Why did the blob base fee jump directly from 1 wei to around 14-16 million wei within 24 hours of Ethereum Fusaka upgrade, with a peak of nearly 20 million wei and an average of about 14.7 million wei, which is about a 15 million fold increase. Why is it rising so sharply? Because the starting point of the original blob gas fee was too low, almost free, it jumped from 0.00000000 1 Gwei (1 wei, super low) to an average of around 14.7 million wei, an increase of nearly 15 million times. You can see from the following picture that Fusaka went viral immediately after its launch. Simply put, to understand blob, Ethereum mainnet L1 is the central hub of the brain, and processing transactions directly is relatively expensive. As a result, the "lucky hand brains" of L2 cheats were born, which process transactions much cheaper. But for safety reasons, these deputies need to submit the data (transaction records) to L1 after processing, and these data are stored on Blobs (large temporary storage, not occupying main network space, but temporarily loaded into memory for KZG verification, and can be trimmed after expiration). The Blob gas fee is the "data availability fee" that L2 provides to L1. Previously, when Dencun upgraded, there was no minimum price mechanism for this fee, resulting in the fee being stuck at 1wei and almost free. The result is that the network is subsidizing the blob, and nodes are working at a loss (providing free KZG verification computing power). This Fusaka upgrade has an EIP called 7918. It is to add a "reserve price" to the blob fee, which cannot be lower than 1/16 of the L1 execution fee this time. Technically, it is called BLOB_BASECOST/GAS-PER_SLOB. That is to say, before this, blob fees could be as low as 1wei, almost free; Now the blob fee is ≥ (L1 execution base fee)/16 (precisely 1/15.258). When the execution cost increases, the blob cost also increases; If it is low, there is also a minimum guarantee fee. Based on historical data, it is approximately 0.01~0.5 Gwei, which is 10 million~500 million times higher than the previous 0.000 00000 1 Gwei. Shortly after the launch of Fusaka, the actual average surged to 14.7 million wei, which is 15 million times higher. The reason for this design is that Blobs not only occupy storage, but also require computation during node verification proof (KZG proof is 15 times more expensive than single point computation). Previously, L2 paid for nothing, but now pay a 'reasonable rent' - at least covering part of the actual cost. This is a good thing for the cost market, as prices fluctuate normally and the network can use "price increases" to control blob traffic, avoiding congestion or waste. In addition, Fusaka also comes with PeerDAS (data sampling), which allows nodes to only check small samples to verify large blob, and the capacity can be multiplied several times. For ETH burning, it is also a good thing, as there is a chance to burn more ETH. Based on historical data backtesting simulations (if launched in June 2025), according to Blockworks calculations, it can burn approximately 8 times more. Bitwise analysis suggests that with the increase of L2 trading volume in 2026, blob can burn an additional 200000 to 400000 ETH within a year, contributing 30-50% of the total burn amount. Of course, this seems to be an optimistic estimate, but it depends on the growth of L2 trading volume.