Recently, many friends have asked me what I think of Dynex due to the 10-fold increase in 3 months. First of all, let me be clear that the price of a project is not necessarily positively correlated with what is actually done, so the risk warning does not mean denying its price (of course I think There are risks at this price now). Next, I will mainly try to demonstrate the possible risks of Dynex at the actual implementation level based on public information.

Dynex is a typical black box project. The founding team is anonymous and the core code is not open source. Someone once revealed that the founder is Daniel Mattes, an Austrian entrepreneur who was punished by the SEC and is good at speeches and packaging. You can also open the official website of Dynex. I saw a lot of unclear concept packaging, such as quantum computing, neural superform, artificial intelligence, smart city, revolution, cutting-edge... these words are straight to the top of the sky.

Dynex defines itself as “the world’s first neuromorphic supercomputing blockchain based on the DynexSolve chip algorithm, a Proof of Useful Work (PoUW) approach to solving real-world problems.”

So first of all it is a public chain, and secondly its core lies in DynexSolve and PoUW.

It can be seen that DynexSolve is its core original algorithm, which magically converts any GPU into a simulated neuromorphic computing chip. However, there is no public paper on such a revolutionary technology, and neuromorphic computing has not yet been published. It has been applied in the real world, including top universities, the US military, Intel and IBM. They are still in the exploratory stage. Has it been broken through by Dynex?

Coming to the concept of PoUW, which is not new, the logic is easy to understand. In the past, POW used a lot of computing power to compete for a meaningless value, resulting in a huge waste of resources. If this value becomes meaningful, It would be nice to have both your cake and eat it, mine to ensure the safe operation of the network, and do some practical things and even earn more returns. Therefore, the concept of PoUW came into being, especially with the popularity of AI in the past two years.

Some scenarios that require computing power can be given to the chain by issuing tasks. After the node completes the calculation and completes the task, the results will be returned to Party A. In addition to the reward of the chain itself obtained by normal mining, Party A can also give the node Remuneration, and most of these parties can also apply it to the national economy and people's livelihood areas such as smart cities and smart medical care, so that the blockchain can truly empower the real economy. It sounds like a perfect ending, but it cannot withstand scrutiny.

For a blockchain, PoUW is too non-standard in the three stages of beginning, process and end. The number and frequency of tasks are non-standard, the difficulty and time-consuming of completing tasks are non-standard, and the correctness of verification task results is non-standard. As a result, it is impossible to stably generate blocks and effectively verify them, and this chain will not have basic usability.

It can be seen on the Dynex official website that 3,300 tasks have been completed, and only 1 is currently in progress. All of these tasks are black boxes and I don’t know what they are doing. There is no entrance for task release. So where do these tasks come from? Come on, what is the current use of such large-scale GPU computing power? We have to find a way to give them work, so we returned to the self-entertainment state of POW.

The logic behind Dynex's rise can be divided into concept and demand, apart from being a banker.

Concept: Take advantage of the popularity of AI and apply the concept of PoUW to reap a wave of narrative dividends.

Demand: With the conversion of Ethereum to POS, a large number of idle mining machines have been generated, coupled with the overflow of Kas, so from the demand point of view, there must be something that can catch these computing power, and then package the DynexSolve algorithm, which claims that any GPU can be transformed In order to simulate neuromorphism to attract computing power.

At present, Dynex is quite controversial. After all, pulling the strings is justice, and there are miners and secondary profit-making interest groups. However, it still emphasizes that the price of the project is not necessarily positively correlated with what is actually being done. Meme coins are not doing anything but they still have prices, so I only discuss and demonstrate my personal opinions from the perspective of doing things, and do not involve any investment advice. I will continue to wait and see. After all, from the perspective of industry development, I really hope that projects like Dynex can succeed.