2026-01-08 09:24:12

Ethereum co-founder Vitalik recently shared some interesting perspectives on what blockchain should prioritize in the pursuit of performance. He believes that rather than solely focusing on reducing latency, enhancing bandwidth and scalability are key to ensuring decentralization and security. The logic behind this is quite realistic: latency is fundamentally constrained by physical and economic factors such as the speed of light, geographic distribution, and resistance to censorship, making infinite optimization impossible.

Vitalik's stance is quite intriguing — Ethereum should exist like a global "heartbeat," rather than competing with game servers for low latency. As for applications that truly require rapid response? He believes the answer lies in L2 and off-chain solutions. In the future, there may even be city-level or highly localized L2s to meet different scenario needs, especially in the AI era where such demands will become more prominent.

ETH-3%

View Original

This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.

9 Likes

Reward
9
10
Repost
Share

Comment

0/400

quietly_staking

· 01-11 08:23

V哥, you're right again this time. The low-latency setup is basically a false demand. We still need to preserve the fundamental principle of decentralization.

View OriginalReply0

SurvivorshipBias

· 01-10 22:52

V神's recent remarks are truly eye-opening; finally, someone dares to say that delay is not a cure-all. L2 solutions should have been prioritized long ago. What's the use of just focusing on TPS? Bandwidth is the real bottleneck. The myth of low latency being exaggerated has been covered up for too long. City-level L2 sounds pretty good; it seems that AI really needs to do it this way. But just to complain, there will still be people who continue to be obsessed with that false sense of speed.

View OriginalReply0

LiquidatedThrice

· 01-10 10:30

V神's words are spot on; finally, someone has exposed the truth about low-latency internal competition. Finally, someone dares to say that latency is the ceiling, and not all chains need to compete at the millisecond level. L2 is the future; the mainnet should be stable and reliable, I agree with this approach. Ethereum acts as the heart, while L2s do the work; the architecture is really well thought out. To be honest, compared to millisecond-level transactions, I care more about fund security. V神 understands.

View OriginalReply0

SpeakWithHatOn

· 01-08 15:09

Brother Dian, you’re not wrong this time. If you can’t win by delaying, then don’t bother delaying. Focusing on bandwidth is the right way. People are inherently greedy, always wanting to have both fish and bear paws. But if you can’t even keep up with the speed, why keep fighting over low latency... This VB approach is actually about facing reality. The L2 solution should have been popularized long ago. Why are so many still fixated on the mainnet? Really. Wait, in the AI era, do we still need city-level L2? That idea sounds a bit crazy, it seems like another infrastructure wave is coming. Compared to those public chains that boast millions of TPS every day, being more practical is still more appealing.

View OriginalReply0

AirdropworkerZhang

· 01-08 09:51

Vitalik's approach is still clear-headed, not comparing with those projects that nitpick and delay. Letting L2 showcase its capabilities is the right way; anyway, I'm just waiting for the city-level solutions to come out.

View OriginalReply0

OnchainDetective

· 01-08 09:42

Wow, V God, your logic is still top-notch. Finally, someone said it out loud. A bunch of projects are still desperately fighting delays, not realizing they're just digging their own graves... The laws of physics are right there, yet they insist on racing against the speed of light? Artificial intelligence generated content, for reference.

View OriginalReply0

ResearchChadButBroke

· 01-08 09:40

V God is still the same, not following the trend of delays. I think it's quite rational. L2s are showing their unique strengths; this is the trend, but can localized L2 really take off? Bandwidth is the key, there's no point arguing about it.

View OriginalReply0

SadMoneyMeow

· 01-08 09:39

Haha, finally Satoshi has spoken. I've always thought that delaying the roll-up was pretty stupid. --- The L2 approach is indeed a way out, but the problem is that the ecosystem is still a mess. --- The analogy of global heartbeat is excellent; finally, someone dares to say that performance isn't everything. --- The idea of city-level L2 is interesting, but it seems it will take a long time to truly realize. --- It sounds good, but the key question is: can we still use it now? The latency issue is still being criticized every day. --- I agree with this logic; giving up on illusions is much better than wasting resources. --- Wait, doesn't that mean the route like Solana will cool off? --- Bandwidth and scalability have been neglected for too long; this priority should be changed.

View OriginalReply0

FunGibleTom

· 01-08 09:37

Agreed, delaying the roll-up is just self-deception. Brother V spoke very clearly this time. --- L2 is the way out; the main chain should prioritize stability above all. --- Physical limitations are right there; no matter how much you optimize, it's useless. Better to think about how to make good use of L2. --- I like the concept of city-level L2; I feel it will be very interesting in the future. --- Instead of comparing TPS with Solana, it's better to design good scaling solutions. It's a matter of vision. --- Basically, don't try to have everything; focus on making the fundamentals solid. --- Sounds reasonable, but there seem to be very few projects that actually do this. --- The AI era will indeed increase the demand for localization. Brother V is looking far ahead. --- Decentralization security > low latency; I agree with this ranking. --- Finally, someone dares to talk about the latency ceiling; there have been too many pseudo-optimizations before.

View OriginalReply0

OneBlockAtATime

· 01-08 09:28

V神's recent ideas are truly brilliant; finally, someone has spoken out. Finally understand why Ethereum doesn't need to compete with Solana on latency; the physical limits are right there. The L2 layering concept is correct; compared to hard-forking TPS, division of labor is smarter. Speed of light can't be surpassed, yet they're still competing on low latency? That's a bit funny. This guy's thinking is really thorough; infrastructure needs to be robust. City-level L2? Sounds promising; localized networks can indeed solve pain points. Finally, some clear-headed voices; don't be fooled by those flashy performance metrics. The core still has to be security and decentralization, after all, those are the roots of blockchain.

View OriginalReply0