FS2004 Boeing 747 Classic

[FSMuseum]

Here's a sneak preview of the forward pedestal area, which itself is part of the much larger (and unfinished) throttle panel. Full previews of that panel to come soon probably as it is nearing finished.

First thing to notice is the INS: it is the standard Simufly CIVA INS, but with a graphical makeover, made possible by Marco's XML bridge. The MSU on the overhead panel is completed as well. I was not satisfied with the default Quarts font, nor could I find a good font that matched the seven segment displays, and I am not able/not smart enough to convert the INS data strings to individual characters which can then pull bitmaps like what I did for the DME gauge, so I had to resort to hand-crafting a custom font to match the displays. The fruits of that effort shown below.

Also shown is the handy performance/speeds handbook, which is based closely on the real United Airlines speed handbooks of the time. The gross weight at the bottom can be selected by clicking up or down on the values, and the speeds will update on the card. These speeds are directly from the United Airlines maintenance manual showing the speeds that should be indicated on these exact cards. The middle section, which is normally blank, currently shows the CG and takeoff stabilizer setting that would be calculated by the Flight Engineer, and then placed on this section in real life in the form of a sticky note. Clicking on the appropriate speed ranges on the card will automatically set the airspeed bugs if you like, or you can set them manually.

That's all for now!

 

[FSMuseum]

The arduous task of updating all of my code to include night lighting bitmaps has begun...

The final product will have all the same night lighting options (flood lights, back lights, etc.) as the real aircraft, though only selectable between on and off as I do not have the patience to make them dimmable!

 

[Jon_aus]

Fantastic!

But one word :

Backup!

 

[FSMuseum]

Fantastic!

But one word :

Backup!

Not to worry Jon, I keep multiple backups that get updated every single day (sometimes multiple times a day), including an online cloud backup. If anything should happen, I am never more than a single day's development behind.

 

[FSMuseum]

Progress continues...

 

[Jon_aus]

Not to worry Jon, I keep multiple backups that get updated every single day (sometimes multiple times a day), including an online cloud backup. If anything should happen, I am never more than a single day's development behind.

Haha I knew you would.


That panel is amazing.

 

[fs10inator]

Wow is your panel work amazing or what? And I'll be honest, I thought the penultimate pic you posted was part of the VC; it's just so crisp!
I've lost count as to how many times I've consumed videos, especially the RFP B742 and Dreamfleet 727 ones, but I do remember hearing about your 747 Classic project, and just seeing this thread, I thought I'd follow along for your progress, though take your time; I'll be looking forward to more ahead!

Also, I first saw you in this thread about bringing Blender models to FS9 (solved by Lagaffe), and was wondering what you'll use to make the 3D model.

 

[FSMuseum]

Wow is your panel work amazing or what? And I'll be honest, I thought the penultimate pic you posted was part of the VC; it's just so crisp!
I've lost count as to how many times I've consumed videos, especially the RFP B742 and Dreamfleet 727 ones, but I do remember hearing about your 747 Classic project, and just seeing this thread, I thought I'd follow along for your progress, though take your time; I'll be looking forward to more ahead!

Also, I first saw you in this thread about bringing Blender models to FS9 (solved by Lagaffe), and was wondering what you'll use to make the 3D model.

Thanks so much! All of the panel is at 1920 x 1080 resolution, and it holds up really well on my 2K monitor. I am creating most of the graphics by hand (such as the panels and text) so that they will be as crisp as possible.

The recent news that Active Sky Evolution will no longer function has really put a damper on my will to use FS9, even though there are still other programs that work for weather. It has solidified for me that FS2004 has limited days before online weather will no longer be usable, and unfortunately a simulator is not as fun without realistic weather.

As such, I will not be developing an FS2004-compatible exterior model as I fear I will lose so much more time to that than I already have to other projects. The panel will continue to be developed until completion, and will be 100% compatible with the CLS 747 model. I am also seriously considering a POSKY-compatible version, just so my panel isn't locked behind an unnecessary paywall for those who don't own the CLS model.

I am still working on a 747 model using Blender 3.6, as I have full intentions of 're-starting' this project for MSFS2020 and probably FS2024 too. I am having to relearn a lot of what I know about modeling, texturing, and programing to make something FS2020 compatible, but hopefully I will have something to show for it soon.

 

[fs10inator]

Thanks so much! All of the panel is at 1920 x 1080 resolution, and it holds up really well on my 2K monitor. I am creating most of the graphics by hand (such as the panels and text) so that they will be as crisp as possible.

The recent news that Active Sky Evolution will no longer function has really put a damper on my will to use FS9, even though there are still other programs that work for weather. It has solidified for me that FS2004 has limited days before online weather will no longer be usable, and unfortunately a simulator is not as fun without realistic weather.

As such, I will not be developing an FS2004-compatible exterior model as I fear I will lose so much more time to that than I already have to other projects. The panel will continue to be developed until completion, and will be 100% compatible with the CLS 747 model. I am also seriously considering a POSKY-compatible version, just so my panel isn't locked behind an unnecessary paywall for those who don't own the CLS model.

I am still working on a 747 model using Blender 3.6, as I have full intentions of 're-starting' this project for MSFS2020 and probably FS2024 too. I am having to relearn a lot of what I know about modeling, texturing, and programing to make something FS2020 compatible, but hopefully I will have something to show for it soon.

Yikes, RIP ASE for FS9...In case of weather, have you tried out FSrealWX lite? It's freeware, and still works with FS2004; heck, I've used it in my last flight two days ago (IL76, BIKF-UBBB). I don't know of any other real time weather utility for FS9 that still works, but so long as this one still chugs along, why not?

A POSKY-compatible version is something I'd be gunning for in no time...albeit partly for the sake of a young Uzbek cargo airline whose name I'll be using in a meme or two*.

Speaking of Blender, this is as far as I can simplify my Blender-to-FS2004 workflow. Tedious, yes, but since I'm more familiar with modeling in Blender, it's how I've been at for the past three years.
1. Blender: modeling, textures, export parts as .obj's.
2. gmax: import .obj's, assign materials, animate parts, export aircraft.

I can't wait to see what you've got to show! Good luck!

*"heh"—Flight Simulation Museum

 

[Jon_aus]

FSGRW still works VERY well with FS9. I've been downloading as much archive wx files as possible with it in case the plug is pulled on that one although there seems to not have any talk of that happening.

 

[FSMuseum]

FSGRW still works VERY well with FS9. I've been downloading as much archive wx files as possible with it in case the plug is pulled on that one although there seems to not have any talk of that happening.

I might ask you to send those along to me somehow if you're willing... I'd hate to completely lose access to realistic weather. I have FSGRW now that ASE no longer works, I actually find I like how it works much better than ASE. Much simpler interface, automatic reloading of AI, and I find the cloud layer depictions to be more realistic. So no regrets there!

Yikes, RIP ASE for FS9...In case of weather, have you tried out FSrealWX lite? It's freeware, and still works with FS2004; heck, I've used it in my last flight two days ago (IL76, BIKF-UBBB). I don't know of any other real time weather utility for FS9 that still works, but so long as this one still chugs along, why not?

A POSKY-compatible version is something I'd be gunning for in no time...albeit partly for the sake of a young Uzbek cargo airline whose name I'll be using in a meme or two*.

Speaking of Blender, this is as far as I can simplify my Blender-to-FS2004 workflow. Tedious, yes, but since I'm more familiar with modeling in Blender, it's how I've been at for the past three years.
1. Blender: modeling, textures, export parts as .obj's.
2. gmax: import .obj's, assign materials, animate parts, export aircraft.

I can't wait to see what you've got to show! Good luck!

*"heh"—Flight Simulation Museum

I'll have to look into that exporting into GMAX process. If so, it might save my AI Boeing 707 I've had lazing about for a couple years now. Thanks for the heads up on that! And as far as the POSKY version of my panel goes, as long as I don't find any major incompatibilities I should be able to make it work. My big worry (which I have not yet investigated) is the flap animations. I use a custom set up in the aircraft config file with four separate flap.n sections, one for the gauges, one for drag only, lift only, and pitching effects only to get the aerodynamics exactly right (or as close as possible, anyway). The CLS model has it's own baked-in flap animations that are independent of the entries in the aircraft.cfg (probably because of flexing wings? Not sure) and I am hoping that the POSKY will be the same (especially since they are made by the same modeler!). If not, it will not stop me making a POSKY compatible version probably, but I would hate to completely lose the flap animations.

 

[Jon_aus]

I might ask you to send those along to me somehow if you're willing... I'd hate to completely lose access to realistic weather. I have FSGRW now that ASE no longer works, I actually find I like how it works much better than ASE. Much simpler interface, automatic reloading of AI, and I find the cloud layer depictions to be more realistic. So no regrets there!

Can do.

 

[FSMuseum]

Alright, it's been a hot minute, some real life things have gotten in the way lately, but I have been working away in the background, and I even managed to get in a week vacation to Washington, DC where I got to see the Northwest 747 cockpit that is in the Smithsonian. I used the opportunity to get a really good look at all of the lights in the cockpit, captured the best photographs and mental snapshots that I could of how the lights looked, and then turned that into simulation nonsense!

It's no secret to anyone with a good pair of eyes that average and typical cameras do a really poor job of capturing how lights actually appear to the human eye. They end up blowing out the lights, creating unrealistic glows and halos, and tend to have lots of hot-spotting where the bulbs of the light are much more visible or blown out (over exposed) than they should be. Of course, up until my vacation, I had never had an actual visual reference for what the lights in the cockpit of the 747 looked like to the human eye, I only had my best guesses as well as photographs.

Now that I have lots of much better visual references, I have tuned my indicator lights to closely match what I remember seeing and what photographs I was able to capture that manage to closely replicate the appearance of the lights.

Below is an example of how I was originally doing the indicator lights. A bitmap with no transparency and lots of artificial camera glow.

The main issue with this approach can be seen in the following image, taken at night with the lights in the cockpit off. With no transparency effects and attempting to simulate unrealistic camera lens glow, the grey engraved front plate of the light cap glows unrealistically because it is all part of one singular bitmap that I was informing FS9 was supposed to be "Bright", as a light would be.

Now, here is what that same light looks like during the day using the new method:

For one, the extreme yellow-ness of the old light has been much reduced, as these lights are amber in color and to the human eye appear very orange. To achieve the effect of a bright amber light that appears orange-ish without compromising the illusion of brightness, I have layered a brighter yellow-ish text layer on top of a very redish colored text layer, with the lower layer being bolded so as to slightly extend beyond the upper brighter layer. This, from even close viewing distances, gives the correct color appearance, while also providing the very slight light-halo (bloom effect) that the human eye sees, without overdoing it like how most cameras see things.

Secondly, and most importantly, the bitmap now has transparency between the cap edges and the text, which means no more unrealistic light-cap-glow at night:

The overall effect is much more convincing than the old method, and the difference is much more apparent with multiple lights next to one another:

All of the lights, green, amber, red, and blue, all now use this technique, and look much much better as a result.

Something that was surprising upon first sighting was how green-ish the blue capped lights appear to the human eye when lit, as in the following image. However, it makes sense if you think about it: the bulbs underneath these caps are warm yellow in color when lit, being older incandescent bulbs, and placed underneath a blue colored cap results in a very turquoise, almost green color. The color and luminosity has been closely replicated below.

And finally, besides just working on the lights, I have been working on one of my favorite subtle visual features of the digital read-out gauges: alpha-mapped shadows and shading on the rotating drum digits. In the image below, the number 1 engine (left-most) N1, EGT and Fuel Flow gauges have shading applied to the digital readouts, while the remainder of the gauges do not. The visual difference is both subtle and quite interesting: the gauges, in my experience, are actually even more readable (not to mention realistic looking) with the shading. All of the drum-digital gauges will receive this treatment.

Alright, I think that is all for now.

Chris

 

[FSMuseum]

One more slight adjustment to the lights, then I'll move on to something else, I promise

This time, something spicy was added to the mix...

That's right, to solve the debate once and for all on the color of these darn lights, I picked myself up a real panel from a 747 on ebay. I even broke apart the lights (very gently and respectfully of course, so I could put them back together easily!) and examined the light filters, the bulbs and their color temperature, and even found some documents from Korry on what the exact chromaticity of the lights should be for various colors.

They all seem to indicate the same thing, as does my own experience playing with the above panel: the amber lights are WAY more orange to the human eye than you might expect if you had never directly observed them before, and the green lights appear very lime green when brightly lit, which is also more or less what I observed when visiting the 747 at the Smithsonian. Attempting to capture photos of these lights (at least, with my high definition camera) seems to consistently result in amber lights with dulled out colors that are far too red-shifted even with careful white balance adjustment, as well as terrible hot-spotting from the light bulbs that I do not observe with my eyes. Green lights actually appear very close to the same in my photos of the lights as they do to my eye with a little adjustment of the camera settings.

A fun aside: to get the lights above brightly lit, I simply wired up an 18V drill battery to the master light test circuit. The light bulbs at 18V are very close to their full brightness which is achieved around 24V. A delightful little electronics experiment with directly tangible results.

After much effort, I think I have finally matched the lights as closely as is possible to what my eyes actually see when observing these lights, within the confines of sRGB colors, various monitors, and also attempting to make a good looking XML gauge format 8-bit bitmap. Phew.

And also, just for fun, here's some red lights:

All for now!
Chris

 

[Jon_aus]

Haha amazing, Chris!

Why do I have a feeling you'll be posting pictures of a real JT9D in your backyard so you can get accurate engine sounds?

All jokes aside, that panel though is looking incredible.

 

[FSMuseum]

Haha amazing, Chris!

Why do I have a feeling you'll be posting pictures of a real JT9D in your backyard so you can get accurate engine sounds?

All jokes aside, that panel though is looking incredible.

Oohh man if I had the money and the space... You better believe I would do exactly that! I really do want to make my own sound pack with all the correct engine harmonics and such but there's not enough high quality recordings I could use to do so, not to mention none of them include data on the RPM of the engines to correlate the sound to, so I guess all of that will have to wait until I can find a spare JT9D at my salary

 

[Puterboy1]

Once finished, will they come with fictional liveries?

 

[FSMuseum]

Once finished, will they come with fictional liveries?

At the moment I have no plans for fictional liveries. I will however be releasing some liveries for the CLS model on my custom paint-kit, and there is a slight chance I may do the same for the POSKY model with that version of the panel as well.

 

[Jon_aus]

At the moment I have no plans for fictional liveries. I will however be releasing some liveries for the CLS model on my custom paint-kit, and there is a slight chance I may do the same for the POSKY model with that version of the panel as well.

How's the work on the FDE going? Any info or any tricks up your sleeve as to what to expect? Two different models will mean two different air files wouldn't it? That will keep you busy!

 

[FSMuseum]

How's the work on the FDE going? Any info or any tricks up your sleeve as to what to expect? Two different models will mean two different air files wouldn't it? That will keep you busy!

Theoretically it's actually pretty simple. The biggest differences will simply be adjusting for the different locations of the visual model origins, and the potential that my particular flight model might break the flap animations on the POSKY model. Otherwise, two different airfiles, but not too much difference between them.