#programming — Public Fediverse posts on home.social

Karthikeyan A K :firefox: @mindaslab · 2026-09-23 · 17:19 UTC

Very nice to find out I'm kinda discovering Lambda Calculus

https://k.libl.in/M

I have no clue what it is 😂😂

#Programming #Logic #Lisp #Clojure

#programming #logic #lisp #clojure

Christian Noll @vnzn · 2026-09-23 · 17:04 UTC

Developers use Macs (heise.de)

https://www.heise.de/en/news/Developers-use-Macs-11462336.html

#Linux #macos #windows #Microsoft #development #programming

#programming #development #microsoft #windows #macos #linux

mnl mnl mnl mnl mnl @mnl · 2026-09-23 · 13:05 UTC

I really enjoy how much pedagogical “engineering” there is in #mathematics books (even very specialized ones), because I think that, besides obv being strongly academic (and thus requiring practitioners to teach), mathematics as a collaborative activity is pedagogical as well (and they’re definitely not the simplest concepts to communicate).

That makes all the maths books about the same / very related subjects very “personal”, with people caring and refining and focusing more on the order of things and how they introduce / anticipate / build upon others, in a myriad of ways.

Coming from #software_engineering of #programming (and amateur CS), which is also about “learning and teaching” when working on a team, I haven’t seen anywhere close to the same attention to “how should I teach something like X” (vs here’s how I describe the thing that is, which arguably should be the code itself. That we even need documentation shows the need for pedagogy, but it isn’t conceived as such).

Part of it is that software engineering is to large part artisanal and subject to the pressures of the development cycle, while maths has the advantage to be perennial (crossing centuries if not millennia, even), but so are many things in software once you squint a bit.

I have over time built up a little arsenal for introducing “the programmer’s monad pattern”, be it either a sequence of examples, or starting with the types, or pointing out “literally the shape of the code” and apply them differently based on who I am talking to, and that highly mathematical construct is maybe the one used in practice that programmers discuss a lot at the “meta-level” (meaning, patterns to teach a pattern). There’s is of course a lot around functions and objects and types and the like, but even here textbooks feel mostly “rote” and don’t have the variety of styles I am starting to appreciate in maths books (I can think of sicp and “the little …” and whys poignant guide and zine culture as going in that direction).

Anyway, morning musings

#mathematics #software_engineering #programming

Driftya @[email protected] · 2026-09-23 · 09:14 UTC

Here is demonstration of Code Meridian trying it on FizzBuzz challenge,

It is to keep the human in the loop when developing with #AI

First I picked the wrong answer and it gives a explanation why it is wrong.

Then I picked correct answer, as you can see we can write a note for the future which is stored in graph.

This way we can always #learn from the code that #LLM generated, but also learn just like studying for a driving license.

#MCP #dotnet #typescript #programming #developer

#ai #learn #llm #mcp #dotnet #typescript

Driftya @[email protected] · 2026-09-23 · 09:14 UTC

Here is demonstration of Code Meridian trying it on FizzBuzz challenge,

It is to keep the human in the loop when developing with #AI

First I picked the wrong answer and it gives a explanation why it is wrong.

Then I picked correct answer, as you can see we can write a note for the future which is stored in graph.

This way we can always #learn from the code that #LLM generated, but also learn just like studying for a driving license.

#MCP #dotnet #typescript #programming #developer

#ai #learn #llm #mcp #dotnet #typescript

Driftya @[email protected] · 2026-09-23 · 09:14 UTC

Here is demonstration of Code Meridian trying it on FizzBuzz challenge,

It is to keep the human in the loop when developing with #AI

First I picked the wrong answer and it gives a explanation why it is wrong.

Then I picked correct answer, as you can see we can write a note for the future which is stored in graph.

This way we can always #learn from the code that #LLM generated, but also learn just like studying for a driving license.

#MCP #dotnet #typescript #programming #developer

#ai #learn #llm #mcp #dotnet #typescript

Driftya @[email protected] · 2026-09-23 · 09:14 UTC

Here is demonstration of Code Meridian trying it on FizzBuzz challenge,

It is to keep the human in the loop when developing with #AI

First I picked the wrong answer and it gives a explanation why it is wrong.

Then I picked correct answer, as you can see we can write a note for the future which is stored in graph.

This way we can always #learn from the code that #LLM generated, but also learn just like studying for a driving license.

#MCP #dotnet #typescript #programming #developer

#developer #programming #typescript #dotnet #mcp #llm

Driftya @[email protected] · 2026-09-23 · 09:14 UTC

Here is demonstration of Code Meridian trying it on FizzBuzz challenge,

It is to keep the human in the loop when developing with #AI

First I picked the wrong answer and it gives a explanation why it is wrong.

Then I picked correct answer, as you can see we can write a note for the future which is stored in graph.

This way we can always #learn from the code that #LLM generated, but also learn just like studying for a driving license.

#MCP #dotnet #typescript #programming #developer

#ai #learn #llm #mcp #dotnet #typescript

jadedctrl :programo: :korektu_min: @[email protected] · 2026-09-23 · 04:14 UTC

I’ve made my first :emacs: Emacs package! :taneŝima_ĝojas:

https://kod.ujo.moe/jadedctrl/bookmark-groups.el

I really do need to split out some of my init.el into more packages, more of my code is pretty generic, like this :gutkato_hihias:

#lang_en #emacs #programming

#emacs #lang_en #programming

jadedctrl :programo: :korektu_min: @[email protected] · 2026-09-23 · 04:14 UTC

I’ve made my first :emacs: Emacs package! :taneŝima_ĝojas:

https://kod.ujo.moe/jadedctrl/bookmark-groups.el

I really do need to split out some of my init.el into more packages, more of my code is pretty generic, like this :gutkato_hihias:

#lang_en #emacs #programming

#emacs #lang_en #programming

jadedctrl :programo: :korektu_min: @[email protected] · 2026-09-23 · 04:14 UTC

I’ve made my first :emacs: Emacs package! :taneŝima_ĝojas:

https://kod.ujo.moe/jadedctrl/bookmark-groups.el

I really do need to split out some of my init.el into more packages, more of my code is pretty generic, like this :gutkato_hihias:

#lang_en #emacs #programming

#emacs #lang_en #programming

jadedctrl :programo: :korektu_min: @[email protected] · 2026-09-23 · 04:14 UTC

I’ve made my first :emacs: Emacs package! :taneŝima_ĝojas:

https://kod.ujo.moe/jadedctrl/bookmark-groups.el

I really do need to split out some of my init.el into more packages, more of my code is pretty generic, like this :gutkato_hihias:

#lang_en #emacs #programming

#programming #lang_en #emacs

jadedctrl :programo: :korektu_min: @[email protected] · 2026-09-23 · 04:14 UTC

I’ve made my first :emacs: Emacs package! :taneŝima_ĝojas:

https://kod.ujo.moe/jadedctrl/bookmark-groups.el

I really do need to split out some of my init.el into more packages, more of my code is pretty generic, like this :gutkato_hihias:

#lang_en #emacs #programming

#emacs #lang_en #programming

Elias Ebner @[email protected] · 2026-09-23 · 00:52 UTC

I folded, I'm looking into emacs (as a vim user).

#emacs #vim #coding #programming

Elias Ebner @[email protected] · 2026-09-23 · 00:52 UTC

I folded, I'm looking into emacs (as a vim user).

#emacs #vim #coding #programming

Elias Ebner @[email protected] · 2026-09-23 · 00:52 UTC

I folded, I'm looking into emacs (as a vim user).

#emacs #vim #coding #programming

Elias Ebner @[email protected] · 2026-09-23 · 00:52 UTC

I folded, I'm looking into emacs (as a vim user).

#emacs #vim #coding #programming

#programming #coding #vim #emacs

Elias Ebner @[email protected] · 2026-09-23 · 00:52 UTC

I folded, I'm looking into emacs (as a vim user).

#emacs #vim #coding #programming

Enalys :dragn_verified: @[email protected] · 2026-09-22 · 22:52 UTC

Can't wait for the AIGen bubble to pop and see all the FOSS projects who embraced the slop machine crumbling to clean the mess they introduced because “Ai Is ThE fUtUrE”.

It will be glorious and funny. :dracthyr_tea:

#AI #FOSS #Programming

#ai #foss #programming

Frontend Dogma @[email protected] · 2026-09-22 · 21:40 UTC

Inside a Software Factory, by @pauliusztin.wsocial.eu (@OReillyMedia):

https://www.oreilly.com/radar/inside-a-software-factory/?ref=frontenddogma.com

#programming #concepts #ai #linklists

#linklists #ai #concepts #programming

RS, Author, Novelist, Prosaist @[email protected] · 2026-09-22 · 16:39 UTC

Yep. Installed Apple's #iPadOS 27 last night. Immediately found interface design faux pas that effect use.

Look at the screenshot carefully. It's showing Windowed App Mode. The two windows to the left (Digital Trends website and Mastodon) are Safari web apps created in iPadOS 27 using Add to Home Screen to ensure nothing from a previous release influences rendering.

Note that the stoplight controls overlap browser window content.
In both of these cases, because of the design of the web page (which should not matter because these are essentially the contents of a Safari tab), the stoplight control covers a link/button!
In the #Mastodon web app window, it prevents using the link/button that it covers.
In the Safari slide-over window, note that there are tabs and caption bar buttons that always push down the content in the browser below the stoplight controls.
Notes, shown an app example, has a stoplight control. It is neatly isolated on the caption bar from the content below. It seems only that the web apps have this flaw.

Opps! The dev design team and Q/A missed that web apps created by the Add to Home Screen feature need a caption bar, or something—for example a web app title?—not to cover web content. Not including one causes a UI malfunction.

While on the topic of overlapping content:

Notice that the magnifier button on Spotlight.
The transparent button overlaps, very slightly, the content of its search field.

What purpose does that serve, other than being unnerving visually? Looks very much like a design error. Or are we showing off how neat button transparency really is?

Last, I brought up the menu bar in the DigitalTrends window.

The web is no longer the top window, but the menu persists. Is this a bug?
Worse (visually anyway), with a long caption name, in white of all colors, when "deactivated" such that the Liquid Glass effect goes away, the white menu is still vaguely visible across the top of the Notes window. Look closely at the full-sized image posted here; squint if you have to: white letters on a white window background!?
Not sure if the caption area of Notes is transparent (though I've used every means to disable the transparency of Liquid Glass) or if the menu is overlapping the caption. Probably the former.

But really, ick!

Come on #Apple! "Crack design crew," this is aimed at you: Embarrassing!

#BoostingIsSharing

#iPad #iPadOS27 #liquidglass #UIdesign #programming

#mastodon #ipados #apple #boostingissharing #ipad #ipados27

RS, Author, Novelist, Prosaist @[email protected] · 2026-09-22 · 16:39 UTC

Yep. Installed Apple's #iPadOS 27 last night. Immediately found interface design faux pas that effect use.

Look at the screenshot carefully. It's showing Windowed App Mode. The two windows to the left (Digital Trends website and Mastodon) are Safari web apps created in iPadOS 27 using Add to Home Screen to ensure nothing from a previous release influences rendering.

Note that the stoplight controls overlap browser window content.
In both of these cases, because of the design of the web page (which should not matter because these are essentially the contents of a Safari tab), the stoplight control covers a link/button!
In the #Mastodon web app window, it prevents using the link/button that it covers.
In the Safari slide-over window, note that there are tabs and caption bar buttons that always push down the content in the browser below the stoplight controls.
Notes, shown an app example, has a stoplight control. It is neatly isolated on the caption bar from the content below. It seems only that the web apps have this flaw.

Opps! The dev design team and Q/A missed that web apps created by the Add to Home Screen feature need a caption bar, or something—for example a web app title?—not to cover web content. Not including one causes a UI malfunction.

While on the topic of overlapping content:

Notice that the magnifier button on Spotlight.
The transparent button overlaps, very slightly, the content of its search field.

What purpose does that serve, other than being unnerving visually? Looks very much like a design error. Or are we showing off how neat button transparency really is?

Last, I brought up the menu bar in the DigitalTrends window.

The web is no longer the top window, but the menu persists. Is this a bug?
Worse (visually anyway), with a long caption name, in white of all colors, when "deactivated" such that the Liquid Glass effect goes away, the white menu is still vaguely visible across the top of the Notes window. Look closely at the full-sized image posted here; squint if you have to: white letters on a white window background!?
Not sure if the caption area of Notes is transparent (though I've used every means to disable the transparency of Liquid Glass) or if the menu is overlapping the caption. Probably the former.

But really, ick!

Come on #Apple! "Crack design crew," this is aimed at you: Embarrassing!

#BoostingIsSharing

#iPad #iPadOS27 #liquidglass #UIdesign #programming

#ipados #apple #boostingissharing #ipad #ipados27 #liquidglass

RS, Author, Novelist, Prosaist @[email protected] · 2026-09-22 · 16:39 UTC

Yep. Installed Apple's #iPadOS 27 last night. Immediately found interface design faux pas that effect use.

Look at the screenshot carefully. It's showing Windowed App Mode. The two windows to the left (Digital Trends website and Mastodon) are Safari web apps created in iPadOS 27 using Add to Home Screen to ensure nothing from a previous release influences rendering.

Note that the stoplight controls overlap browser window content.
In both of these cases, because of the design of the web page (which should not matter because these are essentially the contents of a Safari tab), the stoplight control covers a link/button!
In the #Mastodon web app window, it prevents using the link/button that it covers.
In the Safari slide-over window, note that there are tabs and caption bar buttons that always push down the content in the browser below the stoplight controls.
Notes, shown an app example, has a stoplight control. It is neatly isolated on the caption bar from the content below. It seems only that the web apps have this flaw.

Opps! The dev design team and Q/A missed that web apps created by the Add to Home Screen feature need a caption bar, or something—for example a web app title?—not to cover web content. Not including one causes a UI malfunction.

While on the topic of overlapping content:

Notice that the magnifier button on Spotlight.
The transparent button overlaps, very slightly, the content of its search field.

What purpose does that serve, other than being unnerving visually? Looks very much like a design error. Or are we showing off how neat button transparency really is?

Last, I brought up the menu bar in the DigitalTrends window.

The web is no longer the top window, but the menu persists. Is this a bug?
Worse (visually anyway), with a long caption name, in white of all colors, when "deactivated" such that the Liquid Glass effect goes away, the white menu is still vaguely visible across the top of the Notes window. Look closely at the full-sized image posted here; squint if you have to: white letters on a white window background!?
Not sure if the caption area of Notes is transparent (though I've used every means to disable the transparency of Liquid Glass) or if the menu is overlapping the caption. Probably the former.

But really, ick!

Come on #Apple! "Crack design crew," this is aimed at you: Embarrassing!

#BoostingIsSharing

#iPad #iPadOS27 #liquidglass #UIdesign #programming

#mastodon #programming #uidesign #liquidglass #ipados27 #ipad

RS, Author, Novelist, Prosaist @[email protected] · 2026-09-22 · 16:39 UTC

Yep. Installed Apple's #iPadOS 27 last night. Immediately found interface design faux pas that effect use.

Look at the screenshot carefully. It's showing Windowed App Mode. The two windows to the left (Digital Trends website and Mastodon) are Safari web apps created in iPadOS 27 using Add to Home Screen to ensure nothing from a previous release influences rendering.

Note that the stoplight controls overlap browser window content.
In both of these cases, because of the design of the web page (which should not matter because these are essentially the contents of a Safari tab), the stoplight control covers a link/button!
In the #Mastodon web app window, it prevents using the link/button that it covers.
In the Safari slide-over window, note that there are tabs and caption bar buttons that always push down the content in the browser below the stoplight controls.
Notes, shown an app example, has a stoplight control. It is neatly isolated on the caption bar from the content below. It seems only that the web apps have this flaw.

Opps! The dev design team and Q/A missed that web apps created by the Add to Home Screen feature need a caption bar, or something—for example a web app title?—not to cover web content. Not including one causes a UI malfunction.

While on the topic of overlapping content:

Notice that the magnifier button on Spotlight.
The transparent button overlaps, very slightly, the content of its search field.

What purpose does that serve, other than being unnerving visually? Looks very much like a design error. Or are we showing off how neat button transparency really is?

Last, I brought up the menu bar in the DigitalTrends window.

The web is no longer the top window, but the menu persists. Is this a bug?
Worse (visually anyway), with a long caption name, in white of all colors, when "deactivated" such that the Liquid Glass effect goes away, the white menu is still vaguely visible across the top of the Notes window. Look closely at the full-sized image posted here; squint if you have to: white letters on a white window background!?
Not sure if the caption area of Notes is transparent (though I've used every means to disable the transparency of Liquid Glass) or if the menu is overlapping the caption. Probably the former.

But really, ick!

Come on #Apple! "Crack design crew," this is aimed at you: Embarrassing!

#BoostingIsSharing

#iPad #iPadOS27 #liquidglass #UIdesign #programming

#ipados #apple #boostingissharing #ipad #ipados27 #liquidglass

Coa @[email protected] · 2026-09-22 · 03:44 UTC

Hi check out new posts on Vasnic Community!
https://vasnic.wtnet9.site
#gaming #pcgaming #indiegames #music #art #programming

#gaming #pcgaming #indiegames #music #art #programming

Coa @[email protected] · 2026-09-22 · 03:44 UTC

Hi check out new posts on Vasnic Community!
https://vasnic.wtnet9.site
#gaming #pcgaming #indiegames #music #art #programming

#gaming #pcgaming #indiegames #music #art #programming

Coa @[email protected] · 2026-09-22 · 03:44 UTC

Hi check out new posts on Vasnic Community!
https://vasnic.wtnet9.site
#gaming #pcgaming #indiegames #music #art #programming

#programming #art #music #indiegames #pcgaming #gaming

Coa @[email protected] · 2026-09-22 · 03:44 UTC

Hi check out new posts on Vasnic Community!
https://vasnic.wtnet9.site
#gaming #pcgaming #indiegames #music #art #programming

#gaming #pcgaming #indiegames #music #art #programming

Papel @[email protected] · 2026-09-22 · 01:32 UTC

Hell yeah, managed to get the polygon to change to any arbitrary step between its base/original form and the next!

Looking at the gifs, you might notice that the white head, which is the one being changed, doesn't properly align with the orange (middle), that's because it's only interpolating between the blue (rightmost) form and the yellow form (leftmost). That it still comes very close to the defined 3/4 facing head is a very welcome surprise, though one that will be short lived, as a proper 3/4 to full side facing will have to hide certain features later on.

Took me a while to reach a correctly working function, but the steps are the following:

1. Ensure size of the step (for each vertice, how much X and Y have to move to reach the final point) is filled, it's an array the same size/length as the total number of vertices
2. Get the current step/frame (in my case, i have a total of 20 steps/frames, from 0 to 20, or 0 to -20)
3. The polygon's vertices will receive (final polygon's vertex.x position minus (step for that vertice times the current frame)) - for i in 0..polygon.len(): polygon[i].x = nextpolygon[i].x - (step[i].x * currentframe) -- This "inverts" which polygon is the start and end, but it works.

Next up, figure how to invert the polygon's facing, so it can look either left or right. With that done, experiment with eyes and nose on top of the face.

I might also look how to make a nim macro to make it easier for me to loop over a polygon's vertices and doing basic math with raylib's Vector2 type. #programming #raylib #nim #gamedev #vectorgraphics

#vectorgraphics #gamedev #nim #raylib #programming

.:\dGh/:. @[email protected] · 2026-09-22 · 00:59 UTC

I have to say, MiMo v2.5 and MiMo v2.5 Pro are pretty decent.

These can't beat GLM 5.3 tho.

Anyone have these two fighting?

#AI #LLM #CodeCommand #Xiaomi #LargeLanguageModels #OpenCode #OpenRouter #MiMo #VibeCoding #VibeCode #Programming #SoftwareDevelopment #WebDevelopment #WebDev #Coding #Code

#ai #llm #codecommand #xiaomi #largelanguagemodels #opencode

.:\dGh/:. @[email protected] · 2026-09-22 · 00:59 UTC

I have to say, MiMo v2.5 and MiMo v2.5 Pro are pretty decent.

These can't beat GLM 5.3 tho.

Anyone have these two fighting?

#AI #LLM #CodeCommand #Xiaomi #LargeLanguageModels #OpenCode #OpenRouter #MiMo #VibeCoding #VibeCode #Programming #SoftwareDevelopment #WebDevelopment #WebDev #Coding #Code

#ai #llm #codecommand #xiaomi #largelanguagemodels #opencode

.:\dGh/:. @[email protected] · 2026-09-22 · 00:59 UTC

I have to say, MiMo v2.5 and MiMo v2.5 Pro are pretty decent.

These can't beat GLM 5.3 tho.

Anyone have these two fighting?

#AI #LLM #CodeCommand #Xiaomi #LargeLanguageModels #OpenCode #OpenRouter #MiMo #VibeCoding #VibeCode #Programming #SoftwareDevelopment #WebDevelopment #WebDev #Coding #Code

#ai #llm #codecommand #xiaomi #largelanguagemodels #opencode

.:\dGh/:. @[email protected] · 2026-09-22 · 00:59 UTC

I have to say, MiMo v2.5 and MiMo v2.5 Pro are pretty decent.

These can't beat GLM 5.3 tho.

Anyone have these two fighting?

#AI #LLM #CodeCommand #Xiaomi #LargeLanguageModels #OpenCode #OpenRouter #MiMo #VibeCoding #VibeCode #Programming #SoftwareDevelopment #WebDevelopment #WebDev #Coding #Code

#code #coding #webdev #webdevelopment #softwaredevelopment #programming

.:\dGh/:. @[email protected] · 2026-09-22 · 00:59 UTC

I have to say, MiMo v2.5 and MiMo v2.5 Pro are pretty decent.

These can't beat GLM 5.3 tho.

Anyone have these two fighting?

#AI #LLM #CodeCommand #Xiaomi #LargeLanguageModels #OpenCode #OpenRouter #MiMo #VibeCoding #VibeCode #Programming #SoftwareDevelopment #WebDevelopment #WebDev #Coding #Code

#ai #llm #codecommand #xiaomi #largelanguagemodels #opencode

Vulkan @[email protected] · 2026-09-21 · 22:22 UTC

New Sample – Descriptor Heap

This sample renders textured cubes to demonstrate VK_EXT_descriptor_heap, which replaces the traditional descriptor set and pool workflow with a memory-centric one where descriptor data is written directly into heap memory. It shows new heap-binding functions alongside vkCmdPushDataEXT for push data, so existing shader binding syntax keeps working without a traditional pipeline layout.

https://docs.vulkan.org/samples/latest/samples/extensions/descriptor_heap/README.html
#vulkan #programming #code #gpu

#vulkan #programming #code #gpu

Vulkan @vulkan · 2026-09-21 · 22:22 UTC

New Sample – Descriptor Heap

This sample renders textured cubes to demonstrate VK_EXT_descriptor_heap, which replaces the traditional descriptor set and pool workflow with a memory-centric one where descriptor data is written directly into heap memory. It shows new heap-binding functions alongside vkCmdPushDataEXT for push data, so existing shader binding syntax keeps working without a traditional pipeline layout.

https://docs.vulkan.org/samples/latest/samples/extensions/descriptor_heap/README.html
#vulkan #programming #code #gpu

#vulkan #programming #code #gpu

Vulkan @[email protected] · 2026-09-21 · 22:22 UTC

New Sample – Descriptor Heap

This sample renders textured cubes to demonstrate VK_EXT_descriptor_heap, which replaces the traditional descriptor set and pool workflow with a memory-centric one where descriptor data is written directly into heap memory. It shows new heap-binding functions alongside vkCmdPushDataEXT for push data, so existing shader binding syntax keeps working without a traditional pipeline layout.

https://docs.vulkan.org/samples/latest/samples/extensions/descriptor_heap/README.html
#vulkan #programming #code #gpu

#vulkan #programming #code #gpu

Vulkan @[email protected] · 2026-09-21 · 22:22 UTC

New Sample – Descriptor Heap

This sample renders textured cubes to demonstrate VK_EXT_descriptor_heap, which replaces the traditional descriptor set and pool workflow with a memory-centric one where descriptor data is written directly into heap memory. It shows new heap-binding functions alongside vkCmdPushDataEXT for push data, so existing shader binding syntax keeps working without a traditional pipeline layout.

https://docs.vulkan.org/samples/latest/samples/extensions/descriptor_heap/README.html
#vulkan #programming #code #gpu

#gpu #code #programming #vulkan

Vulkan @[email protected] · 2026-09-21 · 22:22 UTC

New Sample – Descriptor Heap

This sample renders textured cubes to demonstrate VK_EXT_descriptor_heap, which replaces the traditional descriptor set and pool workflow with a memory-centric one where descriptor data is written directly into heap memory. It shows new heap-binding functions alongside vkCmdPushDataEXT for push data, so existing shader binding syntax keeps working without a traditional pipeline layout.

https://docs.vulkan.org/samples/latest/samples/extensions/descriptor_heap/README.html
#vulkan #programming #code #gpu