Bug report - Feature request — Orx: Portable Game Engine

movements depend on framerate

Tang — Sat, 16 Jul 2022 04:26:15 +0000

Low framerate result in low moving speed and vice versa.

How to reproduce:
1) turn off vsync
2) set frequency to 0 in the clock section (otherwise vsync = false will have no effect)
3) set a fixed physics framerate
3) start a game, any game
4) limit all cpu frequency to 500 MHz, and the framerate will drop significantly which is expected. What is unexpected is that object also moves significantly slower.
5) cancel the cpu freq limits, rerun the game, the framerate can easily exceeds 1000 fps on my device and my little hero becomes a flashman.

Since Orx is not a engine especially dedicated for some kind of console platform, I would say this should be a bug. I looked into the code (glfw, orxPhysics.cpp) and confirmed that the physics updating code intends to achieve fixed time step and I counted that the calling rate of b2world->step is correct, which means this part of code works fine. I also confirmed that orxSystem_GetTime gives correct results and main clock dt is correct. So I honestly got no clue, why would physical movements depend on main framerate?

While set orxSPAWNER_KU32_MASK_USE_RELATIVE_SPEED, Spawner should apply scaling prior to rotation

Tang — Thu, 27 Jan 2022 08:22:57 +0000

When using spawners with the "UseRelativeSpeed" flag along with scaling of parent objects, rotation is applied before scaling. This can cause problem when using scaling to flip objects. Taking "The Beginner's Guide to Orx" as an example where we use scaling to flip the hero and set the "UseRelativeSpeed" flag to "spawner" in the "BulletSpawner" section. If a non-zero rotation is applied to the hero, the bullets won't fly in the right direction when the hero is flipped. Just as shown below,

This is because the parent object's scale is applied piror to the its rotation, which can be seen from the code in "orxSpawner.c:823",
orxVector_Mul(&vSpeed, orxVector_2DRotate(&vSpeed, &(pstSpawner->vSpeed), pstSpawner->fPendingRotation), pstSpawner->pvPendingScale);
This problem can be solved by modifying the line to
orxVector_2DRotate(&vSpeed, orxVector_Mul(&vSpeed, &(pstSpawner->vSpeed), pstSpawner->pvPendingScale), pstSpawner->fPendingRotation);
All three similar lines in the switch statement should be modified accordingly. Then the direction of bullets will correctly follow the direction of hero.

Release 1.11 setup.sh fails for new install

orsonb — Sun, 20 Sep 2020 00:10:18 +0000

I just tried creating a new install on Mac from the stable release, 1.11. The setup.sh script failed to download the resource file https://bitbucket.org/orx/orx-extern/get/0487f7600096.zip.

When I go to that website there is no orx-extern folder in https://bitbucket.org/orx.

Everything is fine when installing HEAD, but I don't want to rely on unreleased code for my game.

Can someone either release a 1.11.1 that points to a valid url, or restore the file to the url above?

Thank you

FPS display in debug

funemaker — Wed, 13 Nov 2019 03:58:41 +0000

This is extremely minor, but in case it hasn't been noticed-- I'm using windows VS 2017.

When I create a new project using the init script, I get a d.ini file that has
ShowFPS = true
in the [Display] section. The result is that the FPS does not show when running the debug build since this seems to belong in a section called
[Render]

If I add [Render] and move the ShowFPS line to that section, I then get the frame rate showing when I run the debug build.

Another apparent issue with FPS is that if I set Fullscreen = true in my [Display] section, the displayed framerate seems unreasonable (like 1500). If I set fullscreen to false, it hovers around 120. Not sure if this is something weird with my hardware (laptop with nvidia) but wanted to mention it.

Learning - snags / hangups

seltzy — Fri, 23 Dec 2016 22:38:23 +0000

Hi all. I'm a newbie learning the ropes of orx and am in the process of compiling where I've hit bumps and snags and such. Hopefully this helps improve the tutorials for other learners, though a good lot of this may well be due to my own ignorance.

Before I begin, thank you everyone who has contributed to the wiki and tutorials. It really does help. It's looking much more approachable than it did a few years ago when I first tried using orx.

I'll update this periodically as I come upon new challenges. Here's what I have so far.

My biggest initial hangup every time I've started was what options there were for project setup - it's super flexible but tutorials led me to beleive I should do it a certain way that I didn't like. This is more particular to me since I'm picky about having a good template project setup. I think I've figured out a setup that I like, but seeing a list of possible suggestions to illustrate this flexibility would be helpful to newcomers.
none[strike]Another hangup was the fact that the config file is next to the executable - it would be nice to note that this can be changed in code.[/strike]
The next challenge was understanding how camera and viewports interact. [strike]Seeing a lot of sample config files use display resolution for frustum dimensions led me, someone who had forgotten the nuances involved in selecting frustum size, to think that this was the norm and should be conformed to.[/strike] Having some more pictures in the viewport/camera tutorial that illustrate how the camera behaves in relation to objects would be helpful. Explaining what kinds of things to take into consideration would be good. Explaining what worldspace looks like in the context of orx would also help. Having a solid reference for the user to anchor their mind to is essential for this.
AutoScroll was a temporary confusion. At first config files will appear to be black magic, so the language here may suggest that the camera scrolls with X instead of configuring parallax scrolling behaviour. The Scrolling tutorial on the website explains the many nuances of the configurations for Cloud, Sky, and Camera, but the config file in the repository does not and the explanations in comments are not as descriptive as they could be.
none[strike]The animation tutorial makes frames in wrong order - they increment down each column, not each row.[/strike]
I assumed WatchList automagically hotloaded objects at runtime - didn't occur to me that pre-existing objects wouldn't change - totally magical thinking in retrospect
none[strike]The physics tutorial doesn't really explain the motivations behind there being body and bodypart. I can imagine why, but clarifying their differences and giving a brief explanation of why the properties available to them differ might be useful for some users who are unfamiliar with 2D physics systems. If these are covered elsewhere on the wiki, a link would suffice.[/strike]
In general these tutorials cover only what is needed, which is good for brevity but bad for those who are intuitive and curious. Having links to more detailed descriptions (like wikipedia articles do) would add a significant level of polish to these tutorials.
This is very minor, but having prev and next links would also be nice. I only say this because some browser plugins use those as a way to streamline browsing - I use one such plugin that adds keyboard shortcuts for those. Minor annoyance. Not a huge issue.
The changing direction tutorial doesn't seem to work for idle animations. I think this is because no idle animations are defined. This may be a mistake on my part, but if it's not, please show why this happens in the tutorial and fix it (much like in other sections of the tutorials). It also takes a more code-oriented approach than the official tutorial files. [strike]I think overall suasage's tutorial involves less typing. In general, I think this is what is important to strive for as a programmer. It calls into question what the advantage is to the full config approach for animations. It'd be nice to understand the pros and cons of each. The anim tutorial files are quite large and intimidating by comparison, but if there was a way to crunch that file down a bit that'd be ideal.[/strike]
The property name "Speed" is somewhat misleading. Technically speaking, as a Vector, it describes Velocity. When I think Speed, I think of a Scalar value. Overall this is me being picky about naming conventions. Most people don't know/care about the difference. It mostly struck me as strange, so it didn't kill momentum too much.
none[strike]On the shooting tutorial, there's a minor typo. Where it says "This spawner can shoot BulletObjects. It can shoot one at a time with 0.02 milliseconds between shots" is inconsistent with the config above it; the interval is 0.1, not 0.02.[/strike]
Animation tutorials talk about animations being stored as a directed graph but there's little information illustrating that. Adding a picture that demonstrates what such a graph looks like from the config in the tutorial would be much better than a wikipedia link.
Colliding with the star kills the player's momentum since they are physically colliding with it. It'd be nice to know how to prevent that from happening. Imagine if that happened when picking up rings in Sonic, or coins in Mario.
Just remembered, in the jumping tutorial, the player can actually fly if they repeatedly tap jump. Not a bad idea to point this out and perhaps show people how fix it.
The spawning monsters seem to still be ground sliding by the end of the tutorial. I couldn't really tell much of a difference between before and after setting AllowGroundGroundSliding=false. Same sorta goes for angular damping. Apparently AllowGroundSliding was removed.
The naming conventions for physics types isn't quite what I'd expect. Box is okay I think, but does sphere really make sense in a 2D game? I sorta just guessed it would be circle. Threw me off for a second. Not a huge deal, just a bit odd.
I was completely ignorant of the fact that image space, physics space, world space, and sound space each have their own ratio to pixels. It led to quite a bit of confusion, particularly with the difference between world space and physics space. Although a lot of that is abstracted away from the user, I think it'd help to have that information in one place. This sort of relates back to my problem with understanding the world size. It may be worth having a short wiki page just on this relationship. I think part of this stems from being a user who had been using IDE-like engines with world editors that give you a spacial sense of the world right off the bat in no unclear terms. When you have a world editor that you can prototype in and experiment with engine features these differences become apparent, but with library-like engines the user doesn't have this intuitive knowledge as immediately. To get the same effect, this information needs to be demonstrated in some other way. This gets back to giving the user an anchor to ground themselves in their sense of all aspects of the game world.
I'm pretty sure none of the tutorials use deltatime. I'd suggest registering an Update method that uses a clock instead of using Run. You'll probably need to adjust speeds as well.
It may be wise to set velocity instead of applying impulse when moving characters around. It's better for determinism, which is usually what we want for gameplay.
Double-check hero's survival tutorial. When setting hero's lifetime to 0, there's a possibility of a crash due to it being... well... gone but not exactly null.
In terms of dealing with deprecation problems on the wiki, a solution that's similar to http://en.cppreference.com/ would be very nice. They annotate all behavioural changes to the API with versions/standards that were in use at the time. That way anyone that references it can clearly see what got deprecated when. That might be hard to do with the current doxygen/wiki setup, but something to think about. Having a no-nonsense way of seeing when tutorial and documentation are out of sync with the current state of Orx is essential. #8 should help mitigate this issue, but won't completely solve it. Perhaps if there was a good way to automate the tutorials with unit tests, it'd be easier to see when things get out of sync (at least with some things).
The API dox are really hard to use since the search box doesn't let you search for middle-parts of a word. If I want to find GetString, I need to know what my target result starts with. This may mean typing in orxConfig_GetString, which maybe I didn't know that was I was looking for was under orxConfig. This also makes it impossible to explore the API documentation in a meaningful way without using some sort of command line wizardry on the source.

BUG: Console control keys barf after config reload

enobayram — Sun, 28 Feb 2016 10:36:57 +0000

Hi,

I'd like to file a bug report: While the console is open, if you change one of the .ini files loaded by orx, and therefore cause it to be reloaded, the control keys (like enter, left arror, backspace etc.) stop working. You can still input characters, but you can't, say, delete characters or issue a command. When you close the console (console toggle key keeps working) and reopen it, the control keys start working again until the next reload. I guess the bug is due to the input set reverting to the initial one.

The bug gets quite annoying if you periodically call orxConfig_Save.

Cheers

Incosistenf FX behavior between Linux and Android

enobayram — Fri, 19 Feb 2016 11:34:18 +0000

Hi,

I've just noticed an inconsistency in the way FXs behave on Linux and on Android. Check the following config sections:

[TestText]
Graphic=@
Text=@
String = Test
Position = (-300, 200, 0)
Pivot = center
FXList = @
SlotList = TestSlot
Scale = 4

[TestSlot]
Type = position
Curve = linear
EndTime = 0.5
EndValue = (0,-200,0)

On Linux, a TestText object behaves as you'd expect. It moves up for half a second as soon as it's created. On Android though, in addition to not moving at all, in a strange way, it appears at position (0,0,0). The behavior on Android gets fixed if one adds:

StartValue = (0, 0, 0)
StartTime = 0

to TestSlot.

Without commenting on whether there should be default values on these keys, I think it's worrisome that the behavior is different across platforms.

Cheers

orx 1.7 - I can't get mouse wheel delta

ainvar — Fri, 12 Feb 2016 12:44:32 +0000

Hi all,

it seems I'm not able to get mouse wheel delta from orx.
The same problem compares to orx 1.6 and 1.7.

In my application class, a scroll based class, i have a sequence of calls in the Run() function:

- orxMouse_GetPosition
- orxMouse_GetMoveDelta
- orxMouse_GetWheelDelta
- orxMouse_IsButtonPressed

I can detect mouse movement and button pressures but nothing about wheel delta.

In orxMouse.c of GLFW plugin, in function orxMouse_GLFW_MouseWheelCallback() I see sstMouse.fWheelMove is set with a value corresponding the wheel movement.

The problem is that before I call orxMouse_GetWheelDelta() to get the value, orxMouse_GLFW_Clean() is called by orxClock_Update() and it blanks the value.

Am I missing something or wheel mouse handling is not working at all (and no one used it then no one niticed it )?

Font structure ignoring TextureCorner

sausage — Thu, 28 Jan 2016 10:08:43 +0000

It appears that TextureCorner is being ignored by the font structure.

In my font sheet, I had an 18 pixel gap on the left hand side of the sheet which meant all letters were slightly offset.

Doing:

TextureCorner = (18, 0, 0)

...made no difference so I simply cropped off the left 18 pixels in gimp and all was well.

So this is easily solved, but just reporting this as TextureCorner is listed as available at:

http://orx-project.org/wiki/en/orx/config/settings_structure/orxtext

Why is orxFontGen cropping glyphs?

enobayram — Thu, 17 Dec 2015 15:07:20 +0000

Could you help me understand why orxFontGen crops some glyphs? For instance it's cropping 3 rows from the character 'g' while generating a font sheet with character size 40 with the Sincere Heart Font. I can't figure out why it's doing that and which command line argument is relevant.

Thanks

Feature Request: Improving orxConsole on Mobile

enobayram — Fri, 04 Dec 2015 09:35:53 +0000

Hi,

I was wondering if we could make the console more usable on Mobile, that would be a great help for debugging, which is already harder than it needs to be on mobile.

Right now, I have a debug button in my game that shows the console, and I can use orxKeyboard_Show() to bring the virtual keyboard up, but there are two problems:
1) The command input line stays under the virtual keyboard. Maybe we could alleviate this with a function like orxConsole_SetHeight?

2) OK, this is more of a bug report, the "Enter" key on my Samsung Galaxy Note 10.1 virtual keyboard doesn't issue the command, so basically, I have no way of "enter"ing any command.

The console is already very useful as a logging medium but it would be ten times more useful if we could also enable inputting commands.

As a side question; what do you guys use for debugging on mobile (Android in particular)? Maybe we can share some tricks.

Thanks!

"Console.Enable false" messes up inputs

enobayram — Fri, 04 Dec 2015 02:57:40 +0000

Hi, I'd like to make a bug report. If you open the console using the toggle key and issue a "Console.Enable false" command at the console, the inputs stop working in the game. It's as if the console keeps interrupting them even though it's disabled.

Cheers

Spawner interpolation badly interacting with FX

enobayram — Sat, 03 Oct 2015 15:55:05 +0000

I've noticed a glitch caused by the interaction of FXs and spawner interpolation. If an object has an FX with a StartValue != 0 (or 1 for scale), and if the object gets created with a spawner with interpolation, you can occasionally observe very strange scale and position values for the objects spawned on the first frame. If this is not a known bug, I can try to prepare a minimal setup.

orxConfig_Save is bugged.

Knolan — Thu, 06 Aug 2015 21:19:46 +0000

The parameters given to the save callback seems to be wrong:

static orxBOOL filterAchiev(const orxSTRING _zSectionName, const orxSTRING _zKeyName, const orxSTRING _zFileName, orxBOOL _bUseEncryption){
    orxLOG("Tracking section: %s / %s / %s",  _zSectionName, _zKeyName, _zFileName);
    if (orxString_NCompare(_zSectionName, "AchieveTrack", 12) == 0){
        return orxTRUE;
    }
    return orxFALSE;
}

This prints:

[18:15:58] [LOG] Tracking section: achievements.ini / (null) / (null)
[18:15:58] [LOG] Tracking section: achievements.ini / (null) / achievements.ini

It seems that the filename is being passed to the section name parameter.

WatchList loading too soon, corrupting data

enobayram — Sat, 24 Jan 2015 14:28:04 +0000

Hi, I've been trying to use the [Resource] WatchList feature, but it's giving me trouble.

I've tried to use it for hot-swapping my textures, but any image file that's not ridiculously small gets corrupted when I try to change it at runtime.

I've dug a bit into the problem and realized that the orxDisplay_LoadBitmap (src/display/orxDisplay.c:385) is being called too soon. I think as soon as I click export in GIMP and the data starts to get written to the disk, orxDisplay_LoadBitmap gets called and receives incomplete image data. For .bmp files it manifests itself as an texture that's missing some parts of the image, for .png files, I get

[ASSERT] [orxDisplay.c:orxDisplay_GLFW_LoadBitmapData():1091] [ASSERT] : <(s64Size > 0) && (s64Size < 0xFFFFFFFF)>

I've further confirmed my "loading too soon" hypothesis by trying to first save the image with a different name and then renaming it as the texture. This way, there was no trouble reloading the texture.

All of this happened on Ubuntu 14.04 with a 30.2kB 200x84 .png image.

Android, duplicate touch source ID with stylus

enobayram — Fri, 30 Jan 2015 10:01:02 +0000

Hi,

I've been running my Orx based Android application on a Samsung Galaxy Note 10.1, which has a stylus.

I've, possibly wrongfully, assumed that payload->stTouch.u32ID would be unique for every continuous touch source. Or that at least, I wouldn't receive two orxSYSTEM_EVENT_TOUCH_BEGIN events for the same ID without a orxSYSTEM_EVENT_TOUCH_END event in-between.

This assumption had held true, until I've tried to use the stylus in combination with my fingers on the Orx based app. Apparently, stylus based touch events CAN have identical ID's to those that emerge from non-stylus sources.

Is this a bug, or expected behavior? Is there any other way to distinguish touch sources?

Thanks

Relative include of .ini files

Trigve — Sat, 24 Jan 2015 15:23:43 +0000

Hi,
I really miss the relative including of the .ini files. That is (using hypotetical syntax) when I'm in some dir named "subdir" in some .ini file and would use:

!@some_other_ini.ini@

then "some_other_ini.ini" would be loaded from the current dir ("subdir"). If no file is found, try to other locations defined in resources.

It would really simplify some .ini generation which are coupled together.

Any opinions pro/against?

Thank You

Compile dependencies for current Visual Studio ver

sonicbhoc — Wed, 31 Dec 2014 22:19:05 +0000

Hey, since Visual Studio 2013 has a free community edition, you should take advantage of it and compile orx to target that new version. I tried, but the compiler version is different and box2d didn't like that too much.

Mouse button down for the frame

Trigve — Tue, 06 Jan 2015 12:05:06 +0000

I was searching for the function in the mouse module, to find if the mouse button was down in current frame. I know there exist the orxMouse_IsButtonPressed() but this function does return the actual state of the button (true if the mouse button is pressed).
What I would need is to have function which will return true only when the mouse button was pressed, and false in subseqent frames when the mouse button is held down. That is I want find if the single click has happened.

Is it somehow possible with the current functions?

Thanks

wiki missing info

Trigve — Sun, 28 Dec 2014 08:56:46 +0000

Hi I just want to note some missing/ out of date information in wiki:

-http://orx-project.org/wiki/en/orx/config/settings_structure/orxcamera Camera is missing the "GroupList" description

-http://orx-project.org/wiki/en/orx/config/settings_structure/orxviewport Viewport has now "TextureList" instead(?) of the "Texture"

iarwain screenshot

coosadog — Wed, 12 Nov 2014 17:21:40 +0000

OK, I did as you requested. I never saw any Display_* with values that high, very low as a matter of fact. Now, that said, I couldn't make the scene scroll. The screenshot should be attached.

Crashes in Tutorials

kulak — Tue, 30 Apr 2013 18:29:30 +0000

I got the latest source code.

02_Clock Tutorial Issue

Just use arrow Left or arrow Right and it reproduces every time.


Thread 1, Queue : com.apple.main-thread
#0	0x00000001018d7565 in _orxDebug_Break at /Users/sergei/SWDev/CrossPlatform/orx/code/build/mac/xcode/../../../src/debug/orxDebug.c:306
#1	0x00000001018c3ff0 in orxClock_SetModifier at /Users/sergei/SWDev/CrossPlatform/orx/code/build/mac/xcode/../../../src/core/orxClock.c:1059
#2	0x000000010000119c in InputUpdate at /Users/sergei/SWDev/CrossPlatform/orx/tutorial/build/mac/xcode/../../../src/02_Clock.c:132
#3	0x00000001018c2f2a in orxClock_Update at /Users/sergei/SWDev/CrossPlatform/orx/code/build/mac/xcode/../../../src/core/orxClock.c:555
#4	0x000000010000169a in orx_Execute at /Users/sergei/SWDev/CrossPlatform/orx/tutorial/build/mac/xcode/../../../../code/include/orx.h:346
#5	0x0000000100001432 in main at /Users/sergei/SWDev/CrossPlatform/orx/tutorial/build/mac/xcode/../../../src/02_Clock.c:230
#6	0x0000000102ead7e1 in start ()

Exception State Registers
trapno unsigned int 0x00000003
err unsigned int 0x00000000
faultvaddr unsigned long 0x000000011ccba000

12_Lighting

All I have to do is start the application. Wait for no more that 2 or 3 seconds. Don't touch keyboard or mouse. Mouse is even outside of the window.


Thread 1, Queue : com.apple.main-thread
#0	0x00000001018da565 in _orxDebug_Break at /Users/sergei/SWDev/CrossPlatform/orx/code/build/mac/xcode/../../../src/debug/orxDebug.c:306
#1	0x000000010193eb64 in orxDisplay_GLFW_SetShaderBitmap(void*, int, __orxBITMAP_t const*) at /Users/sergei/SWDev/CrossPlatform/orx/code/build/mac/xcode/../../../include/../plugins/Display/GLFW/orxDisplay.c:3978
#2	0x00000001018db862 in orxDisplay_SetShaderBitmap at /Users/sergei/SWDev/CrossPlatform/orx/code/build/mac/xcode/../../../src/display/orxDisplay.c:361
#3	0x000000010197811c in orxShader_Start at /Users/sergei/SWDev/CrossPlatform/orx/code/build/mac/xcode/../../../src/render/orxShader.c:794
#4	0x000000010197afa1 in orxShaderPointer_Start at /Users/sergei/SWDev/CrossPlatform/orx/code/build/mac/xcode/../../../src/render/orxShaderPointer.c:481
#5	0x0000000101979f5d in orxShaderPointer_EventHandler(__orxEVENT_t const*) at /Users/sergei/SWDev/CrossPlatform/orx/code/build/mac/xcode/../../../src/render/orxShaderPointer.c:145
#6	0x00000001018dc390 in orxEvent_Send at /Users/sergei/SWDev/CrossPlatform/orx/code/build/mac/xcode/../../../src/core/orxEvent.c:379
#7	0x000000010196ea09 in orxRender_Home_RenderObject(__orxOBJECT_t const*, __orxBITMAP_t*, __orxFRAME_t*, __orxDISPLAY_SMOOTHING_t, __orxDISPLAY_BLEND_MODE_t) at /Users/sergei/SWDev/CrossPlatform/orx/code/build/mac/xcode/../../../include/../plugins/Render/Home/orxRender.c:1261
#8	0x0000000101969f05 in orxRender_Home_RenderViewport(__orxVIEWPORT_t const*) at /Users/sergei/SWDev/CrossPlatform/orx/code/build/mac/xcode/../../../include/../plugins/Render/Home/orxRender.c:1987
#9	0x00000001019520c0 in orxRender_Home_RenderAll(__orxCLOCK_INFO_t const*, void*) at /Users/sergei/SWDev/CrossPlatform/orx/code/build/mac/xcode/../../../include/../plugins/Render/Home/orxRender.c:2083
#10	0x00000001018c5f2a in orxClock_Update at /Users/sergei/SWDev/CrossPlatform/orx/code/build/mac/xcode/../../../src/core/orxClock.c:555
#11	0x0000000100002a0a in orx_Execute at /Users/sergei/SWDev/CrossPlatform/orx/tutorial/build/mac/xcode/../../../../code/include/orx.h:346
#12	0x00000001000027a2 in main at /Users/sergei/SWDev/CrossPlatform/orx/tutorial/build/mac/xcode/../../../src/12_Lighting.c:484
#13	0x0000000102eb07e1 in start ()

Console:

[11:26:44] [LOG] 
- 'MOUSE_LEFT' will create a new light under the cursor
- 'MOUSE_RIGHT' will clear all the lights from the scene
- 'MOUSE_WHEEL_UP' will increase the radius of the current light
- 'MOUSE_WHEEL_DOWN' will decrease the radius of the current light
- 'KEY_SPACE' will toggle alpha on light (ie. make holes in lit objects)
[11:26:44] [DISPLAY] [orxDisplay.c:orxDisplay_GLFW_CompileShader():644] Couldn't compile fragment shader:
ERROR: 0:5: '/' does not operate on 'float' and 'int'
ERROR: 0:13: '^' does not operate on 'int' and 'int'
ERROR: 0:13: Use of undeclared identifier 'vNormalizer'


[11:26:46] [ASSERT] [orxDisplay.c:orxDisplay_GLFW_SetShaderBitmap():3978] [ASSERT] : 
(lldb)

Skeletal Animations?

sonicbhoc — Sun, 13 Apr 2014 21:09:24 +0000

I think I might have brought it up earlier, but I'm curious now, and I think it could use its own thread. What's the status of the skeletal animation system? No pressure, but I am curious to know how far you've gotten with this. I'd love to use it in my project. Because of the simplicity of my project, it would be a good stomping ground to try new things. Thanks!

Patch to get orx to compile on OS X 10.9

nikki — Fri, 06 Dec 2013 11:11:54 +0000

Hey!

Downloaded the source as soon as I discovered orx and tried to compile it. Few problems here and there but I fixed them! Mostly it was -isysroot was wrong for 10.9. First I set it to the correct one but then I found simply removing the option from premake4.lua worked too. Also glext.h had a conflicting typedef with the system OpenGL.framework.

Not sure if this is the right way to fix things, but here's a patch:
https://gist.github.com/nikki93/7822094

Most of the patch is actually due to the premake4.lua changing the output gmake Makefiles (they could probably be removed from the hg repos since they should be generated locally anyway...).

Good work on orx, code looks really neat!

adding ellipsoid shapes to orx

willcaliblue — Tue, 03 Dec 2013 18:10:01 +0000

the options right now are box, sphere, and concave polygons, how much work would be involved to support ellipsoid shapes?

i tried looking at the source code of the underlying physics, and was more puzzled after since the physics seem to be for 2d (box2d?) but orx does 3d space, just 2d rendering (from what i understood before).

so, my questions would be:

1- where would an ellipsoid shape be implemented, and any silly-wild-guesses on how hard that would be?

2- what am i not understanding, since orx has 3d objects, but i didn't see 3d physics

ios 7 rendering

tdomhan — Fri, 27 Sep 2013 10:24:54 +0000

Hi there,
I just noticed that on ios 7 there's a black bar on top of the screen, moving all elements down a little. This means that some things are pushed off the screen at the bottom. Also for some reason there's a small green bar inside this bar. Well I attached the screenshot, maybe then it makes more sense.

2D skinning

lydesik — Fri, 18 Mar 2011 10:22:58 +0000

how hard would it be to add a 2D skinning feature to orx?
there is a good sample here

Tutorial 12_Lighting crash on linux

nowolfer — Sun, 07 Jul 2013 00:21:52 +0000

[08:16:49] [LOG]
- 'MOUSE_LEFT' will create a new light under the cursor
- 'MOUSE_RIGHT' will clear all the lights from the scene
- 'MOUSE_WHEEL_UP' will increase the radius of the current light
- 'MOUSE_WHEEL_DOWN' will decrease the radius of the current light
- 'KEY_SPACE' will toggle alpha on light (ie. make holes in lit objects)
[08:16:49] [ASSERT] [orxDisplay.c:orxDisplay_GLFW_CompileShader():752] [ASSERT] :
Trace/breakpoint trap

Is it a bug, or i do something wrong?
I build the code using embedded dynamic, then compile link the source with debug config.
I using Gentoo with Nvidia card.

some bugs maybe！

viavia — Fri, 14 Jun 2013 16:48:52 +0000

hi：

I found some bugs, i'll list them one by one at each floor.

Multiple Render Targets

subr3v — Sun, 31 Mar 2013 16:00:11 +0000

Hi,

I'm currently developing a game with Orx Engine and I have to say I'm very impressed by the overall quality of the code and design of the engine!

However, since I'm implementing something similar to 3D deferred rendering in 2D I was wondering if there's any support for multiple render targets within the engine;
I looked for it in the source but I haven't found anything at the moment.

Long story short: Is there any kind of support for MRT at the moment?
My idea was to add something like orxDisplay_SetDestinationBitmap with an array of bitmaps instead of just one but I wanted to know if this feature is already planned.

(As for my project, I'll probably post about it in a week in the right section because I think this engine deserves more attention )

Random Crashes on Startup

kulak — Mon, 13 May 2013 23:27:53 +0000

I've been having random crashes on startup lately. I think it might be issue with orx. Not sure what to report in this case. Here is what I've got in XCode:

error: address doesn't contain a section that points to a section in a object file

#0	0x0000000000000000 in 0x00000000 ()
#1	0x00007fff85aa53f4 in __CFURLEqual ()
#2	0x00007fff85a88bd8 in CFEqual ()
#3	0x00007fff85a88929 in __CFDictionaryStandardEquateKeys ()
#4	0x00007fff85a7b405 in CFBasicHashFindBucket ()
#5	0x00007fff85a7ac06 in CFDictionaryGetValue ()
#6	0x00007fff85ab65e1 in _CFBundleCopyBundleForURL ()
#7	0x00007fff85ab6020 in _CFBundleCreate ()
#8	0x00007fff84569af4 in islPopulateKLPropertiesFromBundle ()
#9	0x00007fff84568301 in islGetInputSourceProperty ()
#10	0x00007fff845675fb in TSMGetInputSourceProperty ()
#11	0x00007fff8456998b in TISGetInputSourceProperty ()
#12	0x00007fff845654bb in TISCreateInputSourceList ()
#13	0x00007fff84564e50 in SyncHandwritingHotKey ()
#14	0x00007fff84564389 in _FirstEventTime ()
#15	0x00007fff84563e0c in RunCurrentEventLoopInMode ()
#16	0x00007fff84563b94 in ReceiveNextEventCommon ()
#17	0x00007fff84563ae3 in BlockUntilNextEventMatchingListInMode ()
#18	0x00007fff88771563 in _DPSNextEvent ()
#19	0x00007fff88770e22 in -[NSApplication nextEventMatchingMask:untilDate:inMode:dequeue:] ()
#20	0x00000001001403e3 in _glfwPlatformPollEvents at /Users/rom/workspace/orx/extern/glfw-2.7/build/Xcode/../../lib/cocoa/cocoa_window.m:1058
#21	0x000000010009c158 in orxDisplay_GLFW_EventHandler(__orxEVENT_t const*) at /Users/sergei/SWDev/CrossPlatform/orx/code/build/mac/xcode/../../../include/../plugins/Display/GLFW/orxDisplay.c:1116
#22	0x000000010003b950 in orxEvent_Send at /Users/sergei/SWDev/CrossPlatform/orx/code/build/mac/xcode/../../../src/core/orxEvent.c:379
#23	0x000000010003ba4b in orxEvent_SendShort at /Users/sergei/SWDev/CrossPlatform/orx/code/build/mac/xcode/../../../src/core/orxEvent.c:415
#24	0x00000001000b1bdb in orxRender_Home_RenderAll(__orxCLOCK_INFO_t const*, void*) at /Users/sergei/SWDev/CrossPlatform/orx/code/build/mac/xcode/../../../include/../plugins/Render/Home/orxRender.c:2078
#25	0x00000001000254ea in orxClock_Update at /Users/sergei/SWDev/CrossPlatform/orx/code/build/mac/xcode/../../../src/core/orxClock.c:555
#26	0x0000000100002317 in orx_Execute at /Users/sergei/SWDev/CrossPlatform/CastleDefense/../orx/code/include/orx.h:346
#27	0x00000001000021d2 in main at /Users/sergei/SWDev/CrossPlatform/CastleDefense/Source/main.c:293

INI:


[Display]
ScreenWidth   = 800
ScreenHeight  = 640
Title         = Castle Defense

[Input]
SetList = MainInput

[MainInput]
;KEY_ESCAPE = Quit
MOUSE_LEFT = MouseClick

[Viewport]
Camera            = Camera
BackgroundColor   = (210, 180, 140)

[Camera]
; We use the same size for the camera than our display on screen so as to obtain a 1:1 ratio
FrustumWidth  = @Display.ScreenWidth
FrustumHeight = @Display.ScreenHeight
FrustumFar    = 1.0
FrustumNear  = 0.0
Position      = (0.0, 0.0, -1.0)

[Resource]
Texture = resources # resources/explosion

[Scene]
ChildList = Terrain # 
			Tank # 
			Enemies

[Terrain]
Graphic    = @
Position   = (-400., -320., -0.)
Repeat = (10., 10., 0.)
Scale = (10., 10., 0.); @.Repeat
Texture = grass.png

[Tank]
Graphic = @
Position = (-100., -300., -0.1)
Texture = tank.png
Pivot = center
GSide = player
GType = tank

[Enemies]
ChildList = Turret1 # 
			Turret2

[Turret1]
Graphic = TurretTexture
Position = (200, 300, -0.1)
Rotation = 180.0
Body = Box

GType = turret
GSide = enemy

[Turret2]
Graphic = TurretTexture
Position = (200, -300, -0.1)
Rotation = 180.0
Body = Box

GType = turret
GSide = enemy

[Bullet]
Graphic = BulletTexture
Position = (0, 0, -0.1)
Scale = (3, 3, -0.1)
Body = BulletBox

; ******* textures

[TurretTexture]
Texture = turret.png
Pivot = center

[BulletTexture]
Texture = pixel

; ******* clocks

[Clock]
Frequency = 100

[ShootClock]
Frequency = 0.5

; ****** effects

[FlashFX]
SlotList = Flash

[Flash]
; This FX slot plays on alpha with a triangle curve
Type        = color
StartTime   = 0.0
EndTime     = 0.4
Period      = 0.2
Curve       = triangle
Absolute    = true
StartValue  = (255, 255, 255)
EndValue    = (255, 0, 0)

; ****** physics

[Box]
PartList = BoxPart

[BulletBox]
PartList = BulletBoxPart
Dynamic = true

[BoxPart]
Type        = box
Restitution = 0.0
Solid       = true
CheckMask   = 0xFFFF
SelfFlags   = 0x0001
Density     = 1.0
Friction    = 1.0

[BulletBoxPart@BoxPart]
HighSpeed = true

/** Inits the tutorial
 */
orxSTATUS orxFASTCALL Init()
{
    orxViewport_CreateFromConfig("Viewport");
    _scene = orxObject_CreateFromConfig("Scene");
    traverseScene(_scene, InitObject);
    
    orxCLOCK *clock = orxClock_CreateFromConfig("Clock");
    orxClock_Register(clock, Update, orxNULL, orxMODULE_ID_MAIN, orxCLOCK_PRIORITY_NORMAL);
    
    orxCLOCK *shootClock = orxClock_CreateFromConfig("ShootClock");
    orxClock_Register(shootClock, Shoot, orxNULL, orxMODULE_ID_MAIN, orxCLOCK_PRIORITY_NORMAL);
    
    orxSTATUS status = orxEvent_AddHandler(orxEVENT_TYPE_PHYSICS, OnCollision);

    return status;
}

hrtf and 3d audio

lilmike — Sat, 04 May 2013 00:16:51 +0000

Hi,
I am a developer of mainly audio games, games driven almost entirely by sound. Obviously this would include things like pan and volume, but one thing I was looking for an engine with was full headphone 3d audio hrtf support. Basically so you can wear headphones (and sometime use speakers) but it would sound like sounds are coming from all around. Does orx provide this? Is it possible that if not, it can be developed?
Thanks,
-Michael.

maybe another

viavia — Tue, 30 Apr 2013 12:08:40 +0000

hi，again:

in "orx-1.4-srccodesrcioorxInput.c"

and in "orxInput_Update(const orxCLOCK_INFO *_pstClockInfo, void *_pContext)"

there is codes:

......（at the end）
else
{
orxFLAG_SET(pstEntry->u32Status, orxINPUT_KU32_ENTRY_FLAG_NONE, orxINPUT_KU32_ENTRY_FLAG_NEW_STATUS);
}

here ,u want to update if not active when no "new status" ,but why not set off the flag "orxINPUT_KU32_ENTRY_FLAG_ACTIVE"???

this means that when not active, and when not "new status" ,an internal call would not change the status of the input.Is that what u want?

it would not get any problem now,because it's not a public funciton, but may get problems in late release.

i think it's a problem ,do u?

viavia — Sun, 28 Apr 2013 14:47:07 +0000

hello guys:

in "orx-1.4-srccodesrcioorxInput.c"

and in "orxSTATUS orxFASTCALL orxInput_Save(const orxSTRING _zFileName)"

there is codes:

......
orxConfig_ClearSection(orxINPUT_KZ_CONFIG_SECTION);

if((orxConfig_HasSection(orxINPUT_KZ_CONFIG_SECTION) != orxFALSE)
&& (orxConfig_PushSection(orxINPUT_KZ_CONFIG_SECTION) != orxSTATUS_FAILURE))
{...}

i know u want to rebuild a "input" section ,but after "ClearSection" ,don't you need call "SelectSection" to new one?

and u call "HasSection" just behind "ClearSection" ,will only get false as return.

ofcourse ,maybe i am wrong, please check.

a little bug ,maybe not

viavia — Mon, 22 Apr 2013 05:34:01 +0000

in orx-1.4-srccodesrccoreorxCommand.c ,

and in " orxCOMMAND_VAR *orxFASTCALL orxCommand_Evaluate(const orxSTRING _zCommandLine, orxCOMMAND_VAR *_pstResult) " ,

there is "if((_zCommandLine != orxNULL) & (_zCommandLine != orxSTRING_EMPTY))"

did forget a "&"? ofcourse ,& would not cause problem ,but usaually we use && in a "if()" .or this would get any good?

orxDisplay_DrawMesh Android texture coordinate bug

enobayram — Mon, 15 Apr 2013 11:14:10 +0000

Hi Everyone,

First of all, it feels nice to post on the Orx forums after a long break. I hope you're all doing well.

Diving right into the issue: I've recently started using the orxDisplay_DrawMesh function, and it has been working perfectly in Linux. Now, I've tried porting a project using the function to Android, and I observe some issues related to the texture coordinates.

What I do is: while handling the rendering of an object with a texture, I try to draw a triangle to the upper left corner of the screen, and the triangle spans the texture coordinates from 0.0 to 1.0 in both the U and V axes. Just to be sure that the object does indeed have a valid texture, I also return orxSTATUS_SUCCESS from the handler, in order to see the object itself too.

With this very simple setup, I observe the expected result under Linux, where I have the textured object in the middle of the screen, and also the correctly textured triangle at the upper left corner of the screen. When I try the same in Android, the result is the same except that the triangle is uniformly colored. I've observed that the color that the triangle assumes is exactly the pixel at the lower right corner of the texture. I would like to stress that I do observe the correct texture on the object.

Here's what I see in Linux (left) vs. in Android (right)

Here's the rendering event handler:

orxBOOL OnRender(orxRENDER_EVENT_OBJECT_PAYLOAD *_pstPayload){
	const orxBITMAP * _pstBitmap = orxTexture_GetBitmap(orxTEXTURE(orxGraphic_GetData(orxOBJECT_GET_STRUCTURE(GetOrxObject(), GRAPHIC))));

	vector astVertexList;

	astVertexList.push_back({0.0,0.0,0.0,0.0,orx2RGBA(0xFF, 0xFF, 0xFF, 0xFF)});
	astVertexList.push_back({100.0,0.0,1,0.0,orx2RGBA(0xFF, 0xFF, 0xFF, 0xFF)});
	astVertexList.push_back({100.0,100.0,1,1,orx2RGBA(0xFF, 0xFF, 0xFF, 0xFF)});

	orxDisplay_DrawMesh(_pstBitmap, orxDISPLAY_SMOOTHING_ON, orxDISPLAY_BLEND_MODE_ALPHA, astVertexList.size(), astVertexList.data());
	return orxSTATUS_SUCCESS;
}

Note that I'm using the latest Orx version from the Mercurial repository.

I tried to dive into the source code and catch some place where the texture coordinates are set to constant values but I got lost in the code.

Thanks in advance!

Crash with Visual Studio

N_K — Tue, 19 Feb 2013 19:57:26 +0000

Hello,

I've built Orx 1.4 (and the tutorials) from sources using Visual Studio 2010 Express. The compilation went fine, everything built fast and there was not even a single warning.

But for some reason, the engine doesn't work. The plugin sample that comes with the source package (Bounce) starts up, but after the console window comes up, it crashes.

The tutorials are starting up, they show the console window with the instructions, but apart from this, they do nothing, there's no render window.

These things happen regardless of the build type (I've built and tried everything, static/dynamic, debug/release, and profiler).

My machine has an AMD Phenom II CPU, an ATI Radeon HD4200 GPU, and runs Windows 7 64-bit. The prebuilt tutorials from SourceForge work fine. What should I do now?

Feature [Achievements]

brielga — Fri, 30 Nov 2012 15:19:31 +0000

Is there any implementation in relation to achievements?
Something like this would be feasible or useful?
If yes, how long would it take to create?

something like
this

or

this

List sizes break > 2518 items

sausage — Fri, 23 Nov 2012 04:53:53 +0000

This is a follow on from https://forum.orx-project.org/discussion/4937#Comment_4937 where the list limit was raised from 255 to 65535.

Seems to the limit actually breaks at 2518 items in the list.

[Lists]
MyBigList = A #B #C #D #E #F #G #H #I #J #K #L #M #N #O #P #Q #R #S #T #
	A #B #C #D #E #F #G #H #I #J #K #L #M #N #O #P #Q #R #S #T #
	A #B #C #D #E #F #G #H #I #J #K #L #M #N #O #P #Q #R #S #T #
	A #B #C #D #E #F #G #H #I #J #K #L #M #N #O #P #Q #R #S #T #
	A #B #C #D #E #F #G #H #I #J #K #L #M #N #O #P #Q #R #S #T #
	A #B #C #D #E #F #G #H #I #J #K #L #M #N #O #P #Q #R #S #T #
	A #B #C #D #E #F #G #H #I #J #K #L #M #N #O #P #Q #R #S #T #
	A #B #C #D #E #F #G #H #I #J #K #L #M #N #O #P #Q #R #S #T #
	A #B #C #D #E #F #G #H #I #J #K #L #M #N #O #P #Q #R #S #T #
... etc to 2518 items and no more...

orxConfig_PushSection("Lists");
	
orxS32 count = orxConfig_GetListCounter("MyBigList");
	
orxLOG("Total in the list: %d", count);

If you add one more item to make it 2519 items in the list, the count will result in 0 and some of the list will spit out into the log.

I discovered this issue as I started increasing the size of my level map.

output:

[15:51:56] [CONFIG] [orxConfig.c:orxConfig_Load():2771] [game.ini]: Begins th
e processing of included file @../game.ini@.
[15:51:56] [CONFIG] [orxConfig.c:orxConfig_Load():2944] Key <#T #> has no value,
 assign character '=' not found.
[15:51:56] [CONFIG] [orxConfig.c:orxConfig_Load():2944] Key  has no value, assign character '=' not found
.
[15:51:56] [CONFIG] [orxConfig.c:orxConfig_Load():2777] [game.ini]: Ends the
processing of included file @../game.ini@.
[15:51:56] [LOG] Total in the list: 0

[BUG] android-light-demo does not show the "light"

jagd — Wed, 21 Nov 2012 12:10:13 +0000

hallo everyone!

i've tried several opensource engines and find orx really good from it's design and idea. futhermore, it is easy to setup the develop environment.

After playing through some tutorials on the PC, i've installed the both demos for my android.

the "Balls and Walls" demo works.
but for my device (LG Optimus 7) that "light" demo does not show the "light". what i can see, are only the asserts under a gray mask.

built with GLES2, there's only a (logcat-) message from the profiler, which seems unrelated with this problem.

orxProfiler.c:455) Can't pop marker: marker stack is empty.

could someone explain, how to fix it?

TimeTrack problem with uppercase section name

acksys — Tue, 20 Nov 2012 06:10:22 +0000

I was doing some timetrack stuff like this

[O-Planet1@O-MenuButton]
TrackList = T-O-Planet2Tap 

[T-O-Planet2Tap]
0 = Config.SetValue Settings currentPlanet Mars ; works
0 = Config.SetValue Settings CurrentPlanet Mars ; doesn't work

When the section name was capitalized, orxConfig_GetString retrieved a null string instead of "Mars". Otherwise, it worked.

Additionally, orxConfig_GetString retrieved "mars" instead of "Mars" as the value. Not a major problem. I just preferred the upper case value.

commit 3a97578

iPhone assert

acksys — Thu, 08 Nov 2012 05:27:51 +0000

I'm getting this assert when I run to iPod Touch: http://imgur.com/zvzTc

Running in the iPhone Simulator is okay..although the simulator is iOS 6.0 and the iPod is 5.0.1.

hg commit 35fd466

Is it a known issue?

Landscape orientation doesn't work on iOS 5.0 SOLV

acksys — Thu, 08 Nov 2012 07:36:09 +0000

In the latest orx, it's not necessary to do "camera parenting" to achieve landscape orientation anymore. I noticed today this does not work properly on iOS 5.0, though.

My game's orientation appeared correct in iOS Simulator 6.0 when landscape mode is set in the .plist file. When I ran it on the device and iOS Simulator 5.0, the orientation was incorrect.

Is this something that can be fixed for iOS 5.0? If not, I might have to put the camera parenting system back in for my game.

ScrollObject::GetName acting weird

sonicbhoc — Sun, 21 Oct 2012 20:45:48 +0000

Hey Iarwain, I'm trying to get the name of a scroll object - in this case, my menu object. If I use
[code]
std::string(orxObject_GetName(this->GetOrxObject))
[/code
everything works fine. Butif I try to use this->GetObject, the name instead of being "Menu" comes out RT0006. Why is that happening?

Feature Request: Class identifiers for orxObjects

enobayram — Wed, 18 Jul 2012 18:06:52 +0000

Hi,

I've started using scroll, and I find it very useful. One extra feature that I'd really make use of, is somehow enabling object class hierarchies in config files. For instance:

[ObjectBase]
; various object fields...
ClassIdentifier = MyObjectType

[DerivedObject1@ObjectBase]
; various other object fields...

[DerivedObject2@ObjectBase]
; even more different object fields...

With this config file, if I now register the identifier "MyObjectType" to one of my ScrollObject classes, all derived objects will also be created as one.

Motivation: The reason why I'm asking for this, is that I can then start introducing custom fields for my base classes, that get handled correctly by the corresponding C++ type.

Implementation: I think ScrollObjectBinder, the binder that's used when the object type doesn't match any registered class could be made to create an object as usual, and check for the existence of the ClassIdentifier field. If it exists and the value matches a registered class, it could use the

template
ScrollObject *ScrollObjectBinder::CreateObject(orxOBJECT *_pstOrxObject, const orxSTRING _zInstanceName, ScrollObject::Flag _xFlags)

wrapper method of the corresponding type.

Thanks!

Sound (but not music) crashes on Android [SOLVED]

sausage — Fri, 05 Oct 2012 12:32:53 +0000

I have the following two sounds in my config:

[Music]
Music       = music.ogg
Loop        = true
KeepInCache = true
Volume      = 0.7

[ShootEffect]
Sound       = shoot.ogg
Loop        = false
KeepInCache = true
Volume      = 0.7

The following code plays music fine on an Android device:

orxSOUND *pstMusic = orxSound_CreateFromConfig("Music");
orxSound_Play(pstMusic);

The following code crashes an Android device:

orxSOUND *pstShoot = orxSound_CreateFromConfig("ShootEffect");
orxSound_Play(pstShoot);

The error according to catlog is:

[SOUND] [orxSound.c:orxSound_createFromConfig():629] Can't create sound : invalid content.
[ASSERT] [orxSound.c:orxSound)Play():1043] [ASSERT] : <(_pstSound != orxNULL) && (((((orxStructure *)(_pstSound))->u64GUID & orxSTRUCTURE_GUID

As an experiment in my config, I changed my ShootEffect from Sound to Music:

[ShootEffect]
Music       = shoot.ogg
Loop        = false
KeepInCache = true
Volume      = 0.7

And this plays in the game. Only once of course because it is music, but it proves the point that the sound file is fine and that it is locatable by orx.

My SVN version is a month old. hg getting the latest from bitbucket but it will take some time to get it all down. In the meantime, is this a known issue or have I made a mistake?

TimeLine and Commands

iarwain — Tue, 17 Apr 2012 23:45:11 +0000

Hi all,

I've been stalling again with the skeletal animation support, I blame Legend of Grimrock.

Anyway, that's only part of the truth as I'm still thinking on the best way to declare animations, especially when they are hybrid animations (bone + sprite) across children, and try to duplicate as less as possible info (especially the anim graphs).

I haven't found a great solution to make it easy to not only author said hybrid anims but also to trigger them in code. I want it to be as painless as possible...

So in the meantime I've been thinking about the other features I'd like to implement before 1.4 gets released: TimeLines.

I think I'm getting a clear picture of what I want to make for those. There'll be two parts:

a Command module
a TimeLine module

The Command module:

It would take command registration as plain text. Any orx (or user) module can register commands and will only get an array of strings as parameters (argc/argv style).
For orx I'll add a basic set of commands at first with some Object Oriented naming such as Object.Create, Object.AddFX, Object.SetSpeed, etc...

Those commands can then be invoked via code (not sure of the interest except maybe for some scripting support in the future?) but, more important, will be the main listener to events sent by the TimeLine module.

The TimeLine module:

It will be very simple: an array of tracks with the option of adding/removing tracks on the fly, looping and using the owner's clock for timing.
Tracks are simple lists of strings associated to timestamp.

When TimeLine are evaluated, they'll send events with the string in payload. That's about it. So anyone can hook any string to a time line and listen to them for custom sequencing. The interesting part will be that the Command module will also be listening to those and recognize strings that are formatted in a certain way. It'll then execute the command if recognized.

That's about it. That sounds a bit like a lite scripting version, which it is to some extent. That should mostly help with complex creation patterns that spawners can't solve or with menu creation, scene creations, etc.

I'm not sure right now if command should only accept strings and do their own parsing later or if I should centralize that in the command module, making it more complex to write as it'll need to register types for parameter checking, etc.

Any suggestions on all that?

I'd like also to be able to specify a target for commands other than the current object or uniquely named structures but in the worst case scenario I think there's an easy config trick that could rely on the new GUID system to achieve this at runtime.

This actually sounds much more exciting to me than skeletal animation to me as it'll give huge possibilities to orx with even less code to write while still remaining performance-focused.

New animation module

iarwain — Fri, 06 Apr 2012 23:57:26 +0000

Hey all,

Well, I've recently (as of tomorrow ) resumed working on the skeletal animation and I will actually change how the whole module is working by adding channels to animations.
Nothing groundbreaking there but it was definitely a concept missing to orx's animation system.

As I'm doing this, I might as well try to make the definition of animations and animation graphs easier in config. That will break backward compatibility but I think it's really worth it as defining those is really extremely annoying (and verbose) right now.

I'm posting here to get suggestions from all of you. Any idea?

As a reminder, here's what we currently have:

[AnimationSetTemplate]
AnimationList = AnimationTemplate1#AnimationTemplate2#...
LinkList      = LinkTemplate1#LinkTemplate2#...
Frequency     = 

[AnimationTemplate]
DefaultKeyDuration = 
KeyData         = GraphicTemplate
KeyDuration     = 
KeyEventName    = 
KeyEventValue   = 

[LinkTemplate]
Source      = SourceAnimationTemplate
Destination = DestinationAnimationTemplate
Priority    = 
Property    = immediate | cleartarget

Now what do you think of something more compact, like:

[AnimationSetTemplate]
AnimationList = AnimationTemplate1#AnimationTemplate2#...
Frequency     = 

Defining the links directly:

Anim1-Anim2 = immediate|cleartaget|low|medium|high(priority)
...

[AnimationTemplate]
ChannelList = EventChannel # SkeletonChannel # GraphicChannel

[EventChannel]
Type = Event
0.5 = StepLeft (# Value)
1.0 = StepRight (# Value)

[GraphicChannel]
Type = graphic
0.1 = AnimFrame1
0.3 = AnimFrame2
...

[SkeletonChannel]
Type = skeleton
0.2 = Child1Frame1 # ... # ChildNFrame1

[ChildFrame]; This is the transformation node for 1 bone at a given key
Position = 
Scale = 
Rotation =

That's basically it. No more separate link sections, everything will be within the AnimSet section.
And using timestamps as keys for all the channel data.

In the same time I'd add similar syntax for the TimeLine feature which are basically scripts/sequences attached to an object that can triggers events over time. Some of those events will be understood by orx like CreateObject, Transform, AddTimeLine, AddSound, AddFX, etc...

But more details in another post.

Any suggestion is welcome. Sooner than later as my plans are to begin implementing the new anim system this week end.

Change Log File Name

Peso — Fri, 21 Sep 2012 23:29:21 +0000

Hello iarwain,
I think there is a bug in orx when we try to change the log file name.

Function _orxDebug_SetLogFile

/* Had a previous external name? */
  if((sstDebug.zLogFile != orxNULL) && (sstDebug.zLogFile != (orxSTRING)orxDEBUG_KZ_DEFAULT_LOG_FILE))
  {
    /* Closes it */
    fclose(sstDebug.pstLogFile);

    /* Deletes it */
    free(sstDebug.zLogFile);
  }

If we have a previous file name, it closes the file and releases the name string. But if we have not used the log until this point, the log file would not have been open. I think it is the same with _orxDebug_SetDebugFile.

If we call orxLOG before changing the file name, it works.

But then, when we call orxLOG again, now with the new file name, we have another problem.
Function _orxDebug_Log line 455

/* Needs to open the file? */
        if(sstDebug.pstLogFile == orxNULL)
        {
          /* Opens it */
          sstDebug.pstLogFile = fopen(sstDebug.zLogFile, "a+");
        }

The previous file was closed but sstDebug.pstLogFile is not NULL, so it does not open the new file.

I think this is all.

orxObject_IsEnabled -> IsEnabled() for Scroll

Mhyre — Sat, 15 Sep 2012 17:23:13 +0000

Just a quick request here. Scroll has a function Enable(), it would be useful to have IsEnabled() too!