My first attempt at doing some screen recordings - probably need to use a different codec as the one I used resulted in a bunch of compression artifacts. Either way, hope you're as excited as I am for the upcoming topics!
Also be sure to take a for look at the Config. Modern encoding often has excessive values set for things like bitrate but then don't balance the key frames etc. What I mean is , a cursory review of the deeper encoder settings can result in far far better encodes with less wasted space. Of course you might not even b12 able to access such through minimalism GUIs and lack of back end engineering these days but meh.. And this sort of default behaviors go back over 20yrs.. no one ever fixed that aspect of end-user experience
I like the changes you made to your videos, but I especially like the back and forth with your community through posts on TH-cam. Please keep it that way! :)
There is a good video by Inkbox that is called "What is the Smallest Possible .EXE?". This can give a good idea of what is minimally needed to have a runable exe file. Another video by Nir Lichtman called "Making a Very Minimal Windows Executable in C" also has some good information in it.
Yesss I‘m aware, thanks! As far as I know, the smallest possible file is in the double digits I think? A little over 64 bytes? I think I just read that on stack overflow somewhere though… Might consider those techniques for fun some day, right now I just want my output files to not contain the entire standard library for a start :D
I think with this language/compiler CRDT and graph/AST-based CVS (because code is already parsed and always valid AST-graph) also will work good. For online collaborative work, like in Figma. Probably for most cases conflict-free merge could also be done with seamless history (automatic commits into branches - you assign commit messages only to meaningful points in history sequence)
its a bit funny from a perspective of functional programming and the idea of function's purity, lambda calculus with a visualiser, can't wait for the state management
See, if it were me, I'd be so excited to build the tool itself then get bored and quit at the tedium of figuring out all the bytes for a header (and subsequently the ones for the machine code instructions you'll need).
Oh yeah, I'm definitely fighting my way through this project with sheer f*cking will... there is no other explanation for why I'm doing this to myself :D
@@Kronark I cheer you on as a random internet stranger! Honestly I find content like this inspiring, and I want to get back to writing my own dumb shit but I'm currently in "I want to write C++ for windows but I don't want to use visual studio" hell and I have no idea what I'm doing when it comes to configuring clangd for Nvim myself xD
The main benefits are: 1. extremely fast compilation times 2. lowest level control over your output 3. prevention of a whole host of write-time errors It'll become more apparent the more we use it though :)
Question! This might be getting way ahead of myself, but aside from stupid fast compile time, is this also like, 'more safe' way to write programs? As like like less prone to bugs, security errors, and stuff like that?
I'll have to wait and see how those benefits turn out, but it should generally be able to prevent a whole host of bugs that can occur in traditional programming, yes :)
I just found your channel. My question is, Why PE when ELF is probably better documented and even has official source available in the Linux kernel you can check to see how it works. Where as PE you need to just go on documentation. There is no source availability to know how the file is being read by Windows its self short of looking into a system like ReactOS possibly. Your system looks very adaptable and, as you said at the start of the series, could be used to make any file format.
Because I have a windows pc and no linux system available. Additionally, Windows is the dominant OS on the software market, even more so on the gaming market. So it‘s just more sensible to start developing for Windows once I get into the software and game development business. I‘ll obviously support ELF eventually, but definitely only once I can actually run it on a system without having to setup virtual environments for it. May get a Linux laptop next year for that :)
I use Davinci Resolve for all the editing and audio mastering :) It allows me to build presets for all my effects, enabling me to keep the look consistent and makes usually complex animations a little simpler to pull off :D
Either we‘re in the same time zone or you know where I live :o If it‘s the former, that‘s some serious dedication to getting first comment :D Just testing video performance for different upload times right now.
All audio is mastered following TH-cam guidelines to be at around -14 LUFs and a true peak at -1.5 dB - if it's too loud, then you've made it too loud on your end, probably because you just watched a different video that didn't follow the audio guidelines and made their content to quiet... It's unfortunately pretty common :/
One does not simply watch one Kronark video without re-watching the past 5 videos to load the necessary context into your RAM
Very funny. Ha ha.
As it was written, for this is the will of the Omnissiah.
@@Kronark Holy shit. This lore goes deeper than I thought 😮.
My first attempt at doing some screen recordings - probably need to use a different codec as the one I used resulted in a bunch of compression artifacts. Either way, hope you're as excited as I am for the upcoming topics!
Also be sure to take a for look at the Config. Modern encoding often has excessive values set for things like bitrate but then don't balance the key frames etc.
What I mean is , a cursory review of the deeper encoder settings can result in far far better encodes with less wasted space. Of course you might not even b12 able to access such through minimalism GUIs and lack of back end engineering these days but meh..
And this sort of default behaviors go back over 20yrs.. no one ever fixed that aspect of end-user experience
this is a great resource for learning about executables/os, thank you!
Glad you think so!
I've been following this series since the beginning, great work !
You make amazing content that it is a crime to have only this many subs :3
Glad you're onboard mate!
Yeahhhh... my sub count has kind of stagnated to be honest... I've been "almost at 8000 subs" for the past 5 days :D
Congrats dude, thats freaking amazing!
Thanks man! :D
You've encouraged me to take on a passion project about s-expressions and make some content on it; very glad I found your videos.
Sounds awesome! Feel free to send me a link to your content via the email in the channel info, once it's online! :)
Following u since the 4th video.
What i say just simply Genius 🎉❤
Glad you‘re on board! Thank you brother
I can barely follow this… but looks very cool. Crazy what behind a little ‘hello world’
It does build upon all the previous videos, hopefully it'll get a little simpler soon :)
I like the changes you made to your videos, but I especially like the back and forth with your community through posts on TH-cam. Please keep it that way! :)
I definitely will, the feedback is invaluable :D
There is a good video by Inkbox that is called "What is the Smallest Possible .EXE?". This can give a good idea of what is minimally needed to have a runable exe file.
Another video by Nir Lichtman called "Making a Very Minimal Windows Executable in C" also has some good information in it.
Yesss I‘m aware, thanks! As far as I know, the smallest possible file is in the double digits I think? A little over 64 bytes? I think I just read that on stack overflow somewhere though…
Might consider those techniques for fun some day, right now I just want my output files to not contain the entire standard library for a start :D
I think with this language/compiler CRDT and graph/AST-based CVS (because code is already parsed and always valid AST-graph) also will work good. For online collaborative work, like in Figma. Probably for most cases conflict-free merge could also be done with seamless history (automatic commits into branches - you assign commit messages only to meaningful points in history sequence)
"Building Executables without a Programming Language But Good CS Base Knowledge"
Gonna make that the motto around here
its a bit funny from a perspective of functional programming and the idea of function's purity, lambda calculus with a visualiser, can't wait for the state management
Still gotta do some bug fixing before I can showcase the global state management feature :S
See, if it were me, I'd be so excited to build the tool itself then get bored and quit at the tedium of figuring out all the bytes for a header (and subsequently the ones for the machine code instructions you'll need).
Oh yeah, I'm definitely fighting my way through this project with sheer f*cking will... there is no other explanation for why I'm doing this to myself :D
@@Kronark :’)
it's you guys that keep me motivated to do it
@@Kronark I cheer you on as a random internet stranger! Honestly I find content like this inspiring, and I want to get back to writing my own dumb shit but I'm currently in "I want to write C++ for windows but I don't want to use visual studio" hell and I have no idea what I'm doing when it comes to configuring clangd for Nvim myself xD
Hey! I was in the exact same hell! :D
C++ and current build tools / compilers drove me to do this project :P
can't wait for the actual assembly so I can actually understand shit
Might do the mov instruction next week, so your prayers have been heard
i will watch this whole video even though i wont understand a thing
I‘ve been told this one is a little easier to understand, with the screen recordings and such :)
I am so excited for the hello world! (Also I just realized that I forgot to subscribe mb)
It‘s all definitely starting to pick ip the pace slowly but surely :)
And hey, you found your way back here so all good :P
not even a programmer and still watching this
(i won't have to be a programmer with this LOL)
Everyone‘s a programmer to a certain degree :P but i‘ll definitely try to make it easier for everyone
Great project!
But, I can't wrap my head around the advantages of this way of creating a program compared to the old fashioned way for this use case.
The main benefits are:
1. extremely fast compilation times
2. lowest level control over your output
3. prevention of a whole host of write-time errors
It'll become more apparent the more we use it though :)
@@Kronark thanks, for the explanation. I'll keep following your progress.
glad you're on board! let me know if you have any other questions any time! :)
Question! This might be getting way ahead of myself, but aside from stupid fast compile time, is this also like, 'more safe' way to write programs? As like like less prone to bugs, security errors, and stuff like that?
I'll have to wait and see how those benefits turn out, but it should generally be able to prevent a whole host of bugs that can occur in traditional programming, yes :)
I just found your channel. My question is,
Why PE when ELF is probably better documented and even has official source available in the Linux kernel you can check to see how it works.
Where as PE you need to just go on documentation. There is no source availability to know how the file is being read by Windows its self short of looking into a system like ReactOS possibly.
Your system looks very adaptable and, as you said at the start of the series, could be used to make any file format.
Because I have a windows pc and no linux system available. Additionally, Windows is the dominant OS on the software market, even more so on the gaming market. So it‘s just more sensible to start developing for Windows once I get into the software and game development business. I‘ll obviously support ELF eventually, but definitely only once I can actually run it on a system without having to setup virtual environments for it. May get a Linux laptop next year for that :)
Also welcome to the channel :D
Your voice is very familiar, have you video on another channel?
Nope :)
Hay, how do you make these videos? I have always been curious how people put together visuals for technical topics
I use Davinci Resolve for all the editing and audio mastering :)
It allows me to build presets for all my effects, enabling me to keep the look consistent and makes usually complex animations a little simpler to pull off :D
Please decrease the quality and editing time of your videos because it scares me
Request denied
bro 4am are you serious 💀
bro maybe not everyone lives in the same timezone as you
Either we‘re in the same time zone or you know where I live :o
If it‘s the former, that‘s some serious dedication to getting first comment :D
Just testing video performance for different upload times right now.
@@Kronark well, i'm in france, and also i litteraly work at 5am, and the place where i work is a 10min walk away from my house
Have fun at work! You can watch the video on your way there :P
WHAT?
WHAT?!
8:00 that is one mess of connection lines
I‘d like to refer to the comment in the top right corner at around 7:26 :)
Is the intro obnoxious loud on purpose?
All audio is mastered following TH-cam guidelines to be at around -14 LUFs and a true peak at -1.5 dB - if it's too loud, then you've made it too loud on your end, probably because you just watched a different video that didn't follow the audio guidelines and made their content to quiet... It's unfortunately pretty common :/
@@Kronark The audio is good, don't worry