Messages

46

General discussions / Re: Searching for music compositor software

« on: January 31, 2014, 09:00:09 am »

The kind of music should be completely independent of the sequencer used. LMMS is the closest you can get to a professional tool for free as far as I know. To sum up the bad news: as a music producer you'll have a hard time doing things without spending any money - at least if you want some convenience.

If you are willing to spend at least a little money, go for FL Studio. The (still pretty common) rumor that it's a child's toy which can't be used to professionally produce music have been false for years by now. It's extremely powerful, yet relatively simple to use compared to Cubase, Logic and the likes. I will recommend it! However, it's not really useful for live performances at all.

Depending on how you are going to make music, you should also get a MIDI device after all. I personally use a hardware knob and slider panel to input effect automation, it's really a lot more convenient and natural sounding than drawing curves and stuff like that (I've been there). If you are more into preparing samples and sequence them on the fly, go for Ableton and maybe a Launchpad by Novation. Both do not really have a good free/cheap replacement in the same quality.

Music production can easily become a very expensive hobby. It's a shame, but once you fall in love with it you will be ready to spend a bit of money.

47

Java / Re: NetBeans + jSFML, linker error at first project

« on: January 24, 2014, 08:35:02 am »

Sorry for the late reply, I'm not really sure about what's wrong here.

Did you check the directory it's refering to? C:\Documents and Settings\cx3\.jsfml\windows_x86\ ?
It should contain 14 files, DLLs and MD5s.

If so, the message seems to suggest that there's actually something missing on your system. Do you have the Microsoft Visual C++ 2010 Redistributable Package installed? It is required, but I'd expect the message to be different.

48

Java / Re: Is there a more efficient way of updating VertexArrays?

« on: January 24, 2014, 08:32:01 am »

It's the way it's meant to be, so for right now, no.

For JSFML, I chose simple data structures to be immutable, ie they cannot be modified after construction. This can avoid a lot of potential pain when dealing with vectors and colors, for instance, passed as method parameters. I honestly would want to preserve consistency, but what I could do is introduce additional "copy"-like methods that return new instances. That might shorten the code a bit.

I know that it feels a little awkward to create new objects all the time, but that's probably the way of thinking when you come from C or similar worlds? Since you talk about "efficiency", I actually did a few benchmarks to compare this to the paradigm of publicly assignable fields. The performance difference was barely notable - the JVM is extremely efficient dealing with object re-allocation. I should repeat these benchmarks to get some referable data for this.

49

Java / Re: SFML and LWJGL strange bug

« on: December 03, 2013, 09:10:35 am »

I'd love to give you an answer, but for now, I can't. LWJGL support has not really been researched too much yet - the last time I tried it, I didn't see a thing. Therefore, cool to see anything working with LWJGL and JSFML at all!

I tried your FPSCamera's strafing methods and those work fine, so I assume there is something wrong with the depth buffer. Clipping occurs way too quickly despite the settings applied by gluPerspective. You didn't give any context settings to your render window, therefore the context is initialized with zero depth buffer bits. However, passing one with forced depth buffer bits didn't do the trick:

ContextSettings contextSettings = new ContextSettings(8, 0, 0, 2, 0); //8 depth bits, rest is default
RenderWindow win = new RenderWindow(new VideoMode(800, 600), "SFML Game", WindowStyle.DEFAULT, contextSettings);

I will have to sit down at some point and investigate. If you happen to find out something, it would be great if you shared the info.

50

Java / Re: getPlayingOffset() always 0?

« on: December 02, 2013, 08:08:54 pm »

Confirmed, this was a misinterpretation of the method on my part.

51

Java / Test Release #5

« on: December 02, 2013, 08:02:19 pm »

I just uploaded another test release of JSFML which addresses the vertex array limit (it is lifted) and fixes some serious bugs concerning floating point conversion between Java and native code.

It's available here: http://jsfml.org/?page=download

At the moment, there is no way for me to provide a Mac OS X version. 

52

Java / Re: getPlayingOffset() always 0?

« on: November 28, 2013, 07:03:34 am »

Sounds like a bug, I will check on this.

53

Java / Re: Trying to draw a line using VectorArray

« on: November 28, 2013, 07:03:13 am »

Vertices are immutable in JSFML, that's why there is not "setColor". That was a design decision for simple data structure classes, although it has not been too popular for vertices, so I might make an exception for these at some point.

wintertime is right, you can assign a color via the constructor. You can always consult the Javadoc for these kinds of problems.

And wintertime is correct about the primitive type for vertex arrays. You simply created a "new VertexArray()", that will cause every vertex to be drawn as a single point ("PrimitiveType.POINT"). Create it with "PrimitiveType.LINES" and it will connect each two vertices with a line. This is also documented in the Javadoc.

54

Java / Re: Trouble setting the viewport

« on: November 28, 2013, 06:58:55 am »

I will release a new version this weekend.
However, Hiura and myself are currently not able to build Mac OS X binaries, so the update won't be available for Macs. 

55

Java / Re: SFMLNativeDrawer.MAX_VERTICES

« on: October 15, 2013, 05:47:39 am »

Gee, I could have sworn I lifted that limit, but it looks like I didn't! I just pushed a commit that automatically enlarges the buffer as needed. Since no native code had to be changed, here's an updated jsfml.jar that should fix the problem, would be nice if you could give it a try: http://jsfml.org/files/jsfml-vertexlimit.zip

Thanks for reporting! This week I should be packing up another test release to contain another critical bugfix in native code.

By the way: I did not really get JOGL or LWJGL to work with JSFML yet, that topic is up for research. Groogy had it working with JOGL I think, this thread contains his log: http://en.sfml-dev.org/forums/index.php?topic=9223.0

56

Java / Re: Trouble setting the viewport

« on: September 19, 2013, 02:00:04 pm »

"Luckily", I didn't build one yet, because I just found and fixed another critical issue of a similar kind.
Unless I find even more, I hope I can pack up another release this weekend.

57

Java / Re: Trouble setting the viewport

« on: September 16, 2013, 04:13:03 am »

Hello, sorry for the late reply, busy days / weeks / months.

You found quite a simple but serious bug: float rectangles passed to C++ are read as integers, ie all values of your float rect are rounded down to 0. That was probably a copy & paste mistake and I've just pushed the fix. However, there is no quick workaround for you right now. I hope I can pack up another release one of these days.

Thanks for finding and reporting!

58

Java / Re: Trouble setting the viewport

« on: September 10, 2013, 11:36:18 am »

Hi, I cannot see any mistake in your code right now.
If the same works in C++, it might indeed be a binding problem. I will give this a shot later today and report back.

59

Feature requests / Re: Open Watcom Support

« on: August 18, 2013, 06:59:57 pm »

Uff, games like Doom, Duke Nukem 3D and Command & Conquer were built using that compiler. That IS a while ago...

60

Java / Re: GUI for JSFML

« on: August 16, 2013, 06:40:03 pm »

For what I know, nobody has written any yet.