Captain's Log 3.14159265359


#21

I’m away at the moment but will double check when I get back.

Double check complete: both files have existed on the server at the location specified since original posting, just before 6pm EST:

This is why I double checked. I couldn’t be sure I didn’t miss something that easy; but the proof is in the pudding. :slightly_smiling: Is complex physics hull via URL broken?

special_note the .fbx can be seen sitting upon a normal purple box entity, but the .obj fails to import into the PHULL url and tosses the errors seen in the code block above.


#22

Typos, have a close look:
http://metaversecafes.com/images/test_barrier.fbx
http://metaversecfes.com/images/test_barrier.obj <==

This one works:
http://metaversecafes.com/images/test_barrier.obj


#23

I missed that ‘a’ in cafe. Damn it. But yes, correct spelling and the file is in fact where it should be. Sorry. :frowning: I spent an hour searching around once. I was trying to figure out why my logical expression would continue to fail.

The expression was using flase and not false. :joy: Spelling errors aside my claim still remains. The PHULL Rendering does not catch the senario; making it difficult to spot the true problem.

cc: @chris

I will now correct the original link should anyone else click on it.


#24

Here is a nice description of non-manifold objects. Looking at the cuts suggested by the photos you posted, it looks to me as if the three sided cut object has no thickness. Physics engines do not like that because volume of the shape cannot be determined.

http://blender.stackexchange.com/questions/7910/what-is-non-manifold-geometry

Physics engines also require the physics shapes to be convex hulls. Something I wrote a long time ago about that:
http://inworldztech.com/techblog/2012/11/19/vehicles-and-physical-objects-in-physx/


#25

Progression of PHULL is going well. I look forward to taking on many Alphas (and some Betas) in the months to come.


#26

User Interface creation with QML is fast and relatively painless… when it works. Unfortunately, there is a strong disconnect between current JS API and what is usable in QML, and while it was mentioned to us in that thread, it is no less disheartening to lose a planned day of UI development. 1 step forward, 2 steps backward. :frowning:

https://alphas.highfidelity.io/t/ui-suggestions-that-no-one-has-time-for-right-now/10195/12


#27

After getting bummed out on the QML defects I decided to switch gears and work on Soft Attachments… fail. Three crashes later, I’m calling it a day here. Thanks for playing.


#28

QML development has been progressing. I do encourage and hope for an expedited JavaScript API for all QML as it will make for a VERY compelling experience within High Fidelity if executed correctly.

Steady as she goes I stay the course.


#29

PBR integration on the ship will be a time consuming process but will be well worth the end results. The custom shader files have held up nicely and I’m very excited about progress made with XMLHttpRequest inside QML. The ability to navigate undocumented “features” and api calls has tested my merit as a Captain. I struggle daily with what can and what cannot be accomplished, with the written promise that the company focus is all JS calls are going to be available in QML. I very much hope this is the case. Last night I reflected on the talks I viewed many years ago of Kevin Rose speaking with @philip and how I thought it was so cool I could share the same virtual ground with an icon. All I really know is that regardless of what happens in alpha, I’m destined to be around far long after…


#30

Look at https://github.com/highfidelity/hifi/pull/7149 - no idea if it gives some of what you’re looking for in ref to QML + JS. If it does you might be able to 1) have something you want and 2) validate it works correctly so it can get a vote to make it into mainline code.


#31

Continued work on the User Interface has been extremely uplifting the last few days. I’ve managed to dynamically load external QML when and where I need it. Leveraging focus and passing data snatched from the API. It’s empowering. Chat is taking longer than anticipated, but I want it to look good.


#32

Hi @AlphaVersionD , do you want to have a look at this https://github.com/highfidelity/hifi/pull/7247


#33

Thanks for the heads up on the PR @chris! I’ll hook up the GET / SEND .js in combo with the message mixer example scripts and should be essentially done on chat.

I can try this out tomorrow; and it is effectively what I was going to be searching for in the available code base. I was really impressed with the internal hackathon and feel it has established a good bar to reach for.


#34

Functional. QML to JS and JS to QML. However, I do not wish to spoil the fun of showing it in action at the moment. I also commented in GIT.

Stoked.

“we hook up more chat interfaces before 8 am than most people do all day.”

UUID+hash_concat = instant InstantMessenger…

Big thanks to @Jherico for making it possible in source, and shout out to @chris for the alley-oop on PR via GIT.


#35


#36

That token there is the SessionUUID.

@ctrlaltdavid a while ago made his chat script work with the messaging system, but it wasnt integrated to the UI like yours. Might want to swap notes with him on it or check the script out to learn stuff from it :slight_smile:


#37

I figured it out.


A Comparison Between High Fidelity and Sansar
#38

Heh coolio. and yes, I knew it (been an avid commander every now and then in a cutter)


#39

Pushing right along through chat interface checklist. Integration of FontAwesome.ttf within QML has proved vital to the success of emoji and symbols. With support for many UTF-8 codes rendering Vector graphics, I can move on to more amusing activities. Current symbol support… over 600.



+1 for the good guys… :wink:

cc: @Guinevere


#40

Captian ! the log is corrupted with strange characters !