tkyler

Tired of workin'..late night blog: You may recall the picture of the falco cockpit I have on the front page of x-scenery.com. Contrast it with the one seen here, and you'll note all sorts of goodies that will be "up and coming" in 2010. I have spent the better part of the year working on a foundation of techniques and systems to speed up the process of development. The MU-2 was such a exasperating experience to finish up, I knew I had to standardize methods and find ways to reuse the things I'd already done. I can't tell you how many times I've modeled the same entity for each new project. Well I'm quickly nearing a major milestone in my developer tools that I hope to finally capitalize on and finish up some projects. It's easy enough to keep moving on stuff you know and worry about the stuff you don't later...crossing those bridges when you get to them. I crossed too many of those bridges on the MU-2 to ever want to do that again. So the most prominent development has been a plug-in template with a deep tool-set (at least to me it is)for programming whatever I may need in short order. The MU-2 had over 100 manipulators...some tied to default datarefs and others tied to custom ones. Testing those required lots of exporting from blender...and every export from blender meant about an hour of hand editing the exported object file towards the end of the project. This new code template allows me to separate the manipulators from their functionality. I set up a complete cockpit with "empty" datarefs and animate the whole sha-bang. The plugin code has "holes" or "hooks" for the logic to be programmed later. I simply read in the position of all knobs, buttons, switches, whatever and then code in whatever I need to. This saves me the hassle of hand editing, exporting and testing repeatedly. It's much quicker to compile a new plugin and reload the aircraft than to reboot x-plane. So what does that have to do with this Falco anyhow? Well all those techniques are getting tested on this guinea pig...which will go out as freeware when done. What you see in this project is what will be in future products and also what the MU2 will evolve to. Firstly, the instruments don't have that funky bright and uniform look. This is from using ATTR_cockpit_region in the object file. Doing so makes the instruments receptive to the new lighting algorithms. You'll also note the 3D attitude indicator and HSI. This is the HSI from the MU2, but the Attitude indicator was developed for the MU2 and will get back in that cockpit eventually. There's an additional fuse panel with 8 fuses. Why put in more fuses? Well because I'm working on complete electrical simulation, including the fuses..and I'll need the practice dealing with so many circuits by the time I put this into the MU2 simulation as well as other future stuff. The 3D looks the same, but there's a lot more brawn behind the stuff now. Fuses can trip, the alternator can be overloaded and eventually (some unknown time in the future), you'll see failures where you'll have to diagnose the problem from symptoms. The Garmin GTX transponder is almost fully functional now..and will be fully functional by the times its released. How functional? "Go to garmin's website and download the manual " functional. The next thing to try and hash out are custom sounds. Thanks to Ben Supnik for providing a file loader. I just have to work through a few file file path issues, get acquainted with manipulating sounds, then hopefully we'll experience some cool stuff. So 2010 is looking to be a good year of growth for x-plane...at least in the back half of 2010 as more momentum gets built. I suppose one day I should get back to doing some scenery.

Of course the "qualifiers" keep going Jan...so the proper answer is: " generic text with ATTR_cockpit_region ...AND...the lighting mode of the generic text set to "mechanical", which is the default setting". Because almost every generic text in existence represents some type of LED output...it's 99.9% (if not 100%) of the time going to have it's lighting set to glass, in which case, folks would see no problem....but leave it to me to have it set to mechanical and bang my head against the wall. Either / or, Ben confirmed the bug and it'll be fixed for the next version...not that anybody except a bozo like me would ever deal with it. I'll presume your generic text instruments are set to lighting mode "glass". Set them to mechanical and see if you have the same problem.

YEE-HA! I just lost a stupid number of hours trying to solve a problem that turned out to be a bug in xplane and not any fault of my own after all....uh...other than that incorrect XPLM type....*cough...*cough..thanks Jan. Anyhow....the problem is that generic text is not working when used with ATTR_cockpit_region in the cockpit object. I switched back to ATTR_cockpit and everything just worked.

Something weird is definitely going on here. I can't even display something as simple as the tail number in this generic text instrument. I'm going to quit focusing on the code and investigate other stuff.

Thanks Jan. Still can't get it to work with the generic instrument. I can use Sandy's dataref inspector to view the output..and that comes out working great, but just can't seem to get the output into a generic text instrument...maybe you can see something funny. Here's the struct definition: struct xs_timer_data{ int hours; int minutes; float seconds; timer_mode mode; long int timer_int; char timer_text[9]; }; and the struct declartion and memory from the heap: struct xs_timer_data* inRefcon = (xs_timer_data*) malloc(sizeof(struct xs_timer_data)); and the registration: XPLMDataRef myDataRef = XPLMRegisterDataAccessor(myDataRefName, xplmType_Data, 1, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, xs_getTimerChar_CB, NULL, inRefcon, NULL); and the callback: long xs_getTimerChar_CB(void* inRefcon, void* outValue, int inOffset, long inMaxLength) { xs_timer_data* ptr = (xs_timer_data*)inRefcon; memcpy(outValue, ptr->timer_text, 9); return 9; } This says to me....copy 9 bytes from ptr->timer_text to outValue....no offset. Below is a screenshot in sim showing the issue. I'm beginning to think it's not the code, but perhaps my use of the generic text instrument or sloppy UV mapping...I'm having trouble getting ANYTHING to show in that generic text instrument

Well the datarefs works.....but doesn't show up in a generic text instrument. I note that you have some..."+ length" after your dataref in your memcpy line...what is this for Jan?..what's getting added after your text and why? I suspect I need to familiarize myself a bit more with the byte array and length.

Crap....I was registering it as type Int myself....THAT would explain why I was getting an int back....everything else is done correctly I think though I have yet to test. Thanks Jan for mentioning that "extra" bit about the XPLMDataType...a bad habit of cutting/pasting code in this case. AHHHH...thanks again Jan. Worked with that simple data type change. It's always the littlest things.

in a module I have declared a struct: struct xs_timer_data{ int hours; int minutes; float seconds; timer_mode mode; char timer_text[9]; }; and allocated some storage on the heap and have a ptr to this storage. I then stuff the struct with some stuff struct xs_timer_data* inRefcon = (xs_timer_data*) malloc(sizeof(struct xs_timer_data)); inRefcon->hours = 00; inRefcon->minutes = 00; inRefcon->seconds = 00; inRefcon->mode = STOP; strcpy (inRefcon->timer_text, "00:00:00"); I register a dataref with a "read byte array" callback and here's my read callback..write callback is NULL: long xs_getTimer_CB(void* inRefcon, void* outValue, int inOffset, long inMaxLength) { xs_timer_data* ptr = (xs_timer_data*)inRefcon; strcpy((char*)outValue, ptr->timer_text); return inMaxLength; } To be honest, I really don't know what's going on here byte wise and I'm sure that's my problem. I'm not fully a programmer, but certainly feel like I'm walking a glass laden path on the way there. In xplane, this code produces a dataref of type int with the value zero. I just need some help reconciling how the bytes allow xplane to know if something is an int or data..and how this is manifested at the keyboard (i.e. where do I type in such values).

Have you tried the sim Jacoba? My buddy says it definitely looks better than it flies

Greg Mink..the owner of the MU-2 I patterned the default paint livery after...has installed a Garmin G600 in his MU-2 Pay real close attention to the PFD (primary flight display) as he touches down. You'll notice the G600 has "synthetic vision" and it simply looks like he's flying a desktop PC :-P......it's NOT a camera BTW.

Very well worded question Will, this response is primarily for you....and I think I might be able to shed some light...I had to think about it myself. The initial post was, as best as I can remember, a "snap" if you will...having dealt with many other issues. These issues run a bit deeper than can be expressed through action or words...they're ideological if you will, almost philosophical. I tend to key in on character traits through people's actions, not necessarily first impressions mind you, but over time, you can get to know someone through their words and actions. Some would say this prejudice...but since I reserve judgement for a time, I like to just drop the 'pre' and call it 'judice' :-P...because my judgement is then based on concrete knowledge and experiences. So then, sometimes I will see something that looks to be initiated with a certain motivation... driven by a character trait that is offensive to me...indeed I'd argue that it's offensive to many...and sometimes I tend to over-react for the exact situation that triggers my emotions and this is what happened in this case....but I don't apologize for the emotions I feel over the character traits that set me off. At the least, I feel I put this out there for people to comment on, and convey absolute honesty to those around me who would ask...as opposed to sly talk and evasive answers. My other thoughts on this complex topic I will not elaborate here as it is a very long and drawn out topic not suitable to text. I think you might understand Will that my lashing out was not so much at that specific competition, but rather a mentality that runs deeper than just the name that was given to the competition. I sincerely apologize to all the "most popular" winners out there ;-) If any org moderator would care to discuss more in depth why I might be behaving the way I have as of late, then please feel free to PM me if you feel you need to know more.

Will, I can assure you from the bottom of my heart it gives me no great pleasure to do and say some of the things I have. I certainly don't do it out of anger, spite or revenge or personal satisfaction.....but out of what I feel to be necessity....sometimes it just must be done to facilitate a change for the better. I can tell you I will do no more than I believe necessary to see equality restored to the community....but if I have to continue to squeeze a bit harder, then I will....regretfully so. It's been said that I "come crying back to xplane.org from time to time" whenever I go on the offensive...but I have no more desire to promote my own work through the org anymore. I feel confident enough to move forward on my own reputation...but for the sake of others, I'll continue along the path I think necessary to see "x-plane's community" actually be x-plane's community.

I'm done trying to convince anybody of anything. I know what I believe, you wont' change my mind. I'm having success in the path that I'm taking, I'm achieving what I want. I've been told both by Nicolas and now Will that certain behavior is beneath me....well perhaps so...but if people are nipping at my feet...then I have to bend down to swat them away from time to time. If you were aware of even half of the lies and bullshit we have documented on the org administration....you should be ashamed to defend it. We're witnessing such behavior right now as we post....you wonder why that post on the org linking to Javier's video hasn't been deleted? Every other one has....why not this one? Well I know why! Do you Will? ....and if you know the truth and defend it...then that indeed is low. I will continue making the best products I can and promoting the best products for x-plane no matter who makes it...because I want my community to grow.

This is not a feature that xplane currently has in it's default configuration. I very much agree with the difficulty of operating the GPS. Whenever I use my mouse to move stuff in 3D....get the mouse near where you want it and then tap the 'q' or 'e' key a bit. What this does is wrestle the camera control from the mouse and gives it to the keys...specifically 'q', 'e', 'w', and 's'. By tapping q or e...you basically tap the camera slightly to the left or right, but then you'll be able to use the mouse to mess with the GPS. At this point, you either use the keyboard to move the view..or you hit "CTRL-o" to once again have the mouse control the view. It seems like a lot of hassle, but the best I can tell you is that after a while, you get used to it. it's not ideal, but until more navigation friendly options come around, we do what we can. It seems though...that very recently a feature was put in xplane...so that when you use the mouse to move around the view...that whenever you press the mouse button down, the camera stops moving the view while the mouse button is down. This lets you change things in 3D without moving the view. When you release the mouse button, the view is again controlled by the mouse. I am not sure which version of xp this is in, or if it's an option..I do know that it works for me in 9.40.

link Wynthorpe?

I'm not so sure about that Jim. This situation is but a scratch on the surface of what's going on in the background. Nicolas owns the domain x-plane.org, but we own the community...and if the org is the xplane community, then we are it. If things continue the way they have been, there will be a major shakeup soon enough. In my opinion, org management needs to regroup and adopt a business strategy that works with the times.

seems the latest version of x-plane DOES lock the camera view while you have "mouse down" on a manipulator...does it not work with track IR? Also..once you're in the middle of a manpiulation..you can move the camera wherever you want and still change the manipulator until mouse up

Ok..so this video is NOT of the MU-2..but it has implications for the MU-2 and all my future work. One reason I didn't do the ADF on the Mu-2 was because it is actuated via a 3-knob stack. That's just plain hard to work in 3D....but NO MORE! This video (@7MB quicktime) shows an animation of actuating the radio knobs with manipulators and some plug-in code to manage the mess. The important thing here is that the knob is "detented" like the real knob...as you grab and move the mouse, the frequencies will "snap" very easily in conjunction with the animation and you can select your frequency rather quickly. Another important distinction here is that the knobs themselves are the manipulators...well...sort of, but anyhow, the point is that if you can get the mouse cursor on a knob..then you're actuating that knob. No more hunting for obscure hot spots..or clicking on the wrong knob. If you have enough hand/eye coordination to put the mouse cursor on the knob you want, you're set. This will make a 3-stack ADF knob for the MU-2 a breeze. http://dl.dropbox.com/u/948823/TKylerCode.mov

So blender 2.5 alpha has been released. Expect some instability and bugs...it's a totally new interface so you can expect to have to take some time to familiarize yourself with it. I have no idea how it works with the existing scripts by marginal. I'll let you know if I find anything. the interface is extremely customizable..but I'd suggest sticking with the default layout for a while so that when we share information, we're using the same frame of reference. Tom Kyler

Rob...you're right about your point....and you're TOTALLY wrong if you think that I didn't know that already. I'm fully aware people want the airliners..always have. Of course I don't care about winning..what I care about is them labeling something 'best of' when it's clearly not the best of anything. Best artwork? nope...best system simulation? Not by any definition I'm aware of, best flight model? I doubt it... I mean..if the org can call that the "best of"....the next thing you know, they'll be saying that a project took a year when it only took seven months....or that something cost 40 bucks..but they're kind enough to lower the price for you...or that there's only one more spot in the HTML newsletter. What I can't stand are the lies disguised as marketing. ....doesn't that rub anybody besides me the wrong way? How about i tell you my MU2 has a bunch of features it doesn't have....how about I tell you it was originally 100.00..but for you special people...I'll knock it down to 50.00. How about I tell you it has ground-breaking features...when the features were broken ground by someone else earlier (like propsman)...all for the sake of making you think I'm more than I am so you'll give me your money? ...eh I'm just rantin'

I have no idea when x-plane 10 will come out. I think this "revamp process" for the Moo will be a bit more involved than previously thought...isnt' that always the way? Well I am going after the 'troublesome' parts first, like this attitude indicator and the altimeter. You can see the difference "pure 3D" makes. I hope it doesn't bring computers to their knees :-). The wipers are animated and working now also.

So it's been a year with minimal updates...so what makes anybody think that the MU-2 will actually get updated? Well here's the difference. A year ago, I was using "stock" blender scripts by marginal to export out my geometry. In that time, Ben Russell has helped develop some customized versions for x-aviation that support manipulators. If you look at the screenshot below..you'll see the time differences it now takes to export out and hand adjust manipulator parameters. On the MU-2...it was common to export out the cockpit..begin hand editing..get about 20 minutes into it..and realize I had missed something during the export....so I'd have to go back, re-export...hope I didn't miss anything, which I usually did...and then start over again. The MU-2 uses over 100 manipulators I think...I haven't counted them in practice..but there's certainly well over 100 manipulator groups in the blender file...I think Nils with his BK117 is the next highest product at 11 or so. The King air uses about 20 so far..so I'm just getting started, but I don't have to even touch the object file for manipulators. I basically make some hand edits for the lighting groups..which will be in the scripts soon enough also..and the export is done. So not only am I down from about 1 hour of adjustments to 3-4 minutes..the lighting rheostats actually control the correct lighting in the King Air...whereas in the MU-2 it currently does not. So with this increase in speed, redoing the MU-2 cockpit will be much quicker. The downside is I have to rebuild the MU-2 pit from scratch in order to set it up properly to be exported. Once done though, changes will be much much quicker..as will alternative versions of the pit if need be. With some of the workflow changes that have been worked on over the last year, the hope is that we can shorten development time drastically.

Well...one of the 3 projects is my King Air B200. I won't say what the other one's are....but I will say I'm helping improve the quality of a few default aircraft for V10.

This will be a running thread to keep people informed of my update work on the MU-2. I haven't done this before because frankly, the tools and methods to change the MU-2 was intense enough that I haven't been ready to address it for almost a year now. Well with another year of experiences, trials and errors, enormous help from other individuals like Ben Russell, I'm about ready to begin the MU-2 update process on the road to version 2.0. Does this mean I'm going to skip a 1.2 update? possibly. If I think I can redo the whole cockpit quick enough..yes I'll skip right over 1.2 and go for a heavily revamped version 2.0. "Redo the whole cockpit...did i hear that right?". Yes you did. Nils wonderful BK-117 helicopter..available at x-plane.org has demonstrated just how cool a fully 3D pit can be. The MU-2 only used a few true 3D instruments like the HSI because over half of it was already completed in Version 8, 2D type of technology. I always wanted to make fully 3D gauges with clearer numbers and better interaction. The interior lighting model isn't accurate, the fuses aren't even there. Well the tech to revamp all this stuff and bring it "up to par" is in place in x-plane and thanks to Mr. Russell, I have some tools I need to tackle the task. So then, I'm going to rebuild the cockpit, add some more features, clean up the plug-in and then release an update..and here's where you can read about the work. First order of business is to prepare the 3D instrumentation to build the panel. I began with a true mechanical attitude indicator to replace the "lazy" EFIS unit that's in there now. I had wanted to do this originally, but time constraints and lack of resources kept me from it...no more. Do keep in mind that I'm working on 3 projects for 2010 in addition to some contract work. This revamp will start a bit slow, but will build steam rather quickly. Since I'll be reusing so much artwork from the original MU-2, things should progress quite quickly.....so I'll post here simply for folks curiosity as well as my occasional need to just do something other than model stuff.

not exactly sure. I'm about ready to start the upgrades..just have to find that "starting point" between other work. I wasn't ready to tackle the update because it's pretty heavy until just a week or so ago..so I'm planning how to best redo the work. When I begin...I hope to have it done within four-6 weeks.