Fanz
1.7K posts






Today we discuss how accurate validators are in selecting the head chain and how latency affects this. In Ethereum (PoS), validators must see the latest chain (head), then perform a vote (attest) to it in order to receive rewards. Of course, rewards are only obtained if the vote is correct and timely. From here, problems often arise. When latency is high, validators may be late in seeing the head, which leads to delays in voting on the chain, resulting in incorrect votes and no rewards. In another case, validators may vote correctly, but enter the slot too late, causing the vote not to be counted and rewards not to be given. In this case, latency can make you right, but still lose. The rewards you receive are calculated based on a_val (validator accuracy) and a_net (network condition). This means that it is not only the validator’s ability that determines rewards, but also the overall quality of the network. @get_optimum offers an approach to address this. By improving network quality through mump2p, which is about 150ms faster, head vote accuracy increases from 98.6% to 98.8–99.1%. It may seem small, but the impact is significant at the network scale, especially when multiplied across thousands of validators and slots. This also makes the chain more consistent and the network more stable, as there are more correct votes and fewer misses. In the end, user experience also improves. Lower latency is a win for everyone. Validators gain more optimal rewards, stakers see an increase in APR, the network becomes more stable, and the user experience becomes more reliable. Latency does not only affect speed, but determines whether validators can see and choose the correct chain in time. Even small improvements in latency can significantly increase head vote accuracy, which ultimately has a direct impact on rewards and overall network stability.









Gmum ♾ Blockchain biggest problem can be solved with "flexnodes" a key part of @get_optimum • accelerate data distribution • provide a memory layer • overcome network bottlenecks Flexnodes are not just nodes, but an upgrade to the way data moves on the blockchain.



Kalian sekarang bisa coba sendiri simulasi network dari @get_optimum Mereka ngebandingin langsung performa RLNC vs metode lama kayak Reed Solomon di kondisi real (mesh network). Dan hasilnya gak main main: → latency bisa sampai ~10x lebih rendah → bandwidth dipakai ~6x lebih efisien Artinya? data bisa nyampe lebih cepat ke seluruh network tanpa buang resource berlebihan. Ini penting banget di blockchain, karena beda beberapa milidetik aja bisa nentuin siapa yang dapet peluang lebih dulu. Coba sendiri di sini 👇 gmum.cc/simulation-mesh


Optimum is basically a super‑fast RAM for blockchains that runs on RLNC (Random Linear Network Coding). @blockchainjeff @get_optimum @MurielMedard




Web3 can compute. Web3 can store data. But it still cannot use data instantly. That is the missing layer memory Without it apps feel delayed data must be fetched instead of accessed real time interaction breaks DERAM changes that Not by adding more storage but by making data available when it is needed That is what Web3 has been missing


,@get_optimum Components : ⟶ MuMP2P | DeRAM With this combination, some of the classic blockchain problems can be overcome • transactions are processed faster • scalability increases • the network can handle more activity Bandwidth and resource usage become more efficient


It was a wonderful experience to celebrate the investiture of Andrea Goldsmith @SBUprez as president of @stonybrooku and be part of a panel with @TelleWhitney of The Anita Borg Institute, Ayanna Howard of @OhioState , and Jennifer Cochran of @Stanford .








