Solana is dead on the beach, the nuclear cloud arrived from the north and they never finished building the race car. Dead project is dead, this is a dead project. Solana brute-forced their way into speed instead of using clever design and the costs of their brute force approach make the project unworkable. The blockchain blackouts are too frequent to make this is a serious project. Just one blackout would be unacceptable and they have already had two. The reason is obvious, the blockchain time is too short for the nodes to coordinate with each other and the chain is threatened with forking as groups of validator nodes become separated from each other. So they shut down the network and stop making blocks. This has happened twice in the very short history of Solana and will happen again. Because it has to happen, the block time is too short.
400ms block times cannot work, ever. That isn’t enough time between blocks for the validator nodes to coordinate their activities. If you change the time between blocks it linearly affects the transactions per second the blockchain is capable of. If you cut the time between when blocks are added to the blockchain, usually called the blocktime, in half then you double the transactions per second. If you triple the blocktime then you reduce the transactions per second by two thirds. The bizarrely low blocktime for Solana is one part of its overall brute force attempt to solve the speed problem. But it doesn’t and can’t work.
The fundamental activity of any blockchain is validating transactions. In a proof of work system like Bitcoin, this is open to be done by anyone and a new block of transactions is validated every ten minutes. Once a block full of transactions has been validated and added to the blockchain there is a small chance of a reorganization happening and that recent addition becoming irrelevant, it becomes as if it never happened. This is rare and becomes rarer as more blocks are added. After two more blocks have been added after a block most people assume that block, and the transactions within it, are permanent additions to the blockchain. If you paid for something with Bitcoin after two blocks have been added to the blockchain after the block containing your transaction everyone will treat this payment as final, though there is a tiny chance of a reorganization still happening. After a hundred blocks the chances of a reorganization making that block irrelevant are down to being a once every ten years or longer event. It could happen, but people treat it as impossible because it is so rare. This type of event might not happen with Bitcoin again in your lifetime.
In proof of work systems like Bitcoin transactions are validated by anyone in the world, it is a completely open system. Any node can validate a transaction. A node is simply a computer on a network, any computer, or node, on the Bitcoin network can validate a transaction. In proof of stake systems only centrally approved validating nodes can validate a transaction. Any computer on a network is a node in the network, but only an approved list of nodes are allowed to validate transactions in a proof of stake network. Solana is a proof of stake network and has unusually strict requirements for being a validating node, the estimate is you need at least $5,000 worth of computer equipment to meet the application requirements. This is part of their brute force approach to achieving speed.
Since blocktime linearly affects transactions per second and you always want as many transactions per second as possible the obvious choice is to make blocktimes as small as possible. There are two reasons this isn’t done. One is that shorter blocktimes obviously mean more blocks and so the data to keep track of the entire blockchain becomes longer and harder to store. If you cut blocktime in half you double the transactions per second and you double the data being generated per second and holding the entire blockchain history in your hard drive becomes twice as hard. Having nodes store the entire blockchain history is important to security and so this gives a motive for longer blocktimes. The main reason the Bitcoin blocktime is at ten minutes is they want as many computers as possible to be able to work as nodes for the sake of security and decentralization, so they don’t want the size of the blockchain history to become too large.
The second reason blocktimes do not become too short is that the nodes need time in between nodes to coordinate their activities. There is communication back and forth and decisions that have to be made. The nodes on the network are talking to each other as part of the process of choosing the next block and that takes time. If you give them too little time to talk to each other the coordination between nodes can break down. Nodes can become unaware of what other nodes are doing and make uncoordinated decisions. Instead of functioning as one network, the isolated nodes begin to function as more than one network when they can’t talk to each other fast enough to coordinate the choosing of the next block to add to the blockchain and will add different blocks, creating more than one chain. This is called forking and no blockchain can survive it.
So making the blocktime shorter is an easy way to increase transactions per second, but if it's too short it makes it harder for inexpensive computer set-ups to serve as validators and it makes forking more likely. The importance of allowing inexpensive computers to serve as transaction validators is highly debatable and in large part a philosophical argument. Solana obviously does not care about this and that is their decision to make. But forking is not a philosophical argument, it is a practical argument. No blockchain in the world can survive forking, this automatically kills any project. The Solana blocktime is just too short to reliably allow validator nodes to coordinate the choosing of the next block. Eventually, communication between validating nodes will break down and forking will occur. No blockchain can survive forking, this means the project is dead. Any project that can’t stop forking is a dead project, this is not even debatable.
Solana could increase their incredibly short blocktimes but that would reduce their currently very impressive transaction per second numbers and these numbers are the whole reason for Solana’s existence. The second-largest market in cryptocurrency is the smart contract market controlled by Ethereum, the largest is Bitcoin’s position as the store of value. Bitcoin’s position is untouchable for decades if not generations. Ethereum’s control of the smart contract market could end in a year. The smart contract market is the second-largest and is touchable, competitors can take part or all of Ethereum’s market share while Bitcoin’s market share is simply untouchable for decades if not lifetimes. So the main target for investors in new projects is the Ethereum smart contract market.
There are many things wrong with Ethereum, but the most famous thing wrong with Ethereum is it is slow, the transactions per second is just too low, everyone knows that. So Solana aims to be the fastest blockchain ever created, the greyhound which will take business away from the tortoise. The entire Unique Selling Proposition that Solana uses to market itself is that it is the fastest blockchain in the world and this is why customers should switch to it from Ethereum. If Solana slows its transaction per second numbers then it is discrediting its own Unique Selling Proposition, it is discrediting its own reason to exist. Because without speed Solana is nothing.
A simple doubling of Solana’s blocktime to 800ms will cut their transactions per second in half. That means the value of Solana’s Unique Selling Proposition gets cut in half, it is only half as valuable as before according to its own inner logic. And 800ms is still too fast for the nodes to always coordinate with each other, forking will still happen, but not as often. Solana needs a drastic increase in its blocktime which will lead to a drastic reduction in its speed. But speed is the only reason Solana exists, they are in an unworkable situation.
What Solana did reminds me of how you can bore and stroke an engine to double its horsepower. This is how race car engines are made and it is an easy way to get more speed. But the metal fatigue from running the engine is so great that you need a full-time mechanic for the car to keep replacing key parts before they snap. If you have the money and want to race cars it's worth it, but for your daily commute boring and stroking your engine would be insane. Boring and stroking is an unusual, brute force answer to the speed problem only useful for race cars because the disadvantages are too great for anything else. This is what Solana did, they bored and stroked the engine to go faster. But this isn’t a race car, this is a blockchain that is supposed to be running non-stop forever. Bitcoin runs non-stop forever, good blockchains can do that. With the right design, a blockchain never stops. Solana stops. Because it is not a good design, they used brute force instead of clever design to achieve impressive transaction per second numbers.
Solana is a fake project, it pretends to have solutions it does not really have. It is a fake project built for a quick pump and dump. To attract attention it was created with the best transaction per second numbers of any blockchain. Those numbers are real, Solana is the fastest blockchain. Speed numbers are something easy to understand and explain to the public. And the public is already familiar with the fact that Ethereum’s biggest problem is lack of speed. On the surface it makes a compelling marketing case for Solana adoption: Ethereum is too slow and Solana is the fastest blockchain in the world. This is a very simple pitch for the merits of Solana, one that can be explained on an elevator ride, one that the average person can easily understand. These are the kinds of arguments you want for a quick profit, and Solana backers have made billions.
There are more reasons than just too short blocktimes for being against Solana. But this issue goes to the very heart of the matter. Solana has always been about speed. It truly is the fastest blockchain in the world today, they achieved that goal. But they did it through brute force instead of clever design and the brute force solution is causing metal fatigue. They built a race car engine when they needed a Toyota. Because cheap and easy ways of increasing speed produced transaction per second numbers that impress the public, a public which is less able to evaluate reliability issues. But after two shutdowns the public realizes something is wrong with Solana reliability. And there will be more shutdowns to come and eventually they will have to lengthen the 400ms blocktime. That number can never work, part of it gets eaten up by ping time and nodes need time to evaluate data and there are inevitable slowdowns on any network. Eventually, the nodes will stop coordinating and the chain will fork or get shut down. 400ms can never work.
Solana is a business success because it made the investors rich. Most proof of stake projects are built that way. Proof of stake is inherently worse than proof of work at everything except making the early investors rich. There are very applications where proof of stake is the objectively better solution, it is almost always chosen because it is better for a quick profit. Just knowing a project is proof of stake should make you wary. My experience is the projects which try to pretend they aren’t proof of stake when they are even worse. Not even one proof of work project describes itself as being anything other than proof of work. When you are the gold standard for decentralized security you don’t call yourself by any other name. It is always the proof of stake networks that come up with new names to hide what they are. Solana calls itself a proof of history network and claims to be the first of its kind. But it's just another name for proof of stake. Most proof of stake projects are suspect, projects which invent new names for proof of stake are even worse in my experience. Solana is even worse.
In a certain way, I do respect the Solana team. They had the goal of being the fastest blockchain in the world and they achieved it. Some other projects will claim higher numbers but if you look closely the numbers don’t hold up. I am convinced Solana really is the fastest blockchain. They didn’t become billionaires by accident. They had a goal and reached it. But they reached the goal by building a race car when they needed a Camry. Solana’s speed is the product of brute force design choices that will not hold up over time, the speed is not the result of a design breakthrough that other teams did not think of. Anyone can bore and stroke an engine and anyone can shorten the blocktimes. Brute force designs are easy to understand, just hard to keep running. Solana is a dead project and always was, it was a bad-faith effort to achieve speed numbers the public could easily understand and use that for a quick profit. That has happened, and now comes the crying.
Filter By:
