|
Title: Can somebody help me understand SPECTRE by daglabs Post by: lizardbtc on February 20, 2018, 04:17:01 PM I've heard about this way back but only recently started to read more about it.
For those who do n't know what I am talking about check these links: https://www.daglabs.com/ , https://eprint.iacr.org/2016/1159.pdf Tho one thing is not clear to me. I get it in this protocols miners are like going rouge and simply mine as many blocks as they can and thus creating this graph of blocks. The problem that I see is that it is not neccessery to mine blocks which will contain the same transactions in them. As I see this it is just puting more weight on the network in terms of how large this DAG can actually be. Yes it can scale but then to run a node you will have to have good internet speed and a lot of space on your hard drive with some of these blocks being totaly unnecessary added to the network. Can somebody shed some light and explain me why is this proposal good or you belive it is bad, why so? Title: Re: Can somebody help me understand SPECTRE by daglabs Post by: Random Seller on February 20, 2018, 05:49:48 PM I've heard about this way back but only recently started to read more about it. For those who do n't know what I am talking about check these links: https://www.daglabs.com/ , https://eprint.iacr.org/2016/1159.pdf Tho one thing is not clear to me. I get it in this protocols miners are like going rouge and simply mine as many blocks as they can and thus creating this graph of blocks. The problem that I see is that it is not neccessery to mine blocks which will contain the same transactions in them. As I see this it is just puting more weight on the network in terms of how large this DAG can actually be. Yes it can scale but then to run a node you will have to have good internet speed and a lot of space on your hard drive with some of these blocks being totaly unnecessary added to the network. Can somebody shed some light and explain me why is this proposal good or you belive it is bad, why so? This proposal changes the trade off needed to achieve faster confirmation times. Assuming we don’t have lightning or any off chain solutions the classic trade off is between security and scalability. This is because as you lower blocktime or increase block size there comes a point where security is significantly reduced. The proposal changes the trade offs to a trade off between (network & storage) and security. This proposal would allow significantly lower block times which in turn would significantly reduce confirmation time but the cost is larger storage capacity and more bandwidth. In my opinion, this change would significantly increase confirmation speed but lower the amount of nodes due to the higher requirements. But I digress, it really depends on the rate of growth of storage capacity and network bandwidth. |