MtI rev 63 08 nov 2007

Announcements, requests and support regarding the MAYA exporter
User avatar
dougal2
Developer
Posts: 2532
Joined: Wed Nov 15, 2006 8:17 am
Location: South London

Post by dougal2 » Sun Nov 11, 2007 5:36 am

REV 74

- functionalised albedo_texture in order to:
- implemented bump maps
- implemented exponent maps for phong
- implemented diffuse transmitter (anisotropic shader)
- fixed bug with background colour. made mti_EnvironLight work properly (should sunlight appear in this list rather than a separate option??)
- added a check for whether shader.incadescence exists before using it (in case we need to use shaders that don't have it)
- added a function mti_makeIESLight() which makes a small plane parented to a locator for IES light use.
- updated GUI
- changed version numbers in scriptnode - is there a way we can prevent the stupid float rounding error? - use int for version numbers

- code + message tidying here and there. tidied XML output somewhat.

yes, I think we should release this soon - especially now we have texture/bump/exponent/diffuse transmitters/ies in.

please test for a bit, release on monday/tuesday?
Attachments
im1194714906.png
IES light shining through a diffuse transmitter
im1194714906.png (452.17 KiB) Viewed 3329 times

bkircher
Posts: 115
Joined: Thu Nov 23, 2006 6:24 am

Post by bkircher » Sun Nov 11, 2007 5:52 am

I'll check.

I thought I had the rounding error was fixed using a "version < 0.701"
comparison.

I'll try to write a conviniencefuncition to substitute - o.ooo1 < x < 0.0001
with a clean zero sometimes to clean-up the camera output and the Instance matrices.

bkircher
Posts: 115
Joined: Thu Nov 23, 2006 6:24 am

Post by bkircher » Sun Nov 11, 2007 6:00 am

... oh, and monday is fine.

I think sun and backround can appear at the same time, so we should
leave it at the separation, although we could make the wording more clear.

@CTZn: please check the instancing code I mentioned earlier in mti_Main (or was it mti_Gen?).

Can we three takle the camera/world scale issue? That's a real concern, and it would be great if we could solve that quickly.
I'll try to post a summary this evening, maybe we find a solution.

User avatar
CTZn
Posts: 7240
Joined: Thu Nov 16, 2006 4:34 pm
Location: Paris, France

Post by CTZn » Sun Nov 11, 2007 8:52 am

Well bkircher, that is documented...

mti_makeInstances allows the user to create an arbitrary number of instances for one object in a row.

- Create simple Maya locators, place them etc
- Select the geometry to be instanced, then the locators.
- Execute mti_makeInstances and go.

You can select more than 1 geometry, all but first will be ignored. You can select whatever kind of object between the mesh and the locators, they will be ignored. You can update previously assigned mtiInstanceLocator to another mesh that way.

Indeed I need help to integrate it in the gui, I already put an icon in the apropriate folder. I already wrote that I think.

Physical sun in a list is fine !
obsolete asset

User avatar
dougal2
Developer
Posts: 2532
Joined: Wed Nov 15, 2006 8:17 am
Location: South London

Post by dougal2 » Sun Nov 11, 2007 9:56 am

I've started the new MtI documentation, it's a Word file in the repository.

Once all the _correct_ info has been moved into it from OldDocumentation.txt, we can remove the txt file and I'll create an HTML version of the doc for easy viewing.

bkircher
Posts: 115
Joined: Thu Nov 23, 2006 6:24 am

Post by bkircher » Sun Nov 11, 2007 10:59 am

Added mti_makeInstances to GUI with the Icon, changed the layout slightly to make some space for it.
Changed some minor things in mti_ScriptNode.

And finally I've understood mti_makeInstances. :D Nice one.
Will change the way I use locators.

Concerning the documentation:I haven't looked into it, but could a wiki-setup make sense? Judging from how well SVN works here, this could work magically. (I guess not for the first release, as we are close to it, but later?)

p.s.: Did you have a chance to look into the particle instancing thing?

User avatar
dougal2
Developer
Posts: 2532
Joined: Wed Nov 15, 2006 8:17 am
Location: South London

Post by dougal2 » Sun Nov 11, 2007 11:04 am

todo: i forgot transparency settings in the blinn/specular section of the documentation.

User avatar
dougal2
Developer
Posts: 2532
Joined: Wed Nov 15, 2006 8:17 am
Location: South London

Post by dougal2 » Sun Nov 11, 2007 11:34 am

I've just re-arranged the files in the SVN
- moved the Icons, and Copy* folders up a level
- renamed MayaToIndigoSimple -> MayaToIndigo
- removed the old exporter.

please make sure you get a full update before doing any more work.

User avatar
CTZn
Posts: 7240
Joined: Thu Nov 16, 2006 4:34 pm
Location: Paris, France

Post by CTZn » Sun Nov 11, 2007 12:43 pm

Yep, sure Dougal. Thanks a lot bkircher !
obsolete asset

User avatar
CTZn
Posts: 7240
Joined: Thu Nov 16, 2006 4:34 pm
Location: Paris, France

Post by CTZn » Sun Nov 11, 2007 12:43 pm

]Edit:[/b]

I can't read the doc, could it be in a more common format please ? Also I'm not sure that's cross-platform (MtI may run on Linux and Mac). *yelling as usual*

r84:

- Updated my baby so it now respect IES locators.
- Extended the logic of modularity and extracted tools procedures from mti_Main.mel to mti_ModuleTools.mel.

PS: would be cool if you could get mitInstLocators from this particule script, was that your idea bkircher ? What would be the point on having real meshes for Indigo anyway... At first glance I think we could make a much simpler script to get particles coordinates and create mtiInstLocs there. Your thoughts, as usual.
obsolete asset

bkircher
Posts: 115
Joined: Thu Nov 23, 2006 6:24 am

Post by bkircher » Sun Nov 11, 2007 9:19 pm

:shock: I was just commiting my revised version of mti_Main,
were I moved everything unrelated to mti_helperScripts..., and now this... :shock:

:P

I changed some other stuff slightly, will incorporate that into your last update. We could derive mtiInstanceLocators with this, but for the moment I'd prefer a simple step by step solution...

I'm adding a rudimentary help-section window soon, just with notes and mental checklists. No documentation, just notes where normals should point (I find ExitPortals' normal direction counterintuitiv) etc.

Changed settings to accept lower F-Stop values.

User avatar
dougal2
Developer
Posts: 2532
Joined: Wed Nov 15, 2006 8:17 am
Location: South London

Post by dougal2 » Sun Nov 11, 2007 10:42 pm

I've coverted the doc to RTF and comitted it.

bkircher
Posts: 115
Joined: Thu Nov 23, 2006 6:24 am

Post by bkircher » Sun Nov 11, 2007 11:10 pm

Great stuff Dougal,
easy to understand and to the point.

Added some infos on normals and phongE, feel free to change the wording, so that we stick to your style.

- - - - - - - - - -
Added Support for different Environment Light Options, though I haven't tested LatLong yet.

bkircher
Posts: 115
Joined: Thu Nov 23, 2006 6:24 am

Post by bkircher » Mon Nov 12, 2007 3:59 am

the camera setup is a mess, and since we've moved to a new location for the repository, I can't seem to go through the history. I thought there was a working solution this morning and now, nothing is working anymore.

Anyway, I can't find a solution there that works reasonably with a scaled world. Until we have this fixed though, I don't think we have a working solution.

I'll attach some screenshots of one test I did this morning, which looked promising, and some from inbetween, just to show you why I think this is fishy:
Attachments
cam02.jpg
cam02.jpg (52.59 KiB) Viewed 3244 times
cam01.jpg
cam01.jpg (26.81 KiB) Viewed 3244 times
Innenraumquest.jpg
Current Status with GSF 1 vs 0.1
Innenraumquest.jpg (335.69 KiB) Viewed 3244 times

User avatar
CTZn
Posts: 7240
Joined: Thu Nov 16, 2006 4:34 pm
Location: Paris, France

Post by CTZn » Mon Nov 12, 2007 5:03 am

Mmmh yes I'm not sure that scaling is a reliable approach, until we understand what is wrong with cameras. In your demonstration position only is changed (twice ;)), while all distance-based setting should be affected, that is, almost every camera setting. Maybe I didn't get your point right...

- - - -

Again, I give you there my feeling about some Tools, wich I believe belong more to modeling with Maya rather than being necessary to MtI. There is already a bunch of toolBoxes all around the place wich are just shortcuts for one single function accessible through standard menu or key mapping otherwise. That is not serious-looking IMO. Refering myself to the doc:

First Row:
5. Open Hypershade. This one is border line. I'm myself using Ctrl+5 for years as it is an important editor but ok.
9. Toggle undo. Redundant, already in Status Line.
10. Delete all history. Mmh fine for little test scenes, but sometimes you don't want just that because you are exporting converted nurbs with History and you may want to edit them later. So instead of adding a secong icon for "by object History deletion", I suggest we remove this.
Second Row:
2. Toggle face normals visibility.
3. Toggle backfaces visibility.
4. Reverse face normals.
5. Turn smooth shading off.
6. Turn smooth shading on

Nah. Exit Portals should be created with normals visible and that's it. Again that belong to the modeling stage of the scene.

Let's have a single line of icons !

That was my feeling expressed sincerely, please don't get offensed and feel free to argue with it :)

Oh and why is it ?
Copyright © 2007
Last edited by CTZn on Mon Nov 12, 2007 6:26 am, edited 1 time in total.
obsolete asset

Post Reply
94 posts

Who is online

Users browsing this forum: No registered users and 26 guests