This is a transcript of The Resonance from 2025 July 27.
00:00: Start recording. Okay, both are working.
00:04: Hello. Can you hear us?
00:06: >> Hello.
00:07: >> Hello.
00:09: >> Uh, let us know if you can hear us on
00:11: the stream. Yep, we got check. Uh, oh,
00:14: we got prestream blah for check the fox
00:15: out there. Hello.
00:18: Let me just post the announcements as
00:20: well. So just going to post
00:25: uh post this
00:28: post
00:31: that
00:34: and two more.
00:37: Posting this one.
00:49: There we go with announcement posted.
00:50: Hello everyone. Uh hello and welcome to
00:54: another resonance. Uh
00:58: just going to position this. We got
00:59: people in the chat. Hello. Oh my god,
01:01: it's cutie.
01:03: >> Yellow dog man.
01:05: >> Cutie.
01:08: >> We have we have yellow yellow fruxus and
01:11: we have brown fruxus.
01:13: >> Brown fruxus.
01:14: >> Yeah,
01:14: >> it's all fruxuses. We also got purple
01:16: fruxus and team and
01:20: actually I don't like fruxus now.
01:24: Anyway, hello and welcome to resonance.
01:26: Um, for those of you who don't know,
01:29: this is like a weekly stream where you
01:31: can ask pretty much like anything you
01:33: want to resonate. I'm FXUS. I'm here
01:35: with Cyro aka Brown FXUS apparently. Um,
01:39: and we'll be answering your questions uh
01:43: for the next hour or two. Uh if you want
01:46: to ask anything, make sure to put a
01:48: question mark. That way, like uh
01:49: whatever you ask, I'm actually going to
01:51: take this uh
01:54: I'm going to show you. This is you. This
01:55: is how we see you chat. So when you put
01:57: question mark, like it pops on the pin
01:59: messages. That way you know we don't
02:00: miss it. Um so make sure to do that. We
02:04: have these two. Uh I'm going to go back
02:07: to this. Um and yeah, feel free to ask
02:11: anything you want about Resonate. Uh it
02:13: can be anything with the development
02:15: engineering philosophy of the platform
02:17: like even team you know just like
02:18: personal things you want to get us know
02:20: better uh uh for some questions we might
02:22: like redirect you to some other office
02:24: hours like for example our team they
02:26: have theirs on uh when yeah it's
02:29: Wednesday
02:31: uh probably prime is like on Tuesday
02:33: there's the moderation office hours that
02:34: happened pretty much like 90 minutes
02:36: before this one uh J4 did one I don't
02:39: know if JF is doing ones regularly right
02:41: now I have to check check on the discord
02:43: there's like the events tab and you can
02:45: find the different like office hours. So
02:47: there's like plenty plenty of places
02:49: where you can like you know learn more
02:51: as well and for some of the questions we
02:53: might like you know redirect you to
02:55: those so you can get like better answer
02:57: with uh relevant themes but uh either
03:00: way feel free to ask like we'll try to
03:02: do our best to kind of answer um also
03:06: today
03:07: uh usually this goes for two hours I
03:11: depending on how things go I might need
03:12: to like uh end it earlier today because
03:16: like I'm kind feeling a bit nauseous.
03:19: So, we'll we'll we'll see like if this
03:21: kind of goes away, then probably like go
03:23: through the whole two hours, but uh just
03:26: as a heads up, um I might end up
03:30: I might end up like uh needing to end
03:33: this like earlier. I don't expect it.
03:36: So, we'll see. I'm going to try try to
03:40: last as long as I can. Uh I think that's
03:44: everything.
03:46: Oh, we're also going to be um answering
03:49: questions from Discord first. Uh so,
03:53: we're going to go through these.
03:55: Uh once we're done with those, we're
03:57: going to go through all the ones you're
03:58: asking live in Twitch. So, with that,
04:00: let's get started. Uh I'm going to make
04:03: a duplicate and we're going to post this
04:06: here, I think. I don't know if this
04:08: readable, maybe too small. Um
04:13: so Mintshock is asking in typography it
04:15: is considered good practice to optimize
04:17: optimize fonts for different weights
04:19: boldness and ital italicism I don't know
04:23: how to pronounce that how much the
04:24: character leans uh this is often done by
04:26: altering character forms or letter
04:28: spacings effort to optimize eligibility
04:31: uh often these optimizations are baked
04:32: into multiple separate font files or be
04:35: included in one variable font file uh
04:38: this is mainly need which is tricking Uh
04:41: this is mainly needed to preserve
04:42: readability on all situations. Some
04:44: fonts especially ones with serifs. Uh
04:47: the hooks at the start and of strokes in
04:49: some phones tend to be hard to read with
04:51: folks italics program effect by slanting
04:53: a letter. Uh one does support variable
04:56: fonts. Two does by default use poppins
04:59: bold popins italic pop balic when using
05:02: the bold and i tags. If not is there way
05:05: to for to specify separate font files
05:07: for the uses. So right now we don't
05:10: support those. Um that can be extended
05:13: but our system doesn't do that yet. We
05:15: use uh we essentially use um
05:20: like whenever you do bold the way that
05:22: kind of works because we use like sign
05:24: distance fields it actually adjusts the
05:26: parameters for the bolding and the slant
05:28: essentially just slants the quad which
05:30: is kind of simpler to do. Uh but you
05:33: know like like I said like it's not it
05:35: can like affect readability. So if
05:39: that's something like you know like you
05:40: feel is like a issue like I would
05:42: recommend making a GitHub issue or check
05:43: check first if there's like one fority
05:45: and make sure people give it up forward
05:47: because uh that will make it easier for
05:50: us to like you know give it this more a
05:52: priority and extend the font system. Uh
05:55: but right now it will not use that. Uh
05:59: for the variable fonts I have to check
06:00: like with the version of free type that
06:02: we use. It should be like supported
06:04: internally but um I might dig around
06:07: into that.
06:11: Uh next question
06:14: is from Phoenix. Uh
06:17: it's close. Uh Phoenix is asking
06:23: uh do you think uh do you think when
06:25: audio DSP happens it'll be possible to
06:28: make singing based games like shutter
06:30: the wine glass with your voice by
06:32: harmonizing with it natural resonance
06:33: frequency for example I can in between
06:36: getting data from the SP protolex could
06:38: be complicated no it shouldn't be
06:41: complicated that's like the whole point
06:43: is with it being protolex is you can
06:45: have different systems kind of easily
06:47: interact with each
06:48: So, we'll probably have like, you know,
06:50: modules that kind of give you um ability
06:53: to, you know, get like FFT and you could
06:56: like uh do pretty much whatever you want
06:58: with that. Um
07:00: so yeah, like I don't really see that
07:03: being like an issue. Uh the idea of that
07:05: like kind of system is to make like
07:06: interoperability with everything in
07:08: world like very easy. So you can have
07:09: like you can have things in the world
07:11: affect the DSP. Uh so like you know for
07:15: example the environment could change you
07:17: know how the sounds been generated. You
07:19: could like make like an instrument and
07:21: you can you know make knobs that you
07:22: turn around or maybe it's based on the
07:24: hand or what whatever whatever you're
07:26: doing even what whatever environment
07:28: like you want to feed in into the SP
07:30: you'll be able to do that. And similarly
07:32: for any stuff that like results in
07:34: values you'll be able to feed them out
07:35: and you know have it react to it. Uh
07:38: that way people can you can build cool
07:40: like you know audio visual kind of
07:42: experiences. So, um, Titan being an
07:46: issue.
07:49: Next question is, uh, from Shining Hero.
07:54: Uh, Shining Hero is asking,
07:58: uh, you mentioned at one point having
07:59: some cool projects you wanted to work
08:01: on, but couldn't because the splitting
08:02: was taking all your time, energy.
08:04: There's a good chance I won't be able to
08:06: complain half of it, but I'd like to
08:07: hear about it. Anyways, yeah, I have a
08:10: few projects like um that I really
08:12: wanted to do. The one that's on my mind
08:15: a lot is actually kind of related to the
08:16: spliting is being able to use resonate
08:18: as an overlay. So, you could actually
08:21: instead of like, you know, running
08:22: entire world, you'll be able to have
08:23: your dash over any other VR application.
08:26: And then you could also create like sort
08:28: of this augmented reality layer on top
08:31: of other games. So you could like you
08:34: know you could be for example hang out
08:35: in VR chat and you could be like you
08:37: know working on your avatar or working
08:39: on some proto flags or maybe just you
08:41: know want to bring some videos to watch
08:43: uh with the ability for anybody else in
08:45: the same world who has the overlay to
08:47: join that overlay. So you you'll still
08:50: have like you know fully synchronized
08:51: experience. So, it's like, you know,
08:53: full synchronized augmented reality
08:54: experience on top of other games or, you
08:58: know, another thing is like, you know, I
08:59: like to play Minecraft in VR a lot and
09:01: it be really cool just be able like, you
09:03: know, because often times in Minecraft
09:05: like, you know, build like these cool
09:06: places and so on, but like I don't
09:08: really use them to like hang out in them
09:11: because like Minecraft is not really
09:12: like a hangout game. Uh, so it would be
09:15: cool, you know, with like overlay like
09:17: that, especially if there's like
09:18: integration. So, the overlay is actually
09:19: positionally synced. you could actually
09:21: hang out with people like you know we
09:22: built something called Minecraft and
09:24: they just want to hang out for a bit in
09:25: there bring some videos to watch you
09:27: know like do like do some other stuff
09:30: which is it more like as an environment
09:31: and I think that could be like a really
09:33: really cool application that like just
09:35: opens up like this whole new world of
09:38: things you'd be able to do with Resonite
09:41: that like you know nothing else is you
09:45: know providing right now. So that's the
09:47: one that like on my mind like very like
09:49: a lot lately that um I would want to
09:51: like to work on.
09:55: Uh let's see next one.
09:59: Uh next one is
10:02: from goldun gold is asking do you think
10:06: will be able to have geometry nails like
10:07: blender at some point? Um so one more
10:11: thing we have planned is the mesh DSP.
10:14: So that's going to be able to like you
10:15: know generate and process geometry. It
10:18: might not be like you know fully like
10:19: one to one to how they are in Blender.
10:22: It's going to be like its own system but
10:24: like you'll be able to do a lot of
10:25: similar things with it. So pretty much
10:29: yes I would say asterisk
10:32: uh the mesh DSP like the idea is like
10:34: you know you'll be uh you'll be able to
10:36: like you know generate meshes completely
10:37: procedurally. Um so like you know you
10:40: can just generate something from scratch
10:42: you can animate it or you can feed
10:44: existing meshes into it or once we also
10:46: have vertex level editing you could
10:49: actually use it to make a pipeline. So
10:50: you have like you know your base vertex
10:53: mesh and you feed it through like a mesh
10:55: DSP pipeline that does various like
10:57: processing on it and you see in real
10:58: time how it's like kind of changing. So
11:00: for example, we can have a node that's
11:01: denote as does like subsurface like
11:03: subdivision maybe another like
11:05: topologizes another like applies like
11:07: displacement uh stuff like that. You can
11:10: build whatever pipeline you want and
11:11: then you can be doing vertx editing and
11:13: see and seeing it update in real time or
11:15: near real time. Uh because it's one of
11:18: the things you know you do in Blender is
11:19: like you have like your base geometry,
11:21: you apply a bunch of modifiers and then
11:23: you model that way. Um the only thing
11:26: that bugs me with blender is like you
11:28: know it's synchronous like with like
11:30: very heavy modifiers the whole UI is
11:32: kind of becomes laggy at least last time
11:34: I tried it. Uh with the Resonite the
11:37: systems are designed to be asynchronous.
11:38: So like you can still be modifying the
11:40: mesh you know at a higher rate and maybe
11:43: like the the detailed visual will be
11:46: lagging behind a bit but it's not going
11:47: to affect you know your overall frame
11:48: rate and I think that could be pretty
11:50: cool.
11:52: Uh
11:55: the next one
11:58: is uh Corey Kanes is asking, "I'm
12:03: curious if you ever compared libc with
12:05: mpv and other such libraries based on
12:07: fmp. What's the reason you choose libc?"
12:10: So the reason I chose libc is because it
12:13: existed.
12:15: Um the main thing like I think people
12:17: miss about this is like you can't just
12:19: use the library you know on its own. It
12:22: actually needs to be integrated with
12:23: Unity because we want to use it in
12:25: Unity. Uh because the library needs to
12:28: like you know synchronize with Unity's
12:30: rendering pipeline to update the
12:32: texture. Uh and that requires you know
12:36: like higher level of integration. Um,
12:39: and the problem with that is, you know,
12:41: doing that kind of integration, that's
12:42: also quite a lot of work. And as far as
12:46: I'm aware, there's nothing for MPV. So,
12:49: if you wanted to like, you know, use
12:50: MPV, we would have to implement one and
12:53: it might take, you know, at least months
12:55: of work potentially. Um, like VC like,
12:59: you know, there's like a third party one
13:00: that like we've been using, uh, that's
13:02: unfortunately not even developed
13:03: anymore, which is, you know, means we're
13:05: a little bit stuck. the VC actually has
13:07: their official Unity integration that
13:09: people are working on, but it's like
13:10: progressing like super slow and still
13:12: like missing some things. Um, and it's
13:16: been like, you know, in development for
13:18: multiple years at this point. So,
13:22: yeah, it's pretty much like, you know,
13:23: one of those cases is like
13:26: we can't compare, you know, it against
13:28: something that doesn't exist. Lib was
13:31: the one of the few things that existed
13:33: as a choice for this. So our options
13:36: were kind of like you know constrained.
13:45: And next question is
13:49: next question is from Kney. Uh
13:53: K is asking is switching to a new render
13:56: part of the splittening? How much impact
13:57: would it have on performance? Uh so no
14:00: uh that is not part of the splitening.
14:02: The splittening is just separating the
14:04: fruxyagen and unit into two separate
14:08: processes. It's essentially like
14:09: splitting them apart. Um we do want to
14:12: replace the render at some point too,
14:14: but that's a separate project for the
14:16: future. Uh one that we kind of calling
14:18: start calling the snipping because we'll
14:20: snip Unity away.
14:23: I don't I don't think the name will
14:25: stick. But uh yes, this is a per
14:27: project. Um I think it could like impart
14:30: performance quite a bit too because uh
14:32: one of the things we want for the new
14:34: renderer. Uh there's like a few things.
14:36: Um right now with Unity we are using the
14:39: the first pipeline which like allows you
14:41: to have lots of light but there's a fair
14:43: amount of overhead. Uh it uses a lot of
14:45: like memory bandwidth. Um and we want to
14:48: switch to a more modern pipeline like a
14:51: clustered forward. Uh that's going to
14:53: provide a lot of benefits because it can
14:55: render lots of lights even more than
14:57: deferred and it can render them pretty
14:58: much for all materials because if it's
15:01: deferred uh it can only support opaque
15:05: transparent like non-transparent objects
15:08: anything that's uh
15:11: I'm saying it wrong. Uh it only supports
15:14: opaque BBS like materials. If it's
15:17: anything other than that, it needs to it
15:19: falls back to forward and then has like
15:21: limitations where the amount of light
15:23: actually increases, you know, the GPU
15:25: load because it needs to rerender the
15:27: mesh multiple times for every single
15:29: light that affects it. Um,
15:32: so using something like clustered
15:35: forward that on itself should help
15:38: graphical performance quite a lot. Um we
15:40: also want the render to use uh Vulcan
15:43: which is much more like you know faster
15:45: kind of API much more kind of lower
15:46: level that on itself can provide like
15:48: you know another performance boost and
15:51: we want to use modern technologies for
15:52: steroscopic rendering where we cannot
15:55: use you know uh certain newer methods
15:57: with Unity because they're not supported
15:59: like with the combination of pipeline we
16:01: use uh so that itself can also like you
16:04: know help with performance so overall um
16:09: Yes, like I do expect it to actually
16:10: bring uh performance improvements on its
16:12: own, but it is a it is a later project
16:15: for a later date.
16:17: >> And also we can use much better
16:19: anti-aliasing techniques like NSA.
16:22: >> Oh yes. Yeah. There's another huge
16:24: benefit because MSA doesn't work with um
16:27: it doesn't work with uh like deferred
16:31: um but with cluster forward that kind of
16:33: you know brings it back to like where
16:34: you can use it. So you kind of get
16:35: benefits of like, you know, border like
16:37: forward and having lots of lights.
16:39: >> Yeah. That temporal anti-alias thing
16:42: crap.
16:42: >> Yeah.
16:44: And we'll be able to do custom shaders
16:46: because we have control over those,
16:47: which is another big thing that one
16:49: really looking forward to because it'll
16:50: be so much fun to mess around. You can
16:52: just, you know, do stuff with proto
16:53: flags and just have things like change
16:55: and do visuals and real time. It's going
16:58: to be so much fun. Uh, next question is
17:01: from Kaobi. Uh how familiar are you with
17:04: ECS? Uh it keeps writes
17:07: due to some ECS style patterns could be
17:09: useful for the data model rework. Yeah,
17:11: I'm quite like familiar for those who
17:13: don't know it stands for like entity
17:15: component system and it's essentially
17:16: like an approach of like inoscripting
17:18: that like can uh be very high
17:20: performance. Um there's actually some
17:23: parts of resite that kind of use that
17:25: pattern. Um actually the splitening it
17:28: kind of uses you know it uses some
17:30: similar patterns where the data is like
17:33: sent to Unity in like you know kind of
17:36: flat like buffers where um you know the
17:40: buffer is just the chunk of memory
17:41: that's like you know specific class
17:42: track and it just kind of like updates
17:44: those and that's like very efficient. Um
17:46: the trick is with Resonite with the data
17:48: model is the data model is also its own
17:50: system that's like quite unique and it
17:54: has like you know certain features and
17:56: constraints on its own. So we cannot
17:58: like you know just use straight ECS. Uh
18:02: it also doesn't work for everything like
18:04: ECS is great if you have lots of things
18:07: that are all kind of behaving the same
18:09: because you can like you know you can
18:10: make like fill out like big like buffer
18:12: with them and you can just kind of for
18:14: loop over all of that. um have it like
18:16: you know accelerate, have it vectorize,
18:18: you know like be like super fast. Uh but
18:21: like you know you need to have a lot of
18:22: the same thing and there's lots of
18:23: behaviors that like that might like
18:25: depend there might be like a lot of of
18:27: conditional behaviors or you know
18:31: um have like just like more kind of
18:34: heterogeneous kind of composition. So
18:37: what we'll probably end up like having
18:39: is more hybrid system where like some
18:42: components like you know the register
18:44: themselves with some sort of like ECS
18:46: system uh and you know handle their
18:48: updates like that way but the idea of
18:51: like behind the data modeling work is
18:52: actually to use like more you know uh
18:56: more kind of ACS like patterns to
18:58: improve performance. So that's actually
19:00: you know why we want to do the data
19:02: model rework in the first place uh
19:04: because we want to move it like to those
19:06: kinds of patterns and that's one of the
19:07: reasons why it will help like you know
19:10: quite performance because the data will
19:11: be more tightly packed. It will help
19:13: reduce overhead. It will help reduce you
19:15: know cache misses. Um so yeah that's
19:18: that's one of the big parts of the data
19:20: model rework. There's also like parts
19:22: where it needs the update system rework
19:23: for some parts of it. it kind of like
19:26: interacts with each other because with
19:27: Resonate you cannot have the update
19:29: system without the data model. They're
19:31: like very very tightly coupled. Um
19:34: unlike I know with other other game
19:36: engines which makes it like a little bit
19:39: harder to do like one without you know
19:41: the other but yeah generally you can
19:43: expect to see more and more kind of like
19:44: of the less like patterns in the future.
19:49: Uh next question is from missing.
19:53: Uh let's see. There we go. Missing is
19:57: asking uh bit concern over the whole
19:59: payment processor not safe for work take
20:01: down thing. How likely it is that will
20:03: eventually affect resonate since HIOS
20:05: corporate was apparently Stripe the same
20:07: company you guys use? There's two
20:10: questions. Um so one like we're not like
20:13: particularly concerned right now because
20:14: like we we don't sell content. Uh
20:18: because that's one of the things you
20:19: know each IO sells content. Um and
20:23: that's kind of like you know the issue
20:24: like certain types of content like the
20:26: payment processors have issue with we're
20:28: not selling content on our platform. We
20:30: are selling essentially storage uh and
20:34: other kind of like services. So our
20:36: business model is like you know more
20:38: similar to that of like you know Dropbox
20:40: or like one drive or Google drive. So I
20:44: think that kind of makes it like better
20:45: kind of our situation. Um, of course
20:47: we're kind of like, you know, keeping an
20:49: eye on it, but uh right now like I thank
20:54: you very much any sensei for the
20:56: subscription.
20:57: Um, but right now like you know
21:01: I don't think it's going to like affect
21:03: us. Uh, but we'll see. Uh, this like
21:06: kind of concerning you know stuff like
21:07: that is happening. Um,
21:10: but for now end like you know we just
21:12: kind of have to like keep our ear to the
21:14: ground.
21:15: Uh the second question is uh are you
21:18: guys planning on hiring more moderators
21:19: if player base expands? I keep thinking
21:22: about how we'll probably have a big
21:23: influx of trolls and malicious actors
21:24: once the platform becomes something the
21:26: average real user actually knows about.
21:28: It's very easy to make other people time
21:30: hell in public sessions after all. Um so
21:34: the that's actually a good question you
21:35: know for the moderation team and
21:36: actually got asked a few times. So I
21:38: would strongly recommend asking them
21:40: there. Generally yes the moderator team
21:43: moderation team like they have plans to
21:46: kind of onboard like more kind of
21:47: moderators if it's like necessary
21:49: depends you know how fast things are
21:51: going to go. However, there's also
21:53: another aspect that I feel is important
21:55: to mention is like we have self
21:58: moderation tools. Uh and the self
22:00: moderation tools actually give people a
22:02: lot more control because uh often times
22:04: we notice like when there's like you
22:05: know somebody going around trolling
22:07: people, people just b them out of their
22:09: sessions and they're not able to join
22:11: anymore and those people tend to give up
22:14: like pressure quick. So I think just
22:16: having access to those tools will help
22:19: the situation a lot because everybody
22:21: can you know moderate their own session
22:22: and has like you know much more control
22:24: than you might have on in other places.
22:27: Um but either way um we'll you know it's
22:31: one of those things like if uh
22:34: like when whenever like you know that
22:36: kind of growth kind of happens like uh
22:38: we'll we'll do what we can to kind of
22:39: adapt to it. If it's like super fast,
22:42: you know, there might be a period when
22:43: it's going to be kind of chaotic and
22:45: then unfortunately cannot be avoided.
22:47: But uh if it's growing like more
22:49: organically, that makes it easier to
22:50: manage. But do ask this question on the
22:54: moderation, you know, office hours.
22:56: You'll have to unfortunately wait for a
22:57: week because it's like 19 90 19 minutes
23:00: before this one. So you probably kind of
23:01: missed it for this week. Uh but they
23:05: might give you like you know a lot
23:06: better answer on this topic.
23:09: Uh, next question
23:13: is from uh, Palta. Uh, how will there is
23:17: an idea with the new implemented in the
23:18: highly controversial UK online safety
23:20: act? Has there been any thoughts on it?
23:22: Will UK be banned because of the risk it
23:24: carries for you guys personal criminal
23:26: charges and millions of fines quite
23:28: literally? So, we don't really have like
23:30: much to announce like on this front yet
23:32: like we are evaluating this. Um, but
23:35: yeah, not don't have any details for you
23:38: right now.
23:41: Uh, next question is from Mintshock.
23:46: Uh, since it's bigger, I'm going to put
23:47: it here. There we go. Uh, Mrick is
23:50: asking, uh, in addition to question
23:52: above, uh, does stripe really only
23:55: support PayPal and credit cards only get
23:57: to do those two as an option when out?
24:00: This is turn me off of subscribing
24:02: because I don't like to use both. Were
24:04: it possible to implement direct bank
24:06: transfers like SEA? I believe strawberry
24:08: supports sepa but it does not seem to be
24:10: enabled for resonite. Why is that? Uh
24:14: thanks to single euro payments area sepa
24:16: customers can make cashless euro
24:18: payments via credit transfer and direct
24:20: debit to anywhere in the European Union
24:22: as well as number of non-EU countries in
24:24: fast safe and efficient way just like
24:26: national payments. So it does support a
24:28: lot more. Uh the only thing is enabling
24:30: those like there's like you know there
24:32: can be various kind of complications
24:34: like there's like you know paperwork
24:35: there's like you know stuff you need to
24:37: register and do and so on. So that kind
24:40: of complicates those matters but um it
24:43: sounds like separate would be probably
24:44: good option to have what I would
24:46: recommend doing uh make a GitHub issue
24:48: about it you know that you have interest
24:50: in this payment option uh and we'll look
24:52: into it. Um
24:55: so yep pretty much like make it make it
24:58: work for sure please. Um I know like we
25:00: internally we had kind of some
25:02: conversation what we should enable and
25:03: so on and we kind of like going slow
25:05: because like we just went and enabled
25:07: everything now suddenly we have to like
25:08: you know lots of paperwork and taxes and
25:11: other complicated stuff to deal with. So
25:13: we're kind of like you know more
25:14: managing it based on what people what
25:16: payment methods are people interested
25:18: in.
25:22: Uh, and the last question from Discord,
25:26: um,
25:28: last question from Discord is from
25:30: Sheen. Uh, on Blue Sky, you I saw you
25:34: dislike FBX format. Are there benefits
25:36: to importing an avatar in GB format or
25:39: instead of FPX? I am unfamiliar with the
25:42: differences between formats and curious
25:43: is there better choice? Um, yes,
25:46: definitely if you can use Gob. Uh with
25:49: FBX it's always like you know it can be
25:51: a bit of like like you know like a
25:54: roulette whether like it's going to work
25:56: or not. It's a roulette whether the
25:57: materials and textures will import or
25:59: not. Uh problem with FBX is like it's
26:03: kind of like it's a million different
26:04: formats in one and it's not officially
26:09: documented. So a lot of tools will also
26:11: just you know their implementation is
26:13: based on reverse engineering and every
26:15: tool implements it diff a bit
26:17: differently. So you get all these like
26:19: weird like compatibility issues between
26:21: different tools. Like if you export it
26:23: like with this tool with this setting,
26:25: it's going to work in this tool, but if
26:26: you export it with this setting is going
26:28: to explode and your textures are not
26:29: going to import. So like like using GB,
26:33: which is like an open, you know,
26:35: standardized format that generally
26:38: should give you much better results and
26:40: give you much better reliability. So I
26:43: strongly recommend like using it
26:47: with that. Uh the questions uh from the
26:50: discord are now true. So we can get to
26:52: the questions from the
26:55: uh from the Twitch chat. So there's
26:58: quite a few piled in. Uh let's see.
27:05: So first we actually have and also for
27:07: those of you like you know we're just
27:08: training if you want to ask question
27:10: make sure to put a question mark that
27:12: way it pops in our pin messages. Uh so
27:14: we got a question from as twitch 17 and
27:17: it's the classical schnoit
27:19: u which we've kind of like taken to like
27:22: you know using it for like things that
27:24: are like you know kind of frustrating to
27:26: deal with or like you know that like uh
27:29: little like you know pains of like game
27:31: developer life. Uh the schnopet actually
27:34: comes from you know Tom Scott's like
27:36: video where um it actually came to mean
27:41: like you know mean through the memes on
27:43: the internet it came to mean like you
27:44: know that it's something
27:47: where you spontaneously disintegrate
27:49: when you think about it. Uh and I
27:51: actually I have a I have a list I'm kind
27:55: of compiling them. Um
27:59: I since I'm not like feeling super I'm
28:01: not going to go into like the bigger
28:02: ones but um there's one that's not it's
28:06: not even like resonite related. Um and
28:10: it's just kind of like you know like
28:11: when I learned it like you know it's
28:13: just got some psychic damage. So
28:16: apparently the start menu in Windows 11
28:22: it's a web app. It uses a react which
28:26: explains why it's so slow. And like when
28:29: I learned about that I was like
28:33: why why would they do that? They
28:35: literally have like you know their UI
28:36: frameworks that are like fast and they
28:38: made the start menu a web app which it
28:42: explains a lot because to me like it's
28:44: been frustrating because the start menu
28:46: can be super super slow and
28:48: unresponsive. Like I would literally
28:50: sometimes open it takes like few seconds
28:51: to even pop up. I take few seconds, I
28:54: type something, it takes few seconds to
28:55: even respond to it. I'm like, why is it
28:57: so slow? That's like the core feature of
28:59: the operating system. And they made it a
29:01: web app. Um,
29:04: and it just kind of broke my brain and
29:06: like a lot of this is like, you know,
29:09: frustrating because like um
29:13: because like like for me like I like
29:15: I've been using like Windows for a while
29:17: and like it has like lots of flaws but
29:20: like I always like thought like you know
29:21: at least like the core stuff it
29:22: generally works
29:24: and like is an okay system despite a lot
29:27: of the flaws but now like over past
29:30: especially past few years has been kind
29:32: of feeling like you know all the people
29:33: who knew how to make like a good
29:35: operating system or maybe like you know
29:36: decent operating system
29:39: um like they're kind of gone and it's
29:41: just like people who don't know much
29:43: about O development they know like other
29:45: areas and they're kind of you know using
29:47: that in area where it doesn't really
29:50: apply well and like the system is like
29:51: suffering for a lot of that and a lot of
29:53: the core a lot of the core functions of
29:55: the system are now just becoming more
29:57: and more underrealiable and it's getting
29:59: harder and harder to kind of just daily
30:02: drive it. Um,
30:04: and on top of that, you know, there's
30:06: like the stuff like where like they keep
30:08: shoving AI stuff into it and feels like,
30:11: you know, all the AI stuff is coming at
30:13: the cost
30:14: of like, you know, the core
30:16: functionality like, you know, like my
30:17: like audio will not work. The basic UI
30:20: for like volume will not want to pop up
30:22: sometime and it still be like restarted.
30:24: The the file explorer will just kind of
30:26: like the path will freeze and not stop
30:28: like updating, you know. It's just like
30:30: lots of little things that just keep
30:32: breaking that like shouldn't be breaking
30:34: that should have like basic level of
30:35: quality control especially for you know
30:38: company the size of Microsoft
30:41: um and it's just like becoming like it's
30:44: degrading experience and like this and
30:46: this you know the whole thing that like
30:48: they even thought of using React for a
30:53: core feature like that that kind of
30:55: explains a lot and like it like puts me
30:57: like in a situation where I'm I really
30:59: don't want to be using this anymore. I
31:01: want to switch to Linux. There's still a
31:03: few things blocking me. So, I'm kind of
31:05: suffering. Cy managed to escape.
31:09: >> Yeah, escape.
31:11: >> And I hope I will be able to too at some
31:13: point. But yeah, that that's my
31:15: schnopit. And I hope you enjoy the
31:18: psychic damage I've given you with this
31:20: knowledge.
31:23: >> Oh boy.
31:26: you know, so you're you're you're you're
31:28: doing, you know, that's why I'm like
31:30: like when you're working like on the
31:31: stuff like to make stuff work on Linux,
31:32: I'm like, "Thank you. You're paving the
31:34: way for, you know, for other people."
31:38: >> I'm trying. Oh man. Oh man. You should
31:41: tell them about how when you open your
31:43: calendar, your entire computer explodes.
31:45: >> Oh my god. Yeah. And I'm I'm not even
31:46: the only person who has this issue. Uh
31:49: like I've talked with some people who
31:50: literally like encounter the same bug.
31:52: Like
31:54: every now and then if I click in Windows
31:57: I click on the calendar to open the
32:00: calendar because I just want to check a
32:01: date on something or something you know
32:04: um just in the you know taskbar you
32:06: click like on the time and date my PC
32:08: will just start dying and I check and
32:10: like the CPU usage 100% memory usage
32:13: 100% just everything go 100%
32:15: everything's dying for like you know 10
32:18: minutes and then it like you know
32:20: decides to like kill whatever is causing
32:21: head and you know goes away and it's not
32:23: even like visible process that's like
32:25: eating all the memory and it's eating
32:27: all the CPU
32:29: it just does it like and it's been doing
32:31: it like you know now and then and it's
32:34: just like why it's just like
32:39: I don't know it feels like like
32:42: company of the size of like Microsoft
32:44: like they should have better like you
32:46: know quality standards like and it feels
32:48: like they've been kind of dropping and
32:52: I I don't like
32:56: >> Yeah,
32:58: it's not it's not great. I am
33:01: >> Nope.
33:01: >> I Every day I I look I I look back from
33:06: to where I came from
33:08: and I am all the more glad that I
33:10: managed to escape.
33:12: >> Yeah,
33:14: I hope to do the same someday.
33:17: Like at this point it's just kind of
33:18: like it's it's past the point.
33:22: >> I believe in you.
33:24: >> All right. Next one we have tons. I
33:28: actually have a list of those two.
33:31: Um
33:34: uh let's see which one should I do
33:36: today.
33:38: I'm going to save this one for so uh
33:41: typongs. That's also another thing that
33:43: we kind of started doing. It's just
33:44: schnopid in reverse and essentially
33:46: means something you know that that makes
33:47: us reintegrate you know like schnopid
33:50: makes us you know disintegrate like we
33:52: just turn into vapor and typonics like
33:54: we just we just spontaneously like
33:56: reintegrate. So one of those um one of
34:00: the things that like makes me integrate
34:02: are people in our community who you know
34:06: would do see like you know because like
34:08: night has you know lots of flaws a lot
34:10: of things are not complete a lot of
34:11: things like you know are buggy there's
34:12: issues and like you know we get it but
34:16: there's people like you know who see the
34:18: positives you know who see where the
34:19: platform is heading who see what's
34:20: already there and like you know have fun
34:22: with it and you know generally like
34:25: their approach is like you know like we
34:28: have to like we have to we enjoy like
34:29: what is there and there's a lot of
34:31: really cool stuff and a lot of like fun
34:32: stuff and we you know with the flaws
34:35: like we acknowledge them but like you
34:37: know not just let them bring us down um
34:40: you know it's like something to be
34:41: improved in the future um and you have
34:44: fun like you know with what you have and
34:46: you have positivity towards you know
34:48: towards the future and that to me that
34:51: like you know helps like kind of that
34:53: makes me feel like a lot better that
34:54: gives me like energy like you know Um
34:57: like for example you know like uh the
35:00: person who made like the who made like
35:02: the Twitch we're using Ch the Fox author
35:03: he's he's one of those kind of people
35:05: like who like you know always like look
35:08: look at the positives and the positive
35:10: future of the platform and generally
35:12: it's like
35:15: I don't know from developer side like
35:16: this kind of like much better because
35:17: like it it feels like you know it's more
35:19: like we're
35:21: trying to like do do good and like and
35:24: the stuff like we do matters. as it does
35:26: like good uh and it makes me it makes me
35:30: more kind of like you know it makes me
35:32: more like
35:34: happy like you know and gives more
35:37: motivation to actually work and
35:38: improving things and making them better
35:39: for everyone because I know like people
35:42: are enjoying like you know people are
35:44: actually enjoying the platform
35:46: versus like you know there's like people
35:48: who always kind of see everything you
35:50: know as a downside everything's bad and
35:52: if this is fixed there's still tons of
35:54: bad things and and with that like that
35:57: is very draining like you know it feels
35:59: like no matter the amount of work we put
36:00: into things like you know it's still
36:02: going to be bad and it's like it's
36:04: really really hard to work that way so
36:06: like people who do
36:08: see things positively and who enjoy the
36:10: platform you know for what it is that's
36:13: that's that's one of my typonics
36:16: I don't know how to pronounce tonics uh
36:18: but it's one of those things you know
36:19: it's going to make me reintegrate so for
36:21: all of you like thank you for for that
36:24: and thank you Or whenever you say you
36:26: enjoy the platform, like thank you.
36:29: >> It helps.
36:33: >> That makes it all feel all worthwhile.
36:35: >> Yeah.
36:39: >> Okay, let's see. What do we have next?
36:45: Is my
36:47: What happened? My menus are broken.
36:51: >> Oh, you might need to Hang on. Let me
36:54: let me see if I can fix your hands.
36:56: >> You're doing live surgery.
36:58: >> This This happens sometimes for some
37:00: reason where like the context menus will
37:02: suddenly start disabling themselves.
37:04: >> Oh, they probably throw an exception.
37:07: >> Yeah, let me just like open this here.
37:11: So,
37:12: >> good thing I have like a help.
37:15: >> So, the context menu is actually
37:18: >> It's on.
37:19: >> It's on.
37:19: >> The canvas is off. No, the canvas is
37:21: off.
37:22: >> Oh, what? There we go. Thanks. Oh, no.
37:25: There's more stuff that's off. What?
37:27: What? No, it's still not working.
37:30: What? Uh, there's something not
37:32: updating.
37:35: What the heck?
37:37: Something else is disabled.
37:41: Okay. Context menu visual. The canvas is
37:45: on
37:47: the menu.
37:51: >> Arc layout. It's on.
37:54: Something on the context menu itself.
37:59: That is on. I just need to respawn.
38:02: Yeah, I just respawned.
38:04: Uh,
38:06: okay. Uh, this might this might uh mess
38:10: up the stream camera for a bit, so bear
38:12: with me.
38:18: Stand by.
38:31: Oh, looks like the camera survived.
38:33: >> Oh, yeah. Oh, yeah. It's because I I
38:35: think I made it. Uh,
38:38: okay. Let's see. Where was I positioned?
38:42: There we go. This is good.
38:45: Yay. I'm back.
38:48: Okay, now we can go
38:53: >> the other colors of fruxius.
38:55: >> So, Y tools asking, uh, can we get other
38:59: colors of FXUS, too? I've only ever seen
39:00: yellow and brown fruxious. We need we
39:03: need we need more people to have like
39:05: rainbow fiosises.
39:07: Uh, J4 actually did like uh office hours
39:09: like recently, too. Uh, I don't know if
39:11: she's like planning another, but
39:13: probably sometime. Uh, check the discord
39:15: for like the office hours.
39:17: in the events tab there's like more
39:18: information.
39:22: Uh Zank is asking uh are currently
39:25: streaming from the split
39:28: spliting fight version. Uh we're not VR
39:30: doesn't work yet. Um
39:34: it's uh it's coming but there's a lot of
39:37: stuff to kind of proxy. So, it's kind of
39:40: funny because normally I'm like, you
39:42: know, normally I'm like we um I'm like
39:48: controls need to have more buttons and
39:49: I'm like we need just one controller.
39:51: There's just one button because I have
39:53: to like literally rewrite every single
39:54: button and control and make sure it's
39:56: proxied. I've been going through all of
39:58: that and then like I was doing hands
40:00: just before like you know fingers
40:01: transmitting those. It's getting close.
40:04: uh still to kind of wrap up like some
40:06: things but I might might maybe have
40:08: working VR tomorrow maybe. No promises.
40:11: Um but it's kind of funny because like I
40:14: have like Steam VR driver on Unity side
40:16: and when I started working on it I kind
40:19: of commented everything and like
40:20: everything was red. All the thing was
40:22: like you know this code is syntax errors
40:24: and I've been reworking it and now
40:25: there's like only few red bits remaining
40:27: before it can compile again. Uh and then
40:30: I'll have to like test it. So um we'll
40:34: soonish
40:36: but also like if even with that like the
40:38: we couldn't be streaming because like
40:40: one of the things I need to implement is
40:41: blit camera which is used to like you
40:43: know display this the image from the
40:46: camera so that's not working yet. Um and
40:49: it's like you know something I plan to
40:50: implement after like you know we're
40:53: implementing VR because that's usually
40:54: use it with VR. So um
40:59: order of operations
41:02: Wait, where did that thing go? Did I not
41:03: put it in? Uh, next question.
41:09: Next question is from as 17. Uh, how's
41:12: your go Sunday for soon to be Monday?
41:15: Sorry, Monday for me. Um, it's okay. I'm
41:19: just kind of bit I don't know, a bit
41:21: nauseous today.
41:23: I'm tired.
41:25: How about this hair?
41:27: Uh, I mean, I'm doing okay. Um,
41:31: I mean, I'm just kind of chilling here
41:33: on the stream. Uh, I'm excited to I
41:38: mean, I guess I'm excited for uh work
41:44: because uh Fuks put me on the thing for
41:47: making audio work under Linux. And we're
41:49: going to use SDL3's audio system for it,
41:51: which is cool.
41:52: >> Yes.
41:56: >> You can have all run Linux.
41:59: Yeah. Uh, next question is from Zenry
42:02: Dragon. Uh, where is physics on the road
42:05: map? Uh, so we, uh, it's definitely
42:10: like, you know, post split ending thing
42:11: like this to happen after that. Uh,
42:14: there's like a few things are kind of
42:15: high up there like, you know, doing the
42:17: IK, doing the UI, physics, you know,
42:20: collections for protolex. So usually we
42:23: kind of decide the next priority you
42:25: know usually right before we kind of
42:28: work on it. So right now I don't know if
42:30: this going to be the right next thing um
42:34: because there's like you know a bunch of
42:35: others and we usually like look at like
42:37: you know how how many uploads like each
42:39: issue has and we like you know what do
42:41: we need what makes more sense currently
42:43: and we kind of like you know make
42:44: decision on all of those parameters. So
42:48: um it's definitely something we want to
42:49: happen soonish
42:52: but there's not like a specific point
42:54: right now
42:57: if you if you definitely if you want you
42:58: know like physics to happen sooner given
43:02: up vote like on the GitHub I think
43:04: actually has quite a bit so like it
43:06: might be like you know one of the uh one
43:08: of the priorities.
43:12: Next question is from Czech the fox
43:14: otter. Uh Czech is asking I have a
43:17: question today about planned data
43:18: modular rework. Planning talking about
43:20: domain system in past episodes but from
43:22: my understanding the data model rework
43:24: will include more than just that. What
43:27: else is included in that? Data model
43:29: operations are by far the heaviest
43:30: operation when creating content
43:31: resonate. I'm curious if that can be
43:34: mitigated and prove your plan rework. So
43:36: there's like multiple parts to it. Uh I
43:38: actually do recommend checking out the
43:39: GitHub issue because like that one kind
43:41: of goes over the details quite a bit. Uh
43:43: but in short uh one of the uh one of the
43:46: things is like you know reduced overhead
43:48: reducing a little of the data storage
43:50: making things like you know be more
43:51: cache friendly uh reducing storing
43:54: explicitly some data that can be stored
43:55: only like you know temporarily
43:58: um that will help reduce a lot of the
44:00: like overhead. It's also going to
44:02: introduce a concept of like uh like
44:05: continuously changing drives because um
44:08: when you consider data model the data
44:10: model it monitors you know when things
44:11: change so it can replicate them over
44:13: network and it can invoke like you know
44:15: changed events
44:17: um the problem with that is something
44:20: you know say like the IK say the fingers
44:22: every single field of those you know is
44:25: changing every single frame and the
44:27: system is monitoring all of that and
44:29: it's triggering you know events for
44:30: every single one all the time. Uh which
44:35: like you know it makes things work but
44:38: it's not as efficient when you know some
44:40: things will be changing every single
44:42: frame. So one of the things that the
44:44: data model work is going to do is going
44:46: to be able to tell a drive that like the
44:49: or the drive is going to be able to say
44:51: I'm whatever I'm targeting I'll be
44:54: changing it every frame or almost nearly
44:57: every frame. So don't even bother
44:59: monitoring it you know skip all the
45:01: monitoring just assume it's changed
45:03: every frame. Um that way you know it's
45:07: going to tie into you know those systems
45:09: like let's send the events and they
45:11: don't need to be propagating and
45:12: managing all the changes. They will just
45:14: say this is changing every single frame.
45:16: If there's anyone any listener I'm just
45:18: going to invoke it. I don't need to like
45:20: you know keep tracking and keeping a
45:22: list of these things. Um and that's
45:24: going to help for a lot of the
45:25: continuous stuff like you know ek which
45:27: is like modifying a lot of stuff because
45:28: it's changing all the bones all the
45:30: fingers like you know dynamic bones
45:32: those like whenever they write into the
45:34: data model those writes are heavy
45:35: because that triggers all the like you
45:37: know monitoring stuff but the data model
45:39: rework it doesn't need to do that it
45:41: just it writes the value and that's it
45:43: like it's like skips a lot of the
45:45: complexity. Um so that's pretty much
45:48: like you know going to be like one of
45:48: the major optimizations. So those are
45:51: like the two main things for like you
45:53: know um doing that. The other parts are
45:56: like you know making a little bit more
45:57: dynamic too. So you know we can support
45:59: like stuff like domains and so on. um
46:01: that's going to like you know help as
46:03: well like it's going to open up so new
46:04: features but for performance uh the
46:07: reduced overhead is one side and the
46:10: ability to say like you know something's
46:12: changing continuously that's going to be
46:13: another
46:16: and also actually I'm going to mention
46:18: one thing is also proto flags already
46:20: has the system because when I made
46:21: perlex I knew I was going to do this
46:23: kind of a change for like you know this
46:25: kind of pattern for the rework data
46:27: model uh so per flag it actually has
46:30: change tracking to its own. And it knows
46:34: when something changes only occasionally
46:35: and it knows when something changes
46:37: continually. And if it changes
46:38: continually, it skips all the tracking
46:40: for the, you know, um like uh
46:45: like random changes.
46:52: Next question is from J Viden 4. Uh
46:56: Javan is asking in the future could it
46:58: be possible to license or use Forks
47:00: Engine crucially not itself and ship a
47:03: new game on Steam that uses similar how
47:05: Unity games are released? Yes, we would
47:07: actually like to do that at some point.
47:09: Um I think it would be kind of cool like
47:10: you know like we built an experience on
47:12: here and there's already like a few
47:13: there's like you know for example blood
47:15: on clock tower is like one um it's like
47:19: you know very kind of contain kind of
47:21: experience and if you build something
47:22: and you just want to you know publish it
47:24: as its own standalone thing uh we want
47:27: to give you the option to do so will
47:29: probably cost something because like you
47:31: know it's it's a good way to kind of
47:32: build a business model uh licensing the
47:35: engine and having you know your own
47:37: branding on it But it is definitely
47:39: something we want to offer. If you even
47:41: wanted to do something like that now, we
47:43: do offer wide label licenses for
47:45: businesses. Uh there's like some
47:47: businesses you know when they start like
47:49: it doesn't even say night. It will say
47:50: say whatever you know they want. It's
47:52: going to give them custom split screen.
47:54: Um you know like custom branding it
47:57: loads like you know into their own dash
47:58: loads into their own worlds. So that's
48:01: already kind of possible for you know
48:03: business deals. Uh if you have such use
48:05: case you know you can contact us you
48:07: know the business email uh or contact
48:09: form on our uh yoduckman.com website um
48:13: and we could like you know see like if
48:14: there's uh if there's kind of business
48:17: use case
48:21: uh modify arc is asking resite OS um I'm
48:27: actually not sure what's that in context
48:28: stuff like I know some people ask about
48:30: like you know having like optimized
48:32: Linux distro for resite But uh maybe it
48:35: will happen at some point, but we don't
48:36: have any specific plans ourselves.
48:39: >> I mean technically you can make a
48:41: desktop environment out of resite if you
48:43: really wanted to.
48:44: >> You could make it make it the main
48:46: environment. Don't even need like you
48:48: know the other one. Um
48:51: next question is from uh Papoutine. Uh
48:54: what blockers still keep you trained to
48:55: Windows instead of switching to Linux?
48:58: So it's kind of combination of things.
48:59: uh one is like um want to use like you
49:02: know visual studio uh to like do the
49:05: main coding I'll probably have to like
49:06: switch like either which like is paid
49:08: software so like there would be a switch
49:10: and there's also like you know some
49:12: changes with project files for that uh
49:14: the other one is like you know some of
49:15: the unity bits too uh but also like use
49:18: you know software like that doesn't
49:20: exist on Linux right now like uh for you
49:24: know some of the for example
49:25: photoggramometry I need to use Photoshop
49:27: because it has like specific specific
49:29: function for processing photos that like
49:31: I haven't found in any other software.
49:34: Um, and I would like to like drop that
49:36: too because I would also like to drop
49:37: Adobe
49:39: but I don't know any software that has
49:41: like you know that kind of processing
49:42: because they um they have like really
49:45: advanced like D noiseis and that
49:47: actually helps like you know with a lot
49:48: of the scans I do like it makes like
49:50: some of them actually reconstruct it
49:51: would explode
49:53: and it's like I don't have alternatives
49:56: for that one. Uh the other part is also
49:58: like hardware support like um like for
50:00: example use the Viper I and SL has been
50:02: kind of working getting that working but
50:04: you know it's not really out of the box
50:06: experience and and it's actually the
50:09: another part is like I
50:12: right now like I know like bunch of
50:14: people who like you know use Linux but
50:15: like a lot of time like it's a lot of
50:17: dealing with issues uh and get like you
50:19: know you have to spend a lot of time
50:20: just getting stuff to work and I don't
50:24: have time for it right now
50:25: unfortunately. Um, like it's
50:32: it's hard. Like I kind of like need like
50:34: things to get essentially get to the
50:35: state like you know where it just kind
50:36: of works out of the box. Um, but yeah,
50:40: there's like those are like a few even
50:42: like recently I actually couldn't even
50:43: use Linux on my laptop because when I
50:46: tried booting Linux on my laptop that
50:48: I'm actually running on right now and it
50:50: killed the Wi-Fi adapter like
50:52: temporarily like it like when it was
50:55: booting
50:57: um Glitch actually was watching it and
50:59: like you know during the boot sequence
51:01: and printed something which was like and
51:03: he uses Linux you know all the time he
51:05: uses Arch by the way and Um
51:09: he was like, "Oh, I've never seen it
51:10: like print something like that." And
51:11: then like, you know, I think the desk
51:13: cube environment started booting and it
51:15: just kind of exploded and didn't want to
51:17: like work. And then I boot it back into
51:18: Windows and the Wi-Fi adapters was gone.
51:21: So I tried reboothing, you know, tried
51:23: power cycling and it just it wouldn't
51:24: come back like it was just gone. And
51:27: like and the error was like, you know,
51:29: something related to the Wi-Fi adapters.
51:30: So it seems like it put in like some
51:32: kind of weird state. And the only way I
51:34: fixed it is like, you know, I actually
51:36: had to do like a hard power cycle where
51:37: like um and it actually made it come
51:41: back, but it was like, you know, like it
51:42: it just it exploded.
51:45: So, it's I usually run into issues like
51:48: that, unfortunately.
51:50: Um,
51:52: but yeah, there's there's there's just a
51:53: bunch of like software that kind of use
51:56: that like
51:58: doesn't doesn't work, at least not out
52:00: of the box or like there's no
52:02: alternatives for some of it. And some of
52:03: it like I'll see like eventually we can
52:05: like like I would probably like want to
52:07: like kind of do a boot but like I would
52:09: need to switch my development
52:10: environment which is like where I spend
52:12: majority of my time. Uh which is also
52:14: its own kind of process.
52:19: >> Uh
52:21: yeah
52:22: >> plus the VR stuff.
52:27: I also probably wouldn't be able to
52:29: switch like entirely like still like
52:30: dual boot because like um since like you
52:33: know majority of our users is going to
52:34: be using Windows
52:37: we need to test on it make sure like
52:38: stuff works but um
52:42: that's you know right now like I cannot
52:44: like daily drive it yet.
52:50: Uh
52:51: Deco is asking uh
52:55: actually I don't know what this is in
52:56: context of uh is that what happens? I
52:58: wish that was just not loading like dash
53:00: but it mean like context menus that one
53:03: was like something exploded because like
53:05: my death was working and usually there's
53:07: like no whole queue that kind of blocks.
53:10: >> Oh yeah, it was the context menu
53:11: disappearing.
53:14: Next questions from tracking.
53:19: No, my leg.
53:25: Why is it doing such things? There we
53:27: go. Um, next question. Glitch for
53:31: asking, am I orange fus? You are now.
53:36: >> You're orange and gray fus.
53:38: >> Yeah. Remember orange
53:42: is cuter because orange has more yellow
53:46: in it than yellow does.
53:50: >> And I can and I can my tracks not
53:53: working.
53:55: I can demonstrate it live.
53:58: >> Remember orange. It'd be really funny if
54:00: you just stop at remember orange.
54:04: >> Remember orange.
54:05: >> Stop.
54:06: >> Just stop. Like I'm just gonna I'm gonna
54:09: add
54:14: Where is this? Uh
54:17: I'm gonna make a new M actually I was
54:19: going to make quad.
54:22: So you see I made a quad
54:26: and I'll make a material. Actually, let
54:29: me just get rid of
54:31: uh essentials
54:34: tools
54:39: and I'll make an omelet material.
54:45: So, I'm going to make amllet. I'm also
54:47: going to actually going to keep one of
54:48: those as a background. And this
54:51: material, I'm going to make this yellow.
55:07: There we go. So, this is yellow.
55:11: So, you see there's this yellow and I'm
55:14: going to change this
55:17: to multiply.
55:24: And actually this one I'm going to
55:26: change this one too. I'm going to add a
55:27: I'm going to So we're going to start
55:29: with white background.
55:35: There we go. So we have, you know, we
55:38: have white background. Then we have
55:40: yellow that's being multiplied. And I'm
55:43: going to duplicate it a few times. So
55:44: this is one yellow, two yellows.
55:48: Actually, wait, hold on.
55:52: Small adjustment.
55:59: There we go.
56:02: Small adjustments.
56:12: There we go. Okay. So,
56:15: you can see this is set to multiply. So,
56:19: we're multiplying.
56:21: And if we multiply this, why is it
56:23: green?
56:26: Oh no, this is this is failing.
56:29: >> Cuz the world is blue.
56:31: >> Oh, it's messing it up.
56:34: Oh, there we go. Look. Orange. Oh, this
56:36: didn't work as I hoped it would. Oh,
56:40: let's let's let's let's do that again. I
56:42: fixed it. The the blue the blue seeped
56:44: in. It corrupted the result. So, this is
56:46: you know
56:48: duplicating duplicate thing. It's
56:50: orange.
56:51: Yellow multiplied with each other. So
56:54: more yellow
56:56: means it's orange.
57:01: See? Pro proven cute.
57:06: In in short uh orange is more has more
57:09: yellow in it than yellow does.
57:12: Therefore, orange is cuter than yellow
57:14: because if yellow has cute in it, then
57:16: orange has way more cute in it because
57:20: because orange has more yellow than
57:23: yellow does.
57:26: But isn't orange a slight absence of uh
57:30: green?
57:31: >> Yeah, the green. That's why I had to
57:33: make the green go away.
57:38: Uh anyway, next questions from Nookun.
57:43: Um Nukun is asking, "By the way, how did
57:45: you get uh get it so itself could push
57:48: me to the GPU and hand them off to
57:50: Unity? Was it still involved?"
57:54: I don't understand this question.
57:58: >> Asking about how the like mesh upload
58:01: pipeline works.
58:02: >> Yeah. Um there's no SDL involved like
58:05: we're not using SDL right now. Um
58:10: so like Reson itself is not pushing the
58:12: measures to the GPU that's happening.
58:14: Like I'm very confused by this question.
58:18: >> Yeah, Unity is the one doing that which
58:20: is
58:21: >> doing that but like because you're
58:23: saying how did you get Donite itself
58:25: could pull meshes to a GPU but you're
58:27: also saying how do you hand them off to
58:28: Unity?
58:31: I mean like for I'm
58:35: >> you might let's just try rephrasing the
58:38: question please. I don't I don't
58:43: >> I don't know what
58:45: >> uh yeah if you're asking like you know
58:47: context questions you need to include
58:49: context as your question like I we don't
58:51: know like you know
58:52: >> like this like I I don't know what this
58:54: is in context of so you might need to
58:57: include it too.
58:59: same same thing you know over this one
59:00: like I I don't know what what is this in
59:02: context of
59:07: uh
59:09: Nikon's asking will we be able to do GPU
59:11: side panning and such for materials uh
59:13: with custom shader support yes uh like
59:16: once we have that
59:18: um
59:20: you know once we have that and can
59:22: implement whatever functionality you
59:23: want with And
59:29: uh Res is asking is it more techtory
59:31: than more technically? Yeah, that's kind
59:33: of like a little bit how we go about
59:35: things is you know like we have like but
59:36: things and we we try to like be sort of
59:40: more flexible with our priorities. So
59:42: like rather than like you know sitting
59:44: down and being these are going to be our
59:46: priorities you know for the next two
59:48: years um we just decide very close to
59:51: each one because like often times like
59:52: the needs change and as being so able to
59:55: react and do whatever makes most sense
59:57: at a time I feel like we get most out of
01:00:00: that so that's usually kind of like you
01:00:02: know how we can upload like uh how we
01:00:05: approach planning and um prioritizing
01:00:12: asking If I was making so render on
01:00:14: Linux, could you put in for us? Do you
01:00:16: want that?
01:00:19: >> Sure, man. If it that's my payment.
01:00:25: >> I actually get paid.
01:00:30: >> We We don't We don't We don't pay by
01:00:32: pets.
01:00:40: Uh, Nikki's asking, "What do you like
01:00:43: about that world?" Uh, what do you think
01:00:46: could be done better? I really like like
01:00:48: the costics. Like that is a very neat
01:00:50: effect. Um, I like like how it kind of
01:00:53: like, you know, projects and everything
01:00:54: like because it kind of gives it an
01:00:55: atmosphere. Um,
01:00:59: what could be done better? I feel like
01:01:01: like maybe like making it look like
01:01:03: outside is actually water because like
01:01:05: it feels a little bit like you know it's
01:01:07: just nothing there. So adding like you
01:01:09: know say some fog volumes like you know
01:01:12: making feel like there's like more kind
01:01:14: of thick water there that might kind of
01:01:16: help but it's going to also be nitpicky.
01:01:25: Uh next thing uh is uh next question is
01:01:29: from Tool. Uh what led to hiring of
01:01:31: Rusty and Decoy and how official will
01:01:34: the stuff they'll do be like will the
01:01:36: new Rusty tutorial be in hop tub and
01:01:38: does uh just do more stuff on YouTube
01:01:41: and that helps their exposure. Um so for
01:01:45: the for what like this actually be
01:01:47: mostly handled by Koma who's kind of
01:01:49: like leading the marketing. Uh, so you
01:01:52: might want to like ask him. Actually, I
01:01:53: don't know if Cameron is like he's
01:01:55: usually not on the art uh art ones and I
01:01:58: don't think he has office hours right
01:02:00: now.
01:02:02: Um, let me check on Discord. Uh,
01:02:08: do you know if Chroma has office hours?
01:02:12: >> I
01:02:14: don't know. He currently doesn't
01:02:18: he currently doesn't run those. Um
01:02:24: so like the only answer I can like give
01:02:26: you right now is like you know it's
01:02:28: mostly kind of chroma like a site. So
01:02:29: like uh but the main thing is like you
01:02:32: know was kind of looking for like you
01:02:34: know very specific skill set uh that was
01:02:37: needed at the time.
01:02:40: So it's mostly kind of like you know
01:02:41: influenced by that. Uh it's pretty much
01:02:44: like you know Chromos like he's uh
01:02:45: handling certain kinds of marketing and
01:02:47: it was like certain areas where he
01:02:49: needed help with and both Rusty and
01:02:51: Decoy had like you know specific skills
01:02:54: needed
01:02:56: um needed you know to kind of like fill
01:02:58: essentially fill like you know fill
01:03:00: those in um for like you know the
01:03:03: official stuff like whatever stuff
01:03:04: they're doing officially that is
01:03:06: official whatever stuff they're on their
01:03:07: own you know they're still like doing on
01:03:09: their own. So like you know there's kind
01:03:10: of like that separation. And it depends,
01:03:12: you know, what if the stuff is being
01:03:14: made to be like released officially or
01:03:15: if it's like, you know, on their own. We
01:03:17: kind of like, you know, let everyone
01:03:18: kind of keep that separated.
01:03:25: Uh, next question is, uh, Y is asking,
01:03:29: "Does the Reset team have commercial
01:03:30: licenses to avatars? they wear um as far
01:03:33: as I know like like they're it kind of
01:03:35: depends because people have lots of
01:03:36: different avatars but like um just like
01:03:39: because these are like are like personal
01:03:41: kind of use so like we I don't uh for a
01:03:43: lot of them like I don't think like
01:03:45: there's a commercial license required
01:03:47: but also I don't know like status of
01:03:49: every person so um I'm not sure
01:03:58: um modify is asking uh are there Any
01:04:01: plans to hire more software engineers?
01:04:03: Uh, right now, no. Eventually, we would
01:04:05: definitely like to expand the team more,
01:04:07: but right now we don't have any plans.
01:04:16: Nook is asking how what does the mesh
01:04:18: upload to the GPU thing you did recently
01:04:20: actually do and how does it work? Um, so
01:04:26: um
01:04:29: Oh, uh, okay. Okay, so this was like a
01:04:31: thing that was kind of done like you
01:04:32: know a sort of pre-plittening part. Uh
01:04:35: what it does it uses more like lower
01:04:37: level APIs where previously would use
01:04:40: like more highle APIs where it would
01:04:42: like compose a bunch of like arrays pass
01:04:43: it to Unity's API unit would like do
01:04:45: bunch of processing on that bunch of
01:04:47: validation and unit would generate the
01:04:49: actual data that gets uploaded to the
01:04:50: GPU.
01:04:52: uh with a new one we actually using more
01:04:54: low-level APIs where we generate
01:04:58: um
01:05:00: uh like where we essentially generate
01:05:03: like you know the memory layout for the
01:05:06: GPU on our end and we just like you know
01:05:08: give Unity you know essentially just
01:05:11: array of data say upload this to the GPU
01:05:13: also give it like data like you know
01:05:15: this is the layout of this buffer um and
01:05:18: like you know that's much faster because
01:05:20: Unity doesn't have to do the conversion
01:05:21: s we can do them on our end and we can
01:05:23: do them more efficiently. Um
01:05:26: and also it works better over the IPC
01:05:28: mechanism because with the with um the
01:05:33: spliting we don't need to like you know
01:05:35: pass million different buffers we just
01:05:37: pass one buffer uh plus you know few
01:05:39: extra information and like you know
01:05:41: that's makes things much simpler and
01:05:43: this makes also much faster because like
01:05:45: you know we um
01:05:47: we can like generate like skip bunch of
01:05:49: the checks and some of the checks
01:05:50: actually happening twice because we
01:05:52: would do our own channel checks and
01:05:54: validate ations. Uh, and the unit will
01:05:56: do the same ones even though we've just
01:05:57: done them and there was no way to skip
01:05:59: them because we're using higher level
01:06:00: APIs.
01:06:04: Uh, next question is from gamer
01:06:05: delicacy. More pixels or less pixels? I
01:06:08: think you know more pixels.
01:06:10: Oop.
01:06:17: And there we go. I need to reposition.
01:06:23: Next questions from Powderpop. Uh
01:06:26: Powderpop is asking, "I've noticed that
01:06:28: you can air strife and resonate. Is this
01:06:30: intentional? Possibly inspired by
01:06:31: movement in Quake." Um that is kind of
01:06:33: intentional. Like it's it's well you can
01:06:36: kind of configure it like you can
01:06:38: actually change how much control you
01:06:39: have when in air. It's not even striking
01:06:42: strafing specifically. It's like, you
01:06:44: know,
01:06:46: it's like being able to like, you know,
01:06:48: still apply change like forces to
01:06:50: yourself when you're midair. And it's
01:06:52: something you can configure on the locom
01:06:54: motion module. On the physical locom
01:06:55: motion, you can change how much force
01:06:57: can be applied and you can even set it
01:06:59: to zero, which means once you jump, you
01:07:01: cannot change your trajectory. So that
01:07:03: is kind of configurable. It is not
01:07:05: inspired by Quake. It just kind of feels
01:07:08: like it feels kind of better to have the
01:07:09: control like because there's no like
01:07:11: there's no like gameplay reason not to
01:07:13: be able to do that. So by default like
01:07:16: we let people do that but if you want
01:07:17: building a specific game on here you can
01:07:20: configure the locomotion to not allow
01:07:22: that. Uh so you have uh you have like
01:07:26: you know you have the control over that.
01:07:32: Uh Kite VT is asking uh what is the
01:07:35: estimate? Do you have an estimated
01:07:36: amount of time until you think spliting
01:07:38: will be ready? Right now I feel maybe in
01:07:41: 2 weeks. Um, it might be a bit longer.
01:07:45: It's getting close. Uh, but it's also
01:07:48: like, you know, still kind of like
01:07:50: discarded complexity. There's like, you
01:07:52: know, certain bugs that might like end
01:07:53: up like inter rabbit holes. So, we'll
01:07:55: see. Um,
01:07:59: like, you know, things that
01:08:01: >> Yeah,
01:08:01: >> I was going to say the frame timing is
01:08:03: going to be fun to fix, too.
01:08:05: >> I think the one might be like relatively
01:08:07: quick. I have like rough hunch like what
01:08:09: it is, but
01:08:10: >> Okay. But sometimes like things like end
01:08:12: up like being thank you gamer delacy for
01:08:15: a subscription.
01:08:18: Um it kind of it depends like um like it
01:08:23: depends like you know a little over
01:08:24: complexity because like we'll have the
01:08:25: VR support soon and depends like you
01:08:26: know to help like find a lot more bugs
01:08:29: where there be like medium amount of
01:08:30: bugs could be low amount of bugs like we
01:08:32: found a lot of them like you know just
01:08:34: with existing testing but like with
01:08:36: those things like the tricky thing is
01:08:37: like you don't know how much how many
01:08:39: issues you're going to get. Uh so just
01:08:43: at the rate how things have been going
01:08:44: like it feels 2 weeks but might be a bit
01:08:46: longer.
01:08:50: Uh noon's asking have you ever rocket
01:08:52: jump in the resite? Yes I have.
01:08:57: Uh modify arc is asking what make and
01:09:00: model is your laptop? Uh I have the MSI
01:09:02: Titan. Uh forget actually the number of
01:09:05: it.
01:09:06: um
01:09:08: as one of the MSI titans like G7
01:09:12: something
01:09:19: for the for for the bits. My god
01:09:24: for is asking I also heard glitches
01:09:26: cute. Yeah, like proven cute with Twitch
01:09:29: bits which is a big major company that
01:09:32: has those. So like you know it has to be
01:09:34: valid.
01:09:40: Uh
01:09:44: next question.
01:09:48: Uh Johnny Awesome is asking how locked
01:09:51: in are current shaders in Resonate.
01:09:52: Would you consider adding a line or two
01:09:54: to to them to allow using BC7 for normal
01:09:57: maps? Um BC5 BC5 sorry. Um,
01:10:03: actually I think that should just like
01:10:04: there shouldn't be much of a shader
01:10:06: thing. I forget which BC 5 is
01:10:12: uh
01:10:14: >> I forget which one which layout that is.
01:10:16: I think it's like the two two
01:10:18: independently encoded channels.
01:10:21: >> Let's see.
01:10:21: >> Which might just work out of the box. I
01:10:25: don't know. I mean changing that like
01:10:27: that's actually
01:10:29: pretty big change if it needs shader
01:10:31: change because like there's like pretty
01:10:33: much all the shaders that have normal
01:10:35: maps
01:10:36: um they use like unus functions to like
01:10:40: decode stuff. So like we would have to
01:10:42: like make the change deeper or we would
01:10:44: change a lot of the shaders which might
01:10:46: be also trickier to do for some normal
01:10:48: like surface shaders. like you can make
01:10:51: a GitHub request for it. But um
01:10:54: depending on the complexity of the work,
01:10:57: the answer might be, you know, no.
01:11:01: And I don't know like they're not enough
01:11:02: to evaluate that.
01:11:04: >> It looks like BC5 is just two channels
01:11:07: of BC4.
01:11:08: >> Yeah. Yeah, that's kind of what I was
01:11:10: expecting to be.
01:11:18: Uh D is asking uh not much technical
01:11:21: question. Have you entertained some sort
01:11:23: of a single source of truth uh for like
01:11:26: a better word for knowledge management?
01:11:28: Um,
01:11:29: >> what?
01:11:30: >> I have the wiki,
01:11:32: >> but yeah, like I don't I don't know
01:11:33: exactly what we're talking about. Like
01:11:34: at some point like we would like to do
01:11:36: where documentation for for example for
01:11:38: fields is like you know built into the
01:11:41: like engine and our database. So you can
01:11:43: kind of like you know access it very
01:11:44: easily in game and could be considered
01:11:46: like you know the source of truth but I
01:11:49: don't know if that's like you know
01:11:50: exactly what you're meaning.
01:11:57: Uh, next question is from Jay Viden 4.
01:12:00: Uh, earlier you mentioned business
01:12:01: clients can have custom branding. Are
01:12:03: there any other interesting features
01:12:04: that business clients have that normal
01:12:06: steam boats don't? Yeah. So like white
01:12:08: labeling that's like one of those
01:12:10: things. Uh, some business clients they
01:12:12: can ask you know for having their own
01:12:13: infrastructure if they want to host
01:12:15: their own infra uh for separation. That
01:12:17: is a thing you know like we've been
01:12:19: working with some to do. uh this like
01:12:22: you know smaller things like they can
01:12:23: for example have their own update
01:12:24: channels meaning uh you know if the
01:12:27: business is for example running events
01:12:28: they don't want like a certain update to
01:12:30: disrupt you know their event uh so
01:12:33: that's another thing you know that can
01:12:34: be done for business clients and there's
01:12:37: like bunch of like you know some other
01:12:39: like smaller things like you know
01:12:40: support and so on um
01:12:43: I would recommend actually bring this
01:12:45: question to probably primes office hours
01:12:47: because he deals with business clients a
01:12:49: lot oh boy.
01:12:53: He also um developed a lot of these
01:12:56: business features. So he should be able
01:12:58: to give you like much better answer on
01:12:59: this topic.
01:13:05: Uh Gamer Delicacy is asking uh thoughts
01:13:09: on a CH GP Protolex helper.
01:13:11: >> Um probably not going to do that
01:13:13: officially. If if you want to build
01:13:16: something like that on your own, we're
01:13:17: not going to stop you. But I don't think
01:13:19: we'll offer that like as official thing.
01:13:22: Um, it's like one of the tricky things
01:13:25: like that's also like parallel with like
01:13:26: you know stuff like CHTP JPT is like
01:13:29: they tend to hallucinate a lot and
01:13:32: there's actually been a study like where
01:13:33: they found like it only gives you
01:13:34: correct answer like 42% of the time even
01:13:37: with like the you know fanciest model
01:13:40: and that can be you know kind of
01:13:43: dangerous like relying on it like too
01:13:44: much um because it's almost like you
01:13:47: know it's worse than a coin toss whether
01:13:49: you get like you know something that's
01:13:50: correct or not. So
01:13:54: there's like you know a bunch of kind of
01:13:55: concerns with like you know that and
01:13:58: like have any concerns like you know
01:14:00: with like the some of the ethicacy of
01:14:01: like you know the training data and
01:14:03: stuff like that.
01:14:05: >> AI is just terrible for programming like
01:14:09: I see so many people use it. It's so
01:14:10: prevalent in the net field. It makes me
01:14:13: so sad and it's just terrible.
01:14:15: >> Yeah. like people
01:14:18: um
01:14:20: people like trusted like way too
01:14:21: blindly.
01:14:30: Uh next question is from Kite VT. Uh how
01:14:33: did you guys get the idea to add OBS
01:14:35: plug-in for streamer straight from
01:14:37: inside a game? Do you think it's
01:14:38: possible to add Streamlabs plugin too?
01:14:41: Um I don't actually remember exactly. I
01:14:43: think it was just like, you know, when
01:14:45: you're in VR, so you don't have to like,
01:14:47: you know, fuss with your desktop. Um,
01:14:49: it's kind of much easier to like, you
01:14:51: know, have it integrated because you can
01:14:52: have like additional functionality with
01:14:54: it and also you can have like, you know,
01:14:57: just kind of build around it. So, I
01:15:00: don't know, it just seems like a good
01:15:01: thing to do like easier to kind of
01:15:02: control. Uh, the other part is uh for
01:15:06: the Streamlabs, I would say make a
01:15:08: GitHub issue. Uh, we can kind of look
01:15:10: into it. I don't know if they have, you
01:15:12: know, SDK.
01:15:14: Uh, it kind of depends heavily on that
01:15:17: because if they don't have an SDK to
01:15:18: control it, um,
01:15:21: it's, uh,
01:15:26: like it's going to be probably difficult
01:15:28: like if they don't have like any means
01:15:29: to control uh, from other software. But
01:15:33: right now, like, you know, I haven't
01:15:34: looked into it, so I don't know.
01:15:37: Uh, but we actually have all the
01:15:39: questions. Uh, we already we have like
01:15:43: 45 minutes left. So, I think I should be
01:15:46: okay to kind of
01:15:48: finish if there's like more like finish
01:15:50: the whole two hours. If there's more
01:15:52: questions,
01:15:55: but we have no more questions. I need to
01:15:56: ask more questions.
01:15:59: If there's no questions, probably end
01:16:01: early. There's like some topics I kind
01:16:03: of want to get into, but like um
01:16:06: probably start doing that after split
01:16:08: ending is done because like you like
01:16:09: usually on these streams like lately I
01:16:11: don't have that much energy
01:16:13: um because of like all the work
01:16:16: uh
01:16:21: so like we kind of saving them for
01:16:23: later. Oh, there we go. We got we got a
01:16:25: question from Glitch. Uh Glitch is
01:16:27: asking, "Has Mayion been turned into an
01:16:30: instrument yet?" No, I I've I've been
01:16:34: like
01:16:35: I've I've been like uh thinking this up
01:16:37: that nobody did it yet. We need We need
01:16:39: mayonnaise as an instrument. It needs to
01:16:40: be functional instrument on here. I did
01:16:43: have somebody
01:16:45: um
01:16:48: um remember like somebody like you know
01:16:51: like
01:16:52: talking about like scanning a jar of
01:16:54: mayonnaise or something for it, but like
01:16:56: that's about it. I haven't heard like
01:16:58: super much.
01:17:01: Uh, next question is from Nukun. Nikki's
01:17:04: asking, uh, did you see the thing in
01:17:06: Avatar Station and Jupiter Station where
01:17:08: two worlds are connected by airlock with
01:17:10: a button? You've talked about stuff like
01:17:11: that before, but I remember seeing it
01:17:13: elsewhere and I haven't seen that yet.
01:17:17: That sounds pretty cool.
01:17:20: Uh, good one. Uh, modify arc is asking
01:17:24: sire, what uh is the most frustrating
01:17:27: thing dealing with getting the pies
01:17:28: working on Linux?
01:17:31: Oh man. So since we're doing a really
01:17:34: unorthodox architecture where the render
01:17:37: is uh running in
01:17:41: like one proton while like the main
01:17:44: engine is running natively. Uh I found
01:17:48: that there's not actually a lot of like
01:17:51: support for communicating like between a
01:17:54: native Linux process and a wine process.
01:18:00: uh like most people like on the internet
01:18:02: they're like well you just use TCP just
01:18:05: use a pipe just use a local socket or
01:18:07: something and I'm just like no we need
01:18:11: shared memory
01:18:13: uh and so
01:18:16: uh I was like okay well I could just
01:18:19: point it at dev shm which is where all
01:18:22: the shared memory stuff is. Um, but then
01:18:25: that was also kind of janky. And then
01:18:27: the semaphors didn't work. Uh, which
01:18:30: means that like each side couldn't tell
01:18:34: the other side that there was a message.
01:18:35: They just kind of had to keep pulling
01:18:37: for if there was a message. Uh, so I
01:18:39: just made like a little wrapper. But I
01:18:41: suppose the most frustrating part to get
01:18:43: back on track
01:18:45: uh
01:18:46: is managing like the startup sequence of
01:18:50: Resonite in that in that configuration
01:18:53: because what we have to do is we have to
01:18:57: publish the game as if it was a Windows
01:19:00: only game and then the
01:19:04: when it launches it will detect if it's
01:19:07: running in wine aka Proton
01:19:11: um proton is just a modified version of
01:19:13: one
01:19:15: uh and change the startup sequence
01:19:17: accordingly. So we have a little
01:19:19: bootstrapper program that I wrote which
01:19:22: will launch in the place of the main
01:19:24: exe.
01:19:26: Then what it'll do is it'll detect if
01:19:28: it's running one and if it is it will
01:19:30: launch a shell script and that launches
01:19:35: as like a native Linux thing because
01:19:37: when you launch a shell script from
01:19:39: within wine it'll actually spawn like
01:19:42: there's a mechanism to have it spawn an
01:19:44: actual instance of like bash or sh or
01:19:47: whatever you're using for your shell uh
01:19:51: and do like normal Linux stuff with the
01:19:53: shell script. So what that does is it
01:19:55: downloads a native version of .NET for
01:19:58: Resonate to use uh and then launches
01:20:02: Resnite using that version of .NET that
01:20:05: it just downloaded.
01:20:07: Then Resonite will detect if it's
01:20:11: running under Linux and assume that it
01:20:14: will want to launch you know the it will
01:20:17: want to launch the shell script for
01:20:19: renderite which is the renderer
01:20:22: which then uh writes it's a little janky
01:20:27: which then writes the launch arguments
01:20:29: to a file which the bootstrapper then
01:20:32: picks up and spawns
01:20:34: the actual renderite exe. with those
01:20:37: launch arguments so that the
01:20:40: bootstrapper having been launched in
01:20:42: Proton originally with all the correct
01:20:44: environment variable set up will create
01:20:46: a child process that also has the
01:20:48: correct environment set up aka the
01:20:50: renderer.
01:20:54: It is a big old merrygoround that is
01:20:56: crucial in making this configuration
01:20:58: work.
01:21:00: This is pretty cool like you figured out
01:21:02: how to because like like I think this is
01:21:05: going to like you know how like load
01:21:07: better on Linux like you know just like
01:21:09: with the main process being native it
01:21:10: kind of simplifies through things too
01:21:12: but also I think it like makes us like
01:21:14: from at least technological standpoint.
01:21:18: It makes us like like probably the first
01:21:21: platform or application to use this kind
01:21:23: of hybrid architecture where like the
01:21:25: one program has like you know is like
01:21:27: literally a hybrid of native Linux and
01:21:32: Proton for the other part of it. And I
01:21:36: don't think there's like I I think
01:21:38: there's a good chance like we're going
01:21:39: to be the first to like do this kind of
01:21:41: architecture. And I think that's you
01:21:42: know kind of cool. So I'm I'm glad like
01:21:45: you kind of managed to get it working.
01:21:48: Yeah, it's uh it's very unique. Uh
01:21:52: >> if somebody asks us like if if runs
01:21:55: natively on Linux will be like yes and
01:21:59: >> yes no
01:22:00: >> yes no.
01:22:04: Uh next question is from
01:22:07: uh Yummy Deto. Uh today I once again
01:22:12: noticed how easy it would be for me to
01:22:14: program in flux if I could actually find
01:22:16: the nodes that do that do what I want my
01:22:18: own. I think LM that is trained on
01:22:20: resent documentation solely to find
01:22:22: nodes and components would we need to
01:22:24: have. Um no
01:22:26: >> it feels like you just want search like
01:22:29: it feels like like adding like just a
01:22:31: necessary complexity that's just going
01:22:33: to make up information like now and then
01:22:35: and give you you know stuff you don't
01:22:38: need. Um because like the main thing
01:22:41: like we really need is like you know we
01:22:42: need documentation for your nose and
01:22:44: search once you can search it it can
01:22:46: actually search like you know the
01:22:47: keywords and other stuff. So
01:22:51: like I feel like that alone having that
01:22:53: like you know that's going to solve like
01:22:54: pretty much like your problems. Um
01:22:58: again like the issue with like LM is
01:23:00: like you know they can kind of give you
01:23:02: good stuff until they don't and they
01:23:05: will just make up stuff or they will
01:23:06: like not understand stuff. So like it's
01:23:08: like
01:23:10: I find it a little bit like perilous to
01:23:12: like you know depend on them to like
01:23:15: give you because like you're almost like
01:23:17: if that's the only way to kind of access
01:23:19: you know like find the stuff you want
01:23:21: like you know in some cases it's just
01:23:24: going to give you bogus stuff and in
01:23:26: some cases is not going to give you what
01:23:27: you need. Um,
01:23:30: so like I I I don't think that approach
01:23:33: like like I think like just need search.
01:23:42: >> Yeah, that's like like
01:23:44: >> that's what like we would tag the nodes
01:23:46: for for like what categories they're in
01:23:49: and what actions they're related to.
01:23:56: accident.
01:23:58: Uh K is asking when did you start the
01:24:01: mentor program and do you remember who
01:24:03: the first mentor was? So was the thing
01:24:05: that kind of developed over time but
01:24:07: this is actually another thing um
01:24:11: bring this to the office moderation hour
01:24:13: because the mentor program is handled by
01:24:15: the moderation team. So you'll have to
01:24:17: wait until next week because they do
01:24:19: their office hours like 90 minutes
01:24:20: before this stream starts. Um but this
01:24:23: would be like perfect question you know
01:24:25: for their stream.
01:24:30: Um next question is from uh Qualtorium
01:24:35: is asking um are there any file formats
01:24:39: or network protocols other ways of
01:24:41: importing interfacing stuff with
01:24:42: resonate you'd like eventually get added
01:24:44: to platform. So there's like two things
01:24:47: I can immedately think of like one is I
01:24:50: would like to actually open the importer
01:24:52: exporter mechanism and open source part
01:24:55: of it. So everybody can actually write
01:24:57: their own importers and exporters to add
01:24:59: support for whatever formats you need.
01:25:02: Um with that like you know it's not
01:25:04: going to be just up to us whatever
01:25:05: formats we support. if you need support
01:25:07: for super obscure format or if you need
01:25:08: like specific modifications to like how
01:25:10: the import works for formats we already
01:25:13: do support you you'd have ability to do
01:25:16: that. So I think that on itself could
01:25:18: open up lots of possibilities. The other
01:25:21: part I would like to add like a
01:25:23: mechanism that sort of lets you
01:25:24: manipulate the sort of document object
01:25:27: model of the scene over like an um say
01:25:31: like TCP connection like it doesn't have
01:25:32: to be TCP but some kind of like you know
01:25:34: connection. So you could actually
01:25:35: control it from other scripts and use
01:25:38: that to build SDKs. So you could do you
01:25:41: know for example stuff like um you could
01:25:45: like you know build like Unity SDK or
01:25:46: Unreal SDK or Blender SDK and have like
01:25:49: have a kind of interface with lots of
01:25:51: different programs and that itself also
01:25:53: that would be open. So um anybody in the
01:25:56: community could either expand the ones
01:25:58: we provide officially or build their
01:26:00: own. Um, so those two like you know are
01:26:03: things I think could be like very
01:26:04: powerful and could like give a lot more
01:26:06: a lot more power to everyone in the
01:26:08: community to like you know expand
01:26:09: there's like ability to import an
01:26:12: interface.
01:26:17: Uh, next questions uh from a random
01:26:20: meatball. Uh, what is an object or world
01:26:23: that someone has created in Reside that
01:26:25: truly blew you away? I mean there's a
01:26:28: bunch um like it kind of depends on what
01:26:31: like level because like for example a
01:26:32: lot of the say like uh for social worlds
01:26:35: and just like interaction and just
01:26:36: general polish a lot of the team vibe
01:26:38: stuff like their MMC stuff that like
01:26:40: tends to be like excellent there's stuff
01:26:43: like you know um like somebody like
01:26:46: built like a functional like you know
01:26:49: neural train like like some of the
01:26:50: people in Japanese community they made a
01:26:53: functional like you know neural like
01:26:56: network like training world and
01:26:57: demonstration world and it was also like
01:26:59: really cool. I've seen people build like
01:27:01: you know simple like physics engines uh
01:27:04: build like you know tools like for like
01:27:07: like you know pasting like languages.
01:27:09: So, um there's a whole bunch. There's
01:27:13: been like, you know, some games like
01:27:14: even stuff like, you know, say the blow
01:27:16: on clock tower I mentioned. Um that one
01:27:19: is like both like really cool world like
01:27:22: you know like it's very like they put
01:27:24: like so much work into it making it work
01:27:25: but also another thing that blows me
01:27:27: away is like they've been doing it for
01:27:28: like years now like hosting regular
01:27:31: games and the not itself like you know
01:27:34: that blows me away and and we go along
01:27:37: that line. There's also like, you know,
01:27:38: creator jam which is like running right
01:27:39: now and it's been going for even longer
01:27:42: and they've been going every, you know,
01:27:44: every weekend hosting events, you know,
01:27:47: like building stuff and just producing
01:27:48: like, you know, so much things. Uh they
01:27:51: and they even like some of the some of
01:27:53: the stuff they build like, you know,
01:27:55: made it to like uh rain dance, you know,
01:27:58: like another kind of like events. So
01:28:01: there's there's there's a lot like I I
01:28:05: like I could go like you know for a
01:28:06: while like about this like it's like I
01:28:09: don't like picking like you know just
01:28:10: like a single thing. Um
01:28:15: there's just like a lot of stuff like
01:28:16: along the way. It just kind of like
01:28:17: blows me away.
01:28:22: Uh next question is from
01:28:26: uh DJ Sim one. Uh when the time comes,
01:28:30: what will collections look like? Do you
01:28:31: think they will reflect as structures
01:28:33: like JSON, XML, or is there plan to aim
01:28:35: for something even more generic,
01:28:37: flexible? Uh they're going to be more
01:28:39: reflecting like, you know, collections
01:28:40: in C. Uh probably going to have like,
01:28:42: you know, stuff like lists, arrays,
01:28:44: dictionaries, hashmaps, stuff like that.
01:28:46: Ability to iterate over them, you know,
01:28:48: uh access specific elements and so on.
01:28:51: Uh you know, other move elements. Um for
01:28:54: JSON XML, uh there going to be more
01:28:56: like, you know, like parsing stuff. So
01:28:58: like you for example JSON you can parse
01:28:59: it into like you know series of
01:29:01: dictionaries and lists and vice versa
01:29:03: you can like you know convert like that
01:29:05: structure into JSON. Um but like the
01:29:09: JSON like you know this consider like a
01:29:10: conversion point. So you have like you
01:29:12: know collection to JSON and JSON to
01:29:14: collection. Uh so that's going to
01:29:17: convert between the two but like you
01:29:18: won't like you won't like work with like
01:29:21: JSON as the collection itself.
01:29:27: Um,
01:29:29: next question is from Akit Tou. Uh, I'm
01:29:35: sure this was mentioned before, but
01:29:37: roughly how high is the priority is the
01:29:39: IK overhaul? First, the splitting of
01:29:41: course, but any plans what comes after
01:29:43: for as the next big thing? So, the next
01:29:45: big thing hasn't been decided yet. Uh,
01:29:48: IK is pretty high up there, so like you
01:29:49: know there's a chance it might be the
01:29:51: next thing, but also the decision hasn't
01:29:52: been made yet. Uh might do like
01:29:55: something like you know like a molecule
01:29:57: like to kind of like ease the
01:29:59: development but we'll we'll see how
01:30:00: things go. Like no decisions been made
01:30:03: yet.
01:30:07: Oh my god. We got is asking what kind of
01:30:09: rock or mineral would render it be?
01:30:11: Where we got a question like uh we got
01:30:14: it didn't we get like last time?
01:30:18: Uh,
01:30:18: >> somebody was asking about or somebody
01:30:20: was asking what it would taste like.
01:30:24: >> Like pop rocks.
01:30:26: >> Pop rocks. Does it pop your eyes?
01:30:31: >> That anymore. I think we should start
01:30:32: talking about this.
01:30:34: >> Yeah, we should stop talking about this.
01:30:36: You figure it out.
01:30:40: >> Uh, next question is from Ultramarto.
01:30:43: Um would you rather use the full body
01:30:46: you currently have uh
01:30:49: or let's say in future use suit that you
01:30:51: put uh same the way you put on clothes
01:30:54: depends like how bulky the suit is like
01:30:56: if it's pain to get into and if it
01:30:58: restricts movement then maybe not the
01:31:00: suit the good thing about trackers is
01:31:02: you know the light
01:31:04: uh ideally I would like want to use like
01:31:06: something that like does optical
01:31:07: tracking or maybe magnetic tracking so
01:31:10: it works like you could like go under
01:31:11: blanket or something like that and be
01:31:12: comfy. But uh we'll see. It depend. It
01:31:16: depends. Like it mostly depends, you
01:31:17: know, how pain is it to deal with.
01:31:20: >> I want I want Optitra. Give me
01:31:22: Optitrack. I know it's like $40,000, but
01:31:25: give me Optitrack, please. It's perfect.
01:31:27: >> They can they can give you Optitra if
01:31:29: you have $40,000,
01:31:31: >> please.
01:31:33: When a fundraiser for Optitra for Syra.
01:31:36: >> Yes, please.
01:31:40: Uh Jay Widen's asking, "So you've been
01:31:43: answering our questions, but so far you
01:31:46: haven't answered the Elderish textures
01:31:47: questions. What gives?"
01:31:50: It doesn't let me.
01:31:57: Next question is from Death Hammer. Uh
01:31:59: what would you like to do as a
01:32:01: celebration for getting the splattering
01:32:03: done?
01:32:06: not thing for a bit.
01:32:09: >> I think freak wants to become a rock.
01:32:11: >> I just I there's been a thing is like
01:32:13: like there's always so much stuff like
01:32:15: you know like oh possible thing do this
01:32:16: possible thing do that possible thing do
01:32:18: this and I'm just like I just want like
01:32:20: little space to breathe and decompress
01:32:23: like and like not think about things for
01:32:26: a bit.
01:32:27: Also maybe order a pizza or something.
01:32:29: Maybe get some ice cream. I don't know.
01:32:32: Get some fruit. Um, I've had like a pile
01:32:36: of games I wanted to play that I haven't
01:32:39: been able to touch like for a good bit.
01:32:42: Um,
01:32:44: I don't know.
01:32:46: Just kind of like not not have to worry
01:32:49: about like for for just a little bit.
01:32:53: >> I mean, the I know it's like work
01:32:55: related, but a thing that I would be
01:32:57: excited to do is to instrument all of
01:33:00: the engine using
01:33:02: He's
01:33:04: doing things.
01:33:05: >> I know, but it's fun.
01:33:09: I think it's
01:33:16: Can you can do that like when when when
01:33:18: when I'm done when it's out
01:33:22: when I'm not doing
01:33:24: next questions from as 17. uh if you
01:33:27: don't have the bandwidth for molecule
01:33:29: like there's too much on your plate do
01:33:30: you plan on having J4 working on
01:33:32: molecule in the long run since some
01:33:34: parts of his automation? Yeah. So the
01:33:37: automation parts like J4 has been like
01:33:39: working on already sort of like in
01:33:40: preparation for that too. Um
01:33:44: so that's already kind of been
01:33:45: happening. Uh there's like parts that I
01:33:47: kind of been because I I actually
01:33:48: implemented parts of it but I kind of
01:33:50: had to like drop it for other things.
01:33:53: Um, and I do have like very specific
01:33:54: things I want to like do with it and the
01:33:56: way I want it to work. So, um, some of
01:34:00: this like, you know, like kind of
01:34:02: building the tooling for myself.
01:34:06: Um, but a lot of it's like, you know,
01:34:08: like a lot of the heavy stuff is like
01:34:10: because the tooling like the molecule
01:34:13: like for my mind that's mostly just kind
01:34:15: of like gluing things. So like a lot of
01:34:18: the automation for builds that's like
01:34:19: you know where
01:34:21: a lot of the time has to go and J4 has
01:34:23: been kind of doing that. So I like
01:34:25: actually did some of the TIC2 but now
01:34:27: it's almost like J on J4 played.
01:34:32: Uh next questions from I'm moving this
01:34:35: camera again.
01:34:37: Uh modify arc is asking uh why are
01:34:40: inspectors generated by the host? Um
01:34:43: because it's kind of simpler to
01:34:44: implement that way. uh we don't want to
01:34:46: rework them so they're more flexible but
01:34:48: like having them generated by different
01:34:49: users is actually more complicated
01:34:51: because now we have to worry about who's
01:34:53: responsible for like you know generating
01:34:55: and updating it. Um
01:34:58: and like because we have like multiple
01:35:00: users what happens you know if like you
01:35:02: if you spawn an inspector and then
01:35:04: somebody else like you know modifies it
01:35:06: now it needs to go to the host needs to
01:35:08: go to the user now you need to generate
01:35:09: the go to host needs to go to the other
01:35:11: user that kind of complicates things and
01:35:13: then what happens if the user leaves and
01:35:15: needs to like have a system to hand off
01:35:16: the generation and maybe the user like
01:35:19: you know is like now the host is you
01:35:21: know taking care of that uh or maybe do
01:35:24: you want the user who's most recently
01:35:26: interacting with it to generate it. So
01:35:27: it's like more responsive that way. You
01:35:30: know that it carries like a lot of kind
01:35:32: of complications and just like saying
01:35:33: like okay the host is the one who
01:35:35: generates it no matter what. It
01:35:37: simplifies things. Um
01:35:40: and that's pretty much one of the main
01:35:41: reason because you don't have to like
01:35:42: worry about like a lot of these
01:35:44: complexities which like they're solvable
01:35:46: but you need to spend time solving them.
01:35:48: Um and original like inspectors were
01:35:51: like made like you know like kind of
01:35:53: quicker to implement but with the rework
01:35:56: the goal is going to be you know to make
01:35:57: them like uh make it more flexible.
01:36:04: Uh next questions from yummy dev tool. Y
01:36:07: def tools asking
01:36:09: uh so is official flags and component
01:36:13: documentation planned the current v is
01:36:14: basically useless for discovery of nodes
01:36:16: since most pages only describe a few
01:36:17: fields or nothing at all. Um we would
01:36:21: want to have it like you know done but
01:36:22: like people should like work on the w
01:36:24: because like if we even if we do make
01:36:26: like stuff in game um like we probably
01:36:29: source the initial version of it from
01:36:31: the w so like you know like the the
01:36:35: important thing here is like having
01:36:37: official documentation system
01:36:41: doesn't solve the fact that you need to
01:36:42: document it. It just changes when where
01:36:44: they're stored. Um
01:36:48: but like if if if if the documentation
01:36:50: isn't written, it doesn't matter what it
01:36:52: doesn't matter what system you use, you
01:36:54: still need to write the documentation.
01:36:56: So like ideally
01:36:58: um contribute to the wiki like uh that's
01:37:02: that's that's the thing that will help
01:37:03: the most with that.
01:37:08: Uh Satan dub is asking, can you tell the
01:37:10: story behind the bread part of the
01:37:12: birthday badge? Yes. So uh the way this
01:37:15: started is like when when I started like
01:37:17: working on like new sort of concept for
01:37:19: UI uh I was like you know thinking what
01:37:23: are you going to call it um and we
01:37:26: started like using like terminology for
01:37:28: like you know uh terminology for stuff
01:37:32: like you know that has to do with uh
01:37:34: gems you know stuff like minerals and so
01:37:36: on and
01:37:39: um one of the terminology I found is
01:37:42: like you know there's like a lot of
01:37:43: terminology for cuts and you know cuts
01:37:46: of like the gems and I like the you know
01:37:48: terminology like when we have gems they
01:37:51: have facets and that's you know why we
01:37:53: have like facets in the UI that's kind
01:37:55: of named with that theme um and
01:37:59: uh when we're like looking at different
01:38:00: cuts you know there's cuts like
01:38:01: something called like brilliant cut you
01:38:03: know and there's like specific way to
01:38:04: cut like a gem and there's also a
01:38:07: radiant cut which you know the UI has
01:38:10: been named after uh it's called like
01:38:12: gradient UI
01:38:13: because it kind of sounded cool. But
01:38:14: there's also like something called
01:38:16: baguette cut.
01:38:18: Um, and we're kind of joking like what
01:38:21: we just called it baguette UI and we're
01:38:23: like joking about it, but we're like
01:38:24: that's not a it's not actually a good
01:38:26: name, but it sounds funny. Um,
01:38:30: and
01:38:32: you know like um but we're like like you
01:38:35: know like we didn't end up using it for
01:38:36: that for UI name, but we're still like
01:38:38: you know like let's do something with
01:38:39: the bucket. We actually named we named
01:38:41: one of the B like internal bots. We
01:38:44: named it baguette just you know as a
01:38:46: funny and just kind of kept going and
01:38:48: then we're going like we would like to
01:38:50: do like an icon for like when somebody
01:38:51: has like anniversary the way you know
01:38:53: Reddit has a cake day. We wanted to like
01:38:55: you know something similar but you know
01:38:57: something that's not a cake because
01:38:59: Reddit's already doing that. So we're
01:39:00: like what if we do baguette and just
01:39:03: like stick like you know stick a candle
01:39:04: in it. And we're like okay well that's
01:39:06: funny let's do that. And it was like,
01:39:09: you know, one of the use cases where
01:39:10: it's like it's fine for it to be goofy
01:39:13: because, you know, it's just the goofy
01:39:15: thing like it's the anniversary of your
01:39:17: account. And then instead of calling it
01:39:18: baguette, like people started to call it
01:39:20: birthday because it sounds like
01:39:22: birthday, which works even better. And
01:39:24: does you know how this going to end up
01:39:26: like that?
01:39:33: Uh next question is from
01:39:36: KT. Uh will uh K is asking will the
01:39:41: spliting help certain user stream
01:39:43: without getting extreme lag and helping
01:39:45: user not have to enable lowest graphical
01:39:47: settings? So that kind of depends a bit
01:39:50: like on what why they're having issues.
01:39:54: Um spliting it helps reduce the CPU
01:39:57: workload. So like you know the stress on
01:39:59: your CPU is lower from the engine
01:40:00: itself. However, if you're streaming and
01:40:03: you're streaming like you know a camera
01:40:05: that actually you need to render you
01:40:07: know another view and that increases the
01:40:09: CPU load for the rendering part which is
01:40:11: still unity and it also increases load
01:40:14: on your GPU because your GPU now needs
01:40:16: to render your own view and needs to
01:40:18: render your camera view. So now the GPU
01:40:20: is doing a lot more. So if the GPU is
01:40:22: the bottleneck the splittening will
01:40:24: unfortunately not help. if the
01:40:26: bottlenecks your CPU, it should help
01:40:28: because it reduces the overall CPU load.
01:40:30: Um,
01:40:33: but like depending like it sounds like
01:40:35: you're saying if they enable those
01:40:36: graphical settings, um, it might not
01:40:39: help in those cases because from that it
01:40:42: sounds like it sounds like there are
01:40:45: issues, you know, more like the GPU is
01:40:46: not like keeping up with like having to
01:40:48: render two views. So the splitting
01:40:51: unfortunately will not help with that.
01:40:52: But ultimately, you have to try it. see
01:40:55: if it helps or not. Uh it depends on
01:40:57: what specific bottlenecks are.
01:41:01: Um
01:41:03: next question is from modify arc as
01:41:05: well. Uh what is your favorite esoteric
01:41:09: component?
01:41:10: What what esoteric components do we
01:41:12: have? Um let me think.
01:41:17: So maybe actually maybe I have one. Um,
01:41:22: it's going to do create new empty
01:41:23: object.
01:41:26: Uh,
01:41:28: it has to be something of the procedural
01:41:30: stuff.
01:41:32: So maybe
01:41:35: procedural
01:41:38: h I don't know if actually actually h I
01:41:41: don't know if it considered this
01:41:42: esoteric uh but I think can be it can be
01:41:46: procedural texture the simplex texture
01:41:49: 3D
01:41:51: I'm going to go with that one that's
01:41:53: relatively esoteric I've added like
01:41:54: procedural 3D textures because I needed
01:41:56: to test something and I was like I'm
01:41:58: just going to add like one for fun
01:41:59: because it's takes 5 minutes to write Uh
01:42:02: it was this simplex uh procedural like
01:42:04: 3D texture and then people have actually
01:42:07: made a really cool stuff with it uh
01:42:10: which I'm going to show you in a second.
01:42:14: Um people have made stuff like this.
01:42:19: Oh
01:42:20: >> how did I how did I spawn both?
01:42:23: >> No, no, no. I spawned one.
01:42:25: >> Oh, I was going to spawn. So, there's
01:42:28: this
01:42:30: and this is, you know, fully 3D.
01:42:36: So, I'm going to spawn the other one,
01:42:37: too. Uh,
01:42:40: so this one I'm going to make it bigger.
01:42:43: >> Watermelon.
01:42:44: >> Look, this is voluometric.
01:42:48: >> It's watermelon.
01:42:50: >> It's either watermelon vape juice or
01:42:52: like petroleum gas.
01:42:54: >> Yeah. And what's cool about this like
01:42:56: you know this is fully like this is
01:42:59: volutric
01:43:02: and
01:43:02: >> it looks super cool and like it runs
01:43:04: like decently well on like what it is.
01:43:08: >> Yeah, it's running on the CPU.
01:43:11: >> Yeah, this is being generated on the
01:43:12: CPU. It's also like the generator is
01:43:14: like synchronous and I think multi
01:43:17: thread. I forget. You could make it
01:43:19: super big.
01:43:24: ray marched too.
01:43:26: >> Yeah, this is being ray march as well.
01:43:33: So, it's one of those things like it
01:43:34: doesn't have like um
01:43:38: it doesn't have like you know very
01:43:41: it has very niche use cases. So I would
01:43:43: consider this camera as I mostly I
01:43:45: implemented it mostly you know for
01:43:48: testing purposes because I needed to
01:43:50: test
01:43:52: procedure like 3D textures um and also
01:43:56: actually I need to implement like
01:43:57: four-dimensional I need to implement
01:44:00: like four-dimensional uh
01:44:03: simplex noise for uh photon dust and I
01:44:07: was like I can just like you know do
01:44:08: like a I can do like a texture like
01:44:12: this. So, I implemented it and it's
01:44:14: super cool and people do cool things
01:44:15: with it. Uh, there's another version.
01:44:17: This one's kind of also neat like very
01:44:18: like it uses like very like low
01:44:21: resolution.
01:44:23: You can see this is also kind of like
01:44:25: it's 3D. It's voluometric.
01:44:30: So, yeah, that's um I'm going to go with
01:44:33: this one.
01:44:39: Okay. Uh, that's actually all the
01:44:40: questions. We got 15 minutes left still.
01:44:43: So if you got a few more questions, feel
01:44:45: free to ask. Um otherwise we might end
01:44:48: it a little bit early too.
01:44:53: Mater
01:44:57: wolf is saying giving me Star Trek
01:44:58: varcore vibes.
01:45:01: Apparently that's called the bit rate.
01:45:06: Um, check the fox asking, um, actually
01:45:09: what was the original use case for the
01:45:11: univerable box driver? Uh, that always
01:45:14: seemed oddly specific to me. Um, I think
01:45:17: I was building an experience where I
01:45:19: wanted to have like, you know, at the
01:45:21: end of the experience like you would get
01:45:22: like a gift and the gift would like
01:45:24: unwrap itself and would be like, you
01:45:25: know, nice and flashy like, you know,
01:45:27: you get like, oh, it opens up and you
01:45:29: have a thing in it that you can now take
01:45:30: and it's like animation. So, I made it
01:45:32: for that. Um, that's actually one thing
01:45:36: that I'm kind of like avoiding now
01:45:37: because like like like a lot of the
01:45:38: early extension stuff like it was it had
01:45:41: very like kind of monolithic and
01:45:42: specific behaviors and like I've kind of
01:45:45: shifted to like do stuff like you know
01:45:46: that's more generic
01:45:49: um kind of move away from that kind of
01:45:51: like
01:45:52: specific use case approach um generally
01:45:57: and like you know prefer like have like
01:45:59: more complex stuff kind of built in name
01:46:06: uh modify arc is asking
01:46:09: uh how difficult would it be to
01:46:11: implement decoupling each VRI from IPD
01:46:14: so we can emulate chameleon vision so it
01:46:17: would be too difficult but like we would
01:46:19: have to kind of disable certain
01:46:20: optimizations like single pass rendering
01:46:22: because like it kind of you know it has
01:46:25: some constraints on where the cameras
01:46:27: can be like it kind of requires them to
01:46:29: be in certain like position and kind the
01:46:31: VR part kind of controls that. We could
01:46:34: just like render like you know two
01:46:35: separate eyes. The
01:46:38: question there more is like is it worth
01:46:40: like spending time on because like
01:46:43: making the rendering be not one to one
01:46:46: to your eyes is like one of the big no
01:46:48: nos in VR like development. So like the
01:46:51: use cases for that like you know are
01:46:53: extremely niche. So, it's more like, you
01:46:55: know,
01:46:57: probably not going to spend time on it
01:46:58: because there's not really
01:47:01: many like there's not really like much
01:47:03: of a practical use case for that.
01:47:05: >> Do you want to puke?
01:47:07: >> Yeah. Like it's it's kind of like it'll
01:47:10: like it'll hurt your eyes.
01:47:19: Uh, next question. Uh yummy dev tools
01:47:22: asking if resonate was built for
01:47:24: education what would be best school
01:47:26: subject to teach in resonite anything. I
01:47:29: mean that's the thing with education is
01:47:31: like you know it's uh it's universal
01:47:33: like no matter what you want to teach
01:47:36: having the social aspect and having like
01:47:39: you know able to visualize stuff and
01:47:41: like work with stuff it improves all of
01:47:44: it. Like say like you wanted to teach
01:47:46: history, you could literally bring like
01:47:48: you know make immersive environments
01:47:50: like historical environments where you
01:47:52: could like you know watch some history
01:47:54: unfold or interact with it. If you want
01:47:56: to teach chemistry you can visualize you
01:47:57: know molecules interactions you could
01:47:59: make something interactive where you
01:48:00: grab the molecules play with them see
01:48:02: what kind of you know happens you know
01:48:04: stuff like that. If you want to teach
01:48:05: physics you can make virtual physics
01:48:07: experiments. So you get like intuition
01:48:09: like you build intuition for you know
01:48:12: how those physical mechanisms work. If
01:48:15: you want to teach programming you know
01:48:16: like we have protolex so like you know
01:48:18: you could you can program with that. Um
01:48:21: if you want to you know teach like art
01:48:24: we have brushes you know you could like
01:48:25: learn that way. Uh if you want to teach
01:48:28: mathematics you know that's related to
01:48:30: you know is related to like you know
01:48:32: program like protolex programming. So
01:48:34: you could do some of that and you could
01:48:35: also build like you know there's like
01:48:37: beautiful like visualization worlds like
01:48:38: some stuff built by Jim Ambe they made
01:48:41: like really cool mathematical
01:48:43: visualization worlds where you can
01:48:44: actually like see stuff visualized also
01:48:47: like um I remember there's like one more
01:48:48: called like um rose uh curved surfaces
01:48:52: by umbra umbraan
01:48:55: u that's also like excellent you know
01:48:57: educational like world like that shows
01:48:59: you mathematics the recent one in uh
01:49:03: that also on like you know uh an MMC was
01:49:06: the Enigma machine and it was also like
01:49:08: excellent world where you know you learn
01:49:12: some history behind the enigma machine
01:49:13: and they actually made a functional
01:49:15: version that you can use in Resonite and
01:49:17: as you're encrypting stuff with it. It
01:49:20: actually shows you how the current flows
01:49:21: through the machine so you can better
01:49:23: understand how it works you know and
01:49:25: it's like excellent like you can learn
01:49:26: engineering that way. Um there's been
01:49:29: you know underwarded I really like this
01:49:31: um
01:49:33: uh do you want by Axium Wolf like where
01:49:35: you actually learn how Martian cubes
01:49:38: algorithm works and they made like a
01:49:39: functional tool for it too but they also
01:49:41: have like a really cool kind of
01:49:42: visualizer for it or the other one was
01:49:44: like by Japanese community they made
01:49:46: like a geography puzzle where you can
01:49:48: like you know you can be pushing maps
01:49:50: and you know like you can be learning
01:49:52: like you know and reinforcing your
01:49:54: geography knowledge uh in a really fun
01:49:57: way. So like really anything like I
01:50:04: like I I really like like you know the
01:50:05: ESD category for like MMC. If you
01:50:07: actually like search that one there's a
01:50:09: lot of really cools. Um if I just uh go
01:50:14: let me just check ESD. If we check like
01:50:17: MMC and ESD there's like a whole bunch
01:50:19: of stuff. Um
01:50:23: yeah, there's like uh astronomy ones.
01:50:27: Oh, patient journey. That was also like
01:50:29: really cool one because uh um the
01:50:33: patient journey like this kind of like
01:50:35: walked people through like you know what
01:50:36: to expect if you kind of have like you
01:50:38: know certain like operations that you
01:50:41: need to have done and it kind of walks
01:50:42: you through the entire process and makes
01:50:43: it like easier to understand. So really
01:50:47: I feel like there's no limit like teach
01:50:51: everything, teach anything.
01:50:59: Uh next question is from uh Qualtorium.
01:51:02: Uh Quum is asking are there any kinds of
01:51:04: scientific or educational content you'd
01:51:06: personally like to see more of in the
01:51:08: Resonite? Oh yeah, I meant like just any
01:51:10: like like like I said like education
01:51:13: um
01:51:16: uh
01:51:19: education is like you know like one of
01:51:21: my kind of passions and like I like see
01:51:23: like improve on any fronts. I would say
01:51:26: in general um I really like like you
01:51:29: know stuff that's that let's focused on
01:51:32: visualization and like in building
01:51:35: intuition so stuff you know that's
01:51:37: interactive because um you only remember
01:51:40: like if you have if you hear a topic
01:51:42: explained you know there's only so much
01:51:45: to remember if you hear like you know if
01:51:48: you can see some visualization that
01:51:49: helps but being able to like you know
01:51:51: mess with those visualizations like you
01:51:53: know have something you can sort of like
01:51:55: you know poke this thing and see what
01:51:57: happens here and poke this thing see
01:51:58: what happens here and like sort of build
01:52:01: intuition on how things work. That is
01:52:03: the most powerful thing you can do for
01:52:05: learning things. I want to see more of
01:52:07: that like like one of the things you
01:52:09: know for example learning physics um
01:52:12: could be like you know you get
01:52:14: transported to like a laboratory of like
01:52:16: you know Isaac Newton and you're like
01:52:17: his assistant and like you know and you
01:52:20: help him with his experiments so you can
01:52:21: actually understand how some of the you
01:52:23: know how some of the stuff we know now
01:52:26: how was it discovered what kind of
01:52:28: experiments they have done in the past
01:52:30: that like you know let them to
01:52:32: understand what's going on like one of
01:52:34: one of the examples I really like is you
01:52:35: How they discovered, you know, infrared
01:52:38: light is because they would like they
01:52:39: were like playing with like, you know,
01:52:41: prisms and um and they were like, you
01:52:44: know, separate the light into different
01:52:45: colors and they actually put
01:52:47: thermometers, you know, in a row and
01:52:49: they noticed that different each
01:52:51: different um each like color actually
01:52:54: changes the temperature in at a
01:52:56: different rate. And they found like okay
01:52:58: like different different parts of the
01:53:00: spectrum actually carry a different
01:53:01: amount of energy. blood. They also had a
01:53:04: thermometer that was, you know, that
01:53:06: actually rose a lot in temperature, but
01:53:09: they didn't see any light there and it
01:53:10: was beyond, you know, red.
01:53:13: So they kind of figure out well there's
01:53:14: like probably like you know part of the
01:53:16: spectrum we can't see that's you know
01:53:18: now called like infrared that actually
01:53:20: carries a lot of thermal energy you know
01:53:23: and that's how they found out like you
01:53:24: know by doing simple experiments like
01:53:26: you know
01:53:28: uh in the past and I feel like like you
01:53:31: know like teaching things like you know
01:53:33: that way like going through the history
01:53:35: of it and being able to like you know
01:53:37: perform the experiments yourself that I
01:53:40: feel is like would be like the peak
01:53:42: education like how best you can learn.
01:53:44: And actually like this is like one of
01:53:46: the topics I kind of want to talk a lot
01:53:48: like more in depth in one of the
01:53:50: resonances. I'm kind of saving it for
01:53:52: like you know once I'm less tired from
01:53:54: the spliting. Um,
01:53:57: but I feel like this is this is this
01:54:00: approach to education
01:54:02: like you know is like really the best
01:54:04: you can do and there's like a lot that
01:54:06: kind of goes into that because like uh
01:54:08: there's excellent book called free to
01:54:10: learn by Peter Gray that collects a lot
01:54:12: of like research you know around how we
01:54:14: learn um how we learn naturally.
01:54:19: Um, yeah, I'm I'm I'm not going to go
01:54:22: like too deep because like I'm kind of
01:54:23: saving it for like some other time, but
01:54:24: I hope this kind of gives you some idea.
01:54:32: Uh, next question is from modify arc.
01:54:36: Uh, why does use the outdated DS
01:54:38: encryption library for offiscation and
01:54:40: why is offiscation necessary? Um so it's
01:54:44: not using it because like it's part like
01:54:46: one it's the part is kind of old and
01:54:48: also it's not used you know for actual
01:54:51: encryption. Um we're mostly like using
01:54:53: it because like it's quick it's fast.
01:54:56: Um,
01:54:58: and like you know like there's not
01:54:59: really security concerns with it because
01:55:01: like like obuscation unlike encryption
01:55:05: you're it's expected to be reversible
01:55:08: you know just with enough effort and
01:55:10: it's pretty much like you know why it's
01:55:11: there is like just to make
01:55:14: um just to make like you know reversing
01:55:16: some things a little bit more effort
01:55:18: than just kind of being in a completely
01:55:20: plain view. Um in part it's kind of like
01:55:23: the reason is a little bit weird because
01:55:24: like for some of the things we obiscate
01:55:26: them because people will see certain
01:55:28: like data that like is technically just
01:55:30: public data. It's fine data but like it
01:55:34: it will spook people just like seeing it
01:55:36: and we will get like you know requests
01:55:39: you know being like like oh like like
01:55:41: you know this looks scary like you know
01:55:42: this like this scary data string in
01:55:45: there and we're like it's fine. And it's
01:55:48: like, you know, it's not really that
01:55:49: this expect to be like, you know,
01:55:51: private information, but because it
01:55:53: spooks people, we're just going to push
01:55:55: it. The other one is actually for ref
01:55:58: hacking because we use auscation to kind
01:56:01: of like make it bit like make it fair
01:56:03: bit harder to do ref hacking. Um because
01:56:07: the idea is like you know um we want
01:56:10: people to like you know need to like if
01:56:13: they really want to do it we want you to
01:56:15: spend like you know a bit more effort to
01:56:18: do it. Um because when you do that like
01:56:21: like our hope is you know you realize
01:56:23: that like this is something that you're
01:56:26: not meant to do. you're kind of like,
01:56:27: you know, pushing against the system in
01:56:29: a way that you're not supposed to do and
01:56:32: we can't really stop you, but we want
01:56:36: like, you know, people to realize is,
01:56:38: you know, like maybe this is something
01:56:39: that's not intended because if it's like
01:56:42: too direct to do, if it's like too
01:56:43: simple, then like, you know, it it can
01:56:46: give more of an idea like this is an
01:56:47: unintended functionality. But by adding
01:56:49: the obuscation to it, you know, it gives
01:56:52: it like, you know,
01:56:55: like like we wanted to give more of a
01:56:57: vibe was like, you know, no, you're not
01:56:59: supposed to be doing this.
01:57:02: Um, you can like if you really want to,
01:57:05: but you're supposed to. There's like,
01:57:07: you know, caveats and there's like, you
01:57:09: know, this is probably going to break.
01:57:11: This is not an intended behavior of the
01:57:13: system. Uh, and it's kind of like, you
01:57:15: know, why we kind of offiscate it.
01:57:22: Uh and uh let's see. Got uh pretty much
01:57:27: going to wrap up, but um criers are
01:57:30: asking
01:57:32: uh was any vacation? Maybe. We'll see.
01:57:36: As usually stuff kind of comes up, so
01:57:41: we'll see. Anyway, um I think we can
01:57:45: like end it here because we got like two
01:57:48: minutes left.
01:57:49: >> Um unless there's like super quick
01:57:52: question, but uh I don't think like
01:57:54: we've got time for anything else.
01:57:55: >> No, let's just end it.
01:57:57: >> Actually, no, this was quick. Um
01:58:01: >> Monif is asking any upcoming you are
01:58:04: going to um I'm going to be going to
01:58:06: Europe friends this year and also BLFC
01:58:08: very likely.
01:58:12: and one like this little to like s like
01:58:14: travel. So like unless like you know
01:58:15: something happens like there should be
01:58:17: on both and hopefully mF2.
01:58:21: All right, now we can end it.
01:58:23: >> Okay. Uh so anyway, thank you everyone
01:58:25: for joining the stream. Uh I hope you've
01:58:29: enjoyed like you know like u asking
01:58:32: questions and like listen to the answers
01:58:33: and so on. Um
01:58:37: um and like you know like thank you
01:58:39: everyone for like you know supporting
01:58:40: the platform as well whether it's like
01:58:42: you know just being part of it making
01:58:43: cool content or supporting us you know
01:58:45: financially either through Patreon or
01:58:47: Stripe. Uh also if you're on if you're
01:58:49: on Patreon please consider switching to
01:58:51: Stripe because we will get uh you will
01:58:54: get about 10% more from the same amount
01:58:56: of money uh which helps us a lot. Um
01:59:00: so um thank you very much. Uh let's see
01:59:03: if we have like anybody
01:59:08: anybody to raid. Uh I need to load this
01:59:11: up
01:59:13: because like you know different
01:59:15: different streamers like if you're
01:59:16: streaming around this time this might be
01:59:19: uh this might be like a good time to do
01:59:21: it. Uh I see creator jam so probably
01:59:24: send you their way. Um and also like
01:59:28: last thing like you know um if you
01:59:32: uh if you want like know more we'll be
01:59:33: like here like next week and there's
01:59:34: also like other office hours you know by
01:59:36: the team members like throughout the
01:59:38: week. So check out our discord um
01:59:41: there's like you know the events tab and
01:59:43: you can find like when the other office
01:59:44: hours are happening. So thank you again
01:59:46: very much and we will see you next week.
01:59:50: Uh
01:59:52: and say hi to Medra for us. So I'm going
01:59:55: to do raid
01:59:58: and also thank you sir for helping
01:59:59: answer questions too.
02:00:03: >> See you fellas later.
02:00:05: >> See you later
02:00:08: the gather
02:00:11: the uh
02:00:13: to say hi to mother
02:00:16: and they're doing like cool things at
02:00:18: the creator jam.
02:00:20: Okay. Byebye.
02:00:23: Ah,
02:00:26: raid.