Author

Topic: What kind of hash power is needed to confirm a particular amount of transactions (Read 1062 times)

legendary
Activity: 3472
Merit: 4801
You know, if you keep showing up when I disappear people are really going to start thinking these are two accounts owned by the same person.

Ahaha, nice work throwing them off the trail Sat*cough*Danny

The only way anyone is ever going to think that I'm Satoshi is if I delete my first year of posts when I was still learning the basics.
hero member
Activity: 518
Merit: 500
Hodl!
You know, if you keep showing up when I disappear people are really going to start thinking these are two accounts owned by the same person.

Ahaha, nice work throwing them off the trail Sat*cough*Danny
legendary
Activity: 3472
Merit: 4801
Correct.  The only
- snip -

I take a break from bitcointalk for a day to drive around making cash deposits into bank accounts to acquire some bitcoins, and look who shows up out of nowhere and answers all the questions for me.

You know, if you keep showing up when I disappear people are really going to start thinking these are two accounts owned by the same person.
donator
Activity: 1218
Merit: 1079
Gerald Davis
Thanks for the answers. I also read this: https://en.bitcoin.it/wiki/Scalability ... I now understand that Bitcoin can handle 7 tps (= 7 transactions per second or max 604,800 daily) and that this does not depend on network hash rate.

This number was calculated under certain assumptions that might not be true in practice. The size of a transaction depends on the number of inputs (formerly received transactions available to be spend) and outputs. Thus the number of transactions that fit into a single block varies.

Correct.  The only true hard limit is that blocks are currently limited to 1MB.  Changing that will require a fork (and a super majority to support it).  It will also need to be extensively tested and will break existing clients (who will see larger blocks as invalid) so it will require a pretty involved process over the course of months.   The 7 tps is often quotes like it is written in stone but only the 1MB limit is.

The actual transaction rates depends on what the average transaction size ends up being.  Honestly I think the 7tps is a little generous.  It requires the average transaction size to be 238 bytes (1,000,000 / 600 / 7 = 238).   While the mode is probably around 200-300 bytes the less frequent but much larger transactions bring up the average to around 500 bytes.  That would mean an average "full" block having 2,000 transactions or 3.3 tps (2,000 /600 = 3.3 tps).  Note this isn't necessarily a bad thing.   Paying multiple withdraws as a single transaction for example will result in less transactions which are on average larger but the bytes per withdraw is less and thus the block is being used more efficiently.
copper member
Activity: 1498
Merit: 1528
No I dont escrow anymore.
Thanks for the answers. I also read this: https://en.bitcoin.it/wiki/Scalability ... I now understand that Bitcoin can handle 7 tps (= 7 transactions per second or max 604,800 daily) and that this does not depend on network hash rate.

This number was calculated under certain assumptions that might not be true in practice. The size of a transaction depends on the number of inputs (formerly received transactions available to be spend) and outputs. Thus the number of transactions that fit into a single block varies.

newbie
Activity: 36
Merit: 0
Thanks for the answers. I also read this: https://en.bitcoin.it/wiki/Scalability ... I now understand that Bitcoin can handle 7 tps (= 7 transactions per second or max 604,800 daily) and that this does not depend on network hash rate. Hopefully we'll soon come close to those limits and see a hard fork of Bitcoin to a larger max block size Wink
hero member
Activity: 518
Merit: 500
Hodl!
Theoretically yes. Practically no. Because, one AntMiner S3 (450 GH/s) will no be able to find the next block hash for many years at the current difficulty. So, transactions will not be confirmed for a very very long time.

Right, if network participation gradually declined until there was only one Antminer running, it could handle it from there on out.

However, manual intervention with the difficulty is not precluded, i.e. should it be obvious that a conspicuously large amount of hashpower has left the network, 90% or more, then devs could intervene.
legendary
Activity: 1708
Merit: 1036
No. of transaction does not have anything to do with network hash power.

So what you are saying is that it would be just as well if there would only be one, for example, AntMiner S3 (450 GH/s) connected to the Bitcoin network and everything would work just as fine as is now with thousands of miners?

Theoretically yes. Practically no. Because, one AntMiner S3 (450 GH/s) will no be able to find the next block hash for many years at the current difficulty. So, transactions will not be confirmed for a very very long time.
But if the adjustment were not sudden then yes, one Antminer or other device would in theory be sufficient to keep the network running. As stated above there is no inherent relationship between number of transactions and network hashrate. There is a maximum number of transactions that can be included on one block, and the network difficulty is set to adjust to keep a 10 minute average block time.
legendary
Activity: 2226
Merit: 1052
No. of transaction does not have anything to do with network hash power.

So what you are saying is that it would be just as well if there would only be one, for example, AntMiner S3 (450 GH/s) connected to the Bitcoin network and everything would work just as fine as is now with thousands of miners?

Theoretically yes. Practically no. Because, one AntMiner S3 (450 GH/s) will no be able to find the next block hash for many years at the current difficulty. So, transactions will not be confirmed for a very very long time.
newbie
Activity: 36
Merit: 0
No. of transaction does not have anything to do with network hash power.

So what you are saying is that it would be just as well if there would only be one, for example, AntMiner S3 (450 GH/s) connected to the Bitcoin network and everything would work just as fine as is now with thousands of miners?
hero member
Activity: 714
Merit: 500
Bitcoin is designed so it takes 10 min average to confirm a transaction.
legendary
Activity: 2226
Merit: 1052
As the title says - what kind of hash power is needed to confirm a particular amount of transactions?

Let us assume that there would be only 1 TH/s of total Bitcoin network hash power. How many transactions per day (per hour) is able to confirm this kind of hashing power? How does the amount of hashing power correlate with the amount of transactions?

Currently we are aprox. 320 PH/s - how many transactions (confirmed) par day can handle this kind of hash power?

Currently we are aprox. 100k transactions per day - what kind of minimum hash power do we need to handle (confirm) that many transactions par day can handle?

Or is what I'm saying completely wrong and those numbers depends on other factors?

No. of transaction does not have anything to do with network hash power.

Hashpower determines how fast blocks will be found.

Transactions are added under a found block by miners.

How many transaction will be added under a certain block is dependent on the miner, not the hash power. There are instance with zero transaction blocks, i.e. only with the coinbase transaction.
newbie
Activity: 36
Merit: 0
As the title says - what kind of hash power is needed to confirm a particular amount of transactions?

Let us assume that there would be only 1 TH/s of total Bitcoin network hash power. How many transactions per day (per hour) is able to confirm this kind of hashing power? How does the amount of hashing power correlate with the amount of transactions?

Currently we are aprox. 320 PH/s - how many transactions (confirmed) par day can handle this kind of hash power?

Currently we are aprox. 100k transactions per day - what kind of minimum hash power do we need to handle (confirm) that many transactions par day?

Or is what I'm saying completely wrong and those numbers depends on other factors?
Jump to: