TPTB_need_war
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
March 07, 2016, 01:31:26 PM |
|
Side-chains are insecure. DOA.
|
|
|
|
hv_
Legendary
Offline
Activity: 2520
Merit: 1055
Clean Code and Scale
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
March 07, 2016, 01:34:00 PM |
|
Side-chains are insecure. DOA. To the BTC chain itself? IMO it lowers the complexity,...
|
Carpe diem - understand the White Paper and mine honest. Fix real world issues: Check out b-vote.com The simple way is the genius way - Satoshi's Rules: humana veris _
|
|
|
hv_
Legendary
Offline
Activity: 2520
Merit: 1055
Clean Code and Scale
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
March 07, 2016, 01:44:57 PM |
|
Side-chains are insecure. DOA. To the BTC chain itself? IMO it lowers the complexity,... The security is reduced to that of the weakest side chain. So the pegging needs to be transient ! from their Whitepaper: • Two-way pegging using sidechains (currently a federated peg, fully automatic peg subject to Bitcoin improvements)
|
Carpe diem - understand the White Paper and mine honest. Fix real world issues: Check out b-vote.com The simple way is the genius way - Satoshi's Rules: humana veris _
|
|
|
TPTB_need_war
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
March 07, 2016, 01:47:54 PM |
|
Side-chains are insecure. DOA. To the BTC chain itself? IMO it lowers the complexity,... The security is reduced to that of the weakest side chain. So the pegging needs to be transient ! I have no idea what you mean. Chain reorganizations in the weaker chain can cause people to lose their Bitcoins. The chains can get out-of-sync. There is no way for a block chain to securely reference any data point outside of itself. This is fundamentally why Augur and BitUSD can't function without centralization.
|
|
|
|
TPTB_need_war
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
March 07, 2016, 01:52:36 PM |
|
Basically all block chain projects have decided to abandon decentralization and just pretend (whether they realize it or not)[1]. So they can do features that don't work without centralization. Come-from-Beyond say people are happy with that, and the sheep usually are happy to be lead to their slaughter. [1] Note Monero has the most ASIC-resistant hash released thus far. They haven't solved the economic centralization issue, but for the time being afaik I presume mining is reasonably decentralized.
|
|
|
|
Come-from-Beyond
Legendary
Offline
Activity: 2142
Merit: 1010
Newbie
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
March 07, 2016, 02:01:14 PM |
|
Basically all block chain projects have decided to abandon decentralization and just pretend (whether they realize it or not). So they can do features that don't work without centralization. Come-from-Beyond say people are happy with that, and the sheep usually are happy to be lead to their slaughter. I'm sorry that I didn't wrap that post into <subtleSarcasm> tags. I'm flattered that you use my posts as argumentum ad verecundiam.
|
|
|
|
TPTB_need_war
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
March 07, 2016, 02:02:39 PM |
|
Basically all block chain projects have decided to abandon decentralization and just pretend (whether they realize it or not). So they can do features that don't work without centralization. Come-from-Beyond say people are happy with that, and the sheep usually are happy to be lead to their slaughter. I'm sorry that I didn't wrap that post into <subtleSarcasm> tags. I'm flattered that you use my posts as argumentum ad verecundiam. I'm sorry I didn't emphasize that I know you were being sarcastic. But you are also being pragmatic, as you've always astutely been. I don't fault you for that! Even Satoshi failed at decentralization.
|
|
|
|
Come-from-Beyond
Legendary
Offline
Activity: 2142
Merit: 1010
Newbie
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
March 07, 2016, 02:03:51 PM |
|
I'm sorry I didn't emphasize that I know you were being sarcastic.
Your reply has reminded me of Mark Twain.
|
|
|
|
hv_
Legendary
Offline
Activity: 2520
Merit: 1055
Clean Code and Scale
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
March 07, 2016, 02:06:28 PM |
|
Side-chains are insecure. DOA. To the BTC chain itself? IMO it lowers the complexity,... The security is reduced to that of the weakest side chain. So the pegging needs to be transient ! I have no idea what you mean. Chain reorganizations in the weaker chain can cause people to lose their Bitcoins. The chains can get out-of-sync. There is no way for a block chain to securely reference any data point outside of itself. This is fundamentally why Augur and BitUSD can't function without centralization. Yes - agreed, but I mean rather it has no negative effect to the main chain, if you don't care using the side chain.
|
Carpe diem - understand the White Paper and mine honest. Fix real world issues: Check out b-vote.com The simple way is the genius way - Satoshi's Rules: humana veris _
|
|
|
hv_
Legendary
Offline
Activity: 2520
Merit: 1055
Clean Code and Scale
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
March 07, 2016, 02:21:20 PM |
|
Basically all block chain projects have decided to abandon decentralization and just pretend (whether they realize it or not)[1]. So they can do features that don't work without centralization. Come-from-Beyond say people are happy with that, and the sheep usually are happy to be lead to their slaughter. [1] Note Monero has the most ASIC-resistant hash released thus far. They haven't solved the economic centralization issue, but for the time being afaik I presume mining is reasonably decentralized. Some yummy for you on page 15: " RSK aims to be a better payment network. To achieve fast payments, several solutions have been developed: - Use of competition-free block selection (e.g. Hyperledger, Ripple, closed-loop systems) - Use of hub-and-spoke networks (e.g. Bitcoin lightning network) - Use of high PoW block rates Hub-and-spoke networks add new centralization nodes, and require a complete adaptation of client wallets to a new, completely different payment model. Although so this alternative can be easily implemented on RSK, is not the native system for fast payments. RSK adopts the DECOR+ and FastBlock5 protocols, which allow reaching a 10 seconds average block rate that does not create incentives for mining centralization, is selfish-mining free and incentive compatible.
" https://uploads.strikinglycdn.com/files/90847694-70f0-4668-ba7f-dd0c6b0b00a1/RootstockWhitePaperv9-Overview.pdf
|
Carpe diem - understand the White Paper and mine honest. Fix real world issues: Check out b-vote.com The simple way is the genius way - Satoshi's Rules: humana veris _
|
|
|
TPTB_need_war
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
March 07, 2016, 02:44:34 PM Last edit: March 07, 2016, 02:57:19 PM by TPTB_need_war |
|
Side-chains are insecure. DOA. To the BTC chain itself? IMO it lowers the complexity,... The security is reduced to that of the weakest side chain. So the pegging needs to be transient ! I have no idea what you mean. Chain reorganizations in the weaker chain can cause people to lose their Bitcoins. The chains can get out-of-sync. There is no way for a block chain to securely reference any data point outside of itself. This is fundamentally why Augur and BitUSD can't function without centralization. Yes - agreed, but I mean rather it has no negative effect to the main chain, if you don't care using the side chain. I think the insecurity of the side chain can wreck the Bitcoin block chain. If I am mistaken, I request someone to point out why. Please see pages 8, 9, and 12 of the Blockstream side chains white paper. It says that the coins on the Bitcoin block chain can be unlocked by presenting a proof-of-work from the side chain, but that this can be invalidated by a longer proof-of-work. So this means that a lie-in-wait attack on the side chain could allow someone to unlock coins on Bitcoin's block chain, spend them, let others spend them in a fanout of derivative transactions, then reverse the Bitcoin transactions by presenting a longer proof-of-work from the side chain invalidating all those Bitcoin block chain transactions. In short, it seems to me a chain reorganization on the side chain can cause a chain reorganization on the Bitcoin block chain. I am ready to go to sleep, so I am just skimming quickly with my re-reading of that white paper, so perhaps I missed something?
|
|
|
|
tromp
Legendary
Offline
Activity: 990
Merit: 1108
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
March 07, 2016, 02:45:39 PM |
|
[1] Note Monero has the most ASIC-resistant hash released thus far.
Ethash appears more ASIC resistant as it is more memory bound and needs at least 8x more memory.
|
|
|
|
TPTB_need_war
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
March 07, 2016, 02:51:01 PM |
|
[1] Note Monero has the most ASIC-resistant hash released thus far.
Ethash appears more ASIC resistant as it is more memory bound and needs at least 8x more memory. The original version that I reviewed for Charles Hoskinson before he formed Ethereum with Vitalik could be easily parallelized and thus would have superior performance on GPUs. Whereas, Monero's hash includes AES-NI instructions and the CPU is roughly at parity with GPUs. So I meant to imply that Monero has the most CPUs mining.
|
|
|
|
Come-from-Beyond
Legendary
Offline
Activity: 2142
Merit: 1010
Newbie
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
March 07, 2016, 02:53:23 PM |
|
Ethash appears more ASIC resistant as it is more memory bound and needs at least 8x more memory.
Memory bounding doesn't look as a good feature in a world full of FPGAs.
|
|
|
|
tromp
Legendary
Offline
Activity: 990
Merit: 1108
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
March 07, 2016, 03:05:06 PM |
|
Memory bounding doesn't look as a good feature in a world full of FPGAs.
On the contrary. In a world where everybody has an FPGA, DRAM becomes the ideal mining ASIC, with the FPGA making sure that DRAM access is saturated and thus the DRAM is the actual bottleneck.
|
|
|
|
Come-from-Beyond
Legendary
Offline
Activity: 2142
Merit: 1010
Newbie
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
March 07, 2016, 03:07:25 PM |
|
On the contrary.
In a world where everybody has an FPGA, DRAM becomes the ideal mining ASIC, with the FPGA making sure that DRAM access is saturated and thus the DRAM is the actual bottleneck.
You seem to talk about replicating von Neumann architecture on those FPGAs, right?
|
|
|
|
tromp
Legendary
Offline
Activity: 990
Merit: 1108
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
March 07, 2016, 03:10:54 PM |
|
On the contrary.
In a world where everybody has an FPGA, DRAM becomes the ideal mining ASIC, with the FPGA making sure that DRAM access is saturated and thus the DRAM is the actual bottleneck.
You seem to talk about replicating von Neumann architecture on those FPGAs, right? The FPGA directly implements the memory bound PoW, just as an ASIC would, except it doesn't need the ultimate speed/efficiency of an ASIC, since DRAM latency is going to be the bottleneck.
|
|
|
|
TPTB_need_war
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
March 07, 2016, 03:16:01 PM |
|
Readers aren't going to understand what we are referring to with this technobabble about latency bound hash algorithms. Suffice it to say that afaik, Monero has the most CPU friendly hash function deployed. I think it is possible to do better though. Ethash unless they changed it significantly since before Ethereum was conceived is more GPU friendly than CPU friendly. So I presume CPU friendly would be more decentralized, given all other factors being equal. A hash function being CPU-only would still not be sufficient to prevent the mining from centralizing, due to the other economic factors of economies-of-scale (e.g. more hashrate has lower validation and propagation costs).
|
|
|
|
Come-from-Beyond
Legendary
Offline
Activity: 2142
Merit: 1010
Newbie
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
March 07, 2016, 03:19:48 PM |
|
The FPGA directly implements the memory bound PoW, just as an ASIC would, except it doesn't need the ultimate speed/efficiency of an ASIC, since DRAM latency is going to be the bottleneck.
I'm talking about architectures without von Neumann bottleneck. Artificial Neural Networks is an example of such architecture.
|
|
|
|
TPTB_need_war
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
March 07, 2016, 03:24:06 PM |
|
The FPGA directly implements the memory bound PoW, just as an ASIC would, except it doesn't need the ultimate speed/efficiency of an ASIC, since DRAM latency is going to be the bottleneck.
I'm talking about architectures without von Neumann bottleneck. Artificial Neural Networks is an example of such architecture. The algorithm for the "memory hard" hash function isn't going to allow you to implement it with a non-sequential architecture. The parallelism that is attained is to due to reducing memory latency by coalescing memory accesses, or by trading memory accesses for increased computation to maximize the consumption of available computational resources (e.g. on the GPU).
|
|
|
|
|