[QUOTE=helifreak;46766937]But its three times larger than the gif, did you learn nothing?[/QUOTE]
Shut up, you're scaring away the early adopters.
[sp]On a more serious note, I think you might be able to make the huge PNG from the GIF's mostly transparent frames and just place a single PNG with the background underneath each of them, to attain a somewhat similar or maybe even lower file size. This whole idea is still idiotic, though.[/sp]
[QUOTE=Quiet;46766998]Shut up, you're scaring away the early adopters.
[sp]On a more serious note, I think you might be able to make the huge PNG from the GIF's mostly transparent frames and just place a single PNG with the background underneath each of them, to attain a somewhat similar or maybe even lower file size. This whole idea is still idiotic, though.[/sp][/QUOTE]
actually im pretty sure that this is what sublime text did on their homepage for the animated screenshots
[editline]22nd December 2014[/editline]
i remembered wrong but it's similar
[url]http://www.sublimetext.com/~jps/animated_gifs_the_hard_way.html[/url]
I've been working on a package manager for my own linux distro
So far implemented:
1) Installing packages
2) Syncing of package sources
3) Searching for packages
[code](devenv)eyes@eyes-pc:~/dev/wpkgman$ ./wpkgman.py -s filesystem
stable/filesystem-1.0.0:
The basic filesystem for root directory
[(devenv)eyes@eyes-pc:~/dev/wpkgman$ ./wpkgman.py -SI filesystem
Syncing sources....
testing: [=======================================================================================================================================================================] 100% 444.08 kB/s 0.39KB of 0.39KB
stable: [========================================================================================================================================================================] 100% 233.18 kB/s 0.16KB of 0.16KB
Calculating dependencies...
Packages to install (1):
filesystem-1.0.0
Continue? [Y/n] y
filesystem-1.0.0-x86_64.tar.xz: [==================================================================================================================================================] 100% 854.96 kB/s 0.6KB of 0.6KB
Installing package filesystem (1/1)... done
(devenv)eyes@eyes-pc:~/dev/wpkgman$ ./wpkgman.py -s dummy
testing/dummydep-1.0.0:
testing/dummypackage-1.0.1:
A dummy package
testing/dummydep2-1.0.0:
(devenv)eyes@eyes-pc:~/dev/wpkgman$ ./wpkgman.py -I dummypackage
Calculating dependencies...
warning: package dummypackage-1.0.1 is already installed
Packages to install (2):
dummypackage-1.0.1 dummydep2-1.0.0
Continue? [Y/n] y
dummypackage-1.0.1-x86_64.tar.xz: [==============================================================================================================================================] 100% 210.54 kB/s 0.19KB of 0.19KB
Getting URL http://mirror.com/pkg/testing/dummydep2/dummydep2-1.0.0-x86_64.tar.xz failed (404)
Installing package dummypackage (1/2)... done
Installing package dummydep2 (2/2)... failed
1 packages failed to install.
[/code]
Source is here: [url]https://github.com/SunDwarf/wpkgman[/url]
Lately, I've been working on a tool that takes XML documentation generated by Visual Studio, and turns it into Markdown-formatted files ready to put on a GitHub Wiki. It doesn't quite have a feature set like Doxygen or Sandcastle, but it's intended to be a one-click solution, specifically for smaller projects that use GitHub.
Not quite all tags are supported yet, but it can turn something like:
[code]
/// <summary>
/// Creates a new <see cref="Member"/> object for the specified
/// element.
/// </summary>
/// <param name="element">
/// The <see cref="XElement"/> object that contains the documentation
/// for which to create a new <see cref="Member"/>.
/// </param>
/// <returns>
/// A new <see cref="Member"/> object of the type corresponding to the
/// member described in <paramref name="element"/>.
/// </returns>
/// <exception cref="ArgumentException">
/// <paramref name="element"/> has no <c>name</c> parameter, or the
/// value of the <c>name</c> parameter is too short.
/// </exception>
/// <exception cref="Exception">
/// <paramref name="element"/> belongs to a <c>namespace</c>, or the
/// member being referenced is an error.
/// </exception>
/// <exception cref="NotImplementedException">
/// <paramref name="element"/> references a kind of member that is not
/// recognized.
/// </exception>
public static Member Create(XElement element)
[/code]
into this:
[t]http://s.horsedrowner.net/2014/12/1419250743-675x817.png[/t]
I manually put the generated Markdown code on the wiki, so the next step would be to clone and update the GitHub Wiki automatically.
I also don't build a hierarchy yet, so there's no association between members and the classes they're in.
Don't you just love those moments something doesn't do what you want it to at all, but it looks funny?
[thumb]http://i.imgur.com/x0RBRf9.png[/thumb]
[thumb]http://i.imgur.com/5Wfnecy.png[/thumb]
I have no idea what I'm doing :v:
[thumb]http://i.imgur.com/qwRfjvF.png[/thumb]
[QUOTE=xThaWolfx;46769902]Don't you just love those moments something doesn't do what you want it to at all, but it looks funny?
Pics
I have no idea what I'm doing :v:
Pic[/QUOTE]
Looks like a raytracer? Looks like your rays are intersecting the sphere when reflecting if I were to guess. Try increasing (or adding) a small offset along the normal when starting the next ray. This is especially noticeable when using lower precision floating point numbers.
[QUOTE=ben1066;46770305]Looks like a raytracer? Looks like your rays are intersecting the sphere when reflecting if I were to guess. Try increasing (or adding) a small offset along the normal when starting the next ray. This is especially noticeable when using lower precision floating point numbers.[/QUOTE]
No that wasn't it, there was no reflection :v:. The problem was in my color class, using bytes. The addition operator added them together which resulted in some weirdness. When adding the floats of them together and clamping to 1 it worked just fine.
[thumb]http://i.imgur.com/pXbldXH.png[/thumb]
Hey my friend got his computer compromised by a virus and was so kind to try to infect me on steam with it.
It's written in C# and has things like "[I]get_Dreamhack_Karambit_Stained_Souvenir_Knife[/I]" written in it. I'm having trouble decompiling it on Linux though.
Someone want to help out?
[editline]asf[/editline]
I'm not sure what these agrees mean-- reply/PM me if you actually want to help and I'll PM you.
[QUOTE=Naelstrom;46770785]Hey my friend got his computer compromised by a virus and was so kind to try to infect me on steam with it.
It's written in C# and has things like "[I]get_Dreamhack_Karambit_Stained_Souvenir_Knife[/I]" written in it. I'm having trouble decompiling it on Linux though.
Someone want to help out?
[editline]asf[/editline]
I'm not sure what these agrees mean-- reply/PM me if you actually want to help and I'll PM you.[/QUOTE]
You need .NET deobfuscator(s).
[QUOTE=Fourier;46771126]You need .NET deobfuscator(s).[/QUOTE]
Wow that worked, thanks!
[editline]asdf[/editline]
[code][module: ConfusedBy("ConfuserEx v0.3.0")][/code]
That explains a lot, most of the methods are lost despite the deobfuscation, and the few that aren't are dumb to try to read. I guess that ends my poor reverse engineering venture. I just wanted to see what kind of trouble my friend was into.
Thanks anyway.
[QUOTE=Naelstrom;46771362]Wow that worked, thanks!
[editline]asdf[/editline]
[code][module: ConfusedBy("ConfuserEx v0.3.0")][/code]
That explains a lot, most of the methods are lost despite the deobfuscation, and the few that aren't are dumb to try to read. I guess that ends my poor reverse engineering venture. I just wanted to see what kind of trouble my friend was into.
Thanks anyway.[/QUOTE]
They are supposed to be lost because obfuscators try to obtuse things by adding shit.
You must search for methods for networking -> malware has to communicate.
If there is no method to be found, the the code is encrypted somewhere and you need to decrypt it:
You decrypt it so by finding the encrpyted code and finding method that encrypts it. Be carefull not to execute it.
If nothing, then malware uses Steam API to trade items to whatever account he has set to.
They normally from what I've seen encrypt their payload assembly into resources then decrypt and load the module directly into memory. That being said though there are many different variants of it.
I've been continuing my research into facilitating self-organization and emergent properties, and now I've actually got a basic framework that models the processes, and you can actually apply the ideas to a system you're designing and get emergent results!
Most of the work was in working out the overall structure of the universe (it's quite simple actually, just a scale-free network of agents that go through cycles of conflict, cooperatation, specialization and grouping/scaling), and then figuring out the common types of interactions and their properties
Where I'm at right now is figuring out the hard math side as opposed to just the theory; normally this posed a huge problem because the systems get way too complicated to handle with all the insane differential equations that involves. Instead, what I have now involves huge drops in degrees of freedom wherever important things happen, so (tentatively) simulating and simplifying the interactions of agents to move up a level (mostly just for computational power reason, go up even just one level and you'll have to simulate hundreds of agents) could be as easy as integrating a (relatively) simple partial differential equation, and going the other way by differentiating. Kind of an end goal is being able to extract a base equation from some level of a set of systems that already exists, or make an equation for a custom-designed system, and being able to move up and down as much as you want with only the equations. Although that might be too much to ask for, you might not be able to capture all the nuances need I make that work.
Moreover, I realized that there are a lot of complex processes you could very easily simulate with some of the concepts I've developed. One example is a curled chip bag unfolding; that's a result of the internal interactions of agents, and the amount of uncurling can be (probably) linearly represented as a slow release of potential energy. That's fine and dandy on it's own but then all it would take to simulate something stopping that from unfurling is it's interaction with any kind of resistant potential energy, like something just touching the top.
One of the big things in this theory so far is the idea that potential energy is resistance of an interaction with another agent. And basically, you can use it interchangeably between agents/interactions as easily as you can equate most equations of energy to solve for shit
But for now I was able to design a simple simulation involving for collaborating/fighting wedges such that they self-organize. Ive also been trying to look down a level, which has been an interesting experience because I made the wedges as the bottom level, but the more I do with this theory the more I realize that you need to be able to go deeper to get a smoother function (a calculus analogy again). So to get well-behaving upper levels, you need to design the implicit inner workings of things that you'd only normally think to make explicitly defined.
Basically, you can't make a single level of a simulation and expect things to self-organize, you need multiple, simple levels of interaction that create the appearance of complex behavior at higher levels. (Without actually being complicated enough to get lost in chaos.) So it's again very much like the overlaying of self-similar noise in certain coherent noise techniques, but with simple systems of basic interactions instead of scalar noise.
And I'm going to start on actually writing that simulation now that I have time. It'll probably make a lot more sense than this post.
I've had a multi million dollar idea but I'm busy working on something else so I'm going to share it with you. I think there are some scripts that do this out there but nothing accessible.
[sp]Software that removes just the porn from your browsing history.[/sp]
Does anyone know any resources for classic 3d vector graphics programming? the kind where everything is just cleverly drawn 2d lines that look 3d
i love to work with code from other people and with 3rd party libs that produce this:
[CODE]A/libc﹕ Fatal signal 11 (SIGSEGV) at 0xdeadbaad (code=1), thread 24391 (Thread-9134)[/CODE]
[QUOTE=proboardslol;46773028]Does anyone know any resources for classic 3d vector graphics programming? the kind where everything is just cleverly drawn 2d lines that look 3d[/QUOTE]
That's what 3D graphics are.
[QUOTE=Darwin226;46773155]That's what 3D graphics are.[/QUOTE]
yeah but I mean like, just with lines
[QUOTE=reevezy67;46772942]I've had a multi million dollar idea but I'm busy working on something else so I'm going to share it with you. I think there are some scripts that do this out there but nothing accessible.
[sp]Software that removes just the porn from your browsing history.[/sp][/QUOTE]
Private browsing / incognito mode
[QUOTE=thrawn2787;46773294]Private browsing / incognito mode[/QUOTE]
And "forget this site" in history. Unless you have a tendency to spree on several ones.
[QUOTE=proboardslol;46773184]yeah but I mean like, just with lines[/QUOTE]
Then it's even easier, don't need to worry about texture mapping etc!
[QUOTE=reevezy67;46772942]I've had a multi million dollar idea but I'm busy working on something else so I'm going to share it with you. I think there are some scripts that do this out there but nothing accessible.
[sp]Software that removes just the porn from your browsing history.[/sp][/QUOTE]
This reminds me, does anyone remember/have a link to that one site from like 5 years ago that could tell what porn sites you'd visited? I remember it being a thing where people sent a link to a friend, and both you and the friend could see all the results after you clicked it.
I always assumed it had a hidden massive list and could see which links were purple, now I really want to know what made it tick
[QUOTE=Tommyx50;46773430]Then it's even easier, don't need to worry about texture mapping etc![/QUOTE]
Do you know any resources? like any books or anything? I'm really just trying to re-invent the wheel here. I'm not even looking for videogame programming, I just wanna draw 3D shapes with basic 2D lines, like spheres or cubesand be able to shift perspectives around
[QUOTE=proboardslol;46773856]Do you know any resources? like any books or anything? I'm really just trying to re-invent the wheel here. I'm not even looking for videogame programming, I just wanna draw 3D shapes with basic 2D lines, like spheres or cubesand be able to shift perspectives around[/QUOTE]
Look into modern 3D programming with perspecitve projection/projection matrices.
It really is exactly what you want to do.
The only difference is that you'd split the result into lines before drawing (or you could just switch the renderer into outline mode..).
[QUOTE=proboardslol;46773856]Do you know any resources? like any books or anything? I'm really just trying to re-invent the wheel here. I'm not even looking for videogame programming, I just wanna draw 3D shapes with basic 2D lines, like spheres or cubesand be able to shift perspectives around[/QUOTE]
There's a good tutorial of 3d related techniques here: [url]http://www.petercollingridge.co.uk/pygame-3d-graphics-tutorial[/url]
[editline]23rd December 2014[/editline]
[QUOTE=Tamschi;46774030]Look into modern 3D programming with perspecitve projection/projection matrices.
It really is exactly what you want to do.
The only difference is that you'd split the result into lines before drawing (or you could just switch the renderer into outline mode..).[/QUOTE]
He wants to reinvent the wheel, not use an existing renderer like opengl and get lines from it.
Smoke upscaling is going well
Standard resolution boring x direction velocity
[IMG]https://dl.dropboxusercontent.com/u/9317774/smooth%20vx.PNG[/IMG]
Wildly exciting x direction velocity with added wavelet noise to 'upscale' it
[IMG]https://dl.dropboxusercontent.com/u/9317774/upscaled.PNG[/IMG]
Next I need to actually advect the material by the high frequency noise, and then I'm pretty much done, bar optimisation and pretty rendering
The general idea behind this is to take low resolution smoke simulations, and then add high frequency wavelet noise to them to make it look as if its much higher resolution. This will look much nicer than traditional smoke without adding a huge amount of computational complexity
Here's an idea I had: take some character model that's represented with voxel data and standing in a t-pose. Next up you make a low-poly bounding model of that character. You render the low-poly model as you normally would, but have the pixel/vertex shader use the position/rotation data of the polygons rendered to then ray trace towards the voxel data (that is still only in a t-pose) from that.
In theory you could transform the bounding model however you want and the ray traced model should appear to transform with it
Obviously this doesn't account for lighting at all (maybe?) and I have no clue how I would implement it, but I think it has potential to make up for some of raytracing's complexity issues. You'd only ever trace against a single model/data set at a time and you wouldn't need a big data structure to hold all the world's voxel data, also due to the fact that the model you trace against probably won't change much, instancing is a bit simpler
Can you guys think of any other potential benefits/problems with this?
So it takes C 6 seconds to iterate an integer from 0 to 2^32-1. that means (I think) to test every possible value of eulers equation of degree 4:
a^4 + b^4 + c^4 != d^4
would take
a×4+b×3+c×2+d
where
a = 6
b = 2147483648*a
c = 2147483648*b
d = 2147483648 c
seconds, which is, according to wolframalpha
59421121941038485454941323288 seconds
which is, according to wolframalpha
1.883×10^21 average Gregorian years
or
~1.4×10^11 × universe age (~14 Gyr )
i.e: the universe could be born and destroyed 140 billion times before I was able to test every possible positive answer for a 32 bit signed integer
Which logo looks better?
Rate [img]http://www.facepunch.com/fp/ratings/tick.png[/img] for
[img]http://i.imgur.com/EZinvT5.png[/img]
Rate [img]http://www.facepunch.com/fp/ratings/cross.png[/img] for
[img]http://i.imgur.com/nlUIRH0.png[/img]
[QUOTE=Berkin;46775373]Which logo looks better?
Rate [img]http://www.facepunch.com/fp/ratings/tick.png[/img] for
[img]http://i.imgur.com/EZinvT5.png[/img]
Rate [img]http://www.facepunch.com/fp/ratings/cross.png[/img] for
[img]http://i.imgur.com/nlUIRH0.png[/img][/QUOTE]
since they're different aspect ratios, they're sort of incomperable for any one use. Like if you want a banner on a website or devblog, the second one, but if you want just a single logo to use as like a thumbnail of some sort, the first one is better
I like the font in the first better though
Sorry, you need to Log In to post a reply to this thread.
