doing another rust project where i have to pull user info from FP profiles for auth. also had to learn how to use regex. fun stuff
[img]https://dl.dropboxusercontent.com/u/5168294/screencaps/Screenshot%202017-03-29%2016.23.23.png[/img]
[QUOTE=DarKSunrise;52029228]visual studio will fail to install properly if you have certain versions of it installed so you'll probably just want to uninstall everything related to it and reinstall the version you need[/QUOTE]
Its working on the fresh W10 install, thankfully. Also, the WSL is super neat and has made things so much easier. I'm still waiting for clang to compile from the source (wanted to test my ability to do this), but it seems it'll be super easy to test my project with the WSL. VS also seems like it should let me just plug into the WSL and run/test my code there actually, which is pretty cool.
I'm heading home early today though, so I can work on my personal projects (and because this week has left me exhausted already). I'm trying to learn more about template metaprogramming, and I'm trying to use boost::hana to make my rendering system I copy-and-paste between projects be far easier to use (and generally be more adaptable). Things like introspection on user-defined structs seems awesome, but everything I've tried to learn about this so far feels really abstract and hard to grasp. The little I'm able to comprehend and brush against seems really fucking cool, though.
[editline]29th March 2017[/editline]
I can't find a lot of examples of boost::hana in use, either.
Today in reverse-engineering, we learn how not to do mouse input with A4 Engine:
[img]http://imgur.com/AaFSBKg.png[/img]
Yeah, because
[code]
aspect = res_x / res_y;
yaw = mouse_x * sensitivity * aspect;
[/code]
is totally the way to do it.
I've been implementing dynamic resolution scaling into my engine. So far its breaking pretty much everything because of needing to modify gbuffer texture coordinates and everything that needs to use any part of it.
Halving the render scale brings much less than anticipated perf gains, I wonder if this is even worth all the hassle, or if screen space reprojection is a better way to go.
Started building clang last night, had to delete 50 gigs of stuff to make enough room for the build, kept my laptop on all night to finish linking clang itself, maxed out my ram, finally got it built and tried to add templight only to realize that it doesn't work for the latest svn version. Had to revert changes and start a full rebuild. Fucking hell.
script I made while ago to keep track of csgo 10mans my friends and i do.
It parses match from [url]https://popflash.site/[/url] (some free server provider to play our 10mans on)
[url]https://docs.google.com/spreadsheets/d/1YixIensgpaA3xTcQclCO0KmJFrzUsW0PVsG4wg1_Z-A/edit?usp=sharing[/url]
made a script on google docs for first time and pretty proud of it :v:
[url]https://script.google.com/d/1mhl1tP-eLe2AwkGxKfM9qx756svKhSDKffH7rMc5XXbxal9k3qtCA24T/edit?usp=sharing[/url]
[QUOTE=WTF Nuke;52030666]Started building clang last night, had to delete 50 gigs of stuff to make enough room for the build, kept my laptop on all night to finish linking clang itself, maxed out my ram, finally got it built and tried to add templight only to realize that it doesn't work for the latest svn version. Had to revert changes and start a full rebuild. Fucking hell.[/QUOTE]
Even on the 7700k w/ 32gb of RAM on my work machine it was only 10% done after 45m of running at work. I just locked my computer and left it on when I left, but looks like I might have the same issue awaiting me tomorrow :v:
Finished finding the Hand/World Fov, mouse sensitivity and mouse asymetry undoing code injection points, now the game is actually playable on ultrawide/multihead setups!
[t]http://imgur.com/YmIYRMg.png[/t]
[t]http://imgur.com/7XJMa1O.png[/t]
[t]http://imgur.com/HpqGfIM.png[/t]
[editline]30th March 2017[/editline]
Implemented it as a "Flawless Widescreen" plugin. Great fucking tool man.
I wrote a [URL="http://berkin.me/rantdocs/howto/unity-setup/"]tutorial[/URL] on integrating Rant into Unity, if anyone's interested in how that's done.
I'm highly considering writing a commercial Unity plugin for Rant that offers some nice features, like auto-packaging patterns at build time and binding variables to component fields.
oh hey tommy refenes is finally getting the meat world part of super meat boy back up now after some people on here took it down a few years back
[QUOTE=Eric95;52031430]oh hey tommy refenes is finally getting the meat world part of super meat boy back up now after some people on here took it down a few years back[/QUOTE]
A classic, for those who might have forgotten or never heard:
[url]https://facepunch.com/showthread.php?t=1144771&p=33856685&viewfull=1#post33856685[/url]
Whatever happened to swift and shift? Maybe Tommy got his revenge.
[QUOTE=WTF Nuke;52031505]Whatever happened to swift and shift? Maybe Tommy got his revenge.[/QUOTE]
[url]https://github.com/charliesome[/url]
he works at github these days, we have a chat every once in a while
[QUOTE=JohnnyOnFlame;52031085]Finished finding the Hand/World Fov, mouse sensitivity and mouse asymetry undoing code injection points, now the game is actually playable on ultrawide/multihead setups!
[t]http://imgur.com/YmIYRMg.png[/t]
[t]http://imgur.com/7XJMa1O.png[/t]
[t]http://imgur.com/HpqGfIM.png[/t]
[editline]30th March 2017[/editline]
Implemented it as a "Flawless Widescreen" plugin. Great fucking tool man.[/QUOTE]
Release?
got the auth part of my bot working
[t]https://dl.dropboxusercontent.com/u/5168294/screencaps/Screenshot%202017-03-30%2009.38.00.png[/t]
[QUOTE=ryankingstone;52032209]Release?[/QUOTE]
Sure, I'm trying to get in contact with HaYdEn about integrating it into the official repository. In the meantime you'll find it on my gist backups.
Also, here's a sneak-peek at the Metro Last Light fixer:
[t]http://imgur.com/ccJGP26.png[/t]
Need to code the plugin for it now.
[IMG]https://puu.sh/v3HHW/cc711de5fc.png[/IMG]
Runtime API for reflection stuff starting to come together.
My Rant plugin for Unity is coming along nicely. I have automatic packaging and in-editor pattern testing working.
[img]http://i.imgur.com/8tn3ChM.png[/img]
[QUOTE=Perl;52031531][url]https://github.com/charliesome[/url]
he works at github these days, we have a chat every once in a while[/QUOTE]
Well good for him :)
*Crosses feature off list*. Everything below is completely functional, and dynamic (works with 1+ teams)
[IMG]https://i.imgur.com/667XrEc.png[/IMG]
I'm starting to get somewhat tired of working on this game now. Its been like... 2 years, with about a year of solid development, and a year being ill or doing other things (don't join a startup if the lead is on cocaine, folks). Its a long time to have been working on a single project for me, having to do 100% of everything all the time with no input from anyone else. The code is somewhat of a mess (although I've taken steps to address some of the worst offenders of this recently, but this is still the first real thing I've built), and I particularly feel like working 100% solo is hampering my ability to improve at code (because really being able to talk to other folks would definitely lead me down much better design routes both due to their experience, and simple communication). Its also quite isolating to work on a thing by yourself
At this point, I'd really like a job, I've had a particular games company in mind for a long time that I'd very much like to apply to that rhymes with 'spacehunch', although I have no idea if I'd get it :v:
That said, I'm still fairly determined to make this game. I think I'm going to shift my original goal of trying to get something on steam (ie acceptable to a general audience who is picky about missing, weird, or somewhat poor looking features, as well as a minimum of a year of support and updates based on community feedback) to "produce acceptable thing I can show and play with other folks I know" (ie acceptable to a small, technically understanding audience, no expectation of long term support), which is a significantly easier goal that means I won't end up spending a 3rd or 4th year of my post-uni life on this (so you know, I can have a life)
Or maybe I'll just keep making this game for all of time and I'll still be posting this shit 2 years from now. Who knows!
I'm done with trying to get templight to work. It's too much work. Goodbye clang build. Hello 20 gigs of space.
Isn't this from one of you guys?
[url]https://news.ycombinator.com/item?id=13994752[/url]
Third entry on hacker news.
[QUOTE=WTF Nuke;52036075]I'm done with trying to get templight to work. It's too much work. Goodbye clang build. Hello 20 gigs of space.[/QUOTE]
Why not keep it around for clang-tidy? That's still why I'm bothering. clang-tidy is pretty much as nice as static analysis gets for C++, imo.
I submitted my second ever pull request to an open source project and its "mostly" accepted, so-to-speak. It just contains [URL="https://gist.github.com/fuchstraumer/c852d5d9def7e6cdeb62b341839f0d6f"]the simplification algorithm I updated from a github repo I found[/URL]. I keep making small grammatical mistakes, but the main contributors like it enough that they want me to add it to the GUI as an option for the next release, and if it tests well enough and works well enough, it'll be made the default :dance:
The main devs made it clear that it looks really promising, and the algorithm they use now is actually a bottleneck - and has caused a number of bug reports due to simplification errors in complex models. I've not had any of these issues, so hopefully it works for them. Besides being proud of getting code integrated back into the project, I'm mostly just pleased that I was able to contribute effectively to a project that helped me so much since using open-source code in a (temporarily, if I can get our marketing guy to deal) closed-source project was making me feel dirty.
Only problem is that I can't build the project to save my life. At home, Perl throws a shitfit. Here at work, "cl.exe" is somehow not in my path ??
I wrote a long post about what I've been doing with constexprs and lambda and C++17, making the usefulness and speed of my pattern library so much better, but then I discovered a bug in it and spent an hour and a half fixing it, and then CloudFlare ate my post. So here's a basic example instead, a simple key:value list parser
[cpp]
// A slice is just a pair<iterator, iterator>
auto make_keyvalue(const std::tuple<slice, slice> &in)
{
return std::make_pair(
std::string{ std::get<0>(in).begin, std::get<0>(in).end },
std::string{ std::get<1>(in).begin, std::get<1>(in).end }
);
}
auto to_keyvalue(const std::string &source)
{
using namespace tl;
constexpr auto alpha = lambda([](char c) { return std::isalpha(c) != 0; });
constexpr auto digit = lambda([](char c) { return std::isdigit(c) != 0; });
constexpr auto wspace = lambda([](char c) { return std::isspace(c) != 0; });
constexpr auto skip = maybe(many(wspace));
constexpr auto key
= skip + capture(alpha + maybe(many(alpha | digit)));
constexpr auto value
= skip + capture((alpha | digit) + maybe(many(alpha | digit)));
constexpr auto delimiter
= skip + ':';
constexpr auto key_value
= (key + delimiter + value) >> make_keyvalue;
constexpr auto grammar
= many(key_value);
return *grammar.match(source);
}
[/cpp]
Using the function like so:
[cpp]
const std::string str = ""
"first : hello\n"
"second: what\n"
"third: 12056\n"
"fourth :D\n";
int main()
{
for (auto &kv : to_keyvalue(str))
std::cout << p.first << " = " << p.second << "\n";
return 0;
}
[/cpp]
Yields:
[code]
first = hello
second = what
third = 12056
fourth = D
[/code]
I'd love to explain what's going on in detail if anyone's curious to know, but I could talk about this for hours, and I have no idea where to even start. I'm just glad to have gotten this far with it, I've been working on this for years now
Got multi-channel output displayed, I'm thinking of adding a feature to the bindings system to allow channel names to determine the affected property names.
[img]http://i.imgur.com/ccdFEQz.png[/img]
[QUOTE=paindoc;52037644]Why not keep it around for clang-tidy? That's still why I'm bothering. clang-tidy is pretty much as nice as static analysis gets for C++, imo.[/QUOTE]
I'll try to get it running properly on my main PC in a month. I just felt overwhelmed with the dependency hell. I got templight running but now I need to get templight tools or templar to build, and they need qt or graphviz, and one of them needs boost. And then installing qt is another long dependent process. I'm beginning to wonder if open source software was a mistake.
[QUOTE=WTF Nuke;52040108]I'll try to get it running properly on my main PC in a month. I just felt overwhelmed with the dependency hell. I got templight running but now I need to get templight tools or templar to build, and they need qt or graphviz, and one of them needs boost. And then installing qt is another long dependent process. I'm beginning to wonder if open source software was a mistake.[/QUOTE]
That's exactly why I prefer linux for development. Anyway, don't they provide a windows qt installer, so all you need to do is to set up an env for qt root and update PATH to get it up and running? Then in cmake you can just find_package and link to the proper libraries if necessary.
This was linux development.
But installing both qt and boost is really simple on linux... Anyway, I guess you had your reasons to put it on the shelf, I was just surprised by that "long dependent process" part.
I'll try it again later when I have more space, I am running on hard drive fume at this point. Bash for windows also broke with error 0x80080005 so I can't try without a full reinstall, which means recompiling clang again. I glanced over the qt install process and it had a decent amount of steps, although looking at it again it just seems like installing different utilities.
Sorry, you need to Log In to post a reply to this thread.
What are you working on? v67 - March 2017
