<< 07-01-2025 >>
00:01:22FromDiscord<grumblygibson> In reply to @meta_leap "well _not_ compared to": Oh there's also pretty good support for hot code reloading, which might be what you're looking for to get blazing fast iteration builds. Ask around in the game dev channel for more.
00:02:35FromDiscord<meta_leap> In reply to @grumblygibson "Oh there's also pretty": now we're talking! wow you're gettin' me all hopeful here 🥳
00:02:48FromDiscord<System64 ~ Flandre Scarlet> Some progress! https://media.discordapp.net/attachments/371759389889003532/1325977902193574041/image.png?ex=677dc0a7&is=677c6f27&hm=22cfe93b9df86cad92be70061645dc6334430a64a6be85fe6f3157a5f115294b&
00:03:35FromDiscord<grumblygibson> It's not plug and play, but a lot has been done with your use case in mind, and great examples exist already folks in that channel will know.
00:03:55FromDiscord<System64 ~ Flandre Scarlet> @meta_leap I've written this in Nim btw
00:04:31FromDiscord<meta_leap> In reply to @sys64 "<@1131867449021702255> I've written this": nice, an imgui it looks like, is it dear imgui or sth in nim?
00:04:41FromDiscord<System64 ~ Flandre Scarlet> In reply to @meta_leap "nice, an imgui it": Dear ImGUI
00:05:26FromDiscord<meta_leap> In reply to @sys64 "Dear ImGUI": top-notch, this all bodes well for gettin my feet wet here
00:08:44FromDiscord<Elegantbeef> I'm so happy that my compile time for my game is like 4s makes my hot code reload actually somewhat usable 😄↵(@meta_leap)
00:09:43FromDiscord<System64 ~ Flandre Scarlet> In reply to @meta_leap "top-notch, this all bodes": It's a nice language
00:13:23FromDiscord<meta_leap> In reply to @Elegantbeef "I'm so happy that": 4s sounds decent! that's in line with starting a play-in-editor session in say unreal or godot. what's your graphics story, using a c/c++ rendering lib or coded from scratch?
00:13:42FromDiscord<meta_leap> (edit) "decent!" => "pretty good!"
00:18:21FromDiscord<Elegantbeef> Opengl + sdl2
00:19:49FromDiscord<Elegantbeef> Well with assimp and miniaudio, probably could reduce it down if I dynamically linked miniaudio
01:10:51FromDiscord<.tokyovigilante> In reply to @PMunch "Hmm, so `packed` is": Solved it just by casting to a raw `pointer` to pass to the `RenderDrawData` function, not sure why it works, but seems to...
01:11:17FromDiscord<.tokyovigilante> For pointer arithmetic, is this sort of thing the best approach? https://gist.github.com/oltolm/1738289b5ac866ec6a7e4ef20095178e
01:19:14FromDiscord<Elegantbeef> If it's an array no, just use `ptr UncheckedArray`
01:20:17FromDiscord<Elegantbeef> Shttps\://github.com/kaushalmodi/ptr\_math also might tickle your fancy with more features
01:20:27FromDiscord<Elegantbeef> Heh https://github.com/kaushalmodi/ptr_math there
01:26:05FromDiscord<meta_leap> how does one report this sorta thing, surely not as GH issue to nim repo =) `play.nim-lang.org` works with C compiler but keeps `502`ing with C++ compiler
01:26:57FromDiscord<Elegantbeef> https://github.com/PMunch/nim-playground I believe it's here
01:54:27FromDiscord<meta_leap> In reply to @Elegantbeef "https://github.com/PMunch/nim-playground I believe ": sorry deleted my post b4 your reply when I saw it was back up, but thx anyway =)
01:55:29FromDiscord<Elegantbeef> Atleast I no longer look like a raving lunatic
02:39:49FromDiscord<.tokyovigilante> In reply to @Elegantbeef "If it's an array": makes sense, thanks. Its just a bunch of untyped memory, so UncheckedArray[char] might do.
02:40:20FromDiscord<Elegantbeef> `byte` is more accurate
02:40:54FromDiscord<.tokyovigilante> oh yup, ta
03:08:15*rockcavera quit (Remote host closed the connection)
06:06:36*coldfeet joined #nim
06:35:42*coldfeet quit (Quit: Lost terminal)
07:31:57PMunch@.tokyovigilante, I wasn't actually responding to anything, had my own issues with `sizeof` :P
08:17:28FromDiscord<.tokyovigilante> In reply to @PMunch "@.tokyovigilante, I wasn't actually": No drama, I did think it didn't quite make sense, was just being polite 😉
08:19:35PMunchAlways a good policy :)
09:15:17FromDiscord<arnetheduck> In reply to @grumblygibson "<@1131867449021702255> the transpiling to": > the transpiling to C or C++ code is actually very fast.↵not really - it grows with the application size, and fast - this step alone takes several minutes for a medium-sized project like nimbus
09:16:21FromDiscord<arnetheduck> (edit) "In reply to @grumblygibson "<@1131867449021702255> the transpiling to": > the transpiling to C or C++ code is actually very fast.↵not really - it grows with the application size, and fast - this step alone takes several minutes for a medium-sized project like nimbus ... " added "- it will largely depend on how many nim libraries you use"
09:33:17FromDiscord<meta_leap> In reply to @arnetheduck "> the transpiling to": > this step alone takes several minutes for a medium-sized project like nimbus↵ouch! I gotta check out that project's size... which repo is it, GH has too many results for search `nimbus language:Nim`
09:36:43FromDiscord<nnsee> fwiw, I've never had a project so large that the codegen step of compilation takes more than 30 seconds
09:41:00FromDiscord<nnsee> arnetheduck tends to work on some BIG projects
09:41:04FromDiscord<meta_leap> In reply to @nnsee "fwiw, I've never had": well somehow @ElegantBeef reported a build time of 4s for his game project, which I'd hope is more than "Hello Triangle" — @ElegantBeef have a repo for that, or is private? — @System64 ~ Flandre Scarlet same question for your project, and what's the build time for yours?
09:54:11FromDiscord<arnetheduck> In reply to @meta_leap "> this step alone": https://github.com/status-im/nimbus-eth2/
09:58:23FromDiscord<arnetheduck> if you use nim as a glue language, it's decently fast, but if you start actually writing native nim code, compile times go up quite drastically (like most compiled languages)
10:01:57FromDiscord<Elegantbeef> Eh my project is a simple game so it's not that much code really
10:02:46FromDiscord<Elegantbeef> All the deps and everything is only 161k LoC
10:18:52*m5zs7k quit (Ping timeout: 252 seconds)
10:22:31FromDiscord<System64 ~ Flandre Scarlet> In reply to @meta_leap "well somehow <@145405730571288577> reported": Humm 5-10s I guess?
10:24:23FromDiscord<michael.lenz.> On rpi v1 it takes around 20 minutes 😂 because of small ram and swap
10:25:17*m5zs7k joined #nim
10:38:57*beholders_eye joined #nim
11:02:30FromDiscord<jmgomez> In reply to @arnetheduck "> the transpiling to": @meta_leap this is true. But if you are binding a C++ engine you may want to take a look at some tricks we did with NimForUE. Namely to use PCH files for the C/++ compilation and to export/reimport automatically generated bindings to skip the sem of procs bodies by the nim compiler (this should be as fast as IC can be if not faster)
11:04:13FromDiscord<michael.lenz.> Tested now with my hugest project, so it’s about 2.45 seconds
11:19:12*beholders_eye quit (Ping timeout: 246 seconds)
11:21:16*beholders_eye joined #nim
11:36:30*beholders_eye quit (Quit: WeeChat 4.1.2)
11:47:49*beholders_eye joined #nim
12:13:25*alexdaguy joined #nim
12:37:51*anddam joined #nim
12:53:30*alexdaguy quit (Quit: WeeChat 4.5.1)
14:12:04FromDiscord<System64 ~ Flandre Scarlet> I wonder about something↵Would truncating decimals in a float (not strings!) make less errors?
14:44:27PMunchLess errors than what?
14:49:15FromDiscord<System64 ~ Flandre Scarlet> In reply to @PMunch "Less errors than what?": I general, because doing operations with floats can give errors
14:49:20FromDiscord<System64 ~ Flandre Scarlet> (edit) "In reply to @PMunch "Less errors than what?": I general, because doing operations with floats can give ... errors" added "rounding"
15:02:47FromDiscord<tomornot> Hi, y'all! I've usually used arc with Nim for wasm. Does anyone have experience trying orc?
15:02:58PMunchWell drooping information won't give you better precision
15:05:09FromDiscord<Robyn [She/Her]> In reply to @tomornot "Hi, y'all! I've usually": no but I'd assume it'd work fine, try it and see?
15:08:38FromDiscord<odexine> In reply to @sys64 "I wonder about something": You cannot really do this because floats operate in binary and not decimal
15:10:11FromDiscord<meta_leap> In reply to @jmgomez "<@1131867449021702255> this is true.": thx, encouraging stuff indeed! way impressed with your Nim4UE! PCHs is understood, the rest I hope I'll grok better having gone thru the Nim manual in full and with some practice. what's "IC" here?
15:10:45FromDiscord<meta_leap> (edit) removed "better"
15:10:54FromDiscord<meta_leap> (edit) "In reply to @jmgomez "<@1131867449021702255> this is true.": thx, encouraging stuff indeed! way impressed with your Nim4UE! PCHs is understood, the rest I hope I'll grok ... having" added "once"
15:17:54*xet7 quit (Remote host closed the connection)
15:19:19*xet7 joined #nim
15:35:36*xet7 quit (Remote host closed the connection)
15:39:12*xet7 joined #nim
15:42:44*xet7 quit (Remote host closed the connection)
15:44:26*xet7 joined #nim
15:51:51*xet7 quit (Remote host closed the connection)
15:54:48*xet7 joined #nim
16:00:47FromDiscord<michaelb.eth> In reply to @meta_leap "thx, encouraging stuff indeed!": Incremental Compilation
16:04:35*FromDiscord quit (Remote host closed the connection)
16:04:49*FromDiscord joined #nim
16:10:09FromDiscord<tomornot> In reply to @battery.acid.bubblegum "no but I'd assume": Thanks for the reply!
16:15:28FromDiscord<tomornot> By the way, here's a video about a recent project calling into Nim from Odin in a 2-language Wasm app. As I mentioned, I used ARC here, but maybe I should try out ORC sometime and take a closer look at how ORC works. https://youtu.be/RI--EAWScuU
16:15:49FromDiscord<tomornot> What kicks off ORC cycle checking?
16:27:29*xet7 quit (Quit: Leaving)
16:40:41*xet7 joined #nim
16:57:33FromDiscord<bubbly_avocado_86424> In reply to @sys64 "I general, because doing": Good point, maybe this is a way to assure rounding errors don't get in the way with processing the data suggested on screen. https://nim-lang.org/docs/math.html#round%2Cfloat64
17:03:25*xet7_ joined #nim
17:05:17*xet7_ quit (Remote host closed the connection)
17:06:24*xet7 quit (Ping timeout: 246 seconds)
17:14:06FromDiscord<System64 ~ Flandre Scarlet> In reply to @bubbly_avocado_86424 "Good point, maybe this": Wait, round can do this???↵I thought it only rounded to the nearest integer!
17:15:53FromDiscord<bubbly_avocado_86424> In reply to @sys64 "Wait, round can do": rtfm is your friend
17:17:55FromDiscord<System64 ~ Flandre Scarlet> In reply to @bubbly_avocado_86424 "rtfm is your friend": However seems it doesn't fully prevent float errors
17:46:31*ntat joined #nim
17:51:22*coldfeet joined #nim
17:51:47FromDiscord<bubbly_avocado_86424> @System64 ~ Flandre Scarlet what's a float error to you ?
17:52:39FromDiscord<bubbly_avocado_86424> (edit) "?" => "?↵↵https://en.wikipedia.org/wiki/Floating-point_error_mitigation"
17:55:45FromDiscord<System64 ~ Flandre Scarlet> In reply to @bubbly_avocado_86424 "<@380360389377916939> what's a float": sin(x) + sin(x + pi) != 0
17:56:14FromDiscord<bubbly_avocado_86424> yeah well, that's asking for trouble
17:56:25FromDiscord<bubbly_avocado_86424> limit PI = fixed ?
17:56:51FromDiscord<System64 ~ Flandre Scarlet> wdym?
17:57:24FromDiscord<bubbly_avocado_86424> Pi is either limited by the compiler or by the library used, otherwise it is a non-ending calculation.
17:58:27FromDiscord<bubbly_avocado_86424> so, if you find what 'size' of Pi you're able to work with (say 100 decimals) you can account for that in the programming logic
18:00:22FromDiscord<bubbly_avocado_86424> or, as the AI puts it↵↵The issue you’re encountering with sin(x) + sin(x + pi) not being exactly zero is due to the limitations of floating-point arithmetic in computers. The value of pi used in computations is an approximation, not the exact mathematical constant π. This approximation introduces small errors that can accumulate and become noticeable in certain calculations.
18:01:33FromDiscord<System64 ~ Flandre Scarlet> I must be careful
18:01:45FromDiscord<bubbly_avocado_86424> Hilarious, the AI has actually come up with something ↵↵https://search.brave.com/search?q=how+to+work+around+sin%28x%29+%2B+sin%28x+%2B+pi%29+%21%3D+0+in+nim&source=web&conversation=f91457ed2fa3305f0f7fae&summary=1
18:03:23FromDiscord<bubbly_avocado_86424> @System64 ~ Flandre Scarlet i think you're surpassing the requirements 🙂
18:04:43FromDiscord<System64 ~ Flandre Scarlet> In reply to @bubbly_avocado_86424 "Hilarious, the AI has": Ah yeah I also use a tolerence
18:06:18FromDiscord<bubbly_avocado_86424> is the extreme accuracy you appear to be looking for viable in the long term ?
18:09:31FromDiscord<bubbly_avocado_86424> @System64 ~ Flandre Scarlet the TLC you have for this project is admirable, I very much like the attitude
18:11:51*xaltsc joined #nim
18:12:43FromDiscord<sonicsmooth.> I'm using wNim. I have a wFrame holding a wPanel holding another wPanel. If I connect the frame Event_Size then the size event is not propagated to child panels. If I do childpanel.ProcessEvent(event) then the child panel gets the event but size doesn't change. How do I process event in wFrame then let the event continue as normal to the child wPanels?
18:44:19*beholders_eye quit (Ping timeout: 252 seconds)
18:46:10*beholders_eye joined #nim
18:52:21*valms19901 joined #nim
18:52:27*coldfeet quit (Quit: Lost terminal)
18:56:52*zgasma joined #nim
19:28:21FromDiscord<System64 ~ Flandre Scarlet> In reply to @bubbly_avocado_86424 "<@380360389377916939> the TLC you": What does TLC mean?
19:28:31FromDiscord<bubbly_avocado_86424> 😄 Tender Loving Care
19:29:06FromDiscord<System64 ~ Flandre Scarlet> In reply to @bubbly_avocado_86424 "😄 Tender Loving Care": Oh↵Thanks! This is the 5th version of this project
19:40:52FromDiscord<bubbly_avocado_86424> @System64 ~ Flandre Scarlet reminds me of PD/PureData
19:43:43FromDiscord<System64 ~ Flandre Scarlet> In reply to @bubbly_avocado_86424 "<@380360389377916939> reminds me": Not the first time that someone says that 🤣
19:44:03FromDiscord<bubbly_avocado_86424> @System64 ~ Flandre Scarlet is there a repo or blog ?
19:46:39FromDiscord<System64 ~ Flandre Scarlet> In reply to @bubbly_avocado_86424 "<@380360389377916939> is there a": https://github.com/system64MC/Kurumi-XP
19:48:44FromDiscord<bubbly_avocado_86424> In reply to @sys64 "https://github.com/system64MC/Kurumi-XP": with visual style like that, include some screenshots on the repoes 🙂
20:07:01FromDiscord<System64 ~ Flandre Scarlet> In reply to @bubbly_avocado_86424 "with visual style like": I will↵It's just early in development↵But yeah, screenshots would make the project very attractive
20:16:58*ntat quit (Read error: Connection reset by peer)
20:18:50*ntat joined #nim
20:33:24*ntat quit (Quit: Leaving)
20:49:01*valms19901 quit (Ping timeout: 244 seconds)
20:53:24FromDiscord<System64 ~ Flandre Scarlet> New module! https://media.discordapp.net/attachments/371759389889003532/1326292628458700810/image.png?ex=677ee5c3&is=677d9443&hm=cc1f07c76eaea8969edd2b81df5628b746d59f666ecf4ddc8cad9fa5b4fd8d23&
20:53:42FromDiscord<System64 ~ Flandre Scarlet> Additive synthesis!
21:11:41*ntat joined #nim
21:23:01*ntat quit (Quit: Leaving)
21:27:24FromDiscord<michael.lenz.> Add ir support!
21:28:30FromDiscord<michael.lenz.> Tweaking harmonics levels is too hard for having good sounding results
21:29:46FromDiscord<michael.lenz.> I like how ui looks like. Nice design
21:41:18*beholders_eye quit (Ping timeout: 252 seconds)
21:51:28*rockcavera joined #nim
23:04:29*xet7 joined #nim
23:04:44*xet7 quit (Remote host closed the connection)
23:05:39*xet7 joined #nim