Reskinning an existing static model issue

[FlyingsCool]

Hi Don,
So sorry to bother you, but I'm so close to finishing my project. My dad trained at Saufley field in 1957. I've been recreating the field as it was back in the day, and am just about done making a skin for the Carenado T-34 that is close to how it looked for the one he soloed in (140826). I want to reskin Guy Diotte's T-34 static model so I can populate the field with multiple semi-accurate models of the T-34. He was kind enough to include the .mdl with his release. I'm so close I can taste it to getting this project done before my dad's birthday, but I'm stuck. I've created about 21 different skins, putting each set in their own directory. I put the model in the model directory with a model.cfg file. I created an aircraft.cfg file which includes only the first section, one for each skin, like so:

[fltsim.0]
title=T-34B Mentor 005-2S Static
sim=mentor
model=B
panel=
sound=
texture=005
ui_manufacturer=Private
ui_type=Beechcraft T-34B Mentor
ui_variation=140671
ui_typerole="Military Prop - Trainer"
ui_createdby="Guy Diotte"
atc_heavy=0
atc_id=40671
atc_airline=USN
atc_parking_codes=T34
atc_parking_types=M000
description=(C) Guy Diotte

incrementing [fltsim.x] for each one. I created a directory in Flight Simulator X\Aircraft for the model directory, texture directories and aircraft.cfg file. I followed your directions in the post above to the letter as far as i could figure out (I'm a bit confused by all the "model directory" references). I created a directory in Addon Scenery for SAMM to write too. I get as far in the directions as it opens the "Make Object Library" window. I fill in the Object Library Name with by selecting Name from Libraries and select the Aircdraft\Beech T34B directory, click on FSX at the top, and click on FSX at the bottom and fill in C:\Flight Simulator X\Addon Scenery\Beech T-34\Scenery. But the "Please select the models you wish to include" window is blank and, of course, "Save Library" button stays grey

I can't figure out what I'm doing wrong

I included a pic of the relevant screens, and can send a zip of the whole kit and kaboodle. Hopefully you or somebody else can help.

(I included a picture of my dad after his solo just for good measure )


Thanks,

Tom

 

[FlyingsCool]

Ok. After all that, I tried it again this morning, and this time I got it to show something in the "Please select the models you wish to include" window, but it's not what I was expecting. Guess I've got to find and read the manual now and figure out exactly what it's expecting and creating. But at least I've got the program working.

I was hoping to see a list of all the skins in the aircraft.cfg file in the "Please Select the models you wish to include", and that it would create a scenery library of the model with the ability to choose from each of the skins. Maybe I have to hit the "Save .mdl File" for each skin, and then they'll all show up in the "Please Select the models you wish to include" window?

I think I can figure this out, even if I have to make a static model of each plane one at a time, and then combine them all together into a single library.

If anyone wants to jump in with some enlightenment I'd appreciate it, but I'm getting there!

Thanks!

 

[FlyingsCool]

OK! I figured it out!


Well, kind of... each model is named T34B (Static by SAMM) without reference to the texture it came from, but each model is different and that's all that matters! Thanks for this tool. Awesome! Now I can make as many skins for the model as I want. Psyched! And perhaps find a more accurate T-34 model that doesn't kill frames when there's 50 or more of them on the field.

 

[gadgets]

Glad you got it all sorted out.

each model is named T34B (Static by SAMM) without reference to the texture it came from,

I didn't want to force anyone into a rigid naming scheme. So, I provided two text boxes STATIC MODEL Name and Version so the user could utilize whatever scheme he/she chooses. This is covered in some detail at the bottom of page 5 of the user manual.

Fifty highly-detailed aircraft in addition to scenery, AI, and other overheads is likely to bring most systems "to their knees". (Static models are nearly as complex as flying models. Carendo models are exquisitely detailed.)

You also may have missed the second paragraph under MAKING ADDITIONAL VERSIONS OF AND RE-SAVING A STATIC MODEL, which seems to be exactly what you wanted to do. Perhaps that will save you some time and effort if you find a simpler model.

Don

 

[FlyingsCool]

Thanks Don Yeah, I'm glad I didn't have to bother you, too

I'm reskinning Guy Diotte's very simple static model, so it's not bad.

I noticed that for some reason in the conversion process, SAMM left out coloring one of the sets of surfaces. I'm not sure why. It might be something I did in all the work I've been doing. (At first I thought ModelconverterX was the right tool, then I found SAMM and it's doing exactly what I want). I included Guy's original release, and a zip of one of my models including the source I used for SAMM and the output of SAMM (the textures are the same) if you're interested in looking at the issue. When I look at the source model in ModelConverterX, it looks correct, but the model created by SAMM is missing the one of the sets of colored surfaces (the landing gear doors and the mirrored engine nacelle). Not a huge deal. I'll live.

Here's a couple pics of the current state of the airfield. The frame rates have been affected a little, but I'm sure I can fix that through a number of ways (i.e. reducing the number of statics by a few, maybe reducing the textures to 512 from 1024 see what that looks like.).

Thanks again for this wonderful tool

The airfield still needs a ton of work, but it's good enough to show my dad

Tom

 

[gadgets]

I noticed that for some reason in the conversion process, SAMM left out coloring one of the sets of surfaces

SAMM doesn't make any adjustments to textures, the use of which is defined in the aircraft's .mdl file. This is due to a required texture missing from the static's subfolder of the texture folder companion to the static's scenery folder .

Have you checked that those surfaces are textured on the original aircraft. If not, then either the texture is missing from the aircraft archive you downloaded or the components were never textured.

If the original aircraft us fully textured, you should check the static's subfolder of the texture folder companion to the static's scenery folder to determine which texture is missing. You can then copy that texture from the aircraft archive (if it exists). If it does exist and SAMM failed to copy it, please let me know.

Don

 

[gadgets]

Just noticed you sent me the aircraft archive earlier. It appears to be a scenery model, not an aircraft model. So, SAMM probably didn't do anything to the .mdl file other that adjust the location of the textures.

I have also confirmed that all the textures specified in the .mdl file are in the scenery's companion texture folder.

Don

 

[FlyingsCool]

Hi Don, it's not a missing texture. In the model, for some of the surfaces, he just assigned a color to the surfaces. The conversion by SAMM worked for other sets of surfaces (i.e. the canopy), just one set of colored surfaces are missing their color. All the textures are being applied just fine in the SAMM output.

In the zip, I included the mdl I was working with and the one SAMM created. If you open them in ModelconverterX, the source shows the surfaces as colored. (the textures will be missing unless you use the ones from the SAMM model, but that's not the problem). But if you open the SAMM export (in the library directory), some of the surfaces didn't get their color. Given they are colored in the source model, it's a SAMM thing.

(Go through the hierarchy of the source model and you'll understand what I mean).

 

[GaryGB]

Hi Don:

My apologies if this proves to be an already resolved issue, but last time I used SAMM for converting a FS9 add-on aircraft that was used briefly for "sizing" a custom hangar that I was 3D modeling, and SAMM rendered a result of (2) sets of surfaces with Z-Buffer fighting between them.


My tentative conclusion was that SAMM had retained the originally 'hidden' sub-surface of non-smoothed surfaces as well as retaining the top-surface of "smoothing groups" that most FS aircraft have in their 3D model to achive the appearance of a "smooth" fuselage with less actual geometry in the 3D model.

In many cases, the 'faces' of both the original and "smoothing group" surfaces had 'reverse' sides showing, which required extensive manual flipping to ensure the "front" face was pointed at the exterior of the aircraft.

In some cases, either the inner- or outer 'smoothed' surface- may have been missing an applied Material on the "front" face.


FYI: My goal with SAMM at that time was a output of a 3D model that had no internal faces with applied materials intended to be visible as output was only for a static scenery model that would viewed from the outside FS environment.

[EDITED]

BTW: The intended work-flow at that time was to export the SAMM result via MCX to a 3D modeling application for reference when creating the the custom hangar I mentioned above.

Adding another phase of inter-conversion via MCX certainly may have incurred some of the rendering anomalies which I saw at that time, and I do not recall whether use of a MDLDef.XML was specified as required factor in the work-flow of either SAMM or MCX at that time (years ago).

[END_EDIT]


I must confess have not tested SAMM since that time years ago, but I was intrigued as to whether that might also be at least a part of what the OP may be seeing here ?

GaryGB

 

[gadgets]

So, it seems SAMM doesn't address colored materials that don't refer to a texture. Without making excuses (it's probably easy to fix), this is an unusual method for texturing aircraft. To my recollection, this has never before been reported.

But in this case, the model in Beech_t34.zip is not an aircraft. There are no animations. It is a scenery model. So, it's not clear why you are using SAMM. Why not just use the .mdl file as is and simply replace the textures as necessary? I suspect all SAMM did in this case was somehow eliminate the colored materials.

Don

 

[gadgets]

SAMM rendered a result of (2) sets of surfaces with Z-Buffer fighting between them.

My tentative conclusion was that SAMM had retained the originally 'hidden' sub-surface of non-smoothed surfaces as well as retaining the top-surface of "smoothing groups" that most FS aircraft have in their 3D model to achive the appearance of a "smooth" fuselage with less actual geometry in the 3D model.

Hi, Gary. I'm unclear as to the implications of your comment (other than "z-fighting"). But, you do appear to be reading too much between the lines. Basically, all SAMM does is replace animations with static transforms, Otherwise, the .mdl file generated by SAMM is essentially the same as the original .mdl file.

Don

 

[GaryGB]

Hi Don:

Based on your reply above, I can now see that this is probably a subject for a separate thread, and I should properly go back and do another conversion with the add-on aircraft in question to see what issues might still occur.

Perhaps I can then better discern whether any pre-processing of the 3D model might be required by the end user (...me ! ) before actually attempting to use SAMM for output of a static scenery aircraft model of the type I was seeking to achieve.

Considering your clarification above, I now better understand that a high level of fidelity is achieved by SAMM when rendering the original aircraft with only the animations removed, and that removal of the originally 'hidden' sub-surface of non-smoothed surfaces may not have been intended as an option when designing the feature set of SAMM.

As always, I appreciate your efforts to achieve results with attention to precision- and ease of use- in your utilities.

I shall hope to find some way to instead retain only the top-surface of "smoothing groups" that most FS aircraft have in their 3D model to achieve the appearance of a "smooth" fuselage with less actual geometry in the 3D model, to further reduce the demand on the run time rendering engine when displaying such static derivatives of add-on aircraft (...if that is even feasible to do).


GaryGB

 

[gadgets]

Tom, as I suspected, the issue was easily resolved. I've just made a new general release 2.2.07.

In the early days of SAMM, I encountered a few aircraft with untextured components. The diffuse backup color applied by the developer for those areas was sometimes very light and made the untextured areas obvious - especially at night. So, to address this, I changed the diffuse backup colors throughout to medium gray. (As I noted earlier in this thread, using the diffuse backup color intentionally in the display of an aircraft model was not something I had encountered before - until your report.)

The new release leaves the developer-specified diffuse backup color intact. Should the original problem affect anyone else, I'll have to implement an option, In the meantime, I've taken the "easy way out" and just removed the offending code

Don

 

[FlyingsCool]

Why am I using SAMM? Cause I'm a newbie to this static model creation thing

I couldn't figure out how to set up the static model to use multiple texture configurations quickly and easily and plop them in with instant scenery. With SAMM, once I figured out how to use it, it was a total snap and did exactly what I wanted to create a library of models with multiple texture sets in a way I could wrap my brain around and I could process it in minutes once I created the library of texture sets and create a library containing all the models I wanted in it.

And now I can clean things up (thanks for the hint on naming) and recompile it with even more texture sets (not that I think anybody would notice that I've only used 21 different models for 64 planes; you can barely read the ID numbers on the nose as it is unless you're relatively close to the planes).

In the future, I'm going to try to wrap my head around how to use GMAX to clean this model up. It's using a whole 1024 texture to color the prop, when I could just remap that to an unused area in the fuselage texture map, or, heck, leave it black. And I'd like to clean the model up some and make the canopy transparent. It looks like he started with a modern T-6 Texan and modified it a little; I'll cut back the rear of the canopy and shorten the nose and get it more accurate.

I placed I think 64 of these on the field, and it slowed my circa 2012 3.8 GHz 18 GB nVidia GTX 1050 desktop a teensy bit, and my Dell Precision 7520 laptop from work doesn't even hiccup in the least.

The ultimate idea for the field is I'll use mostly static models, and intersperse some AI traffic in between them to add some life to the field. (If you know of a lightweight AI T-34B I'd appreciate it )

And thanks so much for taking the time to modify your program for me. The issue wasn't a huge deal to me, but I thought you'd be interested. And I will take advantage of your fix.

Thanks again!

Tom