|
Author |
Thread Statistics | Show CCP posts - 44 post(s) |
|

CCP Rave
C C P C C P Alliance
8

|
Posted - 2015.02.03 13:42:37 -
[1] - Quote
l0rd carlos wrote:TL;DR: around the lines 200 in main.py object_location_dict[ok] only has one key in the example, that key is "0", what does that stand for and what are "first" and "last" supposed to be? [22:44] max12345 I don't think i can do this without comments or some help with the structuring [22:44] max12345 the thing I'm stuck at right now is that it's looping through some location keys [22:45] max12345 they are sorted by coordinate then by number they by datakey data pairs [22:45] max12345 but I have no idea what the number stands for [22:46] max12345 and they somehow need it for "first" and "last" variables [22:46] max12345 but there is only 1 entry for the number [22:46] max12345 so it'd be the same for the example case [22:46] lord-carlos "Part of the problem is that xrange() is a Python 2.x built-in and I don't think is defined in 3.x. However, xrange() in 2.x has the same functionality as range() in 3.x. Looks like an oversight." [22:46] lord-carlos https://forums.eveonline.com/default.aspx?g=posts&m=5457468#post5457468 [22:46] max12345 yes that's fine, I already changed that, I'm stuck somewhere else. [22:47] max12345 this is almost making me resub [22:47] max12345 :D [22:48] max12345 he should hop on irc :D [22:48] niki who is that awesome CCP Darwin guy [22:49] max12345 idk but he's pretty cool releasing these tools at all imo
[22:53] max12345 around the lines 200 in main.py object_location_dict[ok] only has one key in the example, that key is "0", what does that stand for and what are "first" and "last" supposed to be? [22:55] max12345 I'm sure these are issues he can solve much faster though
:)))))
The keys represent the order of location keys, starting with 0 for the first key.
If there were five keys, they would be 0, 1, 2, 3 and 4 (in order).
Each key actually maps to a dictionary that represents a Bezier (or possibly Hermite) curve. It has the following keys: "start_time": The time from the beginning of the scene to the start of the animation curve. "end_time": The time from the beginning of the scene to the end of the animation curve. "start_value": The x,y,z location of the model at the start of the animation "end_value": The x,y,z location of the model at the end of the animation "control_1": First control point. "control_2": First control point.
first and last should be the first and last of those dictionaries. |
|
|

CCP Rave
C C P C C P Alliance
8

|
Posted - 2015.02.03 13:49:38 -
[2] - Quote
CCP Darwin wrote:Sarmatiko wrote:BTW is the frame capture command works in this build? I've tried "/capturefolder=c:\movie\" but it looks like this command was ignored by Probe. I'm having some trouble getting this to work myself. l0rd carlos wrote:File "/usr/local/lib/python3.4/dist-packages/dae_to_red-0.0.0-py3.4.egg/dae_to_red/main.py", line 289, in _read_axis_keys_from_nodes NameError: name 'xrange' is not defined
Someone in the #eve-dev IRC channel has the same trouble and tries to fix it. Part of the problem is that xrange() is a Python 2.x built-in and I don't think is defined in 3.x. However, xrange() in 2.x has the same functionality as range() in 3.x. I'm not sure what the story is here. Also: I notice that if you use demo.yaml as your scene name, and you by mistake name your yaml file something else (like in my case demo.yaml.txt), it appears to find a demo.yaml that's in the shared cache. You apparently MUST use the extension ".yaml" and not ".yml" or anything else, and then it'll find the file correctly. The shared cache demo.yaml shows the Caldari battleships with default Caldari faction and the example from the docs uses the Guristas faction markings.
Correct. The project was written in Python 2.7. If someone were to upgrade it to be 3.x compatible and send a pull request on GitHub, that would make me happy.
And yes, all yaml files must have the .yaml extension. |
|
|

CCP Rave
C C P C C P Alliance
8

|
Posted - 2015.02.03 18:44:46 -
[3] - Quote
Ragnar Sorlandschip wrote:Traceback (most recent call last): File "setup.py", line 9, in < module > reqs = [str(ir.req) for ir in install_reqs] AttributeError: 'str' object has no attribute 'req' This is what I get when trying to install dae-to-red. Any ideas?
I don't know. Which versions of Python and PIP are you using ? I'm using Python 2.7.9 and PIP version 1.5.6 and it just works. As far as I know parse_reqirements should always return a list. |
|
|

CCP Rave
C C P C C P Alliance
8

|
Posted - 2015.02.05 11:35:10 -
[4] - Quote
Rollo Brinalle wrote:Joshua Foiritain wrote:Two questions, when using _handle_empty_scene to render a scene without a nebula is it possible to change the black BG color to something else? (like green) I noticed the _handle_trinity_setting option but have no idea what kind of options we can change  Also the capture option does not seem to work for me, this is the command line im running: C:\ProgramData\CCP\EVE\SharedCache\probe>bin\exefile.exe /lib=code.ccp /windowed /width=512 /height=512 /scenes=test.yaml /capturefolder=D:\Movies I've tried various folders, existing and none existent ones, different drives, trailing slash, no trailing slash, etc. Nothing jumps out in the error log. Any idea what im doing wrong? The capture did not work for me either also tried various folders etc... Also has anyone figured out how to place a structure? I tried a few stations with no luck. I have it narrowed down to the DNA and tried to use the graphic_type.yaml that has been floating around on the thread with no luck.
There seems to have been an error in the documentation. Fixing it now. The /capturefolder argument should have an underscore in it like so /capture_folder |
|
|

CCP Rave
C C P C C P Alliance
8

|
Posted - 2015.02.09 10:00:34 -
[5] - Quote
Valeo Galaem wrote:CCP, can you say what is happening here? http://i.imgur.com/nkidKCK.png
The crash is happening when the last line of the last keyframe of the first keyframe sequence is parsed. Curve file 'example.red': http://www.pasteall.org/56584
This is the .red file that results when I run the original 'example.blend' file included in the dae-to-red documentation through a Blender script I had to make to get custom animations to not crash dae-to-red (it doesn't like keyframes that don't key on all curve channels).
Looks like every "keys:" element is missing 4 spaces of indentation. |
|
|

CCP Rave
C C P C C P Alliance
8

|
Posted - 2015.02.10 10:43:36 -
[6] - Quote
Fredlah wrote: Yeah they are definitely set to Euler in maya :( I think its an issue with the DAE_FBX export, will poke around and find a different plugin
Try OpenCOLLADA. That's the one I use. |
|
|

CCP Rave
C C P C C P Alliance
8

|
Posted - 2015.02.10 12:50:29 -
[7] - Quote
Fredlah wrote:No, it still looks like I'm having issues... To test, I opened the example maya file, exported with opencollada using the same flags as the example.dae, seems to work fine. I then try and convert the dae file using the pre-compiled script, it does not error in the script, but the resulting .red file simply contains "type: EveSpaceScene", and that's all. here is my example.dae file before attempting to convert it to a .red https://www.dropbox.com/s/6tyvjl2mmscbywk/example.dae?dl=0
I thought it was working earlier as I was not getting an error on conversion, but in reality I was getting empty converted files :(
Looks like the exported data is missing the library_animations tag along with all the keyframe animation data.
This may be a silly question but does the scene you are exporting have keyframes yet? How are you exporting the scenes ? (I usually click File->Export All...) |
|
|

CCP Rave
C C P C C P Alliance
8

|
Posted - 2015.02.10 16:38:46 -
[8] - Quote
Fredlah wrote:Also just trying to play around exporting a camera move, getting the following error when using the dae to red converter: http://pastebin.com/raw.php?i=egvD3uNa
edit: I guess camera animation paths are only supported from the gr2 files anyway. I have a camera animation in maya already to scale, what would be the easiest way of getting that into a gr2 file for probe to load? Would it be possible to get probe to load the movement path of any object in the dae/red export as the camera path? eg I parent dummy geo to the camera, name it 'camera_path", bake it out to keys, export it from maya with my ship movements, so I have a two ship paths and a camera position/rotation path, and have probe use that? eg - [set_camera_animation, 'res:/curves/scene2_objects.red', camera_path]
Correct. Camera animation paths currently only support gr2 files, but adding some kind of support for camera animation is definitely on our TODO list.
In the meantime, I think I have now fixed the titan-shaking issue in a recent commit (https://github.com/ccpgames/dae-to-red/commit/caf9e629ff3a5c5baf296ebabf2dc77bd2d38cd9). (It now works on my machine)
When you're up to it you may want to get the latest version and see if things behave. Let me know if you run into any more issues. I want to see that awesome movie. |
|
|

CCP Rave
C C P C C P Alliance
8

|
Posted - 2015.02.10 16:45:54 -
[9] - Quote
Joshua Foiritain wrote:
I'm wondering if its possible to change the BG color from black to something else (like green) when rendering a scene without a nebula.
Not yet, but it shouldn't be too hard to implement and I can definitely see how that could be useful.
|
|
|

CCP Rave
C C P C C P Alliance
8

|
Posted - 2015.02.11 11:24:59 -
[10] - Quote
Valeo Galaem wrote:
I've pulled your latest revision and it actually makes the animation more jerky. I'm testing it using the example animation after it has been re-baked and re-sampled in Blender. Those poor drones were convulsing!
I don't think I ever did any baking or sampling on my end . If possible, I would like to get my hands on your Collada (.dae) files for debugging. |
|
|
|
|
|