Grumman F4F-3 G-36 Wildcat

[bub]

Ivan,
Try these out. See how it goes. I'll send you a PM with more explanation.
Bob

[Ivan]

Hello Bub,

I will look at them this evening when I get a chance.
Check your email for the F4F-3 and F4F-4 that were originally hosted here.
I have no idea what happened to them here, but there isn't a single released project that I have no uploaded here first.

The projects which I called Wildcat3 and Wildcat4 date back to around 2005 and are not quite up to current standards as you found out when attempting to map textures. I believe that they do have the relocated Internal Virtual Cockpit viewpoint, but that is about it. I was still cooking way too much by recipes back then because I didn't really understand much about SCASM.

I believe the basic model still looks pretty good even by current CFS standards, so the updates really would not be difficult other than perhaps the instrument panel.
The current issues are that I still want to push my Ki 61 to release status first and that my Development Computer is very unreliable and needs to be replaced.

- Ivan.

[smilo]

those early Ivan projects must have been lost
in one of the SOH crashes many years ago.
fortunately, they can be found at the Free Flight Site
along with many other examples of Ivan's work;
http://thefreeflightsite.com/Ivans.htm

[smilo]

the new textures do not work with the model.

[bub]

Smilo,
The reason they don't work is they need to be resized and formatted. I didn't have his email address, so I posted them for convenience. I recolored them as an experiment for Ivan to evaluate. Sorry for any confusion.
Bob

[Ivan]

Hello Bub, Smilo,
This was a first try at applying the textures. I did not need to resize them because they were already 256 x 256.
They actually do not appear to use all that many colours so I just converted them to use a palette of no more than 256 colours as required by Combat Flight Simulator.
The palette sizes vary by quite a bit. Some only have about 10 colours.

Some stuff worked but not everything, so this will need more investigation.
Here are a couple of Screenshots to show what did and did not work and also to give a close up of a section with the new colour scheme.

For what it's worth, between the release of this project and now, I figured out that leaving the texture names as they are generated by AF99 was very cumbersome and that they should all be renamed to Something.BMP.
That is my current practice, but playing with this old project reminds me of why I had to make that change.

Need to get to an appointment soon. Will pick this up again soon.

- Ivan.

[Ivan]

Hello All,

Here is a new Screenshot with the converted texture files with a very small update to make them work with CFS.

The Full Story:

The files as prepared by Bub were RGB format.
The first step was to convert them to use a Palette instead of RGB.
This was actually very simple to do using GIMP.
GIMP can generate a optimal Palette from the colours used in the original RGB image.
I chose to limit Palette size to no more than 256 colours because that is what CFS can take.

Note that in the prior Screenshots, some textures displayed properly and some did not.
The issue was that the images actually used very FEW colours.
Some images have no more than 8 or so colours (I forget the exact count).
What I did to address this was to increase the size of the Palettes where there was a problem in display and now the textures display properly.

Note that this is apparently a much earlier project than I had thought.
I don't think I even included any adjustment for a Virtual Cockpit Viewpoint back in those days and the textures are probably R8 format as they were created by AF99.

Included are updated textures and a couple Wildcat models as a canvas.
Perhaps I will actually re-upload these again without updates, but they need a lot of editing to meet my current standards.
This of course requires a reliable Development Computer.

Thanks for the cool paint job, Bub!

- Ivan.

[bub]

Ivan,
Looking good, nice job on the conversion. I still need to add a few things. I'll finish that up and send them to you.
Bob

[Ivan]

Hello Bub,

There is still an issue though.
Look at the canopy track.
Notice that it is white?
Apparently there is an issue with that area on one of the texture files.
Until I get into my Development Computer, I can't tell exactly what the issue is.
The textures are currently in R8 format and I don't have the ability to examine those on my laptop at the moment.

Instead of sending the new texture files to me, why not upload them yourself?
Include the full package as a repaint if you like or just upload as textures. Your call.
I suggest uploading as a repaint since my own versions are no longer hosted here.

I can tell you that the ordering of the files will most likely change when I update the project for a new release.
The new ordering will make a whole lot more sense.
The ordering is a little weird now because it is the one that is generated by Aircraft Factory 99 and has no relationship to how I originally named them in the source files.
The way I do things now is to renumber the textures according to the order that I have named them in the AF99 source which makes them much easier to find.

Well Done!
- Ivan.

P.S. The way that I changed the Palette was to include a 16x16 image that contains most of the colours of the CFS / FS5 palette before converting from RGB to Palette. That means that the colours in the little image will be included in the new Palette that is generated.
In theory, it should push the Palette out to 256 colours but I generated the image by reducing a screenshot of the GIMP Palette Editor and either the screenshot or the reduction isn't working entirely as expected.

[Ivan]

Hello Bub,
While I was waiting for my Son to finish some of his homework and decided to try a little experiment.

I have the PCX files I used for the AF99 Project, so it was pretty easy to convert them to BMP.
I had to use the old Development Computer to make things easier with AF5PAINT and other utilities.

The process is:

1. For Left-Right or Front-Back Textures, Flip the Top Half of the image left to right.
For Top-Bottom Textures, Flip the Bottom Half of the image left to right.

2. For Left-Right or Front-Back Textures, Rotate the entire image 90 degrees Clockwise.
For Top-Bottom images, no rotation is necessary.

3. Save the result in BMP format.

For Naming Convention, I used the same ordering as the PCX files:
0: Pilot
1: Engine
2: Cowl
3: Fuselage Aft
4: Fuselage Spine
5: Fin
6: Stabilizer
7: Wing Left
8: Wing Right
9: Flaps

So Wildcat3.0AF becomes Wildcat2.BMP
and Wildcat3.1AF becomes Wildcat6.BMP....

After that, I edited the MDL file with a Hex Edior (FrHed) and changed the texture file names.
I am pretty sure that the new file names must be the same length as the old names or the addresses inside the MDL for Jump instructions will be changed.
In any case, there were no apparent errors and the new MDL appears to display correctly with files with a BMP extension.

- Ivan.

[smilo]

i've often wondered why some model texture calls
are not converted to the usable bmps from the start,
so, others can have a go at repainting.
builder's discretion, i guess.

[Ivan]

Hello Smilo,
I would call it more a "Builder's Lack of Knowledge".
That was my excuse for this model.

If a Designer only has Aircraft Factory 99 and Aircraft Animator, then the result will be a model that really is suited for FS 98 and will have a Padlock Cockpit Viewpoint that coincides with the Aircraft Center of Gravity.
Texture file names as generated by Aircraft Factory 99 would be .0AF - .9AF or something along those lines and the image format would be "R8" which is a little obscure.

These days, with all the stuff we have been doing here with SCASM and Virtual Cockpits and such, those MDLs seem ancient, but it wasn't really that long ago. Just look at how many other tools we are using today.

Even with AIR files, the knowledge back then was much less than today. Who would have thought 10 years ago that we would be simulating two pitch propellers using Combat Flight Simulator and be able to tune aircraft performance to the precision we can today?

One odd side effect that I have found is that if the textures are left off the F4F-4 Wildcat, it uses the F4F-3 Wildcat's textures instead. The texture files have the same names at this point but when I redo the models in SCASM, I should probably change that. I just do not think it is worthwhile for me to change the SCASM code for this model when I am going to do some changes in AF99 and regenerate the SCASM code anyway.

- Ivan.

[Ivan]

Hello All,

I did a bit more poking around in this project over the last few days and it is finally starting to make some sense.
The Wildcats were originally finished back in 2005 and released as version 010. That was the only version I believe I ever released.

Sometime afterwards, I used the two Wildcats as SCASM test subjects when I got the emailed SCASM instructions from Hubbabubba. Both had the modifications for a relocated padlock viewpoint for the Virtual Cockpit and the "No Shake" edit by moving the reference point to the aircraft CoG.
Other than that, there were no changes probably because that was all that was in the instructions and I had not thought about what else could be done along with these SCASM edits.

I was very surprised when I extracted the contents from the version 010 ZIP files and the result in the simulator was actually more primitive than the version that I already had.
Another somewhat surprising thing that I found when I examined the AF99 source was that the Wildcat3 and Wildcat4 actually had a difference in the Flaps that I had forgotten about. There is a more obvious difference in the framing of the Canopy as well.

I guess it is time to gather some data and update the project for another release.

- Ivan.

[smilo]

the wildcats were/are typical high quality Ivan models.
now, i anxiously await the new, improved versions.

a side note;
do you know what the issue(s) is/are with your development machine?
is it hardware or something in the operating system?

[Ivan]

the wildcats were/are typical high quality Ivan models.
now, i anxiously await the new, improved versions.

a side note;
do you know what the issue(s) is/are with your development machine?
is it hardware or something in the operating system?

Thanks Smilo,

The Wildcat external models actually won't change other than have BMP textures and texture naming to make editing easier.
They don't have the SCASM Virtual Cockpits, Canopy Frames, Checklist or Panel which all need to be added.
The DP file is almost certainly incorrect in Ammunition weight so that will be updated.
The Flight Model is probably a little off because of lack of knowledge in tuning back in 2005.
I am also debating on creating Propeller Tables for the Wildcats, but that means a lot more work.

For what it's worth, I believe these models were also based on William Wylam's dimensional drawings.
The quality appears to ba a LOT better than the ones for the P-40E.

Regarding the Development Machine:
I believe the problem is both hardware AND software.
The Operating system appears to be slightly corrupted (Windows 98 SR2).
The worst of it appears to be a result of trying to patch Internet Explorer in order to install Visual Studio.
The install didn't work anyway, but seemed to leave a bit of residue and nothing has been completely right since then.
It had already been misbehaving before, but got a lot worse after that.

The hardware problem appears to be an issue with the graphics card / monitor. I am not quite sure which because it is an intermittent problem. It is using a Voodoo 3 3000 (yeah, that old) and I do not have a better replacement. I had a Matrox Millennia card as a spare but that seems to have disappeared after the water leak in my basement and the cleanup afterward.
The disk subsystem appears to be somewhat unstable but I don't know if it is the controller, disk drives, or perhaps an unstable power supply.

We have plenty of spare CPUs around but no spare 3D video cards that would fit an appropriate motherboard.
I do have an old computer in mind, but I also know it will take a lot of setup to get it working with all the software that is on the older Development Machine.

Thanks for asking.

- Ivan.

[Ivan]

Hello All,

I just finished up a rather long session that started yesterday with the purpose of updating the model with:
A Relocated Virtual Cockpit Padlock POV,
Interior Canopy Frame for the Virtual Cockpit,
Tuning the Flap Wells to match the movement of the Flaps,
and renamed BMP texture files.

This was done across three separate computers.
Other than the Development Machine which is still a bit unreliable, the other two do not have a full set of software utilities.

The last few hours were spent trying to figure out why the SCASM code I was changing was compiling but not generating a model that could display in the simulator.

It turns out that the version of MdlDisAs that I am using does not have the ability to create a new MDL file with any reliability.
It can UPDATE an existing MDL but cannot change to a different format.
The problem was that the F4F-3 Wildcat isn't a terribly "heavy" model with a lot of polygons, so it was using the minimum sized MDL file format generated by AF99 and Aircraft Animator.
The data limit is only 64K maximum.

This was fine until I tried to add the Interior Canopy Frame.
This piece is pretty heavy in resources and changed the data size to 67K which needed a new model format.
MdlDisAs offered the option of generating a new MDL with a larger data limit (133K) but the problem was that although the model would build, it would not display in the simulator.

It took me a while to figure out that there was actually no problem with the SCASM code edits that I had made and then it took about another hour to figure out how to work around the inability of MdlDisAs to generate a new MDL with a different format.
As it turned out, the solution wasn't that difficult: I just copied one of the other larger completed MDL files (P-40E Warhawk) and built the SCASM model into a Warhawk MDL file.
I should update my SCASM notes to describe this method.

In all honesty, there is very little visual difference externally.
The Canopy Glass is a bit more Transparent and the Flap Well animation isn't very obvious, but with all the effort, I figure a few more screenshots are in order.

Comments on the Screenshots:
I was in a bit too much of a hurry to do a SCASM update. I should have added some Gun Ports in the Model instead of just using textures to represent them.
The Wildcat seems to look a lot like a Bearcat when viewed from the front.
The Canopy Frame would be less annoying if it were a bit narrower.
I did not change the Gear Animation even though the current sequence is too limited by Aircraft Animator.

Next step is to update the F4F-4 in the same manner and perhaps make the Canopy Framing a bit narrower.

- Ivan.

[Ivan]

Finished editing the Internal Canopy Frames this evening.
Most of the frames were made a lot narrower to allow better visibility.
These are views of the F4F-4 Wildcat.

Does the location of the second screenshot look familiar to you, Smilo?

I am also considering making some minor edits to build the F4F-8 version.

- Ivan.

[smilo]

yup...on a carrier deck.
what better place for a Wildcat to take off?

[Ivan]

Hello Smilo,
I believe the USS Saratoga was your doing or do I recall incorrectly?
I was wondering why I needed so much of the deck until I remembered that normally the carrier would be steaming into the wind and putting another 30 knots or more across the deck.

These Wildcats were not very hot fighters, but I have still doing a fair amount of "Fun Flying" to have a better look at the aeroplane from different angles. Even though the AIR files are a bit old (2008), I probably won't change them very much for a re-release.
The Wildcats were released in 2005.
The AIR files were updated in 2008.
The SCASM editing was done in 2010.
Now it is getting a review in 2018 for no other reason than its resemblance to a repaint by Bub on another forum thread.
The Wildcats are also serving as test subjects to see what I can get done with the computer systems I have at the moment.

- Ivan.

[smilo]

thank you, Kwong, for the opportunity
to give a little shameless self promotion.

yes, with the help of hubbabubba and Ken Atwood,
i did the CV-3 Saratoga off Pearl Harbor
and the CV-7 Wasp Task Force in the Mediterranean.
http://thefreeflightsite.com/Smilos.htm

as a side note, one can add, among other things,
a headwind with CFO weather by Bill Potvin.
as i recall, there's a copy in the Wasp folder.

[Ivan]

Hello Smilo,
That was hardly a "self promotion" at all.
You credited the folks that assisted with the project and gave a link to where others could actually find and download it.
I tried looking for it here but had not thought about the FreeFlight site, so Thanks!

I only have the USS Saratoga. I don't think I ever installed USS Wasp. (I HATE installing scenery!)

Even without CFO Weather, it is possible to get some wind over the bows by aiming the carrier in the proper direction (due North) and setting the wind speed in CFS Weather.

One of these days, I will try to build a Japanese carrier and perhaps the USS Lexington, but it will most likely require a lot of assistance because I don't know a thing about scenery or airfields.

- Ivan.

[smilo]

installing scenery really isn't that bad,
once you get the hang of it.
we got lots of practice back in the multi player days.
as i recall, we used a different scenery for each weekly mission.
if there's any interest,
i'd be happy to explain the process that worked for me.
but, that should be another thread.

as for building a carrier,
there are some tricky aspects.
especially, constructing a usable elevated runway.
remembering how to do it will call for dusting off the old memory banks.
i'm not sure if i'm capable, but when the time comes,
i'll do what i can to help.

[Ivan]

The only thing I actually "know" how to do with a carrier is to build a visual model and I am not entirely certain of that.
I was thinking the subject should be Soryu or Akagi and it would involve a LOT of SCASM work.

This evening, I started gathering information for a Data Sheet on the Wildcats.
I am finding a few differences between the F4F-3 and F4F-4 that I never really accounted for in the earlier flight models.
Has anyone ever tried to build a FM-2 Wildcat before?

- Ivan.

[Ivan]

It just occurred to me while looking at details of the Wildcat's instrument panel that I might need to do some more gauge programming for this aeroplane.
Note that the stock CFS aeroplanes use either MPH or Kilometers/Hour. None use Knots.
This was something that I missed when putting together the panel for the A6M series.
The A6M also used a gauge reading in Knots.

It seems like trying to do things right always takes longer than expected.

- Ivan.

[Ivan]

Hello All,

Note that not all the entries are filled out even though many can be deduced from simple math or similarity in specification to other models.
This was data that I collected when working on the Flight Model for the F4F-3 and F4F-4.

- Ivan.