LobsterDAO 🦞

20 Apr 2025 19:27

we ignore the painful truth of substandard RPC performance. It usually sucks.

LobsterDAO 🦞

20 Apr 2025 19:21

Hmm that's a fair point. I wonder how the DevEx of onchain multiplayer framework is compared to something like spacetime db.

LobsterDAO 🦞

20 Apr 2025 18:58

And as far as the 'real-time' chains are concerned, I am yet to see if their advertised blocktime matches the speed of their RPCs. What's good is a 10ms blocktime if my ping to the RPC is 200ms.

LobsterDAO 🦞

20 Apr 2025 18:12

youtube: https://youtu.be/FCXwHUK0cEw

LobsterDAO 🦞

20 Apr 2025 18:08

State channels are interesting but existing solutions have all had horrible devex in practice. I believe we'll see something along these lines that works eventually.

I'm actually a pragmatist & I think the onchain way is much easier than making state channels good for now.

Eventually we'll have to scale further & state channels could be a good solution for that. An interesting model here is state channels as permissionlessly-sequenced small-scale/ephemeral based rollups (based in the sense that all txs go to DA).

LobsterDAO 🦞

20 Apr 2025 18:04

Real-time does a lot of work here. You can't do FPS on chain for now, granted. (It's also probably the worst fit genre anyway.)

I don't know why people just want to be so cynical & bearish when the kind of stuff is already being done - could use improvements but we have proof MMO & Minecraft onchain works.

Imho, it's actually less work to get very low latency decentralized sequencing than weave a completely novel blockchain/decentralized architecture into existing game infra (but I do think that's very interesting too & would to see stuff being done there, in fact there are ppl like Playmint, topology.gg & more working on this)

Low-latency is literally already on testnets (Lattice, Rise, MegaETH, maybe Monad too?).

Centralized yes. Decentralized doesn't necessarily slow things down (e.g. round-robin sequencing). For really low-latency use cases, you ultimately run into speed of light & you need local sequencers. Which you could totally do if you're willing to delegate sequencing locks on some contracts (the "separate sequencing groups" I mentionned earlier). It's not that difficult - and it's certainly much easier than build an entirely new arch.

Why not ask questions about this instead of assuming the answer is already known?

LobsterDAO 🦞

20 Apr 2025 17:11

I took my sweet time lmao

LobsterDAO 🦞

20 Apr 2025 16:02

Not really. The point is without a centralized relayer, syncing game state across a large pool of players for any kind of real time game is hard, if not impossible. Even if centralized sequencing for rollups gets solved, it won't really bring down the blocktime to 10/20ms. So if that centralization is required anyway, why reinvent the wheel instead of trying to find ways of integrating crypto into existing gaming architecture.

LobsterDAO 🦞

20 Apr 2025 14:57

There is still value in syncing the game state with some decentralized DA layer. But that doesn't mean the whole game logic has to be implemented on a distributed VM.

LobsterDAO 🦞

20 Apr 2025 14:06

I get that often, here's the tea:

1. Ownership - seems like a meme, but if going to create on top of a game, it's nice to have a garantee it's not going to get pulled from under me, esp. when more effort is involved like a mod. There's is swimming in the opposite direction of web2 where shuttering servers is common. Web3 also provides an economic model where open-source works.

2. Blockchain as a fully transparent shared DB + execution is actually the right architecture for maximum automatic modability. Look at the MUD framework if you haven't w/ their world + table system if you haven't. Anybody can register new system & tables on a world w/ full read access to all other tables.

3. Betting is not the only form of economic value in a game.

4. "At the cost of the gaming experience" - that's why we're pushing the boundary ser. There's already a wide range of games where UX could be the same, just held up by a bit of engineering. Stuff like Slay the Spire, Balatro, Darkest Dungeons, Runescape, heck even Minecraft - when can have this today onchain w/ good UX.

LobsterDAO 🦞

20 Apr 2025 10:51

"Bro, it’s my dollars they are in my wallet no one can touch them" bet

LobsterDAO 🦞

20 Apr 2025 10:17

youre an enso competitor?

LobsterDAO 🦞

20 Apr 2025 07:13

We have already developed this in DZap. Not only Aave but you can do any DeFi protocol interaction on same or cross chain.
Currently this protocol to protocol movement is under audit and should go live soon.

LobsterDAO 🦞

20 Apr 2025 07:00

One more sad one:

https://x.com/0xkydo/status/1912522609159286934?s=46

LobsterDAO 🦞

20 Apr 2025 07:00

Even base is complicit w “scams”

LobsterDAO 🦞

20 Apr 2025 19:22

Def a lot of room to improve ofc but frameworks like MUD really shine and show a glimpse of that already

LobsterDAO 🦞

20 Apr 2025 19:10

While it’s true that the tech is not there yet, one of the things that I appreciate about FOCG and the autonomous world meta is how is lowers the barrier of entry for solo/small dev teams to build online multiplayer large scale world like game experiences with less infrastructure scaffolding and operational costs compared to traditional (online) game development

Ofc hard to compare the current state with what’s possible with AAA games but to your point of looking at it from a product angle, one can look at it less as a “killer feature” and more as enablement / democratising access

LobsterDAO 🦞

20 Apr 2025 18:57

Well, as an engineer, I find the idea of a fully onchain game and the tech being worked on to make it possible awesome. But as a product guy, I just fail to see what is the killer feature for something like that. Gaming, atleast modern AAA gaming, is one thing where distributed computing definitely makes more sense than full on decentralization across multiple nodes because as you rightly said, the issue ultimately runs into speed of light. All strategies that there are there to reduce latency for the decentralized system will have an upper bound that will always be lower than a centralized system. Hell even centralized gaming sucks so much that you can't even enjoy a good match with your friends if you are not on the same continent.

LobsterDAO 🦞

20 Apr 2025 18:11

https://www.dustproject.org
formerly biomes.aw

looks like they rebranded but it was basically minecraft 100% onchain

LobsterDAO 🦞

20 Apr 2025 18:07

Minecraft is on-chain?

LobsterDAO 🦞

20 Apr 2025 17:20

Usually it can be solved with state channels. We had a thing called Plasma back in 2017. Also Bitcoin lighting network. Onchain tx throughput is often not needed for these kind of apps, and rarely it makes sense.

LobsterDAO 🦞

20 Apr 2025 17:10

By the way, just yesterday i found what i was looking for. Apparently, i was talking about Farcastle

LobsterDAO 🦞

20 Apr 2025 15:56

That seems like a fully general criticism of rollups rather than games, and to be solved in the same way (w/ decentralized sequencing at first)?

There's some interesting models there, you could have separate sequencing groups per app & what not. We've barely scratched the surface of blockchain design imho (not specifically talking about gaming)

LobsterDAO 🦞

20 Apr 2025 14:52

So honestly all of these arguments are stuff I have thought about before. Around 2023, I was researching on implementing a game engine as a sequencer using the Op stack before unironically I pivoted to AI. It's just none of those arguments doesn't really sound convincing, when it comes to current architecture around gaming.

For example, let's take the ownership argument. Even if you 'own' the right to certain in-game content, it only derives its value from the game itself. So what good is ownership in that case? Because even if you go to the extreme end, have pure P2P game state with some kind of local first architecture (essentially CRDT), it is extremely hard to establish any kind of consensus on the merged state without a centralized component (in this case the sequencer that merges the local state delta). You are ultimately ending up on a similar centralized architecture, increasing complexity of the system, reducing the performance that users experience. To me, it seems using blockchain for the sake of a blockchain without any benefits to show for it.

I concede the economic value point, like marketplace for certain in-game stuff, is one thing where implementing the core mechanisms using blockchains make sense. But honestly, it should be so seamless that users don't even know it is using a blockchain underneath.

LobsterDAO 🦞

20 Apr 2025 10:52

Not exactly, our aim is to give the best output to our users/protocols and aggregating all DeFi protocols in a single place. So everyone is our alley in DeFi, we have integrated Enso but apart from that we do have our own path finder which finds the best path of moving liquidity from one protocol to another in a single txn.

LobsterDAO 🦞

20 Apr 2025 10:50

I mean it’s kinda like the ppl who want to unbank with USDT lol

LobsterDAO 🦞

20 Apr 2025 08:33

yeah but how big is the betting market for games to have provability at the cost of the gaming experience? For user generated content, there is no reason for it to be on chain unless you are gonna have a marketplace for it. the economics part I agree with, especially games that already implement marketplaces for their in-game items

LobsterDAO 🦞

20 Apr 2025 07:04

Actually an interesting comment:

https://x.com/charlottefang77/status/1913628449039847886

LobsterDAO 🦞

20 Apr 2025 07:00

What a shitshow, all around

LobsterDAO 🦞

20 Apr 2025 06:59

https://x.com/vannacharmer/status/1913630106414678290?s=4