What are you working on?

[voodooattack]

[QUOTE=Tamschi;50180364]I saw that, yes.

The issue with not having coroutines is that, at least from what I've seen from Node programs, you usually end up with extremely verbose code if you do anything at least a little complex.[/QUOTE]

That'd be true if you're using callbacks. It's all different when you're using promises, which simplify things greatly.

[larikang]

[QUOTE=Overv;50177639]Try looking at the source code of this:

[url]https://github.com/lokesh/color-thief/[/url][/QUOTE]
Yeah, I saw that before but it looks suuuper complicated. But I just figured out a pretty easy method: I use imagemagick to crop just the top 20px of the image (that's what the taskbar overlaps), quantize it down to 8 colors, and pick the most common color.

That seems to work well as long as the top portion of the image is reasonably uniform in color.
[t]http://i.imgur.com/6iOAgvz.png[/t]

[Tamschi]

Cyclic: Cyclic_1A8C1FA(Reference = Cyclic_265601D(Reference = Cyclic_1A8C1FA))
Cyclic copy: Cyclic_245FB7(Reference = Cyclic_38248FC(Reference = Cyclic_245FB7))

This works now.

[Phygon]

[QUOTE=Sidneys1;50180509]Cool, and well rendered. Reminds me of a project I started a while back. Was going to be a Contagion clone, but based on real-life geography and population statistics:
[t]http://i.imgur.com/KV4K2ng.png[/t]

Unfortunately, the API I used to get geo data only let me get X cities for every country, making smaller countries look denser (e.g., countries in Europe and Africa). My ultimate goal was to heatmap it, but I ended up spending too long figuring out vertex buffers in OpenTK and lost interest :/ I have a bad habit of jumping around between XNA, OpenTK, and MonoGame and having to relearn the paradigms of each every time.[/QUOTE]
That looks neato, I like how you have the network pre-illustrated.

[QUOTE=powback;50179766]Hey! I made something like that! It's not a game though, more like a worm farm.

Each node is randomly placed and connects to X nearby nodes. Each of the nodes has a random key assigned.
A virus is placed on a random node, where it will brute force the connected node keys. Once the key is found, the virus is transmitted, and that node will start doing the same.
So multiple infected nodes can work together to brute force one node, making the virus spread faster.

Nodes are also reformatting now and then, so the node will become clean and unavailable before going back online.

Inspired by [URL="https://xkcd.com/350/"]XKCD[/URL] of course.

[t]http://puu.sh/orqWO/f4479fea5e.png[/t]
[t]http://puu.sh/orr7v/4bf60f6de1.png[/t][/QUOTE]
That's really cool but it also sounds super computationally expensive given that you're literally bruteforcing stuff that you yourself are generating

[editline]23rd April 2016[/editline]

Inspection working well now
[vid]http://i.imgur.com/c3Ga6Tf.webm[/vid]

[Dilbert]

I made a quick and dirty binary tree visualizer today.
[url]https://goo.gl/XBldE5[/url]

[t]http://i.imgur.com/pdRVDXs.png[/t]

[Colliseemoe2]

[QUOTE=tisseman890;50180213]Finally got the smiley selector working.

It was a bitch to figure out.

[vid]https://my.mixtape.moe/xobewu.webm[/vid][/QUOTE]

You're doing god's work, friend.

[Tamschi]

Notable_2BB23B(A = int_189441B(1), B = float_1D364F7(2), c = char_6E8CAF(c), e.E = int_3E2F22E(5), RestoreCount = int_2FA83A4(0), cscss.Value.Value.Value.Value.Value = int_2CEA0CA(9))
File: AAAAB05vdGFibGUBAgMEBQYAA2ludAEABWZsb2F0AAAAQAAEY2hhcmMCBQIAAgk=
From file: Notable_30495D1(A = int_329445B(1), B = float_73673B(2), c = char_EA119(c), e.E = int_83A9E6(5), RestoreCount = int_A0F91B(0), cscss.Value.Value.Value.Value.Value = int_1A8C1FA(9))

Now this works, but I haven't fully implemented it yet. The rest is just filling in blanks though.

[Cyberuben]

[QUOTE=Tamschi;50186312]

Notable_2BB23B(A = int_189441B(1), B = float_1D364F7(2), c = char_6E8CAF(c), e.E = int_3E2F22E(5), RestoreCount = int_2FA83A4(0), cscss.Value.Value.Value.Value.Value = int_2CEA0CA(9))
File: AAAAB05vdGFibGUBAgMEBQYAA2ludAEABWZsb2F0AAAAQAAEY2hhcmMCBQIAAgk=
From file: Notable_30495D1(A = int_329445B(1), B = float_73673B(2), c = char_EA119(c), e.E = int_83A9E6(5), RestoreCount = int_A0F91B(0), cscss.Value.Value.Value.Value.Value = int_1A8C1FA(9))

Now this works, but I haven't fully implemented it yet. The rest is just filling in blanks though.[/QUOTE]

I have 0 clue what you are working on, or what that means

[Tamschi]

[QUOTE=Cyberuben;50186444]I have 0 clue what you are working on, or what that means[/QUOTE]

It's a serialisation library. This is the [I]Stream[/I] interface, but adding support for other formats (JSON, XML, whatever horrible thing else...) is pretty easy and then works for any data you throw at it.

The core library (and the binary formatter) have full support for cyclic references and value types, so it handles anything (as long as there's a format descriptor). [I]cscss.Value.Value.Value.Value.Value[/I] is a deep property path across multiple reference and value types, defined in this embedded format descriptor:

public NoteFormat NoteFormat
    => new MemberFormat<Notable>(
        "Notable",
        () => A,
        () => B,
        () => c,
        () => e.E,
        () => RestoreCount,
        () => cscss.Value.Value.Value.Value.Value)
    {
        Initializer = @this =>
        {
            @this.e = new Embedded();
            @this.cscss = new C<S<C<S<S<int>>>>> { Value = new S<C<S<S<int>>>> { Value = new C<S<S<int>>>() } };
        },
        Finisher = @this => @this.RestoreCount++
    };

Of course you can also register attached descriptors beforehand, which I use for the defaults for the built-in C# types. ([I]null[/I] is handled a little differently: That's the one thing I have a special branch for, though I still need to make that one configurable.)
I'll probably add support for automatic format generation for any [I][Serializable][/I]s, but it would still be better in most cases to quickly use a manual format. (I'll make it opt-in, and it doesn't really affect the core system.)

Essentially: Serialisation, but without any compromises or headaches (other than (currently) speed. It's unoptimised in some regards right now, but I doubt I can quite match the throughput of specialised or individually compiled libraries without [I]a lot[/I] of work anyway).

[editline]23rd April 2016[/editline]

To serialise:

stream.Write(instance.ToNote());

To deserialise:

stream.ReadNote(Code.Default).ReconstructObject();

There are a few optional parameters that let you use custom format sets and serialise multiple top-level objects with shared data in a way that writes them to a [I]Stream[/I] only once each.

[bean_xp]

[QUOTE=Breny;50163380]Some progress on my Alien Carnage remake.
Today was HUD day.
I implemented most of the HUD (no radar yet, because no enemies) and added fancy edges at the sides.

[vid]http://a.pomf.cat/nftvlx.mp4[/vid]

Next up: Doors and enemies.[/QUOTE]

I loved this game, nice work!

[sidneys1]

[QUOTE=phygon;50183975]That looks neato, I like how you have the network pre-illustrated.[/QUOTE]

It's not pre-rendered, but all the connections were computed beforehand (simple voronoi) as well as the primary connections between countries (closest two points between two countries) and nodes that could transmit disease between countries (convex hull points)

All that data is pushed into a couple vertex buffers then rendered out. Every couple of ticks a random 'traveller' A*'s his way from once city to another (potentially spreading diseases)

[quote][video=youtube;xDwDB1XLF6w]https://www.youtube.com/watch?v=xDwDB1XLF6w[/video][/quote]

[Insight]

Posted a page back about how I am getting lasik so heres just a quick update! I know its not programming but figured there are many people who have glasses might find it interesting.

Just had Lasik done yesterday and my vision is already at 20/20 and it is just going to get better over time. My original prescription was 20/80 and I have worn glasses since 1st grade (I hated using contacts).

So if you have ever thought of getting it done, go for a free check up and see if you are a good candidate. If you have any questions you can PM me.

[voodooattack]

Almost 7K views on Medium for my [URL="https://medium.com/@voodooattack/multi-threaded-javascript-introduction-faba95d3bd06#.yfjwanm9a"]article[/URL]. My project went viral overnight.

I'm touched. I thought people hated it from the StackExchange reactions, but apparently not.

[Layla]

[QUOTE=TH3_L33T;50187363]Posted a page back about how I am getting lasik so heres just a quick update! I know its not programming but figured there are many people who have glasses might find it interesting.

Just had Lasik done yesterday and my vision is already at 20/20 and it is just going to get better over time. My original prescription was 20/80 and I have worn glasses since 1st grade (I hated using contacts).

So if you have ever thought of getting it done, go for a free check up and see if you are a good candidate. If you have any questions you can PM me.[/QUOTE]

I really want to get it done, seems too good to be true though.

[Powback]

[QUOTE=phygon;50183975]
That's really cool but it also sounds super computationally expensive given that you're literally bruteforcing stuff that you yourself are generating

[/QUOTE]

Oh yeah, it's really expensive. The goal was to create a semi-realistic network with worms that actually have to work towards cracking nodes instead of just having a countdown or something.

If it was an actual game, I would have done it differently, but it's neat to see how bruteforcing works in practice anyway.

[Insight]

[QUOTE=layla;50187614]I really want to get it done, seems too good to be true though.[/QUOTE]

Feel free to add me on Steam if you want me to go over it in more detail. Millions of people have gotten the surgery done.

What seems to good to be true about it?

[cartman300]

[QUOTE=powback;50187731]Oh yeah, it's really expensive. The goal was to create a semi-realistic network with worms that actually have to work towards cracking nodes instead of just having a countdown or something.

If it was an actual game, I would have done it differently, but it's neat to see how bruteforcing works in practice anyway.[/QUOTE]

It would be cool is each node actually ran some simplified virtual programs, something like Core War except on a network instead of a single computer.

[TrinityX]

[QUOTE=layla;50187614]I really want to get it done, seems too good to be true though.[/QUOTE]

My dad got lasik a couple years ago, his vision was 20/20 after recovery, but got gradually worse again and is now almost as bad as it was before. Just a heads up, it might work for some people permanently, but who knows. Maybe my dad got a bad op, again who really knows...

[Phygon]

Anyone have any recommendations for literature that would be related to generating a network like I'm trying to do? My current method is just kinda forcing it and it's MONSTROUSLY inefficient and disgusting, especially when it comes to calculating distances between nodes to find candidates for connection. I'm reworking it now to be a more efficient tree-based network (All users connect to an ISP and various companies but not other users) but I'm sure that there are better ways to do it.

[awcmon]

c++ compiler is magic

i had a program that took 2 minutes and 55 seconds to finish running

i changed it from debug config to release, all i did, i didn't change any of my unoptimized code

it finished running in 2-3 seconds

[marvincmarvin]

[QUOTE=elevate;50185055]I made a quick and dirty binary tree visualizer today.
[URL]https://goo.gl/XBldE5[/URL]
[/QUOTE]

I think I broke it.
[IMG]http://puu.sh/otk9n/30ebf98c8c.jpg[/IMG]

[zerf]

I created a userscript that replaces the static C syntax highlighting on this forum with autodetected highlighting using Highlight.js

[del]Some things don't work so well, but I'd say it does just as well or better than the static C syntax highlighting as far as readability is concerned.

You can get it [URL="https://gist.github.com/zerfgog/a89986f465f4a6af2963ec91037aa964"]here[/URL]. (Click 'raw' to install. Requires Greasemonkey on Firefox or Tampermonkey on Chrome (or whatever userscript managers y'all use)).[/del]

[editline]wew[/editline]

actually this sucks. don't use it; the default stuff works better (even though everything is treated as C)

[Dilbert]

[QUOTE=marvincmarvin;50189300]I think I broke it.
[IMG]http://puu.sh/otk9n/30ebf98c8c.jpg[/IMG][/QUOTE]

I just didn't bother setting reasonable bounds for the parameters.

[Winded]

[QUOTE=awcmon;50189276]c++ compiler is magic

i had a program that took 2 minutes and 55 seconds to finish running

i changed it from debug config to release, all i did, i didn't change any of my unoptimized code

it finished running in 2-3 seconds[/QUOTE]

Did you run with a debugger the first time? Debuggers do slow down programs, but 2m55s sounds like a very slow debugger. Could also depend on what kind of code you are running

[ECrownofFire]

[QUOTE=TrinityX;50188227]My dad got lasik a couple years ago, his vision was 20/20 after recovery, but got gradually worse again and is now almost as bad as it was before. Just a heads up, it might work for some people permanently, but who knows. Maybe my dad got a bad op, again who really knows...[/QUOTE]

When the technology was newer it definitely led to a lot of degradation over time. Now it should be good for most people.

[srobins]

[QUOTE=awcmon;50189276]c++ compiler is magic

i had a program that took 2 minutes and 55 seconds to finish running

i changed it from debug config to release, all i did, i didn't change any of my unoptimized code

it finished running in 2-3 seconds[/QUOTE]

Does anyone know why this would happen? I'd be very interested to hear an explanation for this behavior.

[larikang]

[QUOTE=srobins;50193210]Does anyone know why this would happen? I'd be very interested to hear an explanation for this behavior.[/QUOTE]
In debug mode the compiler should try to preserve the exact structure of your code as written so that when you debug it you see all the function calls and variables that you expect. In release, the optimizer is free to analyze what your code is actually doing and generate simpler code that has the same outcome.

E.g. doing calculations inline rather than real function calls, reordering statements to make better use of the cache. Stuff like that. It's impossible to say what the real difference is without seeing his code. A big part of why computers are so fast these days has nothing to do with improved clock speeds and everything to do with optimizers being able to find really sneaky ways to rewrite your code to make better use of the CPU.

[Silentfood]

do any of you guys use any sort of todo-list apps? not for professional use as i know about kanban/trello and such. i find it hard keeping on track with tasks i need doing on assignments and having a notepad list seems too manual.

[CmdrMatthew]

[QUOTE=srobins;50193210]Does anyone know why this would happen? I'd be very interested to hear an explanation for this behavior.[/QUOTE]
Also, if you use any kind of standard library functions, those tend to have a LOT of slow debug checks and tend to run dozens of times faster in release mode.

[awcmon]

[QUOTE=Winded;50191714]Did you run with a debugger the first time? Debuggers do slow down programs, but 2m55s sounds like a very slow debugger. Could also depend on what kind of code you are running[/QUOTE]

I just double-clicked the binary in the Debug folder after building it.

It was just a simple markov chain text generator reading from a single input file that was 993353 words long, 6877713 characters long. there's just a lot of putting stuff into maps (didn't even bother to use an unordered map) and vectors.

[editline]24th April 2016[/editline]

[QUOTE=CmdrMatthew;50193325]Also, if you use any kind of standard library functions, those tend to have a LOT of slow debug checks and tend to run dozens of times faster in release mode.[/QUOTE]

this might be the case, since my code's packed with STL