|
ysoliman
Member
![*](https://bitcointalk.org/Themes/custom1/images/star.gif)
Offline
Activity: 85
Merit: 10
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
August 30, 2011, 03:29:59 PM |
|
Great site! I like the stats section!
+1
|
|
|
|
Sukrim
Legendary
Offline
Activity: 2618
Merit: 1007
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
August 30, 2011, 07:37:18 PM |
|
You could offer uploads of blockchains to find more (valid) orphan blocks as you won't get them all on one node.
Also giving sources for your market data might be interesting.
|
|
|
|
piuk (OP)
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
August 30, 2011, 09:30:47 PM |
|
You could offer uploads of blockchains to find more (valid) orphan blocks as you won't get them all on one node.
Also giving sources for your market data might be interesting.
It is my understand that if you have an active connection to the network then you'll receive all blocks, some which may or may not be orphaned later? For historical data if someone could zip me up a copy of their block chain that would be awesome. The market data is from bitcoincharts.com cvs feed, I should add a link to credit them.
|
|
|
|
Sukrim
Legendary
Offline
Activity: 2618
Merit: 1007
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
August 31, 2011, 12:46:34 AM |
|
I'm not sure if bitcoind accepts blocks of the same block height twice (so 2 blocks with the same parent) if there is not a block following up.
A "Y" split might not be found (as your client already thinks it has the "correct" latest block), only if the block you happen to have is orphaned, and the other chain is forced upon you you keep the orphan data - if I remember correctly.
|
|
|
|
Steve
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
August 31, 2011, 04:40:37 AM |
|
Very nice! How are you calculating the cost to mount a 50% attack? Am I correct in reading that you estimate it to be $12.7 billion? That's far greater than what I've estimated in the past. That would be awesome if accurate, but I think that's probably way too high.
|
|
|
|
piuk (OP)
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
August 31, 2011, 11:42:24 AM |
|
I'm not sure if bitcoind accepts blocks of the same block height twice (so 2 blocks with the same parent) if there is not a block following up.
A "Y" split might not be found (as your client already thinks it has the "correct" latest block), only if the block you happen to have is orphaned, and the other chain is forced upon you you keep the orphan data - if I remember correctly.
Ok I see, still it should be sufficient to detect most double spend attacks (assuming they are not targeted at a specific node). I guess this is something i could patch in bitcoind, I need to read more into how blocks are advertised. Very nice! How are you calculating the cost to mount a 50% attack? Am I correct in reading that you estimate it to be $12.7 billion? That's far greater than what I've estimated in the past. That would be awesome if accurate, but I think that's probably way too high.
Should be 12.7 million.
|
|
|
|
piuk (OP)
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
August 31, 2011, 06:27:58 PM |
|
I'm not sure if bitcoind accepts blocks of the same block height twice (so 2 blocks with the same parent) if there is not a block following up.
I've looked at the source and it seems the client will accept two blocks at the same height but it will only relay the best block to other peers. So basically orphaned blocks won't propagate far along the network, which means in order to track them i need to connect to more node. I've added all the big mining pools to my addnode list and also increased my connection limit to 400. It seems to have worked because I got an orphaned block at http://pi.uk.com/bitcoin/143370 which was received after the one which became the main chain. Now just need to work on alerts for transactions which might be double spends.
|
|
|
|
xc
Jr. Member
Offline
Activity: 40
Merit: 4
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
August 31, 2011, 07:03:32 PM |
|
I've added all the big mining pools to my addnode list and also increased my connection limit to 400.
Is there anyway for you to display which mining pools have seen and are relaying a particular transaction? You could then calculate a rough estimate of how much hashing power (and what percentage of the total known bitcoin network) was aggregated behind an unconfirmed transaction. Imagine a table with the major mining pools, their hashing power, and a simple yes/no if they are working on a given version of a transaction. The chance of a double spend after all the major mining pools are working on a particular transaction would be slim to none. This information could help implement Satoshi's suggestions to the so-called "Snack machine problem" and significantly reduce any risk associated with accepting a unconfirmed transaction ( http://forum.bitcoin.org/?topic=423.msg3819#msg3819). Here are some of my posts on the topic: https://bitcointalk.org/index.php?topic=19481.msg253928#msg253928, and particularly https://bitcointalk.org/index.php?topic=19481.msg254559#msg254559
|
|
|
|
piuk (OP)
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
September 01, 2011, 10:54:46 AM Last edit: September 01, 2011, 12:08:23 PM by piuk |
|
I've added all the big mining pools to my addnode list and also increased my connection limit to 400.
Is there anyway for you to display which mining pools have seen and are relaying a particular transaction? You could then calculate a rough estimate of how much hashing power (and what percentage of the total known bitcoin network) was aggregated behind an unconfirmed transaction. Imagine a table with the major mining pools, their hashing power, and a simple yes/no if they are working on a given version of a transaction. The chance of a double spend after all the major mining pools are working on a particular transaction would be slim to none. This information could help implement Satoshi's suggestions to the so-called "Snack machine problem" and significantly reduce any risk associated with accepting a unconfirmed transaction ( http://forum.bitcoin.org/?topic=423.msg3819#msg3819). Here are some of my posts on the topic: https://bitcointalk.org/index.php?topic=19481.msg253928#msg253928, and particularly https://bitcointalk.org/index.php?topic=19481.msg254559#msg254559Interesting, I like this idea. Basically you tally how many times a transaction is relayed to your client which would give you an idea of how far the transaction has propagated across the network. A transaction could then be given a score based on how far it's propagated. If is relayed from an ip of a know mining pool then you can give bonus points based on that pools hashing power. It isn't full proof, but it could definitely help. Theres a couple of things i need to do first before i can implement this.
|
|
|
|
piuk (OP)
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
September 01, 2011, 10:29:27 PM |
|
|
|
|
|
xc
Jr. Member
Offline
Activity: 40
Merit: 4
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
September 02, 2011, 12:57:54 AM |
|
Another twist would be drawing a graph of a transaction's 'score' as you describe it and any transaction fees associated with that transaction. This would give us a window into the current 'market' of what fees miners are demanding.
|
|
|
|
LightRider
Legendary
Offline
Activity: 1500
Merit: 1022
I advocate the Zeitgeist Movement & Venus Project.
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
September 02, 2011, 04:57:52 AM Last edit: September 02, 2011, 05:19:07 PM by LightRider |
|
I would like to see the Bitcoin Days Destroyed metric that is implemented in the ABE software, but otherwise good start! https://en.bitcoin.it/wiki/Bitcoin_Days_Destroyed
|
|
|
|
NothinG
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
September 02, 2011, 05:48:33 AM |
|
Awesome site, one thing I noticed we haven't reached... ![](https://ip.bitcointalk.org/?u=http%3A%2F%2Fgyazo.com%2F46aec9ea3af8e8ec48d0b66045fa0caa.png&t=665&c=mzXWrqkMJ6ZAvw) Why not Gigawatts??? We only need 1.2 Gigawatts and 88mp/h
|
|
|
|
piuk (OP)
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
September 02, 2011, 07:20:28 PM |
|
I've added graphs for some of the most interesting statistics, click on the chart icon on the homepage. The number of transactions seems to have peaked somewhat. http://pi.uk.com/bitcoin/charts/n-transactionsAlso the estimated transaction volume seems have stayed relatively constant. http://pi.uk.com/bitcoin/charts/estimated-transaction-volumeOften when a users sends coins their inputs will be split into multiple outputs, some which go to the indented recipient and some which are returned to user as change. This statistic attempts to calculate which outputs are change and removes them from the total volume. Unfortunately the anomalies you see on the chart are spam transactions, which i need to figure out how to detect. I don't understand this metric, what is it supposed to show?
|
|
|
|
piuk (OP)
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
September 02, 2011, 10:32:12 PM |
|
Design updated
|
|
|
|
bg002h
Donator
Legendary
Offline
Activity: 1464
Merit: 1047
I outlived my lifetime membership:)
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
September 03, 2011, 02:51:23 AM |
|
I've created a site similar to block explorer with a couple of improvements: - Includes orphaned blocks which can be used to track possible double spends
- Estimates the actual volume of BTC transacted (not just BTC sent)
The site can be found at: http://pi.uk.com/bitcoinAll ideas and feedback welcome. well done...and fast code too! DrG 1DrGossc3QidjzgDXzveCAQGiPWsoiDZ8C
|
|
|
|
Stephen Gornick
Legendary
Offline
Activity: 2506
Merit: 1010
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
September 03, 2011, 06:42:22 PM |
|
I don't understand this metric, what is it supposed to show? Your estimated transactions does make an attempt to guess what the change transactions are and remove those from the transaction volume, but if I send coins back and forth all day between two wallets I could inflate your estimated transactions counts fairly easily. BitcoinDays Destroyed would instead know that those transfers back and forth were from recent coins, and thus those transfers would have little impact on the BDD metric, following the initial spend. So what BDD describes is how much spending is from older coins. Well, more specifically, hopefully BDD trends will help to determine if there is hoarding (holding wallet balances steady or adding to them) or spending/liquidating (lowering of wallet balances). This BDD metric too can yield garbage results -- each time Mt. Gox spends from their wallet, for example, BDD will spike but that doens't mean there is any real economic transaction as a result. And it can be abused. Some parties from the Bitcoin Rich List with old coins can try to fake out those speculating on BDD by simply churning their old coins. That's just one property of bitcoin -- we just don't know for certain from the data what is actually happening. Depending on your perspective, the difficulty in measuring Bitcoin activity is either a benefit or a hindrance. BDD is just another metric that might help give a little more insight into Bitcoin.
|
|
|
|
joulesbeef
Sr. Member
![*](https://bitcointalk.org/Themes/custom1/images/star.gif) ![*](https://bitcointalk.org/Themes/custom1/images/star.gif) ![*](https://bitcointalk.org/Themes/custom1/images/star.gif)
Offline
Activity: 476
Merit: 250
moOo
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
September 03, 2011, 06:43:18 PM |
|
why you ban me bro?
nah for some reason cant reach your site from here, but i an when i proxy.
|
mooo for rent
|
|
|
Stephen Gornick
Legendary
Offline
Activity: 2506
Merit: 1010
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
September 03, 2011, 07:38:51 PM |
|
All ideas and feedback welcome. The Summary for a block shows "Previous block" but the value displayed is not the block number but the index, which makes it confusing. - e.g. http://pi.uk.com/bitcoin/143815
|
|
|
|
|