Bitcoin Forum
May 10, 2024, 12:03:10 PM *
News: Latest Bitcoin Core release: 27.0 [Torrent]
 
   Home   Help Search Login Register More  
Pages: [1]
  Print  
Author Topic: Why doesn't the open client process txs?  (Read 1330 times)
doobadoo (OP)
Sr. Member
****
Offline Offline

Activity: 364
Merit: 250


View Profile
May 28, 2011, 09:39:59 PM
 #1

I recall reading somewhere about the BC network that in the future, when mining becomes very unprofitable (say beyond 2014), that transaction fees would support the processing of payments.  I don't get it.  We have big fat cpus sitting there on whatever box is running the client, else how could you send a tx?  why doesn't the idle client utilize a slice of cpu for processing txs? 

Also, why does it take so long for payments to post.  I can ping just about anything in under a second, hell even 100ms.  Why does it take so long for the network to post a single tx confirm?  This is a major stumbling  block to the visa/mastercard network.  They authenticate in seconds.  Will this be better in the future? 

"It is, quite honestly, the biggest challenge to central banking since Andrew Jackson." -evoorhees
1715342590
Hero Member
*
Offline Offline

Posts: 1715342590

View Profile Personal Message (Offline)

Ignore
1715342590
Reply with quote  #2

1715342590
Report to moderator
1715342590
Hero Member
*
Offline Offline

Posts: 1715342590

View Profile Personal Message (Offline)

Ignore
1715342590
Reply with quote  #2

1715342590
Report to moderator
1715342590
Hero Member
*
Offline Offline

Posts: 1715342590

View Profile Personal Message (Offline)

Ignore
1715342590
Reply with quote  #2

1715342590
Report to moderator
There are several different types of Bitcoin clients. The most secure are full nodes like Bitcoin Core, which will follow the rules of the network no matter what miners do. Even if every miner decided to create 1000 bitcoins per block, full nodes would stick to the rules and reject those blocks.
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction.
1715342590
Hero Member
*
Offline Offline

Posts: 1715342590

View Profile Personal Message (Offline)

Ignore
1715342590
Reply with quote  #2

1715342590
Report to moderator
error
Hero Member
*****
Offline Offline

Activity: 588
Merit: 500



View Profile
May 28, 2011, 10:53:11 PM
 #2

I recall reading somewhere about the BC network that in the future, when mining becomes very unprofitable (say beyond 2014), that transaction fees would support the processing of payments.  I don't get it.  We have big fat cpus sitting there on whatever box is running the client, else how could you send a tx?  why doesn't the idle client utilize a slice of cpu for processing txs? 

Also, why does it take so long for payments to post.  I can ping just about anything in under a second, hell even 100ms.  Why does it take so long for the network to post a single tx confirm?  This is a major stumbling  block to the visa/mastercard network.  They authenticate in seconds.  Will this be better in the future? 

The GPU is much better at it; CPU mining is pretty much obsolete at this point.

As for confirmations, transaction awareness happens instantly, even faster than those credit card networks. And the transaction is confirmed much faster than those other networks do, too. What's the problem here?

3KzNGwzRZ6SimWuFAgh4TnXzHpruHMZmV8
phillipsjk
Legendary
*
Offline Offline

Activity: 1008
Merit: 1001

Let the chips fall where they may.


View Profile WWW
May 28, 2011, 10:57:29 PM
 #3

Your CPU isn't as big as you think it is. Most the people "mining" are using GPUs now, which are about 1000 times faster than CPU mining. There are also people experimenting with FPGAs as well, but the cost of electricity is not yet prohibitive for GPU miners.

I think in the long term, mining will stabilize on about 50,000 entities crunching away. This works out to each "entity" winning the lottery (a Transaction block is produced every 10 minutes) once a year.

If you want to trade in Bitcoins, it is easier to purchase them.

James' OpenPGP public key fingerprint: EB14 9E5B F80C 1F2D 3EBE  0A2F B3DE 81FF 7B9D 5160
smooth
Legendary
*
Offline Offline

Activity: 2968
Merit: 1198



View Profile
May 28, 2011, 11:08:16 PM
 #4

I recall reading somewhere about the BC network that in the future, when mining becomes very unprofitable (say beyond 2014), that transaction fees would support the processing of payments.  I don't get it.  We have big fat cpus sitting there on whatever box is running the client, else how could you send a tx?  why doesn't the idle client utilize a slice of cpu for processing txs?  

Also, why does it take so long for payments to post.  I can ping just about anything in under a second, hell even 100ms.  Why does it take so long for the network to post a single tx confirm?  This is a major stumbling  block to the visa/mastercard network.  They authenticate in seconds.  Will this be better in the future?  

The GPU is much better at it; CPU mining is pretty much obsolete at this point.

Yeah, but many of the machines running the client have GPU's as well.  It would be better for the network if every client ran a GPU miner by default, even at a low duty cycle like 1-3% (which could be changed in the preferences).  It would encourage uptake of the transaction client as well.  People don't really know the difference between 1 in 1000 and 1 in a 1,000,000.  If there is some chance of having free money drop into your account, that will motivate people.

Now we have the long tail working for us instead of mining flowing where it is most efficient (one miner, generally).

No one has implemented GPU mining in the client, but there isn't really any reason why it can't be done.

If we get to the point when GPU's are forced out by some magic hash brownie chips, we can think of something else.

doobadoo (OP)
Sr. Member
****
Offline Offline

Activity: 364
Merit: 250


View Profile
May 28, 2011, 11:10:18 PM
 #5

Your CPU isn't as big as you think it is. Most the people "mining" are using GPUs now, which are about 1000 times faster than CPU mining. There are also people experimenting with FPGAs as well, but the cost of electricity is not yet prohibitive for GPU miners.

I think in the long term, mining will stabilize on about 50,000 entities crunching away. This works out to each "entity" winning the lottery (a Transaction block is produced every 10 minutes) once a year.

If you want to trade in Bitcoins, it is easier to purchase them.

So your saying that the cpu's running the  client can't handle the network load...?   I had some one send me a small payment..  took like 10 minutes to see it in my client.  How is that faster than the CC network?

Also, i don't htink that i'm talking about mining.  I am talking about network overhead.  Why does the network "need" miners, or people to receive a tx fee.  Just who does that 1 bc fee go to... (bc is bitcent...BC is bitcoin)

"It is, quite honestly, the biggest challenge to central banking since Andrew Jackson." -evoorhees
smooth
Legendary
*
Offline Offline

Activity: 2968
Merit: 1198



View Profile
May 28, 2011, 11:16:06 PM
 #6

Also, i don't htink that i'm talking about mining.  I am talking about network overhead.  Why does the network "need" miners, or people to receive a tx fee.  Just who does that 1 bc fee go to... (bc is bitcent...BC is bitcoin)

Mining and processing transactions are the same thing.  That's how Bitcoin processes transactions.  People are too stuck on the whole "free Bitcoins" aspect of mining, but that's not the reason mining exists.  The system would work just fine without any free Bitcoins, although it probably would have been harder to bootstrap without the gold rush.  (Though overall I'm not sure this was a good thing.)
phillipsjk
Legendary
*
Offline Offline

Activity: 1008
Merit: 1001

Let the chips fall where they may.


View Profile WWW
May 29, 2011, 12:26:08 AM
 #7

Yeah, but many of the machines running the client have GPU's as well.  It would be better for the network if every client ran a GPU miner by default, even at a low duty cycle like 1-3% (which could be changed in the preferences).  It would encourage uptake of the transaction client as well.  People don't really know the difference between 1 in 1000 and 1 in a 1,000,000.  If there is some chance of having free money drop into your account, that will motivate people.

This can lead to a nasty surprise for those with low bandwidth caps: even if you don't do a lot of hashing, the protocol will use a lot of bandwidth if it ever really takes off.

Quote
Now we have the long tail working for us instead of mining flowing where it is most efficient (one miner, generally).

No one has implemented GPU mining in the client, but there isn't really any reason why it can't be done.

If we get to the point when GPU's are forced out by some magic hash brownie chips, we can think of something else.

If bitcoins become very popular, every world government, many large corporations, and some early adopters will be in the "mining" business. If Transaction fees become a problem, mining pools may see a resurgence just to take advantage of the "long tail" to process "free" transactions.

So your saying that the cpu's running the  client can't handle the network load...?   I had some one send me a small payment..  took like 10 minutes to see it in my client.  How is that faster than the CC network?

No, I was not saying that. I was saying that few people want to wait over a year to win the "lottery". Their CPU time can be put to other uses, or electricity can be saved by putting the machine in suspend mode.

It is faster than the CC network because apparently transactions on not reconciled until the end of the day. If a user has an account with a trusted payment processor, it can be made to appear the transaction goes through immediately, even if it takes all day to be confirmed.

The 10 minute delay is actually too short for some use cases: if you are colonizing Mars, you are too isolated to use Bitcoin.

James' OpenPGP public key fingerprint: EB14 9E5B F80C 1F2D 3EBE  0A2F B3DE 81FF 7B9D 5160
error
Hero Member
*****
Offline Offline

Activity: 588
Merit: 500



View Profile
May 29, 2011, 12:36:27 AM
 #8

Mars will have to have its own block chain. No way around it until the speed of light is broken.

3KzNGwzRZ6SimWuFAgh4TnXzHpruHMZmV8
smooth
Legendary
*
Offline Offline

Activity: 2968
Merit: 1198



View Profile
May 29, 2011, 12:44:00 AM
 #9

This can lead to a nasty surprise for those with low bandwidth caps: even if you don't do a lot of hashing, the protocol will use a lot of bandwidth if it ever really takes off.

This is independent of whether you hash.  Low bandwidth requires a lightweight client, not a regular node.  The standard client doesn't do that currently. 
phillipsjk
Legendary
*
Offline Offline

Activity: 1008
Merit: 1001

Let the chips fall where they may.


View Profile WWW
May 30, 2011, 03:26:48 PM
 #10

Yeah, but many of the machines running the client have GPU's as well.  It would be better for the network if every client ran a GPU miner by default, even at a low duty cycle like 1-3% (which could be changed in the preferences).  It would encourage uptake of the transaction client as well.  People don't really know the difference between 1 in 1000 and 1 in a 1,000,000.  If there is some chance of having free money drop into your account, that will motivate people.

With the recent attacks on mining pools, I have reversed my opinion on this. Rumor has it that many of the "DeepBit" miners don't know or care how to switch to solo mining should the need arise. The problem of users with very little knowledge is only going to get worse.

I now think that it is prudent for the "thick" client to default to 1-3% CPU/GPU usage. The clueless will provide a baseline with their "long tail" in return for possibly wining the lottery by actually processing a block. Obviously, it should be possible to disable this low CPU usage: for example, when the computer is isolated from the Internet or is running on battery power.


James' OpenPGP public key fingerprint: EB14 9E5B F80C 1F2D 3EBE  0A2F B3DE 81FF 7B9D 5160
Pages: [1]
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!