SFML community forums
General => General discussions => Topic started by: Laurent on January 26, 2010, 10:52:13 am
-
Hi
I'm going to release SFML 1.6, which is mainly a bug fix release. Is there anything that you would like to be improved/fixed in this release? Did I miss something?
-
Is there anything that you would like to be improved/fixed in this release? Did I miss something?
Is refreshing while window resizing fixed yet?
-
No, I can't solve this issue. Windows runs its own internal loop while the window is resized, and all I can do is to catch a special event that is triggered by this loop. Unfortunately I can't do anything with it.
-
Hi!
1.6 (it's trunk right?) currently fails to build on kFreeBSD like that:
make[2]: Entering directory `/home/christoph/proj/deb/sfml1.6/src/SFML/Window'
g++ -o Context.o -c Context.cpp -W -Wall -pedantic -I../../../include -I../../ -O2 -DNDEBUG -fPIC
In file included from ../../SFML/Window/Joystick.hpp:71,
from ../../SFML/Window/WindowImpl.hpp:33,
from Context.cpp:29:
../../SFML/Window/Linux/Joystick.hpp:95: error: 'ABS_MAX' was not declared in this scope
Context.cpp:35: warning: '<unnamed>::Dummy' defined but not used
make[2]: *** [Context.o] Error 1
make[2]: Leaving directory `/home/christoph/proj/deb/sfml1.6/src/SFML/Window'
make[1]: *** [sfml-window] Error 2
make[1]: Leaving directory `/home/christoph/proj/deb/sfml1.6/src/SFML'
make: *** [sfml] Error 2
-
Thanks, it's fixed :)
-
I don´t know if it´s already fixed but when I am using font I get this memory error. Maybe it should be fixed :)
-
OGreat :) builds now fine. I ight come around to do some tests later then :)
-
I don´t know if it´s already fixed but when I am using font I get this memory error
Can you tell me more about this error?
-
I don´t know if it´s already fixed but when I am using font I get this memory error
Can you tell me more about this error?
It says something like this, I shall translate the best I can.
program-exe - Programfail
The instruction on "0x6903a33a" referd to th memory on "0x00000054". It didn´t perform a memory arrangment. This error returned: The memory could not be "read".
It pops up everytime a close the window when I draw use/draw a font.
-
No, it's not fixed. I can't even reproduce this bug on my computer anymore :?
-
Any chance of sf::Image.CopyScreen() producing a flippedY image being fixed?
-
Nop, sorry.
-
Nop, sorry.
What about some kind of "WaitEvent()" that blocks untill an Eventis added to the Loop?
-
Added in SFML 2. SFML 1.6 is just a bug fix release.
-
Is is so that SFML 1 is now finished (bug fix only), and any changes will happen only in SFML 2?
If so, then I would like more graphic documentation.
I am thinking about the section of the tutorial that explains how SFML draws things on the screen, in particular.
But actually, the whole tutorial (which seems to be the main documentation) could do with graphical representations of how things work, throughout, to help those of us who are graphic learners understand better.
And thanks again for making SFML! You rock!
-
Is is so that SFML 1 is now finished (bug fix only), and any changes will happen only in SFML 2?
Absolutely.
If so, then I would like more graphic documentation.
I am thinking about the section of the tutorial that explains how SFML draws things on the screen, in particular.
But actually, the whole tutorial (which seems to be the main documentation) could do with graphical representations of how things work, throughout, to help those of us who are graphic learners understand better.
I'm not sure I understand 100% what you mean. Can you explain more, or give an example?
-
Maybe he means screenshots.
-
No, not screen shots... :) What does a screen shot of a programming library look like? Is it a photo of a screen with the source code on it? :lol:
What I mean is graphical illustrations of how things work.
Let's take rendering a sprite, for instance.
How many coordinate systems are at work here? How do they relate to each other? What am I doing to which one when I set Sprite.SetCenter, or any of the other possibilities SFML gives me?
How do I get from that drawing of a teddy bear that I made in paint, to that scaled, centered, and rotated 30 degrees counter-clockwise around its tail version in the SFML view port? How do I know in advance whether its nose will be sticking out of the screen?
A couple of nice illustrations could explain all this nicely, I believe.
For an example of how such an illustration might look, take a look at chapter 3 "Multiple coordinate systems" in 3D Math Primer by Fletcher Dunn and Ian Parberry, if you have that available.
But that is not the only subject that could do with graphical illustration. Another nice thing would be an overview of everything that is SFML, and how they go together. A graphical illustration of all the parts of SFML, and their relations.
Pretty much every part of the documentation could be improved with imagery.
It may not mean much to you, but many people learn things faster, gain a deeper understanding, and remember things better, if there is a simple graphical illustration to go along with the text. They are not meaningless fluff, but graphical explanations.
I really believe this could open up your library for many more people, especially beginners.
-
Oh, and I have one more request:
If at all possible, I think it would be best if the many compiler warnings during debug compile could be taken care of.
I know they can be suppressed, but should they have to be?
And if I take the steps to suppress them, could other warnings, that I might want, be suppressed, too?
Not being the expert, I really don't have a clue. So I just let the compiler moan, even though I really don't like it.
I think this is very confusing, especially for beginners, and I know I have found it to be.
-
If at all possible, I think it would be best if the many compiler warnings during debug compile could be taken care of.
I know they can be suppressed, but should they have to be?
What warnings? There's no warning in SFML, except in the external libraries that I include directly to the source code (and I won't fix those).
-
Here is a small selection of the warnings I get in Visual C++ 2008 Express:
1>sfml-system-s-d.lib(Platform.obj) : warning LNK4099: PDB 'vc90.pdb' was not found with 'C:\libs\SFML-1.5\lib\vc2008\sfml-system-s-d.lib' or at 'C:\Documents and Settings\Administrator\Dokumenter\Visual Studio 2008\Projects\B2BodyMaker\Debug\vc90.pdb'; linking object as if no debug info
1>sfml-system-s-d.lib(Clock.obj) : warning LNK4099: PDB 'vc90.pdb' was not found with 'C:\libs\SFML-1.5\lib\vc2008\sfml-system-s-d.lib' or at 'C:\Documents and Settings\Administrator\Dokumenter\Visual Studio 2008\Projects\B2BodyMaker\Debug\vc90.pdb'; linking object as if no debug info
1>sfml-system-s-d.lib(Sleep.obj) : warning LNK4099: PDB 'vc90.pdb' was not found with 'C:\libs\SFML-1.5\lib\vc2008\sfml-system-s-d.lib' or at 'C:\Documents and Settings\Administrator\Dokumenter\Visual Studio 2008\Projects\B2BodyMaker\Debug\vc90.pdb'; linking object as if no debug info
I get 71 of those in my current project. But if they come from external libraries, then of course you shouldn't fix them. :)
I am very, very happy with SFML, by the way!
-
These are linker warnings about debug symbols not being found. I don't know why some people get them, it doesn't happen on my computer.
-
That's weird - which compiler are you using? And do you set it up in a special way?
-
I'm using Visual Studio 2008 professional, and I do nothing special.
-
SFML 1.6 will allow to draw drawable objects to images (being Image a RenderTarget)?
I did a quick look at the trunk snapshot, I see a reference to a RenderImage class on sf::Image, but I don't find it.
-
No, sf::RenderImage is available only in SFML 2. I removed all the references to this class in the current trunk, I forgot to remove them a long time ago.
-
These are linker warnings about debug symbols not being found. I don't know why some people get them, it doesn't happen on my computer.
Well, because they aren't in the package. Probably they are on your computer, because VS automaticly creates them, but you didn't add them to the archive you distribute.
But it is really simple to resolve those warnings. Simply rebuild the SFML project with VS.
-
Well, because they aren't in the package. Probably they are on your computer, because VS automaticly creates them, but you didn't add them to the archive you distribute.
I don't understand how debug informations work in VC++. I'm almost sure that I've already used debug libraries, and those libraries contained their debug symbols directly, not in a separate PDB file.
I also tried to remove every PDB from my SFML directory, and I still got 0 linker warning.
I really need to figure out what's going on with PDB files and debug symbols :)
-
maybe its a problem between different minor versions of visual studio...
-
Well, I'm new here, but will the libraries be compiled with GCC 4.4.x? Could they be?
-
Did you have seen this? http://www.sfml-dev.org/todo/index.php?do=details&task_id=149&tasks=&project=1&due=12&status[0]=
-
Thanks! I had not seen that, and it certainly answers my question :)
-
sfml 1.6.
when i can download from main page?
i dont want from svn. ;)
-
Very soon.
-
Will smoothing be fixed so it shows the true background color, not the one that is set to be hidden?
Really need this.
-
Will smoothing be fixed so it shows the true background color, not the one that is set to be hidden?
Which bug is it? Can you explain?
-
well say i have background color Magenta, CreateMaskFromColor to Magenta also. Smoothing turned off it is jagged and don't look too good. Smoothing turned on and there will be a line with Magenta color around the object. This also happens when using anti-aliasing from drawing programs, but that is understandable since their anti-aliasing isn't meant for having any other background than what is there. It seems to me though that SFML should be able to handle it.
-
Smoothing is a bilinear filter which calculates a color from its neighbours. So it's the normal behaviour to get some magenta blended with your non-transparent colors.
By the way, you should have given me the link to your post (just found it) instead of duplicating your problem in two discussions ;)
-
Looking forward to 1.6 :)
-
Is there anyone who could build the linux64 binaries for SFML and CSFML (using <repository>/tags/1.6) please?
Thanks :)
-
I can do so when I'm home, just plain make + taring up?
-
just plain make + taring up?
Yep :)
I forgot to mention that I also need the compiled samples (make sfml-samples).
Thank you.
-
ok building.
does that in- or exclude CSFML python-sfml (or others?)
[EDIT]
Also, do you want full packages or just the binaries?
-
does that in- or exclude CSFML python-sfml (or others?)
It includes CSFML, like I said in my first message. That's all, I don't manage the other bindings.
Also, do you want full packages or just the binaries?
Just binaries, I can build the packages on a 32 bits machine ;)
-
OK should have read again more carefully. It's now going there:
http://random.christoph-egger.org/sfml-1.6-amd64.tar.bz2
http://random.christoph-egger.org/sfml-1.6-amd64.tar.bz2.gpg is a signature to check the archive's integrity
-
Thank you very much ;)
-
Out of curiosity, will we be seeing a Mac-release this time again? From the forums it doesn't appear as though a new Mac maintainer has been found.
-
For 1.6, yes. For 2.0, not yet :(
-
Congratulations on 1.6 Laurent, keep up your incredible work!