At 6 am on August 25, 2020, Filecoin officially started the space test on time. At 9 o'clock in the morning, IPFSBase and Cointelegraph Chinese jointly witnessed the smooth operation of the Filecoin space test. This event invited Feierke CEO Xie Jinbin, Interstellar Storage CEO Allen, Interstellar Continental CMO Shen Yingzhe, Leiying Technology CEO Tommy, and flying guest VB Exchange CEO Chris to witness the completion of the first packaging of the space test.

The following content is a selection of exciting answers for the event:

Interstellar Storage CEO Allen:At the core of mining, everyone is currently paying attention to the amount of storage per 10,000 yuan per hour, and I believe that the more important thing is to seal up valuable data. The official purpose of launching Filecoin Discover is also to expect miners to store valid data, not to refresh Just store some junk data. To store valuable data, the efficiency of sealing will be better than that of garbage data. Which miner can use Discover data storage first will have an advantage.

Starland CMO Shen Yingzhe:The amount of coins produced depends on 30% of the hardware, 30% of the team's technical strength, and 40% of the mining machine cluster architecture scale and operation and maintenance costs. Customers are also concerned about how many coins can be mined from the miner in a day after every 10,000 yuan invested, which is the most critical.

Thunderhawk CEO Tommy:There are many core factors in mining. I will divide them into two categories. One is related to the mining machine, from the storage scheme of the storage machine to the calculation scheme of the computing group, hardware selection and scheduling scheme. The elements other than mining machines are the official economic model, the overall volume and scale of each miner’s mining machines. I personally pay more attention to the price/performance ratio, starting from the cost of hardware and the price/performance ratio of the base. At present, the hardware cost and operation and maintenance cost of various miners are also similar. I am more concerned about the gap in packaging efficiency. The same hardware cost may be doubled in packaging efficiency.

Feierke CEO Xie Jinbin:The core point of mining is security. No matter how efficient your storage is and how cost-effective it is, if you can’t achieve security, it will be a waste of money in the end. Another thing is the stability of the sealed data. It cannot be said that it was fortunately sealed for a few hours, but the pledged tokens were deducted due to the power outage. In general, referring to the previous guests, the hardware cost determines when the user will pay back the cost, and the packaging efficiency should focus more on how many parallels can be performed at the same time, and more parallel mining can be carried out at the same efficiency. will be better.

According to the experience of the previous test network, which miner is most likely to be the first to complete the packaging and win the first mine?

Starland CMO Shen Yingzhe:The goal of Interstellar Continental is to keep three and compete for one. It is still uncertain how many dark horses in the market will mine together. As for which miner can complete the packaging first, it is still uncertain.

Thunderhawk CEO Tommy:According to the results of the latest Calibration test network, to be the first to complete the 32G package requires a little bit of luck in addition to relying on one's own strength. Dataline and Renren mines have a fast storage speed and the size of the mining machine is not bad, plus some luck, so the results in Calibration look very impressive, and they can win the first block outside the unofficial node.

Feierke CEO Xie Jinbin:The space test situation is more complicated this time. The official still needs to release a version 0.5.2, and some miners directly insist on packaging at this stage. The situation this morning is not easy to say.

Interstellar Storage CEO Allen:It is really difficult to predict today. Judging from the plan, the best packaging efficiency is 3970X. Someone optimized it to about two hours, but from the perspective of hardware plan and physical efficiency, this plan is the best. Pomegranate is used This is the solution, and there are four potential dark horses still using this solution, but it has not yet surfaced. However, if the mining machine is large in size, this solution may not be the most suitable.

Ordinary investors use Filscan or Filfox to check the latest Filecoin mining information. What are the most important points to pay attention to?

Thunderhawk CEO Tommy:You can view the mining efficiency of each node, the effective storage computing power growth rate, and the encapsulation time of a sector on Filfox. The data on the chain cannot be deceived. It is strongly recommended that everyone pay attention to the Filecoin block browser. The block rewards and lucky values ​​vary greatly among different mining pools, and mines with high lucky values ​​will be very competitive.

Interstellar Storage CEO Allen:The interface of Filfox is very good. First of all, you can see the ranking list of effective computing power. You can focus on the mining machine equivalent, which can be viewed with a 20% discount. Theoretically, the growth rate of computing power is steadily upward. The effective computing power of large-scale miners is very high, but their luck value is not very high, but some small-scale miners have relatively high luck value.

What is the purpose of setting up three 240T effective computing power nodes for the official node each time the test network was reset several times before?

Feierke CEO Xie Jinbin:Like many blockchain projects, Filecoin also set up 240T of so-called zero nodes or seed nodes. This is to prevent the computing power of some miner nodes from increasing sharply, controlling the network or causing impact on it in a short period of time. The official setting The three 240T nodes are to hold the initiative in their own hands during the testing phase and ensure the initial security of the entire network.

Starland CMO Shen Yingzhe:Why do the officials set up three nodes instead of four or two? I feel that the official Filecoin understands Chinese culture very well. The so-called one life is two, two is three, and three is all things, so the three nodes are very good. Haha, just kidding. Why is it 240T? We can simply calculate 240*1024 and divide it by 3, which is about 9000, which actually represents the number of sectors, and the official believes that the operation of such a number of sectors will be relatively stable.

After the Calibration testnet is running, many leading miners cannot accumulate effective computing power. What is the reason?

Interstellar Storage CEO Allen:As far as I know, some so-called leading miners have not actually participated in Calibration’s test network. Everyone is preparing for the real test, so they did not participate. These problems will no longer exist after the participation of real top miners, and I believe that in the future, all technical gaps will be smoothed out, although everyone is currently operating according to their own plans. Therefore, the effective computing power cannot be accumulated only because many top miners did not participate, or stopped after half of the participation.

At the same time, as for the reason why leading miners do not participate, first of all, because the main network is approaching, many leading miners want to focus on preparing for the main network and the test network in August, rather than making a fuss about Calibration. And some of them have already done marketing after running the data on Calibration. After the goal is achieved, what needs to be done is to continue to upgrade the solution and operational capabilities on this basis, and at the same time save hardware costs, so there is no need to continue to participate.

Thunderhawk CEO Tommy:I think Calibration is very necessary to participate. Our Thunder Eagle performance on Calibration, although the packaging speed is outstanding, but there will always be insufficient pledged coins. Although this problem has long been resolved, it reflects the importance of participating in Calibration. sex. For mining pools with large-scale hardware, it is still necessary to test and run to check for leaks and make up for vacancies. I don't think that's going to happen in this space race, because it's a prize test after all, so everyone will come and participate.

Each 32G sector needs to pledge one FIL. Does the official pledge model design have its inherent rationality?

Starland CMO Shen Yingzhe:Since the official economic model has not yet been finalized, I personally think that such a design is actually unreasonable. Because no one can say for sure how many coins a sector can produce, so the amount of coins produced by a sector is uncertain, so it is too egalitarian to pledge one FIL per 32G sector.

Feierke CEO Xie Jinbin:The original intention of this design is that someone used a script to roll out the coins from the faucet of the Calibration test network, and the result was that everyone lost all of them, which also caused several resets of Calibration to a certain extent, and also caused many miners to not have enough coins. Mortgage. Up to now, the number of mortgage coins has been adjusted until now one 32G sector is mortgaged. I don’t think this will be the final solution. It is a temporary value set to ensure the stable operation of the entire test network , the final plan still has to wait for the finalization of the final economic model.

Discuss the importance of fast packaging speed for Filecoin mining. If the packaging speed is slow, can volume be used to make up for it?

Starland CMO Shen Yingzhe:First, I think packaging speed is an important but not necessary condition. No matter how fast the encapsulation speed is, it still faces Proof of Time and Space, so many people see that the amount of node data in the test network is large but the production of coins is very low. This is because the Proof of Time and Space has not been completed.

Second, I personally judge that the utility of fast or slow packaging depends on the economic model, so in the current situation where the economic model has not yet been determined, we do not know whether it is better to be faster or slower. A good company should be able to provide both high packaging speed and slow down.

Thunderhawk CEO Tommy:I think that the lack of speed can be made up with volume, but we need to be clear about the source of volume, which is usually your users. From an investor's point of view, I am willing to invest in miners with high packaging efficiency, so that your return on investment cycle can be guaranteed.

Will the mining income of Filecoin fall into a spiral cycle where the strong are always strong, and can miners that are not in the top five still get a share of this mining wave?

Interstellar Storage CEO Allen:Although we are talking about the storage industry, I think this trend will be the same as Bitcoin computing power mining, and the strong will always be strong. As the volume increases, it will squeeze the survival of small-scale miners, causing small miners to report to the group to keep warm. This must have aggravated the centralization of Filecoin, but to what extent it will affect the ecology of miners will not be known until the economic model is finalized. At the same time, I think that miner nodes that are not ranked in the top five can be tied together to ensure stable income and output.

Feierke CEO Xie Jinbin:From the official point of view, they do not hope that after Filecoin goes online in the future, there will be a situation where the strong are always strong, which will intensify the degree of centralization, so I think the official will add some mechanisms to limit large miners in the economic model. The official attitude can already be seen in some of the current mechanisms, including the 180-day redemption time of the mortgaged currency, and the 20-day linear release of the mined currency. At present, even miners ranked in the middle can still eat meat and drink soup, but the final situation still depends on the economic model.

During this space test, how should users who have not purchased mining machines choose?

Starland CMO Shen Yingzhe:Rational investors should follow the 334 investment principles. The so-called 334 refers to: 30% of the cluster architecture ranks 30% of the software code, 30% of the mining pool's scale, stability, and operating capacity of 40%. Therefore, the currency production capacity is the result of a comprehensive indicator. Although hardware is very important (mainly in the early stage of the project), but now the requirements and configuration of hardware are becoming more and more transparent, so all major miners are working hard on software, ranking, code and operation. But these also only affect plus or minus 5% of your coin production. The real comparison after the mainnet is launched is your stability and security, the scale of the mine, etc., and these factors account for 40% of the currency production! Therefore, the amount of coins produced is the result of a heavy operation, so customers must care about the cost performance in investment, depending on whether your investment is allocated to the production and purchase of machines by miners (such projects have a very slow return cycle), or a certain percentage They have been allocated to the various links I just mentioned.

Interstellar Storage CEO Allen:The first is to look at the number of machines and the computing power generated, which can be measured quantitatively. The main cost in the future will still be in hosting and operation. Just like the cost of Bitcoin mining is put into the machine in the early stage, and the cost of electricity is mainly in the later stage. Where the electricity cost is low, the cost will be low and the cost will be paid back quickly. In the space race, investors should pay more attention to the number of miners' machines and the amount of coins produced to calculate power, and calculate the production capacity of a unit mining machine based on these data. At the same time, measure the ability of future custody and operation, which involves your long-term rate of return. The last point depends on the fluctuation of the currency price. When the currency price falls, due to the mortgage mechanism, miners cannot easily get out (due to the degree of punishment). Beware of the "Xisha model" in the market, that is, some projects do not invest funds in operations, hosting, and code development, but instead purchase machines and promote cloud computing power. Such projects often have hidden murderous intentions. Investors should be careful to identify them.

Let me reveal some of Filecoin's "little shady scenes". The first is cloud computing power. Generally, you don't know how many machines are running behind the project that sells cloud computing power, and no machines are delivered to investors, but only T The number, so it is difficult to know whether the stock you bought is a machine or a certain hard disk of a machine. And you will find that their hardware costs are also uneven. Regardless of technical operations, the performance of all hardware running under the same conditions is the same and there is not much difference. Therefore, the output value behind the purchase of cloud computing power by investors will be How much is actually difficult to determine.

Thunderhawk CEO Tommy:At this stage, as the mainnet is approaching, I hope that everyone will still have to look at the block browser to see a lot of data and indicators, such as packaging speed. Filefox has actually done a good job. From this we can see that the major The performance of miners can actually find out who will perform better, especially for retail friends, we should also pay attention to diversify investment in the miners we like, and it is best not to all in. The block explorer I have always emphasized cannot deceive people, so we must learn to look at the indicators on the block explorer.

Feierke CEO Xie Jinbin:I think it’s more about watching, listening, and learning. Look at the block explorer and learn to compare the above indicators and data. Don't just get in the car with a smack of all in. Also learn to use Du Niang to search for various relevant information to understand basic information. Then learn to use a block explorer. Now there are many excellent self-media channels, such as our IPFSBase, haha, let me make a small advertisement for you.

Next, the flying guest, VB Exchange CEO Chris, will chat with us about the repeated postponement of the Filecoin space test. What timing are we waiting for?

VB Exchange CEO Chris:Since the beginning of this year, the popularity of Filecoin has been increasing, and it is also a popular search keyword in major search engines. The first eye-device that Filecoin launched will have a certain impact on the confidence of early miners. It also has a certain impact on investors who have been holding a wait-and-see attitude since the beginning of the year. But I think it is normal for some large-scale projects to postpone, such as Ethereum and EOS, which we are familiar with. The extension of Filecoin is really within the acceptable range for me. We can also see that it is safe for the mainnet to go online This is good news for investors. Second, a certain degree of delay will also give swinging investors more opportunities to enter the market, and will also increase their popularity. At the same time, this year’s bull market in the currency circle has a lot of incremental funds flowing into the currency circle, which also gives Filecoin more More capital support also makes us look forward to the future value of FIL. Coupled with the rigid demand for IPFS storage, Filecoin combined with it, integrated social resources, followed the development of the times, and created social value, so I think FIlecoin is a serious outlet.

Feierke CEO Xie Jinbin:Blockchain projects or Internet products will always have various BUGs before they go online, which are unpredictable, and often when you solve one BUG, ​​another BUG will be triggered. This is a normal situation, so the delay must be delayed. To a certain extent, it is also a process of constantly improving BUG. For miners, the operation is too cumbersome, and the domestic network is complicated. In fact, there are always problems in the adaptation between the two parties. It cannot be simply said that the delay is whose problem.

Interstellar Storage CEO Allen:The developers of foreign projects, especially the American project Filecoin, are often very perfectionist. Their thinking is very jumpy and logically coherent, but they don’t pay attention to implementation, which will lead to rapid project development but a long time to implementation. Another point, from the perspective of the entire ecology, there were few nodes in the early stage of the project. After the testnet was converted to Lutos in 2018 and 19, more on-chain governance and code layer development were started. This is an open project. Expecting global developers to participate together, his own team has some minor problems, and the joining time and volume of global developers will also affect the progress of the project. Another point is that its code layer determines that its development will not be fast. Only the Filecoin mining machine has been changed to three versions. I think the current configuration has reached the top and cannot be increased, or the mining machine requires further increase. It won't be that fast either. The continuous modification of the code layer also makes everyone's mining machine schemes constantly change, so there are still many problems in the code layer at the core.

Starland CMO Shen Yingzhe:In fact, I have also talked about this with the official. If the official puts robots in Asia, the mainnet can be launched very early, but the official insists on the route of globalization to be placed in Europe, so Asian miners, which are the main force, especially It is difficult for small and medium miners to solve network problems, especially VPN problems. So repeated delays have a lot to do with this.

Thunderhawk CEO Tommy:In fact, the postponement of the space race is a good thing. It gave us miners enough time to prepare. Although there are many bugs along the way, for example, the bug this morning is stuck in the 101st block. But the real test of the space race is the readiness of the miners. The postponement of the space race is like preparing for an exam. It gives us more time to prepare, so that the miners have more time to do their homework and reduce the number of problems and bugs in the space race. The probability.

During the event, the first to complete the 32G sector packaging is t01101 Lianpai Technology, and the second to complete the 32G sector packaging is t01013 Thunder Eagle Technology. As of the end of the event, the unofficial node with the highest ranking is t01155 Dataline&Diande. It is also the first node to complete 1T packaging and the first unofficial node to generate blocks.