
 |
July 05, 2015, 01:03:38 AM |
|
I guess we could estimate it by looking at the ratio of empty blocks to non-empty blocks produced by F2Pool.
If someone wants to tabulate that data, I'll update my post.
If you by empty you mean coinbase-only, then in the last 27,027 blocks (basically since jan 1st 2015), f2pool-attributed blocks: 5241, of which coinbase-only: 139 For antpool, this is 4506 / 246. Not sure if that's all the info you'd need, though. See also: Empty blocks [bitcointalk.org]is there a way for you to tell what % of blocks have been full over the last 3 wks and compare that to prior going back to say Jan 1? i'd include those in the 900+ and 720-750 kB range as being full.
|
|
|
|
|
thezerg
Legendary

Activity: 1246
Merit: 1018
|
 |
July 05, 2015, 01:31:40 AM |
|
If the txn input states that block B is the UTXO then the invalid proof is simply to supply B, right? That's one way to do it, however even this can be shortened. Right now with all the blocks < 1 MB it's not really a big deal to supply the entire block to prove that the referenced transaction doesn't exist, but it'd be nice to not require the entire block especially for when blocks are larger. By adding a rule to new blocks that require all the transactions to be ordered by their hash, you don't need to supply the entire block to prove that the transaction doesn't exist. It would be good to have that ordering requirement in place before blocks are allowed to grow to make sure that fraud proof size is bounded. Makes sense... I'd recommend a quick line or two in your blog to explain that: "In order to reduce the size of the fraud proof needed to show that a transaction input does not exist, additional information must be added to Bitcoin blocks to indicate the block which is the source of each outpoint used by every transaction in the block. A node can provide the source block to the SPV client to prove or disprove the existence of this transaction. But with a few more changes we can provide a subset of the source block. This may become very important if block sizes increase. "
|
|
|
|
|
|
TPTB_need_war
|
 |
July 05, 2015, 01:35:33 AM Last edit: July 05, 2015, 02:30:24 AM by TPTB_need_war |
|
The block size debate is iatrogenesis — any 'cure' is worse than the illness.
Freemarket, in 2010 everyone could buy thousands of Bitcoin for almost nothing, what hindered it was, besides being relatively unknown at that point in time, few people actually believed cryptocurrencies could be a thing, with Monero its almost the same, the difference being its swiming in a sea of shitcoins and not many can see its potential, its the second Cryptonote coin, the first being heavily premined, and it was launched with a MIT licence, there is absolutely no merit to claims Monero stole anything, its like saying Ubuntu stole code from Debian, or that Apple stole from FreeBSD, so even though Monero market cap is low, few people will actually bother buying a large stack because it is not a 100% certain bet, but its clear there is nothing close to Monero as Zerocash/Zerocoin is vaporware and Bitcoin sidechains are like dragons.
My point about my personal preference where I employed the word "clusterfuck" to describe the hundreds of Cryptonote clones and that Monero's marketing (on these forums) to some extent had to vilify other CN clones in order to assert its dominance of CN clones, is instead I would have preferred to add features to CN that would naturally assert dominance over other CN clones. It felt to me like Monero used strong-armed community tactics instead to gain more critical mass than the other CN clones, yet not so much capabilities innovation (rather a lot of refinement which I assume includes a lot of fine grained performance innovations). And I am nearly certain this (lack of outstanding capabilities other than the on chain rings) is why Monero is not more widely adopted and will be the downfallstunted growth of Monero (and I say this with specific knowledge of capabilities that I think will subsume Monero very soon). And that is precisely why I would not prematurely release those features in a whitepaper for 1000s of clones to go implement simultaneously. And yet people criticize me for not spilling the beans before the software is cooked. The marketing battle is not against the other "shitcoins" thus differentiating Monero from shit. Rather the battle is against Bitcoin core on who is going to own the chain that most of the BTC migrates to. Also most of the interest in altcoins is not ideological, but rather speculative. We are in a down market now until BTC bottoms this October, so only getting mostly ideological investment in Monero, not speculative fever. This will turn after October, but it might be too late for Monero depending on the competition that might arise interim. However, I tend to think Monero will get a big boost after October in spite of any new competition, because it is a more stable codebase. As smooth pointed out, the greatest threat to breakage is in implementation error. It would behove Monero to be the first CN coin to apply my suggested fix to insure combinatorial analysis of partially overlapping rings can't occur. P.S. CN is very important.
|
|
|
|
Peter R
Legendary

Activity: 1162
Merit: 1034
|
 |
July 05, 2015, 01:46:56 AM |
|
Am I being sensitive or is this an unnecessarily spiteful reply from Greg Maxwell?: ... You've shown that you can throw a bunch of symbolic markup and mix in a lack of understanding and measurement and make a pseudo-scientific argument that will mislead a lot of people, and that you're willing to do so or too ignorant to even realize what you're doing.
|
|
|
|
ssmc2
Legendary

Activity: 2002
Merit: 1040
|
 |
July 05, 2015, 01:50:41 AM |
|
Am I being sensitive or is this an unnecessarily spiteful reply from Greg Maxwell?: ... You've shown that you can throw a bunch of symbolic markup and mix in a lack of understanding and measurement and make a pseudo-scientific argument that will mislead a lot of people, and that you're willing to do so or too ignorant to even realize what you're doing.
He sounds bitter.
|
|
|
|
|
|
laurentmt
|
 |
July 05, 2015, 01:53:15 AM Last edit: July 05, 2015, 02:30:20 AM by laurentmt |
|
What this shows is that since the subtracted term, τ (1- Pvalid), is strictly positive, the miner's expectation of revenue, <V>, is maximized if the time to verify the previous block is minimized (i.e., if τ is as small as possible).
Actually, <V> is also maximized if P valid == 1 (or P valid as close as possible to 1). How to reach this result ? My humble proposal: make a deal with a few mining pools. Participants will never push invalid blocks to others participants. Blocks received from the cartel aren't checked before hashing a new block. Conclusion: As the average blocksize gets larger, the time to verify the previous block also gets larger. This means that miners will be motivated to improve how quickly their nodes can perform the ECDSA operations needed to verify blocks or that they will be more motivated to trick the system. EDIT: Am I being sensitive or is this an unnecessarily spiteful reply from Greg Maxwell?
Well, he seems a bit upset for now  but I think his message is close from what I've tried to suggest with my comment. We must analyze all the possibilities before jumping to a conclusion which backs our initial hypothesis. The point is valid for all of us, whatever our opinion on this blocksize issue.
|
|
|
|
|
thezerg
Legendary

Activity: 1246
Merit: 1018
|
 |
July 05, 2015, 02:06:05 AM |
|
Am I being sensitive or is this an unnecessarily spiteful reply from Greg Maxwell?: ... You've shown that you can throw a bunch of symbolic markup and mix in a lack of understanding and measurement and make a pseudo-scientific argument that will mislead a lot of people, and that you're willing to do so or too ignorant to even realize what you're doing.
Its a strategy (implemented unconsciously by many) to limit participation to the select few. Unfortunately it tends to create a situation where only similar personalities contribute which is where we are today with the core devs, Gavin excepted. I read his 21ms validation number but its weird because I was wondering just weeks ago why it was taking so long to sync a measly week of blockchain data and came to the conclusion that either the P2P code is complete garbage (compared to bittorrent for example) OR the validation cost is high (given my fan speed, I assumed it was validation). And if validation is so fast, why would these pools have custom code to skip it? It will be interesting to look at stats gathering mode he mentions.
|
|
|
|
|
Peter R
Legendary

Activity: 1162
Merit: 1034
|
 |
July 05, 2015, 02:17:54 AM |
|
Am I being sensitive or is this an unnecessarily spiteful reply from Greg Maxwell?: ... You've shown that you can throw a bunch of symbolic markup and mix in a lack of understanding and measurement and make a pseudo-scientific argument that will mislead a lot of people, and that you're willing to do so or too ignorant to even realize what you're doing.
Its a strategy (implemented unconsciously by many) to limit participation to the select few. Unfortunately it tends to create a situation where only similar personalities contribute which is where we are today with the core devs, Gavin excepted. Thanks. That makes me feel better. I read his 21ms validation number but its weird because I was wondering just weeks ago why it was taking so long to sync a measly week of blockchain data and came to the conclusion that either the P2P code is complete garbage (compared to bittorrent for example) OR the validation cost is high (given my fan speed, I assumed it was validation). And if validation is so fast, why would these pools have custom code to skip it?
This is the empirical fact that motivated me to perform the analysis. I was surprised how many "defensive blocks" we were seeing. Furthermore, based on these numbers: ...in the last 27,027 blocks (basically since jan 1st 2015), f2pool-attributed blocks: 5241, of which coinbase-only: 139 For antpool, this is 4506 / 246.
it looks like the average time these pools are mining empty blocks is 16 seconds (F2Pool) and 35 seconds (AntPool) before switching to non-empty blocks. Like you said, why are these numbers so big if processing the blocks is so fast?
|
|
|
|
|
TPTB_need_war
|
 |
July 05, 2015, 02:38:54 AM |
|
actually, there's 2 names for what's he's done.
one, from a moral standpoint, and one from a legal standpoint. i'll let you figure out what those names are.
Unethical and extortion? (my pops is an attorney perhaps I inherited some of it ... and now you know why I won't go on Skype with you ...)
|
|
|
|
cloverme
Legendary

Activity: 1512
Merit: 1059
SpacePirate.io
|
 |
July 05, 2015, 02:57:50 AM |
|
Am I being sensitive or is this an unnecessarily spiteful reply from Greg Maxwell?: ... You've shown that you can throw a bunch of symbolic markup and mix in a lack of understanding and measurement and make a pseudo-scientific argument that will mislead a lot of people, and that you're willing to do so or too ignorant to even realize what you're doing.
I agree with part of the quote. Also, the blocksize debate needs to go away by people stopping throwing 2 cents in a problem Gavin has a good solution for. What we really need solutions for... -Greater adoption with ease of use (Generation x's and earlier are still using paypal because it's easier) -China mining pools are greed based and harm bitcoin and the blockchain -When people think of bitcoins, they assume drugs and other shady activities Enough.... sorry, no idea who you are, so no offense, but tired of seeing these blocksize posts show up on Reddit.
|
|
|
|
|
|
TPTB_need_war
|
 |
July 05, 2015, 03:03:28 AM |
|
Let τ be the time it takes to verify a typical block and let T be the average block time (10 min). The fraction of time the miner does not know whether the most recent block was valid is clearly τ / T; the fraction of the time the miner does know is 1 - τ / T = (T - τ) / T. We will assume that every miner applies the same policy of producing empty SPV blocks before they've verified, and blocks of size S' after they've verified.
Under these conditions, the expectation value of the blocksize is equal to the expectation value of the blocksize during the time a miner doesn't know, plus the expectation value of the blocksize during the time he does know:
Seffective = ~0 [(τ / T)] + S' [(T - τ) / T] = S' [(T - τ) / T] (Eq. 1)
The time, τ, it takes to process a block is not constant, but rather depends linearly** on the size of the block. Approximating the size of the previous block as S', we get:
τ = k S'
....
QED. We've shown that there exists a limit on the maximum value of the average blocksize, due to the time it takes to verify a block, irrespective of any protocol enforced limits.
Your egregious mathematical error (myopia) of course is that you assume k is the same for all miners. And this is why you totally miss the centralization caused by your nonproof.
|
|
|
|
|