The core logic of Filecash technology development is to lower the barriers to participation and introduce a cache layer for actual storage services to implement database applications.
Original title: “Demystifying Filecoin’s first pre-fork, how does Filecash inherit the ideals of IPFS? 》
Written by: Alex and Noah, the former is HPOOL CEO, the co-sponsor of Filecash community, the latter is the core developer of Filecash
On September 29th, HPOOL CEO, Filecash community co-sponsor Alex and Filecash core developer Noah were a guest at the Deep Chain Academy online salon, focusing on the theme of “Demystifying Filecoin’s first pre-fork, how does Filecash inherit the ideals of IPFS?” share it. The following is the sharing record, which has been sorted out by Shenzhen Chain Finance.
Sharing session
Alex: I am very happy to be here today with you all. My name is Alex. I came into contact with BTC in 2011, and started mining in the blockchain industry at the end of 2011.
In 2013, we established Yibit with some early players in the industry. At Yibite, we have done all the areas that can be touched, including portals, mining machines, exchanges, and blockchain projects. Because we have been paying attention to the underlying technologies and consensus protocols, such as Ethereum and EOS are more focused on the application field, but we discovered in 2015 that the storage-based consensus protocol is very valuable, and judged that the next 5-10 years will be The track where the consensus protocol is stored.
This track is divided into two categories, one is the attempt to remove component service tools based on the underlying consensus of the blockchain, including Sia, and the other is the storage application based on consensus protocols like Filecoin Exploration. Although the two have different paths, they end up in the same way. They are both explorations of storage protocols.
In 2017, we established HPOOL, which specializes in projects storing track formula agreements, and received investment from Fenbushi Capital and Wanxiang Blockchain Lab.
Therefore, we paid attention to Filecoin very early and found many problems in it. This is why there is a Filecash fork project today. In fact, even without Filecash, there will be other fork projects. This is an inevitable trend.
Noah: Hello, everyone. I am very happy to receive an invitation from DeepChain Finance today to join the AMA on Filecash. I am one of the core developers of Filecash, and everyone privately calls me Lao Yang.
In 2014, I paid attention to blockchain technology. At that time, I felt that blockchain technology is a new and very attractive technology. At the same time, I also paid attention to the first IPFS protocol released by the Protocol Lab.
Before entering the circle, I mainly led many large-scale projects in the field of mobile technology and network security, covering big data collection and analysis, distributed computing, and network security. I also studied some public chains. In the earliest ALGO project white paper After the release, I led a team to implement the project based on the content of its white paper.
At the same time, we also studied virtual machines and the development of virtual machines. For Ethereum, our agency team also reconstructed it and studied how to interact with Ethereum and ALGO.
In addition, I have in-depth research on blockchain modularity, P2P, consensus, virtual machines and cross-chain communication.
When doing blockchain development, I have also been following the progress of the entire Filecoin community at the same time. When it changed the algorithm at the beginning of this year, the community had already expressed opinions on the entire technical route. At that time, someone had already expressed the fork. the meaning of.
Therefore, the origin of the Filecash fork did not appear suddenly, but the repeated delays of the entire Filecoin, and the official random changes to consensus, algorithms, and economic models, which caused dissatisfaction in the community.
I also had some heated discussions in the community.
question Time
Deep Chain Finance: Excuse me, Mr. Han of MIX Group recently announced that it would fork Filecoin, which caused heated discussions. Why did Filecash fork Filecoin and what is the background of the fork?
Alex: The basic background of the fork is actually the disagreement caused by the inconsistent voices and interests of the protocol laboratory and the community.
Moreover, the fork itself has become a very basic inevitable phenomenon in the blockchain industry, including the previous BTC, BCH, and BSV. Why does the blockchain industry have such a unique bifurcation phenomenon? Because the openness and inclusiveness of the blockchain is different from the original Internet. Most software on the traditional Internet is closed source, and it is difficult for latecomers to iterate and innovate better on the original code.
The point of divergence of interests is that the current economic model is too protective of the interests of the capital and allows the miners community to take over.
The divergence of the voice of the community lies in the idea that the protocol laboratory is too corporate to operate, ignoring the demands of the community, and many centralized KYC determination rules similar to super ROOT permissions (such as: centralized authentication data set: super computing power, manual KYC order certification miners The regional division of the space race proceeds, and the temporary modification of the rules in the big miner test is too arbitrary), which is unlikely to occur in the previous blockchain world. It is not the community barrier created by public chain code or node governance ideas.
Satoshi Nakamoto hopes that the distribution of mining on the six continents of the world is consistent, but IP cannot be verified by algorithms. Although the Bitcoin team uses centralized manual KYC to verify orders, there are still various ways to deceive official.
Including his manual KYC order to verify the valid data set, here are actually some achievement projects. This may perform well at the capital level, because capital does not care about these processes. He may see that in the end he has signed contracts with many such companies, but these contracts are not real business demands, but artificial ones. With super authority, this super authority may change in the future. At least for now, the whole situation we see is that the official effective data set is ten times the output of the data set that everyone has brushed, which is equivalent to having an official center. Power to generate super miners.
Also in this big miner test, because many detailed rule changes were too arbitrary, the officials themselves did not think about it carefully, and then the loopholes in the mechanism were generated, which were used by deliberate miners. However, the government directly declared the miners invalid through corporatization. Of course, it may not be implemented due to pressure. However, this matter also caused great harm to the entire miner group.
As early as in the Filecoin forum, there are various anonymous IDs (Zhu Yuanzhang, Song Jiang, Song Hai, Song He, etc.) constantly criticizing the official various as expressions of dissatisfaction. It is guessed that these IDs are mostly miners, but the official is too corporate. Operation, standing up with real names and worrying about being blocked by the authorities, daring not to speak up, and forking requires one or more neutral enough initiators to start challenging the protocol laboratory head-on to balance the excessive power of the protocol laboratory. The fork came out under such a background.
Deep Chain Finance: Excuse me, for the blockchain industry, “consensus” affects the development prospects of a project to some extent. Will the fork of Filecoin affect the development of the entire distributed storage ecosystem?
Alex: The fork itself is a consensus of the community. From the history of BTC to the present, there have been many cases of forks that can be used as a reference. Any project with relatively large influence has almost been forked.
And at present, almost any influential and relatively large project will fork. I actually talked with a friend for two or three hours last night. We all agree that the fork itself is fully in line with the spirit of the blockchain. .
Why is the fork the spirit of the blockchain? Because a public chain should not be controlled by a single organization, but after open source, anyone can challenge it, initiate differentiated competition, and initiate different voices.
And if a public chain is strong enough and good enough, it will not be afraid of any challenges. New challenges will only make the original chain better, not worse.
The openness and inclusiveness of the blockchain industry has made bifurcation an inevitable phenomenon in the blockchain industry. Bifurcation is a concrete expression of the community’s different ideas for the development of a project.
Filecash can become an effective Filecoin competition network, which can effectively balance the power of Filecoin’s corporate operation. At present, the protocol laboratory will have root super permissions similar to Filecoin, which was not allowed in public chains in the past.
On the whole, the emergence of forks has a positive effect on the entire ecology than it has disadvantages.
Deep Chain Finance: Noah, what are the technical differences between Filecash and Filecoin?
Noah: In the process of discussing the Filecoin fork in the community, one of the most important attacks is actually Filecoin itself, because the current requirements for hardware have actually become higher and higher, and the ability of the entire mining participants to develop software The requirements are getting higher and higher.
The core logic of Filecash technology development is to lower the barriers to participation and realize the vision of “Let the world’s idle hard drives participate in storage” in the Filecoin white paper. For this purpose, some algorithmic modifications have been made.
1. Upgrade the P1 core algorithm, and upgrade the sha256 algorithm to sha512; because the Sha256 algorithm is integrated in AMD’s latest processor, and then such an algorithm will make AMD’s latest processor have an extreme in the mining process. Big advantage. In the early deployment of mining software, we found that it may take 30 hours to perform P1 calculations with Intel machines, but with AMD processors, it may only take four hours to complete the whole Calculation of P1.
This makes it difficult for Intel’s large number of idle machines to participate in Filecoin mining, because even after participating, Intel’s output is only 1/10 or 1/8 that of AMD processors, which is completely uneconomical.
2. Modify the sector size to 16G to reduce the memory footprint; because the official sector size is now 32G and 64G, when performing a basic calculation, the user’s computer also needs at least 64G of memory. For users, such a memory requirement cannot be achieved at all, nor can it meet the requirements of such a hardware configuration, so we reduce the memory usage in this part, so that more home users can participate in early mining activities.
3. Modify the P1 part of the calculation layer number to 5; because in the P1 calculation, the official setting is 11 layers. In fact, we have studied some POC projects before, including some early storage or encryption projects. In fact, there is no need to do so many layers of encapsulation, and reducing the number of encapsulated layers will also reduce computational overhead.
Therefore, the core logic of our entire technical roadmap is to lower the threshold of participation in the entire ecosystem, so that a large number of idle mining machines and family customers can participate in the entire network and provide consensus for the network. We believe that we are providing network consensus. In terms of the dimension, users should not only have enough computing power, but also make the computing power sufficiently dispersed, so that more people can participate in order to expand this consensus more widely.
At the same time, in Filecash’s technical roadmap, the cache layer L2 for actual storage services is planned, and the cache layer is used to implement database-like applications. The implementation principle of L2 is to reduce the communication pressure of the data network through the node mechanism. Some hot data can be directly placed in the node database, and the Filecash network becomes easier to use through the multi-level storage framework.
The overall technical differences are the above points. These are the largest ones, and there are some small technical differences, which can be refined in some subsequent chats, and then shared with all miners.
Deep Chain Finance: First of all, Alex, what are the characteristics and advantages of Filecash compared to Filecoin?
Alex: I personally prefer to talk about features and differentiation. As for the better and the worse, it is completely determined by different governance ideas. All forks must have a basic point that is ambiguous with the original chain. It started with a fork, but in the long run it is an independent project with different development paths on the same track. All previous forks have become independent projects after the baptism of time, and have become an increase in the market.
Although Filecash seems to be challenging Filecoin head-on, in fact it has its unique differentiated competition from the beginning, and it has increased the diversity of the storage track.
In terms of development thinking, Filecoin is a planned economy + corporate operation is currently more like Xunlei’s player cloud. Filecash is a market economy + public chain community operation which is closer to the fundamental spirit of the blockchain. The original chain and the fork have been based on Two completely different routes to their own development.
The main differences are as follows:
Difference 1: Filecash is friendly to AMD and Intel at the same time. Compared with AMD’s worldwide inventory of intel equipment, Filecash has an absolute advantage. The current design of the Filecoin mainnet’s AMD advantage in disguise makes Intel’s cpu calculation too disadvantageous, blocking the stock and mainstream server market.
Difference 2: Filecoin sealing cost is too high, the time is too long, and the storage cost is too high. Filecash will reduce the sealing cost and reduce the storage cost of the entire network. Assuming the same storage of 1GB content, the Filecash network will be cheaper and faster than the Filecoin network .
Difference 3. Governance is community-based, which relatively reduces the pressure of miners’ hardware equipment and economic model mortgage pressure. From governance to economic models and technical features, it is more friendly to communities and miners.
Difference 4. Remove the super power of the artificial kyc part in Filecoin, Filecash uses a consensus algorithm to determine the fairness of the public chain, not a company ruling.
Deep Chain Finance: May I ask Noah, in terms of technology, what are the characteristics and advantages of Filecash compared to Filecoin?
Noah: In the above question, I have already mentioned the core features of Filecoin compared to Filecoin. Here is a detailed explanation of the core features below this question.
In fact, let’s take this apart and look at it. Filecoin and IPFS are actually two sets of things. IPFS is a storage community based on a principle similar to P2P. Then, Filecoin is actually an incentive layer grafted onto the IPFS ecology.
In fact, the ease of use in the IPFS ecosystem is very high, that is, the speed of the storage network we are talking about is actually very fast, whether you are uploading or downloading, you can achieve such a full network of yours Speed, but because of an incentive like Filecoin, in order to maintain network security and maintain consensus security protocols, the laboratory has done a lot of work to meet security and consensus, so here is There are some new problems.
The biggest problem is that when we use these P1, P2, C1, and C2 packaging processes, the entire upload and download users need to go through this process. According to the hardware configuration of the server with the highest performance we currently use, it takes three hours for the user in the whole process, and it may reach dozens of hours in the long time.
In this process, users not only consume network bandwidth, but also consume a large amount of computing resources. At the same time, they also need to wait for a very long time in this process, and they may also face the risk of failure.
Under the current Internet rhythm, this speed can only meet a very small number of application scenarios. For example, we store some very cold data, for example, we store some photos from ten years ago, or do some pure cold The backup can be placed in the Filecoin network, but once it rises to the commercial network scenario, Filecoin may be very difficult to satisfy, because this Filecoin is a completely decentralized network, and all nodes are distributed. Moreover, when the weights of the nodes are the same, it is impossible to build a large number of commercial scenarios, just like on Ethereum. Ethereum’s smart contracts can only support very few applications to run on it.
So Filecash reconstructed the network in this, and put forward the concept of cache, that is to say, Layer 2 built in Filecash.
Using Layer 2 to solve the problem of low availability, then at this time, we have redefined the two layers of the network. The bottom layer of the network is the network built by Filecoin, which is mainly to maintain the underlying storage. At the same time, to maintain the security of the ledger.
Then, to support an application in the framework of Layer 2 is to perform consensus security in the underlying network, as well as the maintenance of the security of the account book, so as to ensure the operation of the entire core of the network. On Layer 2, we have done more development of the partial application layer, including the optimization of virtual machines, and the optimization of network speed, etc. In order to make the application run better, there are similar We have also conducted a lot of discussions regarding the support of thermal data that needs to be used in the application process.
So in a comprehensive way, the core feature of Filecash is that it solves Filecoin. The current network only has consensus, but this problem is not applied.
DeepChain Finance: Excuse me, it seems that the forked project has not developed as well as the original project. How can Filecash face Filecoin, so that more miners and investors can join Filecash and grow the ecosystem?
Alex: We can directly take out past cases to observe this problem:
- BTC-BCH original chain and fork market value ratio: about 10:1 in the first year
- ETH-ETC original chain and fork market value ratio: about 5:1 to 10:1 in the first year
- BHC-BSV original chain and fork market value ratio: about 2:1 to 1:1 in the first year
- Uniswap-Sushiswap original chain and fork market value ratio: about 1:1 in the initial stage
Although it is difficult for forks to challenge the original chain, from the observation history, forks, challengers, and imitators may not be equal to the original original chain, and the shorter the consensus in the previous cases, the easier it is to be challenged. The more problems you have, the easier it is to be replaced. Of course, there will be a lot of challenges and problems, and the latecomers are all based on the secondary development of the original chain’s technology accumulation, and they take a lot of money in research and development. In recent years, the actual performance of the forks is significantly easier to be accepted by the market than previous projects.
The miner group itself is still a relatively market-oriented entry. The value bearing of the public chain itself needs to be able to support the business cycle. The mining industry will naturally prosper. It should not only focus on the single point of mining. Mining is the infrastructure of the public chain and is part of the cycle. All, we still have to rely on commercial applications to land. The later forks need to use a community-based development path to solve the hot data storage and unstable storage prices that have not been achieved by the Filecoin network.
Deep Chain Finance: Excuse me, Filecoin officially announced the approximate date of the mainnet launch on September 28. When is Filecash expected to launch the mainnet?
Alex: Filecash is expected to go live on the mainnet on 14-15.
Deep Chain Finance: May I ask Noah, how do users participate in mining?
Noah: The initial code of Filecash is inherited from Filecoin. The code and operation logic are basically the same as Filecoin, and the deployment and operation methods are exactly the same as Filecoin. The current code requires a certain technical foundation to participate in mining. The software package for one-click mining will be developed in the following time. At that time, all equipment that meets the hardware configuration requirements can participate in mining.
