SpacetimeDB • Hacker News Board

Re: SpacetimeDB

#51

Post by duxup » Wed, Apr 09, 2025, 5:00 PM UTC

aeturnum wrote: post #28

SpacetimeDB fits into a genre of tech that I want to call "universe brain reorgs" - structural re-arrangements that might unlock wild performance increases. The challenge with these is that they require devs to re-align their thinking and it's hard to disentangle if "the problem" is that the technology isn't being fully embraced or if the promises of the tech don't work for your use case. Other techs in this category…

I feel like a lot of the "needs visionaries?, stuff has a pretty high chance of having not much real value.

That's not a judgment on SpacetimeDB, but it is kinda how it plays out with these kinds of things.

Re: SpacetimeDB

#52

Post by pphysch » Wed, Apr 09, 2025, 5:01 PM UTC

gafferongames wrote: post #40

cloutiertyler wrote: post #37

Earlier quoted context omitted.

We can and plan to handle the complex lag masking netcode as well. We've just focused so far on games that don't need very sophisticated lag compensation

Planning to support complex lag masking netcode as a TODO is not a good sign.

Seems too biased and harsh. The best lag masking and clientside prediction systems I've experienced as a player are those that are closely integrated with the idiosyncratic gameplay mechanics. I would never expect SpacetimeDB to have a one-size-fits-all solution to this, nor for it to be the central focus of their product.

Re: SpacetimeDB

#53

Post by gafferongames » Wed, Apr 09, 2025, 5:02 PM UTC

hesdeadjim wrote: post #41

$0.28/gb for net traffic on the Maincloud is pretty brutal.

Yeah this alone makes this a total non-starter. Do the math at 100k CCU

Re: SpacetimeDB

#54

Post by cloutiertyler » Wed, Apr 09, 2025, 5:11 PM UTC

gafferongames wrote: post #40

cloutiertyler wrote: post #37

Earlier quoted context omitted.

We can and plan to handle the complex lag masking netcode as well. We've just focused so far on games that don't need very sophisticated lag compensation

Planning to support complex lag masking netcode as a TODO is not a good sign.

TODO does not imply no plan or that we haven't thought deeply about the problem. Prioritization is imperative when you're building an MMORPG and a new database with a small team for both.

Re: SpacetimeDB

#55

Post by calebm » Wed, Apr 09, 2025, 5:13 PM UTC

Minor critique - the logo is almost exactly the same as Grok's

Re: SpacetimeDB

#56

Post by cmrdporcupine » Wed, Apr 09, 2025, 5:14 PM UTC

weego wrote: post #4

This looks really interesting, and bitcraft looks cute. I did hit a personal bias: when I saw "maincloud" as a phrase to do with pricing my heart sank because my brain went "they've wedged in some crypto shit!". Turns out no, it's just a naming thing for hosting credits.

There was a lot of crypto-adjacent talk in earlier versions of their marketing and product design. Without it actually being anything actually crypto-adjacent. Analogies drawn to ethereum etc concepts, I guess because they figured somehow there was an audience that would get those references and it would allow them to understand the product better?

I have no idea why. I was hired to work on very early versions of it, and left very early on in part because I wanted nothing to do with that angle. IMO it's a liability.

Re: SpacetimeDB

#57

Post by xixixao » Wed, Apr 09, 2025, 5:14 PM UTC

The “server in DB” is also the Convex model, but Convex is more tailored towards JS full stack apps:

https://www.convex.dev/

Re: SpacetimeDB

#58

Post by ffsm8 » Wed, Apr 09, 2025, 5:31 PM UTC

barrell wrote: post #48

theLiminator wrote: post #45

Earlier quoted context omitted.

> Node What's your reasoning behind this? Just being able to run the same code on frontend and backend?

When node first came out, it was pretty revolutionary. Being able to almost instantly start a JavaScript thread from a single file that could support realtime experiences (a la socket.io et al) without a build step felt pretty paradigm shifting

> could support realtime experiences

Node supports realtime applications? Really? Isn't it garbage collected? Or does the term mean something different in nodejs? I've only come to know the term in the context of time guarantees (i.e. if it's scheduled to run in x, it is guaranteed to run at that time) and that shouldn't apply to nodejs I think? When you schedule something there, I believe it'll run on the first free thread after the timer relapsed - which entirely depends on the load of the system

Re: SpacetimeDB

#59

Post by trollied » Wed, Apr 09, 2025, 5:32 PM UTC

gafferongames wrote: post #24

pmx wrote: post #19

Earlier quoted context omitted.

This is theirs https://bitcraftonline.com/ - not "released" yet but they have alpha testers on it apparently.

Would you seriously consider using any technology built by people who are making their first game?

You've got to start somewhere. Dogfooding makes sense.

Re: SpacetimeDB

#60

Post by anon7000 » Wed, Apr 09, 2025, 5:35 PM UTC

ffsm8 wrote: post #58

barrell wrote: post #48

Earlier quoted context omitted.

When node first came out, it was pretty revolutionary. Being able to almost instantly start a JavaScript thread from a single file that could support realtime experiences (a la socket.io et al) without a build step felt pretty paradigm shifting

> could support realtime experiences Node supports realtime applications? Really? Isn't it garbage collected? Or does the term mean something different in nodejs? I've only come to know the term in the context of time guarantees (i.e. if it's scheduled to run in x, it is guaranteed to run at that time) and that shouldn't apply to nodejs I think? When you schedule something there, I believe it'll run on the first free…

Realtime in this context probably just refers to the end user experience — multiple people connecting over web sockets to the same backend and seeing the same thing in “real time”.