Page 2 of 4

Re: MtI 3.6

Posted: Fri Feb 22, 2013 12:38 pm
by ior
I did not meant to put pressure on you.

I only reported things that seemed to me important to work with MtI.

Go with your pace and work with calm, I hope you do not want to give up, cheer up :wink: .

For now the most important thing that I can say It should be solved is the problem with the moving scale pivot that still exist in newly created scenes. It exist in the versions of MtI after mti_beta_2_test_8e that is the version that does not change scale pivots at export time even if they have weird dimensions.

See you soon.

Re: MtI 3.6

Posted: Fri Feb 22, 2013 2:35 pm
by CTZn
You don't have to feel guilty about anything ior. That would be totally unfair since bug reports are mandatory in order to enhance a software.

No finished render issued from a Maya scene has been exposed for years; as a consequence it seems a commonly accepted idea that MtI can not export scenes to Indigo.

MtI may be updated, so much about the implementation of new Indigo features. In all cases this MEL exporter is a dead end, as far as users expectations go.
Antichamber wrote:Raw persistence may be the only option other than giving up entirely.
Been there, since day one.

My special thanks go to gzavye, Lemo, and ior for his continued support. And Doug Hammond (dougal2) obviously, since he was first to put the exporter for Maya on good tracks.

Not to forget the user 'Big Fan' who played a great role in my determination ^^

Be not affraid of the void, for exporter writers may emerge (people deeply API-able). I never meant to prevent them to. It has been for about three years since I sent my resignation mail to Glare Tech, arguing that someone competent should prevail. For the sake of the company.

I was not able to raise myself at the rank of a programmer, however I want to thank Glare Technologies for the opportunity. I've been learning a lot on others, on myself, and on more trivial topics. I meant to become a wary consummer of images and technology, and these purpose were fulfilled.

On a personal note, I have other boundaries to repell rather than digging into human artefacts such as computers and their softwares. For instance, the MS compiler never worked correctly on my rig (it took a Windows update in order to have its browser windows to open, to start with. Dude.) and I wanted another initiation to c++ than fixing a compiler (or an OS or a PC).

Enough of the hard way !

To all: I'd rather give you that choice: my PayPal account is

I won't vanish any soon. See you :)

PS: my next (second) MEL project might be a level editor for Eufloria :mrgreen:

Re: MtI 3.6

Posted: Mon Feb 25, 2013 1:30 am
by ior
Hi CTZn.

I respect your options but I would say that your job with MtI was not a bad job.
CTZn wrote:In all cases this MEL exporter is a dead end, as far as users expectations go.
You say that the exporter did not work properly but it does the job very well except for this moving world scale pivot problem, that I found out now that appeared in the transition from mti_3.2.5.2 to mti_3.2.12.0.

As far as the groups on groups problem, it was a minor bug that was solved by not having groups with the same name.

As for constrains, I do not expect that every aspect of maya get to be supported spontaneously.

I continue to think that MtI is properly done and you contribution was as good as it could be and maybe a little more support from users and working with another person would had help you to tackle with the job.

I ask you despite the fact that you decided to move to another thing if you can still help me pinpoint the moving world scale pivot problem that appeared as I said in mti_3.2.12.0, that should help you to know what was implemented in that version that make this behavior to start. If that problem goes away MtI becomes a very good working exporter for indigo.

What do you think?

See you soon.

Re: MtI 3.6

Posted: Mon Feb 25, 2013 2:33 am
by CTZn
Good points ior, thanks a lot for the kind words.

Raw persistence led me to burn out twice, and I was just smart enough to prevent it to happen a third time. I will not be trying anything this hard against myself anymore.

One could say that I am another aspect of MtI's dead end ;)

But, since you pinpointed one issue precisely enough I will be looking into it the next weeks. TBH, I do need a cooldown and I will manage it authoritatively.

See you here.

Re: MtI 3.6

Posted: Tue Apr 02, 2013 8:42 am
by ior
I found out what is done to change the world scale pivot:

As I render the scene the values XYZ of world scale pivot become the XYZ values of the rotate pivot (translate), multiplied to the values XYZ of scale, plus the XYZ values of the previous world scale pivot.

When I render the scene again, as the word scale pivot changed and can be bigger, the world scale pivot is going to be multiplied again with the scale and so on until the world scale pivot is so big that makes objects disappear.

Re: MtI 3.6

Posted: Tue Apr 02, 2013 9:06 am
by CTZn
So you don't want to let it die and still want me to make it better huh !?

I'll upload a new version this week, concatenating the last changes and fixes I did before I gave up.

To be honest I have little understanding on transform matrices, that's the issue I am the less likely to address. Because you have helped enough ior will be looking into it however.

At some point.

Re: MtI 3.6

Posted: Wed Apr 03, 2013 1:36 am
by ior
Please do not let it die! :(

Re: MtI 3.6

Posted: Wed Apr 03, 2013 1:12 pm
by CTZn
Think of it as "opening the position for the right person".

I want Maya users to get full access to Indigo, and I can not be proven wrong on this one. I gave brain, nerves and this through time.

Nor Glare Tech or me are interested in leveraging a few bucks a year from this. Them all deserve more and better, including you ior.

Take it easy. Spread the word.

Update b4 the next week, I'm still on my lazy slope.

Re: MtI 3.6

Posted: Wed Apr 03, 2013 10:58 pm
by Zom-B
CTZn wrote:Think of it as "opening the position for the right person".
So are the sources of actual MtI open? If that would be the case this could encourage some people to "take a peak" into it... maybe...

Re: MtI 3.6

Posted: Thu Apr 04, 2013 12:51 am
by CTZn
Zom-B wrote:
CTZn wrote:Think of it as "opening the position for the right person".
So are the sources of actual MtI open? If that would be the case this could encourage some people to "take a peak" into it... maybe...
The zip is the source. MtI is 100% scripted, including "pynodes", to use blender's terminology. There's not one closed bit.

The license type is Creative Commons. Anyone behaving like (s)he owns the code is a disqualified scum deserving to face a flamethrower. Unless if found to be brainless.

The only code that remains within is from Doug Hammond (dougal2) for 80% of the render launch procedure.
The rest is from Adam Penalver (me, CTZn). That's over 99% of the product off the top of my head.
About 0.1% of the code was provided by Glare Technologies (quaternion generation, exponent <-> glossiness conversion).

Users want a profesional coder, wich implies a brand new product. They will not get fooled on this one.

Re: MtI 3.6

Posted: Fri Apr 05, 2013 11:43 pm
by CTZn
patch file for mti 3.6, see readme within

Re: MtI 3.6

Posted: Sat Apr 06, 2013 2:27 am
by CTZn
CTZn wrote:patch file for mti 3.6, see readme within
wait I just figured out how it works:

download the latest mti installer
extract on disk
copy the patch files over.

Judiciously. Right. It's a patch.


Re: MtI 3.6

Posted: Sat Apr 06, 2013 7:39 pm
by CTZn
Changes list:

+ Added to (n)particle instancer (use the Indigo Editor):
full random rotation
per particle rotation (rotationPP) (and radius iirc). Suitable for dynamic simulations.
+ fixed keyframed objects with no rotations
+ changed default splat filter to radial
+ updated and exposed network rendering functions
+ fixed Maya freezing when linking to an external material using comments in ISL
+ fixed redo previous render
+ fixed Editor UI for layeredTexture nodes.
+ fixed exponent conversion to roughness.
+ fixed coating thickness. User input is now effectively in micrometers instead of meters previously.
+ changed the pixel threshold to adapt to the final image sampling level. Previously the threshold was applied against the frame buffer size.
> is it a fix ? double-check with random tris

Re: MtI 3.6

Posted: Sat Apr 06, 2013 7:44 pm
by CTZn
Note on MtI:
* by:
* Adam Penalver (CTZn)
* Arne Oog (aka arneoog) (no concurrent contribution)
* Benedikt Kircher (no concurrent contribution)
* Doug Hammond (Dougal2)
* Guillaume Maurras (no concurrent contributio)
* Matt B. (MattTheMan or ThatDude33) (no concurrent contribution)
* Shawn "L" (no concurrent contribution)
As one can tell many are the people who can contribute. At this day though, no one can claim for a retribution for it no more.

If you want something to happen, make the move.

Re: MtI 3.6

Posted: Tue Jul 16, 2013 1:24 am
by CTZn
well huh, here's an update.

fixed geometry pivot bug
fixed material ID camera option
added depth camera option
added dithering option in Frame Buffer
enhanced Render Modes UI (+wip)
added random seed control in Threads
shadow pass dormant (shadow_caster XML unsupported)
shared shader data

stringArrayRemoveDuplicates(plus the previously mentionned);