Andreas Jung

Back on XP

2007-04-08T11:28:00.000+02:00

I've been running VISTA for the past few weeks, but now I have switched back to XP.

VISTA was simply too slow and unreliable (e.g. the file system). Every program was running way slower than on XP. But the main reason is the lack of a propper anti-virus software which is compatible with VISTA.

I was using the LiveOneCare trial. However, when the trial period ran out, I didn't feel like to buy the full subscription, because the security features didn't really convince me. For example, all MS software was recognized as trustworthy software by the firewall- this is not exactly true for security features such as WGA. Actions such as "firewall approval" required adminstrator permission ("fading screen"), which is nearly stalling the system for fullscreen-applications hitting the firewall.

Another aspact is the fact, that I didn't manage it to find a desktop-mail-client which is capable of synchronizing with Lycos (no problem on XP with OE).

Finally, the VISTA UI is not an improvement at all. I'm quite open minded towards "colorful UI", but VISTA is simply too much.

So ... I'm back on XP with good old free "Antivir", working mail client and acceptable UI.

Blog Makeover

2007-03-21T12:43:00.000+01:00

There are some major changes going on in my life planning right now, and I thought it's a good idea to take this occasion to do a little makeover of this blog.

Changes include:

sidebar re-structured: more cateogries, newsfeed now clearly recognizable, links updated
added "Best Of" categories for "Photos", "FS Development" and "Song Reviews" - this should help to find the remarkable posts between all the trash-posts (Please note: my early posts from 2005 still sound a little bit non-adulty, because they were influeced by school days; anyway, I'd like to keep them, because they are a great reminiscene of this part of my life)
AdSense added to take advantage of all those people which use the "Next Blog" button

I will shorty post some news regarding my life planning. It's just a matter of some organizational issues to be done.

Yet Another **NG Flightsim

2006-09-25T10:39:00.000+02:00

Well, it's true that FSNG is basically dead. OTOH, I have worked in the term break on another flightsim, which was built from ground-up.

I have promised myself not to announce yet another flightsim project without having something to download.
Unfortunetly, the term break was busier than I thought, and now that it's coming to an end with the winter-term just about to begin, I have to face, that I didn't manage it to come close to a milestone that would justify a file release.

But because there won't be much time left in the winter term for programming stuff, I've decided to show you some screenshots, to document that this sim is really there, so here we go... http://sourceforge.net/project/screenshots.php?group_id=169340.

Now some background information.

The source code is fully GPL'd, so you could download the current branch using subversion if you want. Anyway, I doubt it will be of any use for you, because there are no scenery files included, and I have never tried to run a flight without a scenery installed, I guess the sim would probably crash. There is a SRTM-converter-pipeline included, but I doubt anyone except me is able to use this command line utility.

Features implemented yet:

Terrain mesh and overlay, runtime-loading in a thread, no scenery stuff/autogen/water/whaterver yet; just shape and noisy color.
Module pipeline for vehicles.
The NPSNET Flight Model implementation from FSNG, altough still some bugs in it (vibrations).
Incorporated the JSBSim flightmodel into the module pipeline, altough JSBSim doesn't fit into the module pipeline very pretty, and JSBSim itself is not very pretty IMHO. So I probably drop this feature again.
SRTM-to-scenery-converter-pipeline.
Joystick input.
Realtime sky according to Preetham and Hoffman.
Fully self-contained .NET/MDX GUI library to have forms in fullscreen mode. Works independently from any rendering API, so an OpenGL renderer could be written as well.
Property system similar to the PropertyGrid built on top of the GUI-API.
Shades of a math- and algorithm library (the point is, that all algorithms and datastructures go into a separate libarary, so they can be used in other projects as well).
Language customization.
...

The project is not far from the state of FSNG presented on my website. Note that CSNG has been built from ground-up again - there is no FSNG stuff in it. I have made some things better that I have learned from FSNG. For example, I have used MDX 1.0 again, and not MDX 2.0 beta ... what a lucky decision looking at the state of MDX 2.0 now.

The "C" in CSNG stands for combat. Not that I'm much into the combat thing, but there are some advantages: Combat simulators have limited maps ... this simplifies the whole data organization and data reloading stuff ... now I use tiles and I cannot fly near the poles ... :) .

OK, my good old FSNG website will stay where it is, because it's some kind of "monument" and it looks quite good. Maybe I'll replace it, when I have a downloadable version of CSNG in place. But that won't be too soon from now.

XNA Game Studio Express Beta Impressions

2006-09-15T11:28:00.000+02:00

I had a short look at the XNA Game Studio Express Beta a few days ago, so I thought it might be a good idea to state some of my impressions.

Let's start with the good aspects...

The framework seems to be designed very well and careful. It seems to fit the .NET guidelines very well.
This entire "game loop thing" is moved to a higher level. The programmer now has to derive his game-class from the Framework.Game class and implement some methods related to the game loop, e.g. "Draw". So all this stuff takes away the pain from the programmer trying to create a perfect game loop, which is not always fun, in particular in the .NET CLR ("Application.DoEvents").
Device management has been simplified and moved to a higher level. Some traditional terms such as "lost device scenarios" have been revised and there is less witchcraft involved during device creation. This practically means, that large parts of the device creation work have been automated; "lost device" is now "ensure device"; switch to fullscreen is now as simple as calling "GraphicsComponent.ToggleFullscreen()". I think this was way overdue.
Maybe this sounds radiculous, but XNA would be a perfect candidate for a multi-platform game environment. Why? Well, the XNA namespace contains really everything - not only graphics, sound and input, but it also has it's own IO-classes for data storage (!).

So let's come to be points I don't like...

XNA is yet focussed on C# in VS Express environment. This means, that I cannot use other .NET languages, altough this shouldn't really be an issue by definition in the .NET CLR. I also cannot use use my non-Express VS - I have to use the Express edition.
XNA has it's own project type (application and library). I cannot be sure, wether I can use other .NET class libraries in an XNA application or not. This could become a big issue. I could - however - compile my libraries into XNA format, but only *if* XNA supported C++/CLI as 2nd .NET language.
XInput could be the biggest mistake. Yes, it allows me to access mouse and keyboard under Windows, and the XBox360 gamepad under both - XBox and Windows - environments. But what about all the other controlers under windows? One cannot expect that everyone buys an XBox360 pad for his Windows PC, but keyboard- and mouse-devices are not sufficient input methods for some game types. So PLEASE make XInput as general as good old DirectInput.
I don't get this component stuff (GraphicsComponent etc.) ... VS Express displays these components as icons in the forms designer. This doesn't make sense. If they want me to "draw" my games, then they should at least enable the UML class designer in VS Express.

Well, XNA does some things very well. XNA will be a serious option for some game projects. It is not an option for a flight simulator project, simply because of XInput being the only input-API. It could be a serious candidate for a racing simulator (for example).

Wether to use XNA or not is a tough decision. In case you use XNA, you are pretty much bound to XNA and MS, and there is no way back and you are pretty much not flexible. In case you decide to go with XNA anyway, you have a rich game creation toolkit that will make your life easier. I think it's the best game programming API to date, and I don't see any other API reaching a comparable level, because all these APIs are designed with backward compatibility in mind. XNA is a fresh new API without a legacy burden.

In case you really want to go with XNA, you'll feel like Alice in XNA wonderland.

DAFIF to be removed from public access Oct. 1, 2006

2006-08-14T10:32:00.000+02:00

DAFIF will be removed from public access on October 1, 2006. It would be a good idea to create a backup copy of the database if there are any intentions to use the navigation database in the future.

Atlas Of The Universe

2006-08-14T10:14:00.000+02:00

I came across this website, http://www.atlasoftheuniverse.com/. It features the ability to zoom from solar system scale "up" to the borders of the visible universe. There is a wealth of maps for each different scale.

There are also some interesting questions answerd that exceed the scope of discovery channel, without requiring a degree in astro-physics; e.g. that there is no such thing such as the center of the universe, and that the size of the visible universe is not necesarily the size of the whole universe.

Now with curved winglets...

2006-07-06T23:56:00.000+02:00

Check out this beauty...
http://www.airliners.net/open.file/1070915/M/

Now this is fancy.

IDontGetIt.NET/CLI

2006-07-02T23:50:00.000+02:00

Just had a strange problem programming in C++/CLI. I didn't really understand why it was nearly impossible to find a solution. Am I the only one to try this?

So, I want to implement the generic ICollection for my own (generic) collection:

generic<typename T> ref class ItemCollection : ICollection<T>

Let's have a look at the generic ICollection interface:

generic<typename T> public interface ICollection : IEnumerable<T>, IEnumerable

So as we see, we have to deal with one old-fashion enumerable- and one generic enumerable interface. This first defines the method "IEnumerator^ GetEnumerator()", while the generic one defines the "IEnumerator<T>^ GetEnumerator()".

We all know, that we cannot overload functions that solely differ by return-type. However, as the generic ICollection inherits from both IEnumerables, both GetEnumerators are in fact present in the generic ICollection.

Don't ask me why this can exist at all. The problem comes up when I try to implement the ICollection. I have to implement both versions of the GetEnumerator, but when I do this, I have two functions differing only by return type. BANG.

How to deal with this? Not really any ideas, lots of C++ compiler errors such as the one with the two functions only differing by return types, or methods not matching certain signaturs etc.

I had a look at the C# way, because the C# refactor does the implementation on it's own:

public IEnumerator<T> GetEnumerator()
System.Collections.IEnumerator System.Collections.IEnumerable.GetEnumerator()

Well OK, this looks reasonable by explictly having the "targeted" interface in the method name. But still no idea how this looks in C++/CLI, only silly compiler errors. I looked up the signature in the MSDN reference of some collection classes, but this didn't work out as well.

So, after numerous trial-and-errors attempts, this came out:

public:
virtual IEnumerator<T>^ GetEnumerator();

private:
virtual Collections::IEnumerator^ _GetEnumerator () sealed = Collections::IEnumerable::GetEnumerator
{
return GetEnumerator();
}

No need to mention that I'm seriously unhappy with this. This is everything, but definitly not elegant. I couldn't think of it being that hard to implement ICollection.

OK, it's late, no more time to think about this problem.

Series

2006-05-28T10:39:00.000+02:00

Check these outstanding photos over at Airliners.net: http://www.airliners.net/search/photo.search?&countrysearch=Albania&daterange=2006&keywords=chris%20lofting

Obviously, a photographer called Chris Lofting did some tour around Albania to take some breathtaking photos of the local Air Force. Especially those Chinese out-of-service F-6 aircrafts in the underground-tunnel are worth a click.

ILA 2006

2006-05-20T14:01:00.000+02:00

Got up yesterday at 0430 for a 400km drive across Germany heading Berlin - it's ILA time.

Surprsingly, there was a static display of a USAF B-1 this year. Although this aircraft is already some decades old, it was one of the highlights, because the airframe is simply beautiful. The photo does not give a good sense for the size of this beast.

The main reason to visit this year's ILA is the A380 of course. What a beauty. More on the A380 later.

Red Bull had a huge airpark. Between all those well known Red Bull aircrafts such as the B-25, I found this Cobra helicopter in Red Bull livery quite remarkable.

The C-17 is a regular ILA visitor for years. It's always thrilling to stand in front of this giant, imagining it flying. This thing does 20000 feet-per-minute descends with reverse-thrust, according to the legend.

Also lots of russion aviation technology this year. Besides the regular heavy metal, there was also some ground effect "plane". Nice little toy.

Some classic airframes, such as Messerschmidt 108 and 262 (the first jet plane), Spitfire and Corsair. The maneuvers of the ME-262 were simple,but the jet sound was just amazing.

Next was some German metal (I'll leave the Kruppstahl-jokes up to the reader). The Eurocopter Tiger did some loopings and stuff, what is quite spectecular for a helicopter. You simply have to believe, that I did not rotate the image.
The joint presentation of German Air Force and German Army was the radiculous part of the day. First, some special forces roped down the helicopter (just took minutes) to get Mr. Ward Ambassador out of the German embassy in country XYZ.
Then, the first C-160 Transall did a flyby, and the second one did a "Sarajevo" combat approach, which was definitly not 12° steep according to the photo. I guess they didn't want to take a risk, because the wind was quite strong this day.
While the ambassador was getting inside the C-160 on the runway, the helicopters did some woozy maneuvers to confuse the enemy.
Then, the C-160 took off in formation with the helicopters, while 80's synthesizer music played in the background. No comment ... oh, too late.

This was followed by the so called Airbus Parade, consisting of A318, A340 and A380. All three did some circling and a low-speed flyby.
Most shocking was the lowspeed flyby of the A380 - this giant nearly stood still in the air with an amazing angle of attack.
And it's true - the A380 is really silent.

Meanwhile, the Sentry took of. Note the elevator asymmetrie - did some visitor mess around with the roll trim in the cockpit?

Junkers Ju-52 taking off - strange light: absolutely dark in the background, but sunny in the foreground.

The Mig-29 was the highlight of the day, of course. This one's lining up and powers the engines up and goes back to idle three times - too arrogant, but so lovely ;), the crowd went mad, everyone was excited. The maneuvers were as breathtaking as usual - unbelieveable, how they incorporate the spins into their flightpath.

The last show for me was the formation flying of the Patrouille Suisse. Altough I found it becoming boring after the fifth formation flyby. Nevertheless great skills.

Rumble In The Jungle

2006-05-01T10:15:00.000+02:00

Sort of busy these days. Only very few hours left this week for private activities. Most days were like getting up/university/coming back home in the evening/doing exercises in the evening/bed.

Well, that's just justification, and most likely NOT interesting. The only other activity I did this week is to create...

This New Song

(Progressive Tech Tribal House)

Have a look or not. :)

P.S.1: I'd really love to work on FS related stuff, but it's simply so few hours per week that there is no real progress.

P.S.2: I think I have found a good sky model for FS use. But I think I'll attempt implementing it prior dropping announcements, reviews or FS-Developer entries. Anyway, it might be good idea (depending on the context of good) to re-check FS-Developer periodically. There are really some sporadic updates every now and then (more frequent than blog-entries here).

Good one

2006-04-22T00:54:00.000+02:00

muHAHAHAHA!!!111

Someone grabbed the domain http://www.fsng.com/. That's sooooo cute. Imagine: someone paid money for this. And now he is waiting for me to go begging for it. Lol.

uhh yeah ... happiness each and everywhere...

2006-04-18T20:28:00.000+02:00

Science and research brought us some great innovations:

automotive technology
world piece
and now it's time for this

Sorry, but some people really need to check out a meat grinder from inside IMO. Dudes...

Worth a Posting

2006-04-17T16:15:00.000+02:00

Well, a not-so-important posting is still better than a no-posting, so here we go.

If I leave the easter stuff out-of-scope, I'm pretty busy at the moment. I have an additional 5 hours per week at university besides my default 40 hours per week plus 10 hours of driving. International readers: that's lots of time in Germany - people here go on strike for weeks, because they have to work an additional one or two hours more on their less-than 40 hours week.

What I want to say: I had some spare time left this weekend, and this is what I did:

DAFIF is a public domain navigation database by the US military, and this is the respective enroute chart of the gulf region including Iraq, Iran and parts of Afghanistan, Turkey and ex-Soviet Union.

I did a poster print on 3x3 pages. This took quite some time. The chart is a huge PDF which is composed from vector data. Rasterizing all this data takes roughly 30 seconds. Printing this stuff on 9 pages took me roughly 5 hours. I guess the printer driver rasterized the whole document for each data chunk which is sent to the printer. So there were something like 5 minutes between each printed row.

Why? I love to explore the world using all kinds of free datasets - Landsat, SRTM, VMAP, etc.. There are so many beautiful features on the world that the average dumb citizen is not aware of:
-Richat structure - 50km circular feature in western sahara, visible from space.
-Europe: large infrared spots are good indicators for military training areas as well as open pit mines.
-Iraq - south-east border: a square feature as large as a little state.
-Northern siberia, polar sea coast: looks more alien than any other place in the solar system.

Back to the Iraq enroute map.
I find it very remarkable, that ex-Soviet Union is more reddish than Iran. I think allied (or friendly) states should have a different color. Interesting...

Please tell me that this is a fake!

2006-04-06T00:31:00.000+02:00

http://www.ezprezzo.com/videoclips/plane_landing.html

!!! www.FS-Developer.info !!!

2006-03-31T23:04:00.000+02:00

YES!!! My little top-secret project of the last week is now unveiled.

www.FS-Developer.info

FS-Developer.info is my attempt to create a database, which is solely dedicated to flight simulator development.

Other than knowledge collections such as www.vterrain.org, FS-Developer concentrates on these topics, which are important for flight simulation only.

The database is still built up, so there should be more entries following over the next few days. The best part is: everyone can submit entries!

I don't want to go too much into detail. Just a have look yourself.

Would really love to hear some feedback. Use the comment function of this blog or the methods available on the website.

Would also appreciate some outside support :).

Volume Fog

2006-03-24T00:00:00.000+01:00

It's already known among a limited group people, that I like to compare real life aviation photos with flight simulator photos in order to find the important visual clues that make up the difference between sim/reality.

Now I came across a nice reference scene for volume/altitude fog: http://www.airliners.net/open.file/1021518/M/

I have already some experience with this effect from an unmanaged demo in 2004. I will not link it here, because I found out that it doesn't run on my GF6 (worked well with GF4/5). In this demo, I added an altitude-factor besides the distance factor to the calculation of fog density.

The other option would be to use real fog volumes, by calculating the fog depth through front/backface-substraction.

Another New Song

2006-03-23T23:15:00.000+01:00

Well ... let me recite myself:

Well, there needs to be an industrial counterpart for my latest acid production,
so this is supposed to be an industrial/acid crossover.
My plan was to create something with processed drums, so this song is really dirrty as there are lots of Scream4s in it.
Of course you are free to tell me, that the production is dirrty - but it's my intention.
Same for the arrangement: Of course the song is boring for the first few bars - but that's industrial - it's about slamming drums.
However, I can promise that you'll be recompensed for listening on - bar 65 (2:18) features BIIIGGGG ENERGY and the breakdown is really beautiful IMO.
I felt like falling in trance when listening several times to this song during the
production.
Musically it's nothing great - only one chord - it's an industrial thing - it's the stylistic devices that makes up the atmosphere.
So hopefully you can enjoy.
Copyright notice: Image/Sample are from NASA. NASA generally does not claim a copyright on it's multimedia content.OTOH, this does not mean, that I am the copyright holder, that's why I don't upload it at Soundclick, but on my own webspace instead.

Interested?

Check out the MP3 download: http://www.aj-productions.de/music/OrangeSoil.mp3
And the reviews: http://www.reasonstation.net/songs/songinfo.php?song=46093

New Song

2006-03-20T19:44:00.000+01:00

Yep, that's right. New Acid House production.

Listen here: http://www.soundclick.com/bands/songInfo.cfm?bandID=459773&songID=3663271
Reviews here: http://www.reasonstation.net/songs/songinfo.php?song=46012
Cover art here: http://www.aj-productions.de/music/AcidProgressiveExcessive.jpg

Enjoy if you like acid!

Semi Show-Off then...

2006-03-14T20:32:00.000+01:00

Well ... sorry for the lack of the updates.

I'm right in the middle of my term break. Unfortunately, the tests are written INSIDE the term break at my university, so I am busy learning stuff.
Two tests have been successful, but the most difficult one is still ahead, so I should be learning all day theoretically.

Of course this is just theory, so I have some time for programming left.
Nevertheless, time is very limited, and my original flight simulator FSNG is basically dead, because it didn't really survive the port from Visual C++ Express Beta 2 to the final version, as well as the port to MDX 2.0.

Recently, another idea struck my mind. One of my early dreams was to create a Space Shuttle simulator, i.e. a modern version of this, this, or this game. To be exact, I am talking about the last one, i.e. an approach simulator.

Landing the space shuttle is very challenging from the pilots point of view - 3 minutes full of action. I guess the difficulty level can be compared to a Sarajevo- or Kai Tak IGS approach (having attempted neither of them ;) ), although I guess the NASA people wouldn't attempt to de-orbit the orbiter with known windsheers in the area.
There is a nice video out there that shows the approach from inside the Shuttle Training Aircraft - I hope I can find it somewhere later this evening. UPDATE - Here it is: http://www-pao.ksc.nasa.gov/shuttle/countdown/sts112/sta-vid.htm.

Well ... programming a shuttle landing simulator has some advantages over a multi-purpose simulator such as MSFS:

no need for a global scenery - Kennedy Space Center is sufficient
no background-reloading - because only the landing site is simulated
just a flat plane, no global coordinates - not really a hill at KSC/Florida, the runway is the origin
so no complicated collision detection required
easy internet distribution, because no need to ship a global scenery
etc.

And: I have 50% of the code sitting around on my harddrive already, dubbed FSNG. So a large portion of the project consists of simply refactoring stuff.

There are already some parts working. I have "inserted" the old analytical sky model by Preetham et al., and I have improved the terrain/sky transition, by chosing the sky-color as fog-color in the shader of the terrain. Have a look:

I really like the diffuse transition from sky to terrain - no clear line. I can imagine diving from the orbit into the soup of the earth atmosphere.
By accident, the time of day is evening in this image, so it looks a little bit like Huygens arriving at Titan IMO ;).

Here is a view of runway 33 at KSC. The vehicle assembly building as well as launchpad 39a/b can be seen right of the runway. Of course there is no 3D model yet.

The fog is a little thick down there, but it's just a matter of tweaking some constants in the shader.

Now you will probably ask yourself: where does this guy have the satellite imagery from?
Well, there is a great website out there: http://eol.jsc.nasa.gov/sseop/clickmap/. It's a collection of astronaut photography from various missions. Like most NASA stuff, NASA does not claim a copyright on the photos, so they can be freely used according to my understanding of the law.

Of course, these images are not perfect - they are not 100% top down, images have different color balances, some portions might be covered by clouds, and the the point-of-view is different for each photo. These problems are not un-solveable though:

This image shows the March CTP of MS Expression (Acrylic) - a great software IMO. I have inserted something like 8 images (each around 3000 pixels wide), where each image resides in one layer. Then I sorted them by quality. In case there is a cloud somewhere, I use a soft eraser to remove the cloud - so instead of a cloud, the previous layer shines through and replaces the cloud. One has to match all images, and this is REALLY difficult, because all images have been taken from different angles.
I also had to match the color-balance and -contrast - also very difficult. Not to mention all the life-effects for 8 of these huge images, making editing very slow. But in the end, I succeeded I think. With smooth transitons and water masked (not 100% done yet), it looks really good. Exporting this beast of Expression document to 4096x4096 only took 30 minutes ;).

Until now, I really enjoy this project (working with code that already proved functional, using Expression/Acrylic, coding shaders). Hopefully, this one has a better future than FSNG.

Really hope to show more stuff soon.

Industrial...

2006-02-18T20:55:00.000+01:00

Stumbled by accident over the website of CountZero.
You can download impressive calendars with industrial photography from there.
Very nice!

Tastes best with Industrial Music.

Quick'n'Dirty MDX Feb06 Review

2006-02-09T09:56:00.000+01:00

It's a little bit like christmas, when you wake up, and see a new release of DirectX in your feedreader.

The best way to check the changes, is to fire an existing project through the compiler with the new assemblies installed.

This was quite flawless - the compiler did not complain except DirectInput::InputList is now removed. So in case you need to enumerate DirectInput::Devices or it's object instances, you have to use System::Collections::ObjectModel::ReadOnlyCollection. I didn't understand it anyway, why MDX brought it's own list classes with it before (with such a powerful framework in the background).
Well, the current state is what counts, and these ObjectModel classes look like a very clean list implementation for a class library, because they have been designed exactly for this scenario.
In fact, I wasn't even aware of these classes before, and I will consider to use them extensively in the future, when designing libraries (DirectlyInput anyone?).

Next, the issue with DirectInput::Device::GetBufferedData() is now solved, as this call appears again in the class definition. I haven't used it yet though.

Direct3D compiled without complains, i.e. it looks like there are no big changes.
The project also executed without problems, so it looks like runtime behavior didn't change as well.

Nevertheless, I will continue to design my DirectlyInput wrapper, because other than DirectInput, this class library will provide a high-level way to communicate with DirectInput.

New song Again and Again and AiAiAi.......!

2006-02-06T00:34:00.000+01:00

Yes, another song again. This time it is house, to be exact: Sunny summer sound with caribbean style rhythm.

There is even cover art this time:

Listen at Soundclick: http://www.soundclick.com/bands/songInfo.cfm?bandID=459773&songID=3453697
Reviews at ReasonStation: http://www.reasonstation.net/songs/songinfo.php?song=45071

Have fun!

Two New Songs

2006-01-30T16:04:00.000+01:00

YES!

Two new songs of mine at soundclick.

One from the dark side:
http://www.soundclick.com/bands/songInfo.cfm?bandID=441088&songID=3418396

And one for the dance site:
http://www.soundclick.com/bands/songInfo.cfm?bandID=459773&songID=3418306

ENJOY!

NOT to __hack DirectInput

2006-01-27T23:44:00.000+01:00

Little updated concerning my previous post.

I decided to go with /clr:oldSyntax and the old C++ with managed extensions syntax, known for it's double __underscores.

This way, it really works to create a decent wrapper of DirectInput without PInvoke.

Besides back-acclimatization to the old __syntax, this little project is steadily progressing.

I am asking myself how to distribute the project. I think I will choose SourceForge, because of it's simplified project hosting abilities and audience. So I will have to make it open source, what might not really be too bad for me.

So, which license to choose? I have in mind, that the compiled assembly should be shared freely, as long as I remain the copyright holder. In case somebody modifies the source code, he has to distribute the modified source code. Not sure, I think it was the LGPL matching my needs!? Any suggestions?